VIRTUAL ORCHESTRATION - The Fate of the Art?

[gungnir]

I was just thinking that it might be beneficial to all those interested in attempting to realistically emulate a full orchestra with GigaStudio that is may be productive for forum members to occasionally provide mp3\'s of mockups from the available literature for others to audition. It seems to me that this should be a very relevant topic on a GigaStudio Sample Library Discussion forum, since in directly involves the use of the currently available libraries as well as the other elements involved in using them (sequencers, applied reverb, midi control, etc.)

I\'m not advocating a competition as to who does it best (although this would become quickly apparent), because the results would obviously be a reflection of your studio gear and sample libraries. But the capacity of the current libraries to perform in this regard, and the development of the next phase in symphonic articulations, should be extremely relevant topics in this particular forum, shouldn\'t they, or is my thinking in need of massive recalibration?

Shouldn\'t many GigaStudio forum members want a chance to test their approach and equipment in emulating symphonic works, given the time and money they have invested. If forum members were interested in partaking in a few small projects, it would then allow them to directly see what the state of the art in virtual orchestra actually is, and how their own setup and approach currently compares. Is this not the essence of what many involved in movie scores and classical orchestrations and compositions are visiting this type of forum for, acquiring the knowledge and latest tools to create the most convincing orchestral realization possible?

So, I decided to use myself as a guinea pig, and subject myself to the scathing wrath of the barbarian horde that inhabit this forum. But if you read between the lines accurately, you will correctly conclude that it would actually please me much more to hear somebody else clobber my rendering (outlined below), as this would obviously give me the necessary insights into what I must do to improve things, cause I\'m not going to stop trying.

It is unfortunate that other forum user projects seem to fade away without actual results, but I can\'t believe that this type of activity is not an ongoing event on a GigaStudio forum. Let me put it this way. I have tried to create a setup where I have a permanent virtual orchestra set up in Giga stations awaiting instructions from my sequencer (in my case I usually use my \"modified version\" of Sibelius) and controller keyboards. There is a substantial investment involved (Garritan, Vitous, Dean, etc, etc.) and one could easily apply the metaphor that I have a bunch of virtual musicians sitting around inside a pc box waiting for the next gig (pun intended) when the conductor (sequencer) decides to give them a workout/practice session. As such, it\'s no big deal for me to take on a few bars from the classical literature (or movie scores if one can get them) and do a work up. It\'s supposed to be fun and satisfying to do this. It is the essence of what my studio should be able to easily do. If you have things set up optimally, don\'t you agree that this type of activity should not represent a bothersome challenge, but instead only a quick little workout that should yield important information about the state of your virtual orchestra\'s adequacy, as well as prompt discussions on topics such as the current practical deficiencies in both sequencers and sample libraries, etc.

Because I feel that extended string passages are particularly difficult to emulate, I decided first to sequence a good old fashioned romantic piece (Brahms Third Symphony, Third movement) that features long string lines. You may not like Brahms ,etc. but should this really matter, you want to give your virtual orchestra musicians a workout and I assure you, they do like Brahms. Next time, forum members may want something by Shostakovich, Bruckner, John Williams, Ravel, Peter Vasks, who cares, it\'s the results that matter.

So, should anyone be interested, I would gladly send an mp3 of this mockup by email, as well as a midi/Scorch/Sibelius file should you need it, or a few bars of the sheet music if necessary. I\'m unfortunately not currently set up for posting mp3 on the net, but if someone would volunteer a temporary residence on their site, it would be great because it would be easier for forum members to audition the files. I also could send along a mp3 of a small part of an actually recording of the piece for comparison purposes, if this is ok in regards to copyright, etc. (could someone advise? thanks).

I\'m certainly not suggesting this is a particularly well-done rendering of this music. It does provide however, a reasonable template for discussion and how one might approach making improvements to get extremely realistic results. As I suggested, I felt this particular piece was a bit more ambitious because it didn\'t rely so much on attacks, but instead sustains and phrasing dynamics. (Also I wanted to demonstrate to those skeptics who believe Sibelius cannot function as an effective sequencer that they may be approaching things the right way, it is lightning fast and you never have to leave it classy environment, but that\'s a whole other story).

If GigaStudio sample library forum members feel that this approach is not worth the effort, well that\'s valuable information also. There may be much more practical ideas for involvement in the development of our common goal of realizing convincing symphonic orchestrations and movie scores. But a lack of interest in pivoting your own virtual orchestra investment against the real thing to allow objective comparison on a site dedicated to Giga sample libraries will, in itself, also reveal much about the actual state of the art in symphonic realizations. But that\'s also another whole topic.

Happy arpeggios and sorry for the extended length of this post.

[Hardy Heern]

[Hardy Heern]

Me Me Me Please....Thanks. 0.5Mb Max.

[J. Whaley]

YES!!! This is very much needed. I think sequencing a convincing orchestral mockup is the hardest thing a programmer can do. I do tons of professional work, but I\'m stumped when it comes to making an orchestration sound good. don\'t know why, just can\'t do it. I have the Peter Siedlechek (sp?) library, and if I had to say today, I\'d say I hate it. But their demos are so awesome... how\'d they do it? I wanna know and I think an online forum is the best way. It\'d be real appropriate for Northern sounds or someone to allow us to post somewhere. But until they do, I\'d be happy to let you post a mockup on my site. Does anyone know how to give access to a directory, but not the entire site? Thanks

[donnie]

I\'m game and I\'ve actually suggested something like this before. The only way to learn how to make a create recording is to look at the midi file and see what has been done to it.

Send me the file.

donnie@dssoundware.com

[composer22]

Originally posted by donnie:
I\'m game and I\'ve actually suggested something like this before. The only way to learn how to make a create recording is to look at the midi file and see what has been done to it.

Send me the file.

donnie@dssoundware.com<font size=\"2\" face=\"Verdana, Arial\">Yes donnie has adn with GOOD reasons...in comparing libraries...I would STILL like to have a contest sometime though given a same MIDI file and talent

[MartinL]

Great idea!

I\'d like to hear the file. ********** Thanks.

Martin

[ninriggs]

Gungnir,

I would really like to hear the file too i would be very interested to look at the midi file if that is possable?

Many thanks

[gungnir]

Possibly a realistic and technically achievable goal for virtual orchestration (for the purposes of composition and some movie score mockups) is to sound like a third-rate real orchestra, playing late at night after just returning from the local pub.

But even to do this, the approach to sampling and sequencing will have to change significantly. Will a new string library (actually that pdf file announcement has been available for some time now) really allow virtual orchestration to forge ahead to where it must go, or prove to be an exercise in redundancy due to its content?

At this stage do we really need more trills, pizzicatos and sound effects? Aren\'t the \"sound effects\" we require now in a new string orchestra collection divisi sections, undulating string tremolos (on adjacent strings), muted viola detaches, pp attacks, and portamento slides at several different tempos. Some espressivo detache and sustains at different dynamics will be certainly welcome and lets hope that Sonic Implants have concentrated on this crucial aspect.

I have send along the mp3\'s by email to those requesting them including those who sent private emails. Hardy Heern and J. Whaley will you post your email addresses here so I can send them along to you as well.

The point of this whole thing is very simple: we can only evaluate the state of the art of virtual orchestration by copying pieces that will can easily compare to real symphonic ensembles. There is no point in using people\'s compositions to evaluate the current calibre of library sounds, because there is no reference point on which to judge. In many cases compositions are a function of the axiom \"you don\'t control your sampled sounds, they control you\".

Thank you for your interest in this project. This type of involvement on the GigaStudio forum is the only way virtual orchestration will become a true viable option in the near future. I believe that this is the type of thing a Giga Sample Library forum should be doing continuously: rather than talking about sounds, let the sounds do the talking!

Cheers, Gungnir.

[J. Whaley]

I think this sounds great, is there anyway we can get a FTP site to post to? My email will not handle tons of MP3s.... anyone got suggestions? For now though, since I wanna hear it, send it to me.

jeremy@w3productions.net

Thanks gang. I look forward to some great info coming out of this

[Craig_L]

gungnir,
You can post up to 30 Megs for download through http://briefcase.yahoo.com/ (\"http://briefcase.yahoo.com/\") and allow people to access it. Hope this helps.

I\'ve posted a few orchestral demos there with midi files some time ago. Try accessing the site - my particular address is http://briefcase.yahoo.com/duoartc. (\"http://briefcase.yahoo.com/duoartc.\")

Look forward to hearing the Brahms.
Craig

[Hasen]

I heard the files gungnir and I have to say they\'re incredibly similar! I think the biggest giveaway however were the Miroslav celli, they have a strange quality to them and I think you would have been better off with just GOS IMO.

Which reverb did you use btw?

I think this test has been very imformative especially as the MIDI file was available for examination. That way we can see how much editing and cc was used.

Personally I would prefer to see these tests with film scores rather than classical \'cos I think that is what most of us are more interested in writing ourselves.

[Mark_Knecht]

gungir,
Please send me a copy. mknecht@controlnet.com

Thanks,
Mark

[Mark_Knecht]

Hi,
Either would be cool with me, as long as I can buy a CD for comparison. Personally I\'d rather listen to classical.

Mark


Originally posted by Hasen:
Personally I would prefer to see these tests with film scores rather than classical \'cos I think that is what most of us are more interested in writing ourselves.<font size=\"2\" face=\"Verdana, Arial\">

[Lancelink]

Great idea!!! Please send me a copy.

lance840@pacbell.net

[Hasen]

Originally posted by Mark_Knecht:
[QB]Hi,
Either would be cool with me, as long as I can buy a CD for comparison. Personally I\'d rather listen to classical.

Mark<font size=\"2\" face=\"Verdana, Arial\">Well ok I didn\'t realise that. I would be happy to see both then. As long as its not exclusively classical I\'m happy. images/icons/smile.gif

[Mark_Knecht]

Hey, I\'d be excited to see either one done well!

Take care.


Originally posted by Hasen:
[QUOTE]Well ok I didn\'t realise that. I would be happy to see both then. As long as its not exclusively classical I\'m happy. images/icons/smile.gif <font size=\"2\" face=\"Verdana, Arial\">

[mschiff]

gungnir,

I am so glad that you brought up this topic. I\'ve asked some questions about it several times and gotten good answers, but nothing definitive.

At work, I have set up an orchestra emulation using outboard synthesizers (12 of them) and I want to do the same thing with Gigastudio.

My boss is a classical composer, and had chosen the sounds that he liked best for each instrument from a number of different synthesizers and we set them up so he can compose in Encore. He is not looking for a simulation that provides the best recording of his pieces, since he has them recorded by a real orchestra, but he needs something to \"proof\" the score, and he wants it to sound as good as possible. I have all the major Giga libraries here, so it would be extremely helpful to have a list of patches that you feel work well for your orchestra simulation.

I would very much like to receive your files at martin@stevegoldmanNOSPAM.com. Take out the NOSPAM before you use the address.

You can hear one of the pieces done with the outboard synthesizers (an older setup) at www.mp3.com/stevegoldman. (\"http://www.mp3.com/stevegoldman.\") Listen to the midi version of \"Quicksilver\'s Salvitude\". There is also an orchestral recording of the same piece there.

Thanks!

-- Martin

[J. Whaley]

Well, I think it\'s obvious this is going to be a long topic. What\'s the odds Northern sounds would give us our own forum? Anyone know how to do this? Garritan had one... I don\'t see it today for some reason, but it was here.

Ran across this link a couple days ago. Some real interesting stuff... some of you might want to check it out:

http://www.keyboardmag.com/features/rupertgw/index.shtml (\"http://www.keyboardmag.com/features/rupertgw/index.shtml\")

I\'ll post more when I get Giga working right!!!

Has anyone else had a crappy experience with Peter Siedlechek library? or is it just me? They\'re demos are awesome but I can\'t work with the samples for the life of me.....

[gungnir]

Mschiff, I tried the link but it wasn\'t working. I will discuss the sounds a bit tomorrow, no time now. Hasen, you may be hearing a bit too much solo cello rather than the actual Vitous samples, the reverb is just the GigaStudio on board NFX1. Thanks Craig_L for the suggestion re Yahoo, the files have been posted to http://ca.briefcase.yahoo.com/landvaettirsampo. (\"http://ca.briefcase.yahoo.com/landvaettirsampo.\") For those I haven\'t emailed yet, please give that address a try.

Is the concept of a convincing virtual orchestra a myth? There are many talented people out there with huge investments in equipment and experience, does a few bars of Brahms intimidate you? Can the sample library developers themselves not see this as an opportunity to demonstrate a very reasonable practical application of their libraries. Sonic Implants? Some espressivo samples to blow gungnir\'s mundane attempt off the forum. The silence is deafening.

[mschiff]

gungnir,

For some reason the period at the end of the url was included in the link. It should be http://www.mp3.com/stevegoldman (\"http://www.mp3.com/stevegoldman\") with no period at the end. I tried it, and it does work.

I got your files, and the simulation sounds excellent!

-- Martin

[gungnir]

The Yahoo site is something new to me and it appears that, even removing the period at the end of the link, (http://ca.briefcase.yahoo.com/landvaettirsampo ) the files are \"empty\" even though there is megabyte content listed, with both my link and Craig_L\'s link. I may be doing something stupid (sorry), so rather than frustrate people, it may be easier just to email me at island@tallships.ca should you want the files in question.

Mscliff, sorry for the delay in responding to your query regarding GigaStudio patches, etc. but I thought it would be more fruitful to hold off in the hope that there may be a few attempts at the Brahms and then use that as a platform for hands-on discussion. Like I said, I find it a little hard to believe that on a worldwide forum, where a huge amount of discussion is based on purchasing and using \"Orchestra\" libraries, that some people wouldn\'t want to give a few small symphonic sections a go for the sake of creating a ongoing constructive dialog on the forum.

I very much enjoyed Stephen Goldman\'s Salvitude piece, a fresh optimistic energy in a modern (neo-romantic?) style with a wonderful approach to the ending, tasteful twist and turns along the \"cadential pathway\" to the conclusion. Mr. Goldman represents a perfect example on someone who would benefit greatly from the ability to create more realistic virtual renderings, and believe me, it is possible to forge ahead in the art. I have several other orchestral mockups, featuring woods, etc. vs. strings, but it appears so far that many people seem to wish to remain in the seclusion of their studios, harboring what they may perceive as their own \"trade secrets\", rather than being involved in actively developing the art.

Yes, the midi mockup by Mr. Goldman just won\'t do, its five years behind the times and all sorts of things can be done for massive improvements if you have the larger giga libraries. If a constructive active dialog doesn\'t happen on this forum, I will email you with some suggestions (not that I am any sort of an expert, but I am dedicated to the concept, cause it\'s not going to go away).

Regards, Gungnir

[Craig_L]

Gungnir,
My link is http://briefcase.yahoo.com/duoartc (\"http://briefcase.yahoo.com/duoartc\") without the period on the end, then just go into the folder which says \"music\" and there should be five files, 4 X mp3\'s and a midi file - some Wagner Ring motives with different reverbs and some Tchaikovsky fron the 4th Symphony for pizz strings. Yes, I couldn\'t get through with your link. Craig

[donnie]

Originally posted by gungnir:
Is the concept of a convincing virtual orchestra a myth? There are many talented people out there with huge investments in equipment and experience, does a few bars of Brahms intimidate you? Can the sample library developers themselves not see this as an opportunity to demonstrate a very reasonable practical application of their libraries. Sonic Implants? Some espressivo samples to blow gungnir\'s mundane attempt off the forum. The silence is deafening.[/QB]<font size=\"2\" face=\"Verdana, Arial\">Gungnir,

Don\'t be so quick to jump to this conclusion. Doing a mock up like you are talking about does not happen overnight. I would love to have 4 hours during my day to do a mock up but I just don\'t so I have to do little spurts on it when I can. I am working on some things though.

Donnie

[mschiff]

gungnir,

I agree with Donnie. This is the sort of thing that could take a week or two for someone that does not already have a virtual orchestra set up. Even though I already have one (outboard synths) here at work, I don\'t currently have it set up so I can record from it, though we just got a new console and I will have that capability very soon.

Be patient. I think there is a lot of interest and this can be an ongoing discussion.

-- Martin

[Tokyo Joe]

Craig,

I get the following message when I try and access the files at both sites:

\"This folder is currently empty.\"

Any ideas?

Thanks in advance.

[RickD]

Craig,
I checked out the Yahoo link, and your folder was empty. I\'d like very much to check out the files. Could you email them to me? If so my email is:
rick.dortignac@verizon.net

Thanks,
Rick

[gungnir]

Ah, now, we are starting finally to get somewhere, the issue of time. Even though I will not discuss it until several people have made the effort to do a mockup with their huge investment of gear (that is supposed to perform a task after all), I will tell you that, once you have configured GigaStudio with a modified format of Sibelius (yes, Sibelius, not dinosaur programs like Finale, Overture, etc,.and certainly not Cubase and other gizmo programs that look like the dashboard console in my car), you can do a mockup like this in about two hours, of which most of the time is spent monitoring your results with the original, not programming. But most importantly, you never leave Sibelius and the process is actually enjoyable!

I am willing to back up this claim regarding Sibelius with a competition with others sequencing on alternate equipment. We would make sure we all had the same sheet music, then, at random, a third party would declare what bars were to sequenced, and then we would go at it and see who, in a given time span, created the most realistic results. But to do this we need people with the same libraries, and that\'s why this thread is now getting a bit frustrating.

Here, on a worldwide sound library forum, there seems to be little interest in applying the currently available string libraries to a few bars of music to give your investment a quick test run. One would assume that many of the people endorsing Gary\'s fine library (where would we be without it) peruse this site occasionally as Giga users, and to throw together a few bars of string music shouldn\'t be any big deal with the stacks of equipment they have.

I am trying, and will try for only a few more days, to create some templates upon which practical discussions on the utilization of the orchestral libraries constantly discussed on this forum are featured. If the many people who visit this forum (there are more than you think) don\'t wish to partake, it should be of interest to forum members why. Possible reasons:

1. You don\'t have the time. (But doesn\'t your job (or hobby) depend to a significant extent of the development of virtual orchestration. Don\'t you wish to excel in this area? Why are you reading this if its not? Are you sure you\'re using the right tools so that doing thirty bars of music is a time consuming, negative experience?
2. You don\'t have the sheet music, midi file, or audio file of an original recording. (I\'ll send you all this if you wish by email. Is a small section from a CD recording ok re copyright, etc? Can someone advise? Thanks).
3. You don\'t have the sounds. (Well, this is a legitimate excuse as long as you aren\'t spending your precious money on software gizmos extras rather than sounds. Sample libraries should be a priority).
4. You don\'t like this old fashioned music by dead people. I don\'t want to use this type of syrupy classical instrumentation in my compositions, I want kickass action cues and brass power with percussion. Well, we could do this next time. But, what\'s wrong with a few bars of expressive strings. Even in film/tv scores, the director may, on rare occasions, actually want you to write a passage evoking emotions from a little higher up on the brain stem, you know, after he\'s decided to hold back on the killing and hacking for a few moments.

A radio announcer the other day described this is one the most respected and poignant melodies in the
repertoire. Even if you don\'t like music from this period, I doubt if it will give you a headache, and
doubt if it is the real reason.
5. It can\'t be done and virtual orchestra is a marketing myth.

I apologize for not being able to provide the mp3 files to forum members in a quickly accessible way. If someone would like to put them up on there site for a few days, I could send them to you via email.

[Tokyo Joe]

Please send me the files and I\'ll post them on my site for a couple of days.

************************

[Hasen]

Originally posted by gungnir:


4. You don\'t like this old fashioned music by dead people. I don\'t want to use this type of syrupy classical instrumentation in my compositions, I want kickass action cues and brass power with percussion. Well, we could do this next time. But, what\'s wrong with a few bars of expressive strings. Even in film/tv scores, the director may, on rare occasions, actually want you to write a passage evoking emotions from a little higher up on the brain stem, you know, after he\'s decided to hold back on the killing and hacking for a few moments.<font size=\"2\" face=\"Verdana, Arial\">My comment about film music was not to suggest that I wanted action music - far from it. Film scores can invoke any emotion and are written for that purpose. I was just remarking that I would prefer film scores rather than classical because I am more familiar with them. There style is very different and one that I prefer.

I get the impression from your post that you have not heard many film scores...

[Duncan Brinsmead]

I think there are a few issues that make it difficult, though not totally impossible, currently to recreate the full expressive sound of an orchestra. One can quickly sequence something, but it takes a lot of time and work to make it sound good.

To me the biggest single issue is achieving a fluid expressive legato. Most sustaining instruments do not simply jump pitch to new notes, but fluidly slide in a variable manner. This is hard to do well with current sampling technology. Another problem with smoothness is the repetition of dynamics in recorded samples.. for example if each note in a passage has a repeated swell or too sharp attack it will sound unnatural and mechanical in a legato setting. Adding expression is difficult. I use a breath controller, but it is not really sensitive enough and so for the best results I usually need to edit the breath control for every single note by hand. This is very time consuming. I\'m currently working on a rendition of the Adagietto from Mahler\'s 5th symphony, and getting a good legato line is killing me. Using pitch bend for the slides is working, but the method I\'m using is also very tedious.(although it\'s starting to sound pretty good)

The second biggest issue is the dynamic range of the instrument sound. Instruments when played loudly have very different tonal and attack characteristics than when played piano. Many libraries do not contain enough of a dynamic range. I\'ve heard several sequenced symphonies where the crecendos end up sounding like someone just fiddling with your volume control. One can mimic tonal ranges on some instruments by dynamic filtering( i.e a lowpass filter on a ff horn sample will make it sound softer ), however such filtering often sounds unnatural. For non-solo instruments one do crossfade layers, which also allows for good swells. For expressive playing it is especially important to have a sound that tonally changes as breath pressure changes. Setting instruments up in this manner is currently doable, and many libraries are now coming with a fuller dynamic range.

Another issue is expression in tempo.. not just from bar to bar, but from note to note. If one insists on playing notes to a metronome then the tempo track should be edited down to the level of at least a quarter note to get the right push and pull of the meter. Delay is an important way that strong beats are emphasized. Without it things tend to sound sequenced. Tempo should be fluid, especially in slow music.

One subtle issue is rapidly repeated notes. The sameness of recorded samples can result in an unpleasant machine gun effect. Having a couple of extra sets of samples and substituting is one way of dealing with this problem( and again sample developers are starting to provide extra samples ). This can also help with rapid up-down bow sorts of passages.

If the samples are very dry, then it is important to have a high quality reverb to make them sit properly in a mix. However this is not such an issue if the samples have a little room ambience. After all.. there were some great recording done 50 years ago using reverbs of the day. The results still sounded like a real orchestra, despite fairly close mics and inferior reverbs.

Some people make a big deal over the cleaness of samples, quality of mics etc. However I don\'t think that this is too important when it comes to making natural sounding instruments( unless the developers have done extra processing on the sound ). The samples should still sound like recordings of real instruments, no matter how bad the equipment is(hiss and distortion are annoying, but these artifacts would only make it sound like an older recording, rather than synthetic ).

To me what makes an instrument sound real is expression: control over tonal range, legato and timing. These are also the primary elements that a performer brings to a score, without which it sounds like a machine.

Duncan

[Tokyo Joe]

Here are the files:

http://home.sc.rr.com/tokyojoe/brahms/ (\"http://home.sc.rr.com/tokyojoe/brahms/\")

[gungnir]

Hasen, I wasn\'t reacting to your comment, my snide remark was because I sometimes get discouraged with film music because it\'s so full of overused cliches and boring trendy formulas; have you noticed that its almost always the same guy, with the emphatic deep voice, that is used in the trailers of many of the new movies (the guy used on Lord of the Rings). He\'s everywhere, or so are the musical ideas in most scores.

In regards to not having heard many film scores, I have just now returned from catching Spiderman, the motivation being primarily Danny Elfman\'s score. It was incredible, this guy is now playing the orchestra like Paganini played his violin. He seems to bend it, and make it pulsate in pure magical fashion. I also harbor an underlying resentment regarding movie scores because it seems that music usually functions in such a subservient manner. I would like to see more movies that used music as a character in the film, not as just another special effect. It\'s been done, there are directors out there who are clued into the potential power of the music-visual complex, for example Poledouris\' Conan the Barbarian. Then of course, some movies provide an impetus for the creation of true art of the highest order (e.g. Schindler\'s List). But in general, the qualities of deep emotive inspiration are lacking because there just isn\'t the talent to do it. No wonder John Boorman had to go back to Wagner to conjure the ominous foreboding power needed in the Excalibur opening. Just look at the sheet music of that Siegfried motif, a tritone infused with the energy of triplets on top of triplets in the violas and cellos. Now that is the stuff of real visual music.

Duncan, you raise very practical and relevant points but I just don\'t approach things that way. I believe you should use midi controllers only as a last resort. Think of what you are spending all your time and money on. All sorts of software and standalone gear modules (gearbage?) that is used primarily for the manipulation (distortion?) of real life sounds. People compress, layer, delay, time stretch, pitch shift and mess around with attacks, decays, envelopes so that you are more or less guaranteed that the end result can not sound anyway near real and expressive, but instead like a facsimile of something real. It\'s similar to all the visual hybrids of reality we see in the movies, which somehow, despite their technical achievement, seem unnatural, we sense something is wrong, almost perverse, like genetic engineering.

If the samples were recorded correctly in the first place, providing the user with the emotion and vibrancy required to phrase a passage convincingly, and also with the proper ambience, much of this time consuming foolishness with midi and gearbage would not be necessary. You would be free to spend your time creating music, and to spend your money on the tools of your trade, all available sample libraries. If a sequencer didn\'t have to perform midi actuated miracles to make the samples sound somehow more realistic, you may finally be free from wasting your valuable time reading the latest user manuals (loser manuals?)

There is another approach, and that is to make the constant manipulation of the best samples available in multiple dynamic layers extremely easy to do. For the purposes of sequencing, not playing in the parts, I use a method that is extremely efficient and very satisfying, and real fast. The only reason I ever use Cakewalk or Cubase is to diagnose midi data problems. This approach is really for stand alone scores and playback via Giga, not meshing with audio or syncing to film, etc. But as I mentioned earlier, the only way we can start discussing the details is with a constructive dialog with several of the major players on this forum to sequence the same passage of music so that the pros/cons of the various approaches will become apparent.

Best regards, gungnir

[mschiff]

Gungnir,

The best samples in the world will sound mechanical if the midi that drives them is mechanical. Since Sibelius does not allow for the editing of the midi velocity or duration of each note, I don\'t understand how you can get natural (human) sounding performances that match YOUR vision of how the notes should sound. I can see how you can use the Sibelius \"humanize\" functions, but that is someone else\'s idea of how the notes should be handled. I agree that Sibelius is by far the best scoring program as far as the interface and creating the score, but it\'s ability to play the score seems VERY limited.

The other issue that you seem to gloss over when you are figuring how long it will take to accomplish the creation of a virtual orchestra in Gigastudio, is that you have to audition and choose the sounds that you will use. This alone could take days. I\'m sure you didn\'t accomplish it in a couple of hours. You\'ve been collecting samples and auditioning them for some time (I would think). The fact is that you have already done that portion of the task, but it is up to the rest of us to accomplish it yet.

There are a number of people here that could make the brahms sound like a million $, but it wouldn\'t be with a virtual orchestra like you describe, it would be in the manner that they have previously described that they work with layering samples and working with a sequencer, not a scoring program. That is a much different task.

-- Martin

[Mel Tron]

I agree Sibelius needs a lot more in the Midi department to be really effective however:

I went to a Synphony Concert tonight; Brahms and Dvorak.

Sat about 40-50 feet back in the concert Hall 1st Balcony. IT occurred to me that:

This Orchestra could be sampled with the correct approach i.e. Vibrato, Expression, ALL the playing styles minimum 6 to 8 dynamic layers Groups and Pairs and Solos. etc. Realistic Stereo Imaging 24 bit 96 khz Digital Recording. IF ONE PERSON HAD THE VISION it might just work. All the instruments TEMPLATES would be identical; e.g. similar velocity levels, key switching etc. to create an easy Sequencer Template spread out over 4 GS systems on 64x4 Midi Channels.

I think it would be relatively easy to do. About 50 or 60k for the players. 10k for the Engineers and Gear and about 6 months of programming time. The big advantage of doing this would be that the instruments would all be sounding in the Same Hall and if the quality was there you would only need 1 Library rather than mixing and matching and getting nowhere and using a Reverb like Altiverb to try and unite the Sound. The key to this approach is in getting the right Stereo Image and position to begin with. I think it would all fall in place after that with the right approach.

[Craig_L]

Gungnir, Tokio Joe and RickD

Sorry about that - the link will work now - had to upgrade to the premium account - Yahoo briefcase use to allow free access.

I will check out the Brahms on Tokio Joe\'s link.

Craig

[LHong]

What is \"a virtual orchestra set up\", could you or someone explains it more what need to be done?
How do we get its setup?

Best regards,
Long


Originally posted by mschiff:

gungnir,

I agree with Donnie. This is the sort of thing that could take a week or two for someone that does not already have a virtual orchestra set up. Even though I already have one (outboard synths) here at work, I don\'t currently have it set up so I can record from it, though we just got a new console and I will have that capability very soon.

Be patient. I think there is a lot of interest and this can be an ongoing discussion.

-- Martin<font size=\"2\" face=\"Verdana, Arial\">

[J. Whaley]

Gungnir, if I may... I think you\'ve got some shtick about something. No one\'s against this forum, on the contrary, I think tons of people are for it. So... let\'s get to work... now it\'s going to take a while for people to get stuff together. But I think a good start would be an evaluation of current libraries.

I for one have the Advanced Orchestra Library from Best Service, and I think it sucks. The demos are AMAZING. Check the demos out if you haven\'t heard them.
http://www..com/cd_mp3_demos/106811.mp3 (\"http://www..com/cd_mp3_demos/106811.mp3\")

http://www..com/cd_mp3_demos/119195.mp3 (\"http://www..com/cd_mp3_demos/119195.mp3\")

2 different Demos. I think they are very cool and make a great synth orchestra, certainly worth the price of the library, the problem is I have yet to be able to create anything remotely similar to this. I really want to do a mock up of Ravel\'s piano concerto (mov. 2) but I can\'t get the sounds together to make it work. So, maybe this library sucks. I\'d like to know what other users think. If they like the library, then we need to figure out why we can\'t make it sound good (I can\'t make it sound good).

I think we should do this with each commercial library, list the pros and cons, then that gives us a jumping ground. If we\'re all shooting form different directions it won\'t make any sense and the data we have won\'t line up.

As for Sebelius.... I doubt it\'s the most definitive program on the market. Maybe it is... i don\'t think so. I use logic for sequencing and finale for Notation, i\'ve worked with sebelius, but have no reason to believe it\'s the standard for MIDI playback... but rather very good reason to believe it\'s NOT the standard.... but this forum should be Sequencer independant. It should be general Orchestral sample playback and \"virtual orchestra\" mock up info. And YES.... I think there is such think as a virtual orchestra... I just havn\'t discovered it yet...

Okay dudes... let\'s start defining the commercial programs... and again, does anyone know how we could get a dedicated forum on this site for this subject? Cause it would be a lot easier to work if we could have seperate threads for each library, and specific topics images/icons/smile.gif I\'m so excited about this guys, let\'s get the word out, I\'ve already told 3 other dudes to come over here who I know will add a lot images/icons/smile.gif Later

J-

[KingIdiot]

Mel,..IMO you\'ll still need reverb (dynamic control like CC11 will control the room sound as well). well...maybe not at 8 dynamic levels....but could you see the polyphony demands if you tried crossfading between layers images/icons/wink.gif

If one person did it all, it would sound consistant and fantastic IMO. It would take one recording session to sound consistant tho, and it take many hours to record ALL articulations images/icons/smile.gif

ITs alot more money than you\'d expect to do it ALL. Which is why no one\'s done it yet. And even if someone does, it will be \"one sound\" not all the sounds one can get from different recording techniques. So in the end, not everyone will be happy images/icons/smile.gif

BTW J, the thing you need to remember about those demos is alot of it is the phrases. AO is fantastic for those and layering in. Also you need to be able to mix to get it to sound that good images/icons/smile.gif

Listen to stuff posted by Maarten Sprujit. Some of the stuff he posts is AO. Also there are a few other composers who\'s names I\'m forgetting at the moment (Giovanni...something) that use AO with some extremely great results.

I used to think AO was horrible, but I\'m finding myself coming back to it from time to time for phrases and licks.

I personally dont like the sound of the string section sustains, but they can still be useful.

[Hasen]

Originally posted by gungnir:
Hasen, I wasn\'t reacting to your comment, my snide remark was because I sometimes get discouraged with film music because it\'s so full of overused cliches and boring trendy formulas; have you noticed that its almost always the same guy, with the emphatic deep voice, that is used in the trailers of many of the new movies (the guy used on Lord of the Rings).<font size=\"2\" face=\"Verdana, Arial\">That guy\'s cool. I wish I had a voice like that. images/icons/wink.gif


In regards to not having heard many film scores, I have just now returned from catching Spiderman, the motivation being primarily Danny Elfman\'s score. It was incredible, this guy is now playing the orchestra like Paganini played his violin. He seems to bend it, and make it pulsate in pure magical fashion. I also harbor an underlying resentment regarding movie scores because it seems that music usually functions in such a subservient manner. I would like to see more movies that used music as a character in the film, not as just another special effect.<font size=\"2\" face=\"Verdana, Arial\">Music in films is used in many different ways to achieve many different results. It can be used to accentuate something as well as to suggest something. Having it exist as a character in a film is interesting but not always desireable.


But in general, the qualities of deep emotive inspiration are lacking because there just isn\'t the talent to do it. <font size=\"2\" face=\"Verdana, Arial\">I don\'t think that is true at all, there are many wonderful film composers.


No wonder John Boorman had to go back to Wagner to conjure the ominous foreboding power needed in the Excalibur opening. Just look at the sheet music of that Siegfried motif, a tritone infused with the energy of triplets on top of triplets in the violas and cellos. Now that is the stuff of real visual music.

Best regards, gungnir

<font size=\"2\" face=\"Verdana, Arial\">The reason John went with Wagner\'s scores is because Trever Jones\' original main theme score was awful. Its not the first time Jones has had to be replaced by another composer, Randy Edelman had to be called in last minute to score Last of the Mohicans.

Its part of the fact that the music is written for film that makes it good. Classical music is written for music\'s sake but movie music is written for a specific purpose. I\'m not saying one is better than the other but I think we should explore both areas.

[ninriggs]

Its a pity that Bill Brown isnt on this forum,
His musical skills seem to be amazing.
I wonder what feedback he could give us all.

A lot of his game work is still done on samplers.
(rogue spear, rainbow six)

I only wish i could write strings like he does

images/icons/smile.gif

[J. Whaley]

King idiot... you\'re not half the idiot I am images/icons/grin.gif

Yes, Advanced Orch. is certainly best for the phrases and articulations.. but who wants to sit around and compose a piece of music by digging out phrases and articulations? and more importantly, that won\'t work if you\'re trying to do a real orch piece, not your own piece concieved for AO.

IMHO: AO perc is good, not the greatest cause it lacks in selection, but good. The harp is good, again, could use more variety, but it\'s good. But the other instruments are so hard to play, they\'re anti-functional. Who wants to sequence a string pad, then go back through and decide which notes need a \"initial attack\" and which notes don\'t. Just doesn\'t make sense. And it seems to me that the \"long\" string samples don\'t move. They\'re very static sounds. apparently they have compiled more complex sounds for Giga sampler, but to trade in my akai version they want me to pay half the price of the AO library... well I don\'t like the one I have, who thinks I\'m gonna want to pay another $400 for something I don\'t expect to use? Anyone else feel this way about AO? I\'d love to learn to use it well, but I just think it\'s poorly laid out.

Where can I find demos from the dude you mentioned?

---The greater idiot!

[csduke]

Originally posted by mschiff:
The best samples in the world will sound mechanical if the midi that drives them is mechanical. Since Sibelius does not allow for the editing of the midi velocity or duration of each note, I don\'t understand how you can get natural (human) sounding performances that match YOUR vision of how the notes should sound. <font size=\"2\" face=\"Verdana, Arial\">Yes, I would also be interested in knowing how quantized and/or automatically humanized midi can sound natural/real since 1) humans do not play quantized and 2) the playing variations from a quantized absolute is very context specific.

Software is not very good at identifying context. Even if you solve the problem of identifying context in music (is this a solo, is it a doubled solo, is it at the end of a phrase, …) there are thousands+ of contexts for the software to be programmed for. Come to think of it, I would be happy to program such a system but would need to be paid by the hour ;-)

Another aspect of midi manipulation as it supports a convincing sound is that these manipulations have to deal with library specific matters. For example, all sustained trumpet samples are not the same (between libraries) and hence require midi different manipulation. You know what I mean.

[Duncan Brinsmead]

If the samples were recorded correctly in the first place, providing the user with the emotion and vibrancy required to phrase a passage convincingly, and also with the proper ambience, much of this time consuming foolishness with midi and gearbage would not be necessary. <font size=\"2\" face=\"Verdana, Arial\">I think there is an important point being missed here. One can\'t put the \"proper\" expression into a single note. Expression is something that happens at the level of an entire composition, not a note. It is like saying that if we just get enough samples of great actors saying all the vowels and consonants with GREAT expression then when we synthesize Hamlet out of the sounds it too will have expression.

The reason musical instruments such as the violin are difficult to master is not because they are vastly inferior to the keyboard, but because they offer many more dimensions of control over the sound. It takes time to master all the controls, but the design of most classical instruments have evolved over hundreds of years to have amazing control. Any attempt to automate the expression resulting from this control will ultimately sound dull, because there is no higher intelligence driving it.
The only hope of getting something equal to a living breathing orchestra( with a hundred VERY skilled musicians ) is to improve the man machine interface to something with the subtle response of a violin. Doing all the tweeking in post is just too time consuming to be practical.
Perhaps someday we can write performance programs that interpret the music, but for it to sound as good as real humans it will have to be as intelligent. (if it doesn\'t have true intelligence a performance may sound OK for a few minutes, but quickly becomes boring)

Duncan

[KingIdiot]

J,

I know now one\"wants\" to do thed work involved in sequencing an orchestral peice, but honestly its what needs to be done.

MIDI orchestration is a skill itself. Composing another one, Mixing another. Tweaking samples is even becoming one of its own for me.

You need to be good at the first three to make a DENT into getting a MIDI orchestra to sound good. Its jsut that plain and simple. You wont EVER get it to sound real out of the box. There is jsut too much nuance in an actual orchestra. Legato playing itself has several issues that aren\'t addressed in ANY libray. Short fast passages have so many variations/possibilities that it will take more than a variety/random amount of samples to make them sound real (it will take some intelligent programming and a sampling style no one has done yet).

A one man MIDI orchestrator needs to be good at a few things, and has to do alot of work to get their piecees to sound good.

Like I said no one \"wants\" to do it, but the people who\'s pieces sound FANTASTIC, are the people who do the work.

About phrases and AO, yah they are likmiting, but using them as layers underneath or over actual sequenced parts help add the realism. Thats what they are there for. Fast runs up to a note for example.

Also about the strings, yes the sustains are very stark, they also dont have great tone IMO. but they can still be useful. Try layering the solo violin over the violin section, it can add alot of dimension. Same with all the other solo instruments.

Obvciously, I hear it all the time, \"I dont want to have to layer stuff, I just want it to sound good\". Tough images/icons/smile.gif

[Duncan Brinsmead]

Something that has bothered me( and I\'m as guilty as anyone ) is our attempt to mimick the sound of an orchestra, rather than recreate the emotional impact of an orchestra. Frequently I\'ve listened to demos that sound roughly like an orchestra, but are lifeless and empty to listen to. Presumably with all our powerful computers and sequencers we should be able to make symphonies of sound that go beyond what an orchestra can do.. why limit the sound pallet to a late romatic period orchestra?
It should not be \"that sounds exactly like a cello\", but \"that is as expressive and affecting as a cello, but with sounds beyond the range of a cello\".
By attempting to mimick an orchestra, rather than embracing the new potential of our new computer instrument, are we not being nearsighted?
If one listens to \"Switched On Bach\" it shows what could be done with the early synths. In many respects I do find this performance much less listenable than on period instruments, but it at least added a new element to these works that nobody had heard before. It was also much more successful in record sales than any synthetic orchestra rendering I know of ( lets ignore movie sound tracks here.. people buy these in part because they liked the movie ).

[mschiff]

King,

Sequencing with samples, and/or synths is a completely different animal from creating an orchestra simulation. You are completely correct that it takes a lot of work to do the former, but to do the latter, it takes only the work of setting it up to begin with. Once it is set up, then you can use it to compose with till the cows come home. For a composer, the simulation does not have to be perfect, or even that good. It just has to simulate the way an orchestra will play the piece, so that you can \"debug\" the score before you hire a real orchestra to play it. Granted it should also be pleasant to listen to, since you will listen over and over, but it will not necessarily sound the same as a piece that had been meticulously sequenced.

-- Martin

[Hasen]

( lets ignore movie sound tracks here.. people buy these in part because they liked the movie ).<font size=\"2\" face=\"Verdana, Arial\">I don\'t. It doesn\'t affect my decision at all as to whether I like a soundtrack but I see your point that a lot of people feel this way. I doubt its the case with most of the people on this forum though...

Classical music is generally a lot easier to simulate than modern film music so this test won\'t be that conclusive IMO.

[composer22]

And, as good as musicians CAN be, remember that there are plenty of BAD performances as well..

So it goes both ways.

Personally, IMHO the best approach is alittle of both...similar to King\'s suggestions:

Use layered approach - hire small set of string players, solo winds etc and add middle voices with samples, or thicken the voices...

[KingIdiot]

Duncan,

I agree. I\'m sure there are a few people doing this tho. The chatter on this board doesn\'t represent that tho.

mschiff,

Of course. If used as a template/tool for cranking out \"IDEAS\" then yah its not much work to do. One must keep focusing on the composition and that its not going to sound real.

Which brings me to another point. Percieved realism affects what one hears in the \"quality\" of the samples, when they listen to demos. ITs not he sound of the samples themselves, but how real they sound that many people listen to subconciously.

Anyhow, my point was more about the AO demos. It takes a bit of understanding of MIDI, and a bit of work to get demos to sound liek that (tho they arent that great).

If you are writing for an orchestra, then use samples as mockup tools to check harmonies and balance and such. If the MIDI is your final piece, then it will take WORK to make it sound more realistic. Thats just plain and simple.

its like the amount of work it took each string player to learn to play the strings. One needs to spend time learning to make samples sound good.

[Duncan Brinsmead]

Classical music is generally a lot easier to simulate than modern film music so this test won\'t be that conclusive IMO.
<font size=\"2\" face=\"Verdana, Arial\">Perhaps you say this because you are trying to create film music. I would disagree, especially if you include late romantic music. I\'ve heard stuff that comes close to John Williams soundtracks. However I have never hear anything remotely close to something like Mahler\'s second symphony.. listen to the final movement from one of Bernstein versions and see if you can imagine it being done on a computer. Or listen to the Sibelius violin concerto. There is a strong bias here towards soundtrack scoring, however keep in mind that this music was meant to underscore dialogue, action and imagery. It is not that hard to have big whonking bass drums with tons of reverb and blasting horns underscoring dramatic action. It is much harder to create a performance of a delicate Mozart chamber work that will bring those with experienced ears to cry at its perfection and beauty. I have great admiration for the creativity and timing involved in works like the Loony Tune\'s soundtracks.. they work wonderfully with the images but are not intended to be listened to on their own. As such, I think one can also tolerate performances that are not as gripping. I have sometimes enjoyed music at a movie only to get the album and be disappointed with how dull and contrived it seemed on repeated listenings. The story is not entirely within the music. Of course some movie music does stand well on its own(no need to bomb me with examples)

[gungnir]

Duncan, upon re-reading my response to your excellent hands-on approach to trying to solve some of the inherent problems with virtual orchestration, it appeared that I was passing over them as if I was some type of guru on the topic. This was not intended, it was very late at night and I was in a rush, sorry, I didn\'t intend to appear flippant about the practical topics you raised. It\'s obvious you have thought about these issues and have considerable experience. Thanks for taking the time to share these issues with us, this is after all the practical type of dialog I am hoping may finally be discussed on this forum.

What I really mean is that I believe there should be less reliance on midi and more emphasis on constant manipulation of the available sample articulations.


To me the biggest single issue is achieving a fluid expressive legato. Most sustaining instruments do not simply jump pitch to new notes, but fluidly slide in a variable manner <font size=\"2\" face=\"Verdana, Arial\">How true. That\'s why I though the Brahms passage was a good choice to start a dialog. That\'s why you don\'t find many extended passages on manufacturer\'s demos. And that\'s why the problem deserves some discussion on user forums.

Another issue is expression in tempo.. not just from bar to bar, but from note to note. Excellent point.


Some people make a big deal over the cleaness of samples, quality of mics etc. However I don\'t think that this is too important when it comes to making natural sounding instruments. <font size=\"2\" face=\"Verdana, Arial\">Yes, it\'s the old quantity over quality hangup, technically perfect recording of lifeless samples that rampant out there. Even Vitous doesn\'t come close. But this should be the topic of another discussion cause its so big and important.


One can\'t put the \"proper\" expression into a single note. Expression is something that happens at the level of an entire composition, not a note <font size=\"2\" face=\"Verdana, Arial\">I see what you mean, but I think a lot of it can. I believe this must be the direction we go cause midi manipulation will ALWAYS sound fake. You can approach sampling with the idea of capturing notes within a certain pre-determined context. If you need a mp flute vibrato swell, it can be recorded at three different tempos for example. Sure, the process will be ambitious, but I, for one, would be willing to pay for a flute library that gave me 80% of the most common articulations so that I don\'t have to worry about playing with controllers any more. I play flute myself, and its only time that preventing me from having a go at it.


The best samples in the world will sound mechanical if the midi that drives them is mechanical. Since Sibelius does not allow for the editing of the midi velocity or duration of each note, I don\'t understand how you can get natural (human) sounding performances that match YOUR vision of how the notes should sound <font size=\"2\" face=\"Verdana, Arial\">Other than real time playing of a fast passage into a sequencer, why anybody would anybody ever use velocity as trigger for samples or sample layers? What does the speed of an initial attack have to do with summoning the required dynamic layer of a note as it organically must change in time?


The other issue that you seem to gloss over when you are figuring how long it will take to accomplish the creation of a virtual orchestra in Gigastudio, is that you have to audition and choose the sounds that you will use. This alone could take days. I\'m sure you didn\'t accomplish it in a couple of hours. You\'ve been collecting samples and auditioning them for some time (I would think). The fact is that you have already done that portion of the task, but it is up to the rest of us to accomplish it yet <font size=\"2\" face=\"Verdana, Arial\">A workup like the Brahms sequence should take no more than two hours, IF the midi data is already loaded. If it takes longer, your system may not be setup properly. I have already said that I\'m willing to prove this with a random sequence, but only if several others are willing to do so. However, it seems that practical applications of sounds rather than talking about sounds seldom happens on this forum. Just look at what happened to Donnie\'s Psalms thread a while back, where\'s the reverb thread now?

Although I didn\'t really want to discuss sequencers on this thread, I will say this. Sibelius, once modified to interface directly and completely with GigaStudio (including registry changes), is, even with its limitations regarding direct realtime input, is, for the purposes of creating symphonic realizations, the best sequencer you can use by far. I am not saying situations where you syncing to audio or playing in a lot of parts. I have neither the interest or the talent to play in all the parts of a symphonic score. I just want to quickly get an idea coming out of my speakers with a 85-90% realistic rendering.

Anybody who purchases sample libraries should obviously become aware of the relative strengths and weakness of the articulations, and use them accordingly. For example, I usually automatically default to Vitous for espressivo lines, Gary\'s attacks for fast passages, etc. It\'s all built in, waiting; every articulation, every dynamic sample, every library sitting ready to be fired and used optimally in the expressive context of the phrase. All within Sibelius, no more god-awful gizmo gearbage intermediate programs, GigaStudio stations waiting for Sibelius. The most positive and elegant composition environment in the world evoking the most appropriate articulations in the most powerful sampler in the world (at least currently) with lightning fast direct input. You input the notes you want in a phrase, then you dab in the passion you want each note to yield at the same time. You hear the result immediately by pressing a button. I haven\'t glossed over the issue of time, indeed I\'m attempting to do just the opposite when I started this thread, it is of pivotal importance and is the one the reasons why virtual orchestral is currently a myth and why nobody on a worldwide forum about sample libraries does a mockup of a few bars of string music. It\'s because, despite all the money they\'ve spent on libraries and gearbage they can\'t justify the time it takes them to do a simple peripheral project to shut me the hell up.


I think it would be relatively easy to do. About 50 or 60k for the players. 10k for the Engineers and Gear and about 6 months of programming time. The big advantage of doing this would be that the instruments would all be sounding in the Same Hall and if the quality was there you would only need 1 Library rather than mixing and matching and getting nowhere and using a Reverb like Altiverb to try and unite the Sound. <font size=\"2\" face=\"Verdana, Arial\">Mel Tron, you\'re talking my language. Now we\'re getting somewhere. All I can say is, your post is music to my ears! I think however you could do it for much less money than that, but that\'s a whole other topic.


I really want to do a mock up of Ravel\'s piano concerto (mov. 2) <font size=\"2\" face=\"Verdana, Arial\">One of the most poignantly beautiful pieces ever written, excellent choice but with AO, I\'m sorry, forget it. But it still is a good library to get started with, and has a few excellent articulations. Many of us started with that library, don\'t feel like you made the wrong decision.


I use logic for sequencing and finale for Notation, i\'ve worked with sebelius, but have no reason to believe it\'s the standard for MIDI playback... but rather very good reason to believe it\'s NOT the standard.... <font size=\"2\" face=\"Verdana, Arial\">You\'re right, out of the box, Sibelius as a sequencer is stupid. They even brag about it on the sight. They stand up with helmets on their heads shouting,
Sibelius is a notation program, it is not a sequencer! <font size=\"2\" face=\"Verdana, Arial\">The reason that they wear helmets when announcing this is that they find opportunity knocking so hard on their heads is distracting.


The reason John went with Wagner\'s scores is because Trever Jones\' original main theme score was awful. Its not the first time Jones has had to be replaced by another composer, Randy Edelman had to be called in last minute to score Last of the Mohicans <font size=\"2\" face=\"Verdana, Arial\">I don\'t agree. Jones\' original score has some beautiful pieces that are evocative of the period, with medieval instruments and textures. Igraine\'s dance was of immense importance in conjuring the rhythmic erotic context of the scene, and most of the rest of the score is quite beautiful, although very rare (available from a small company in Texas I think). Jones got sacked on Mohicans cause he tried to do what he perceived as the honorable thing, stating that he felt Mann was using too much music! His contribution is much more atmospheric and appropriate than Edelman\'s synth textures.


Another aspect of midi manipulation as it supports a convincing sound is that these manipulations have to deal with library specific matters. For example, all sustained trumpet samples are not the same (between libraries) and hence require midi different manipulation. You know what I mean. <font size=\"2\" face=\"Verdana, Arial\">Well, actually they don\'t, not if you have taken the time and trouble (a lot unfortunately) to configure your setup as it must be configured. All libraries respond to all midi messages exactly the same, Sibelius evokes the same response from every sample from every library so that all articulations can be instantly manipulated in the same manner with a mouse click.

All of this discussion is about making things real easy to immediately evoke the necessary sounds to quickly express your musical ideas in a very positive, hassle free environment. I\'m not suggesting for a moment that I can quickly do a mockup of a large movie score (Thomas J. has mastered that). I\'m saying only that it can be much faster to create much more expressive
scores for tradition symphonic works within the usual orchestral context.

Gungnir

[Hasen]

Originally posted by Duncan Brinsmead:

Perhaps you say this because you are trying to create film music. I would disagree, especially if you include late romantic music. I\'ve heard stuff that comes close to John Williams soundtracks. However I have never hear anything remotely close to something like Mahler\'s second symphony.. listen to the final movement from one of Bernstein versions and see if you can imagine it being done on a computer. Or listen to the Sibelius violin concerto. There is a strong bias here towards soundtrack scoring, however keep in mind that this music was meant to underscore dialogue, action and imagery. It is not that hard to have big whonking bass drums with tons of reverb and blasting horns underscoring dramatic action. It is much harder to create a performance of a delicate Mozart chamber work that will bring those with experienced ears to cry at its perfection and beauty. <font size=\"2\" face=\"Verdana, Arial\">I\'ve heard several mockups of classical pieces which sound almost identical to the original but have never heard any mockups of film music that come even close. Could you direct me to these John Williams score mockups that are so realistic?



I have great admiration for the creativity and timing involved in works like the Loony Tune\'s soundtracks.. they work wonderfully with the images but are not intended to be listened to on their own.<font size=\"2\" face=\"Verdana, Arial\">I entirely agree. A lot of film music is atonal and not really listenable at all. It depends entirely on what\'s happening in the movie; the music can be pulled in and out constantly making it of course unlistenable.
The fact of whether they are listenable or not is not the issue. Its whether we can recreate the sound.
Recreating a subtle classical piece is not that hard because samples do quiet very easily. Its emotional/powerful they don\'t so much. Getting power and emotion out of your samples is what its really about.

[Hasen]

I don\'t agree. Jones\' original score has some beautiful pieces that are evocative of the period, with medieval instruments and textures. Igraine\'s dance was of immense importance in conjuring the rhythmic erotic context of the scene, and most of the rest of the score is quite beautiful, although very rare (available from a small company in Texas I think).<font size=\"2\" face=\"Verdana, Arial\">I said the main theme was bad, not all the music he did.


Jones got sacked on Mohicans cause he tried to do what he perceived as the honorable thing, stating that he felt Mann was using too much music! His contribution is much more atmospheric and appropriate than Edelman\'s synth textures.<font size=\"2\" face=\"Verdana, Arial\">According to an interview with Randy, he scored just about all the music for the film in the end so anything you\'re hearing is probably his work.

[Duncan Brinsmead]

I\'ve heard several mockups of classical pieces which sound almost identical to the original but have never heard any mockups of film music that come even close. Could you direct me to these John Williams score mockups that are so realistic? <font size=\"2\" face=\"Verdana, Arial\">I\'ve heard some posts from regulars to this forum that get close that big film music sound(don\'t have the links handy). However I don\'t think this is something really distinct from classical music, unless you are talking before the romantic period. Berlioz, Mahler, R. Strauss and Wagner were all into a big sound. I really see much of the orchestral film music as a sub-genre out of the classical tradition. Most of it is not nearly as original as what went before. I might also add that much of film music is currently done with samples. (Currently there is no one that I\'m aware of listening to classical orchestral music performed on synths). There was a mockup of a score for a Dinosaurs(I think) that came very close to sounding like the full orchestra rendition( perhaps someone here remembers the link, or at least the composer).
At any rate, I\'m not talking about sound or color, but rather expression. When people listen to a great piece of music being performed by a great orchestra and conductor they expect a level of nuance and drama in expression that is quite high. The performers are actors and the the score is the play. It is not enough that the actors simply sound like humans.. they must act the meaning of the play. It is not about color and tone so much( although this is more significant in impressionist works ). It is this element that I find missing in most stuff that uses sequencers.
Duncan

[composer22]

Originally posted by Duncan Brinsmead:
...There was a mockup of a score for a Dinosaurs(I think) that came very close to sounding like the full orchestra rendition( perhaps someone here remembers the link, or at least the composer).

<font size=\"2\" face=\"Verdana, Arial\">That was JNHoward...and if you do diff\'s between the two files Live vs. MIDI, you will find they are BOTH exactly the same...e.g. they are BOTH live recordings... images/icons/frown.gif

http://www.keyboardmag.com/features/jnhoward/index.shtml (\"http://www.keyboardmag.com/features/jnhoward/index.shtml\")




At any rate, I\'m not talking about sound or color, but rather expression... <font size=\"2\" face=\"Verdana, Arial\">Yes, and add *NOT* perfect...the non-linear/chaotic elements is missing from MIDI...perhaps thats where some AI make prove useful, but people have tried and we are still not there yet.

Moral: save your money and hire an orchestra

[Duncan Brinsmead]

Was that for sure about the two versions being identical? I thought that one had some instruments that were sampled( as opposed to the full score ).

Yes, and add *NOT* perfect...the non-linear/chaotic elements is missing from MIDI...perhaps thats where some AI make prove useful, but people have tried and we are still not there yet.
<font size=\"2\" face=\"Verdana, Arial\">The problem here is that people sometimes confuse intentional and random variation. When we listen to a midi file that has been randomized it initially sounds more human. However upon further listening the ear picks up the fact that all the random bits are not adding content, but simply randomness. If one adds enough expressive control to the instruments the connection to the human will by its nature add a little randomness, but I don\'t think that is what really makes it human. I do agree that some things, like waveforms that exactly repeat over a short period, can sound quite unnatural and will distract the listener.

[LifeForceExplorer]

I\'ve watched this thread with interest - from something of an amateur perspective. I\'m a fifty year old weekend warrior /classical composer - with no music deadlines and no films to write scores for. I use GigaStudio and Sibelius also - having studied the software market til I was maxed out and glassy-eyed from analyzing all the wonderful but whiz-bang, rock n roll driven gear and software programs with interfaces that look like control panels for an F-16 jet. My ambition is to use GS/Sibelius as a practical tool for small ensemble classical composing and \"virtual performing\" and recording. I have enough friends who are serious amateur musicians that I can occasionally, when I\'m lucky, have my work played at chamber workshops the real way. But this is rare, of course, and so I\'d like to make the larger portion of my work, which will probably never be played live, as realistic and expressive sounding as possible at the computer - mainly just for my own simple pleasure and growth as a musician.

This dialog brings two issues to mind:

1) Practical issue #1 - Most people making the real progress in this field are professionals doing it for a living - often with deadlines and a career at stake and often tied to films or games or other commercial applications. Therefore, there is actually a curious but understandable reluctance to share trade secrets here. And I really mean it when I say it\'s understandable. This means that someone like me (an amateur who makes a living 9 to 5 in a field wholly unrelated to music) cant REALLY usually learn many advanced techniques here (except for a few professional individuals here who happen to enjoy sharing their professional knowledge)! And I certainly dont begrudge the other professionals. Look at all the years theyve put into developing the unique, high quality sound that is their PRODUCT and their LIVING and their PRIDE. Absolutely understandable. And there is still lots of GOOD and HELPFUL information here for the GS novice getting to the intermediate level like me.

2) The GigaStudio / Sibelius combination seems potentially very powerful to me also. Obviously, gungnir, youve taken it much farther than I have. But where would you stand on posting the technical essentials on how youve been able to get a more natural sound from Sibelius with GigaStudio? I wouldnt begrudge you one bit if you said thats your own hard-earned trade secret of sorts with a value to you (financial or even just as a matter of pride) that you dont wish to just lay out for free for others. But we can save a lot of time here if nearly everyone\'s bottom line is that \"the way I achieve my sound is something I\'ll keep in my own private business or artistic briefcase.\"

If you and enough others, however, would like to share that knowledge, I wonder if there are enough people interested to justify a Sibelius/GS forum here - or somewhere else. It seems to me that there are quite a few people working with these two programs together now. And some people, like yourself gungnir, have apparently become quite expert with them. Such a forum topic might reduce some of the endless variations that GS topics can spin off into - because of different computers and gear and sequencers and sound editors and versions and on and on. Just the Sibelius > GigaStudio interface period as the main topic - with related sequencer discussion and other spin-off topics of course, but still mainly a Sib>GS focus. Does this seem like an interesting forum section idea here to anyone else? I notice that the Sib/GS threads seem to often jump right up to a dozen or more entries pretty fast here lately. Or maybe there are just five people people here who are keenly interested but a great majority who are not.

These are just some related thoughts, gungnir, that I hope you dont mind me appending to your original virtual orchestra post. Perhaps these questions also help to sort out whether your original proposal would lead, as you say, to 1) a sort of Mp3 \"competition\", or 2) to a direct sharing of information with a common purpose. If it\'s just a competition, lets call it that and just enjoy that for what it is. I always like a good competition - that\'s what the North German organists used to do for fun until old JS Bach started blowing everyone away and became the Tiger Woods of the organ. Me, I\'m not even in the PGA of orchestral computer music yet, so I\'m just enjoying watching the real pros from the crowd for right now and working on my serious but amateur computer-music game at home on weekends.

[PatS]

Originally posted by composer22:
</font><blockquote><font size=\"1\" face=\"Verdana, Arial\">quote:</font><hr /><font size=\"2\" face=\"Verdana, Arial\">Originally posted by Duncan Brinsmead:
...There was a mockup of a score for a Dinosaurs(I think) that came very close to sounding like the full orchestra rendition( perhaps someone here remembers the link, or at least the composer).

<font size=\"2\" face=\"Verdana, Arial\">That was JNHoward...and if you do diff\'s between the two files Live vs. MIDI, you will find they are BOTH exactly the same...e.g. they are BOTH live recordings... images/icons/frown.gif
</font><hr /></blockquote><font size=\"2\" face=\"Verdana, Arial\">The two files that I downloaded soon after they were posted online (dated 1/24/2001) are noticeably different from each other, both in size and sound. The MIDI version sounds quite nice, but you can tell it\'s a MIDI mockup almost immediately.

Feel free to e-mail me at pshove@ghs.com, if you\'d like a copy of either file or both. The \"live\" version is 1,914KB (1.86MB), and the MIDI version is 1,943KB (1.89MB).

Pat

[Hasen]

Originally posted by PatS:

The two files that I downloaded soon after they were posted online (dated 1/24/2001) are noticeably different from each other, both in size and sound. The MIDI version sounds quite nice, but you can tell it\'s a MIDI mockup almost immediately.<font size=\"2\" face=\"Verdana, Arial\">Thanks for stepping in there Pat. I heard the original mockup as Simon Ravn posted it on his site once the other one had become corrupted, and yes, it is obvious which one is real and which one is the mockup.

The most realistic mockup I have ever heard is Hans Zimmer\'s \'The Gladiator Waltz\' which is essentially the same track as \'The Battle\' from the original soundtrack. Its virtually identical but Hans used his special custom orchestra samples recorded with the London Symphony Orchestra. I want to know what can be done with the commercially available sounds.
I agree with you Duncan that the challenge here is to get emotion and reality into the mockups. As I recently found its not just the reality in the performance of the samples but the acoustic hall they are all made to exist in, ie the reverb.

[gungnir]

In the consumer driven world of modern art, this language will slowly die leaving us with the infantile musical language of morons, the modern movie score. <font size=\"2\" face=\"Verdana, Arial\">Correction, the end of that sentence should obviously read \"as ominiously demonstrated in many modern movie scores\". There are several superb soundtracks out there as we all know which play a huge integral role in the success of the motion picture, and make major contributions to our musical culture. Upon re-reading my post I discovered what would be an absolutely absurd statement.

[Hasen]

Originally posted by gungnir:

Quote from the liner notes on the Excalibur CD; \"Unlike other films where an original composition is thrown out in favour of a classical temp-track (2001:Space Odyssey, notably), composer Trevor Jones was contracted with the knowledge that Boorman intended to score the film primarily with classical music. Opera by Wagner and Orff based on the stories of Arthur and the Grail had appropriately been chosen as themes for the major characters, and Jones was to create the supporting musical atmosphere for the film. Boorman went one step further, however, and asked Jones to compose both a main theme and an end credit suite to be used in the event that specific performances of the classical music he had chosen were unavailable for licensing.<font size=\"2\" face=\"Verdana, Arial\">I stand corrected. images/icons/smile.gif


Well, Randy\'s telling a fib. Most of the first half of Mohicans (you know, the popular part that everyone buys the cd for) was written by Jones, unless there\'s a misprint on the CD I\'m holding. As for Edelman\'s heroic entry to save the score at the last minute, may I offer a few quotes from music scores sites:
- What remains consistent, though, is the belief that Jones\' first two tracks are the highlight of the album.
- The main title theme is so magnificent and captivating that Jones would use mutations of the theme in
Cliffhanger (1993) and the television film Cleopatra (1999).
- Edelman\'s music doesn\'t fit with the epic proportions of the film.
- Also at issue is the incompatibility of the two parts of the album release. People generally pick and choose which half of the release they enjoy better, but many will also agree that the two halves just don\'t work well together. Jones\' themes are so orchestrally sweeping that they make Edelman\'s music sound like simple filler material.<font size=\"2\" face=\"Verdana, Arial\">Well what Randy said in an interview it that basically he did virtually all of it despite what everything else says. Who knows what to believe but I\'ve no doubt Randy is a better composer than Trevor.


Er, you are the same person who stated that \"I get the impression from your post that you have not heard many film scores...\" ? I listen, but usually grimacing in pain.<font size=\"2\" face=\"Verdana, Arial\">You\'re listening to the wong ones then. Of coures there is bad film music just as much as there is bad classical music, you just have to find the good ones.


Well Hasen, I would argue that this voice is not cool, but the voice of death. His voice is the voice of monoculture, corporate incest and the symbol of the new tyranny that is insidiously gaining ground every day. Haven\'t you noticing how things are going, how there is an averaging trend everywhere. Have you perceived how cars are looking all the same, that news services are using the same essential sources and in many cases are owned by the same companies, how, with globalization, the seedy strip malls and sleazy businesses from North America are infecting the rest of the world and bringing it\'s vulgar language of business, English, with it. Political parties are getting harder to differentiate as they read the polls and spew their corrected superficial rhetoric, the days of the week have become indistinguishable as the malignant greed of corporations have stolen the common sense idea of having one day of the week to chill out and walk with some kids in a park that has less street and construction noise than other days.
Glodal warming is averaging out the extremes in weather, music is by and large nothing but a formula with all its sure to sell permutations, celtic rock, country rock, soft rock and the most hideous concept of them all, world music, the ultimate averaging device. And what we\'ve been watching up on that movie screen is being decided by businessmen more than by artists with each passing day...*snip*...<font size=\"2\" face=\"Verdana, Arial\">Er...lighten up a bit, its just a guy with a cool voice...ok a \'voice of death\' as you put it. images/icons/smile.gif Who would you suggest does the voiceover?

As for Sibelius as a writing tool with GS, I can see the attraction. I imagine its not that unsimilar to \'tracking\'. I used to track orchestral pieces and its enjoyable. You just put the notes where you want them and then adjust the volumes etc until it sounds as real as you can get it. I don\'t think it can compete with actually performing the parts though like with a sequencer. You get so much realism in a few seconds that would take you hours of editing to achieve.

[KingIdiot]

Actually last I remember there was more than \"one\" guy doing those voice overs. One of them was the guy who did the voice for Optimus Prime in the Transformers Cartoon.

When I hear his it brings a smile to my face images/icons/smile.gif

I personally like voice overs taken from monologues in the movie.

and now...I\'m WAY off topic images/icons/smile.gif

[Hasen]

And how cool is Optimus Prime? You certainly can\'t argue with him. images/icons/smile.gif I think Orson Welles voice as Unicron might be even better mind you.

You\'re right though King, its Peter Cullen and he frequently does voices for commercials/trailers etc.

As for the topic in hand, if anyone can do a mockup of the Love theme from Superman then I\'ll...well I don\'t know what I\'ll do but I\'ll be impressed to say the least! images/icons/smile.gif

[composer22]

gungnir

Can I use your rambling monologue for a two act one person play I wanna do Off-Off BWay next year...

Or perhaps as an audition piece for an autobiography about Wagner that I am up for in L.A. next month?

images/icons/tongue.gif

[gungnir]

Hey, you try and save the world on an sample library forum and get no dam respect!

But, ok, it\'s a deal, but it\'s three acts, I get 10% at the door, and if there\'s any scenes with
the starring actress, I\'m called in.

I\'m working on the sequel now.

Gungnir

[composer22]

Damned, and i was looking forward to those steamy scenes with Cosima...

[tomhartman]

Originally posted by J. Whaley:
King idiot... you\'re not half the idiot I am images/icons/grin.gif

Who wants to sequence a string pad, then go back through and decide which notes need a \"initial attack\" and which notes don\'t. Just doesn\'t make sense.

---The greater idiot!<font size=\"2\" face=\"Verdana, Arial\">Well, this is the problem, or has been for me.

I have Gary\'s strings, Roland strings, etc, and with a library with multiple articulations and it becomes a joke trying to put together something credible. You can sometimes have 10 midi tracks with different articulations on each just to play a line that goes by in 5 seconds. With clients expecting work to be completed by ironclad deadlines, even if they were the greatest samples ever heard by man, it is SO time consuming to put together a large, or even not so large, string passage that competes with the real deal that it becomes impractical. Half the time, I\'ve gone and grabbed a quick Roland string sample when I know with time I could do better with Gary\'s strings, for instance (not picking on Gary now..just a for instance) because it just takes to darn long. This is the same for brass, or for the VOTA choirs...all of it is a giant headache. I know that when you get as good as KINGIDIOT on programming this stuff things move faster. But I don\'t want to be a programmer. I don\'t want to TWEAK samples in an editor. I have no time for that stuff. I could use that valuable time to try to compose another version of a scene of sequence, rather than loosing all creative flow by squinting at numbers on a display. I\'m a very good composer with a lot of experience doing real sessions with real people, and about 20 years of working with MIDI mockups, so I\'m not green to it all. But even with talent and experience, it takes WAY too long to do something really impressive....at least, it takes ME way too long.

Alan Parsons was interviewed some years back about synth gear, and basically was saying he didn\'t care how powerful the box was, it was completely useless to him if he needed a programmer to get to the sounds he wanted out of it. It\'s the same thing here. It should not take 8 midi tracks to play a phrase. I don\'t know how it can be technically done or managed on GIGA or some future device, but for heaven\'s sake we need to be able to play a phrase that has a broad legato, a couple of short staccato notes in the middle, and end on a note that slowly adds a gentle vibrato as it holds out without going through the gyrations we currently must to achieve something that simple.

For me it\'s easier to tell clients they will need $15,000.00 or more for the spot and hire real players. I wish it weren\'t the case. Anyone here having similar thoughts?

[composer22]

I look at what I have and work toward:

1) Reverb
2) Articulation variety in a phrase
3) Variety of timing (being silightly late or on the beat, rubato etc)
4) Variety of pitch (not always playing 440)
5) Having more than one sample libr per instrument either as a group (a2) or per player to model slight tibre differences.
6) Residual sounds (release samples, finger scrapes on a string etc...) that a player makes.

Any others?

[gungnir]

The reason John went with Wagner\'s scores is because Trever Jones\' original main theme score was awful. <font size=\"2\" face=\"Verdana, Arial\">Quote from the liner notes on the Excalibur CD; \"Unlike other films where an original composition is thrown out in favour of a classical temp-track (2001:Space Odyssey, notably), composer Trevor Jones was contracted with the knowledge that Boorman intended to score the film primarily with classical music. Opera by Wagner and Orff based on the stories of Arthur and the Grail had appropriately been chosen as themes for the major characters, and Jones was to create the supporting musical atmosphere for the film. Boorman went one step further, however, and asked Jones to compose both a main theme and an end credit suite to be used in the event that specific performances of the classical music he had chosen were unavailable for licensing.


Its not the first time Jones has had to be replaced by another composer, Randy Edelman had to be called in last minute to score Last of the Mohicans. According to an interview with Randy, he scored just about all the music for the film in the end so anything you\'re hearing is probably his work. <font size=\"2\" face=\"Verdana, Arial\">Well, Randy\'s telling a fib. Most of the first half of Mohicans (you know, the popular part that everyone buys the cd for) was written by Jones, unless there\'s a misprint on the CD I\'m holding. As for Edelman\'s heroic entry to save the score at the last minute, may I offer a few quotes from music scores sites:

- What remains consistent, though, is the belief that Jones\' first two tracks are the highlight of the album.
- The main title theme is so magnificent and captivating that Jones would use mutations of the theme in
Cliffhanger (1993) and the television film Cleopatra (1999).
- Edelman\'s music doesn\'t fit with the epic proportions of the film.
- Also at issue is the incompatibility of the two parts of the album release. People generally pick and choose which half of the release they enjoy better, but many will also agree that the two halves just don\'t work well together. Jones\' themes are so orchestrally sweeping that they make Edelman\'s music sound like simple filler material.

Er, you are the same person who stated that \"I get the impression from your post that you have not heard many film scores...\" ? I listen, but usually grimacing in pain.


That guy\'s cool. I wish I had a voice like that. <font size=\"2\" face=\"Verdana, Arial\">Well Hasen, I would argue that this voice is not cool, but the voice of death. His voice is the voice of monoculture, corporate incest and the symbol of the new tyranny that is insidiously gaining ground every day. Haven\'t you noticing how things are going, how there is an averaging trend everywhere. Have you perceived how cars are looking all the same, that news services are using the same essential sources and in many cases are owned by the same companies, how, with globalization, the seedy strip malls and sleazy businesses from North America are infecting the rest of the world and bringing it\'s vulgar language of business, English, with it. Political parties are getting harder to differentiate as they read the polls and spew their corrected superficial rhetoric, the days of the week have become indistinguishable as the malignant greed of corporations have stolen the common sense idea of having one day of the week to chill out and walk with some kids in a park that has less street and construction noise than other days.

Glodal warming is averaging out the extremes in weather, music is by and large nothing but a formula with all its sure to sell permutations, celtic rock, country rock, soft rock and the most hideous concept of them all, world music, the ultimate averaging device. And what we\'ve been watching up on that movie screen is being decided by businessmen more than by artists with each passing day, and those businessmen have decided that your cool voiced man has the formula for success, to suck you into seeing another business venture designed to make sure you don\'t think, and with a mindless, knee-jerk superficial soundtrack to go with it. Music is a language, and like any other language, the nuances of its full expressive potential comes only through its cultivation by immersion with takes time. In the consumer driven world of modern art, this language will slowly die leaving us with the infantile musical language of morons, the modern movie score. I actually had to get up and leave the theatre rather than listen to any more of James Horner\'s squelching in the storm scenes in Perfect Storm, but then again I deserved it by going to that movie in the first place.


But where would you stand on posting the technical essentials on how youve been able to get a more natural sound from Sibelius with GigaStudio? <font size=\"2\" face=\"Verdana, Arial\">I\'m not sure, LifeForceExplorer, I\'m hardly motivated and I\'m not sure the approach would be for everyone. You have to be proficient with the editor, and with a wave editor like Wavelab or Soundforge and willing to spend some time at setting things up. I\'m not saying its any huge panacea, its just that I am absolutely addicted to the Sibelius working environment and hate having to leave it and will sacrifice some elements such as easy realtime input to achieve it (it can still be done, but its stupid and awkward). But for me, and the type of stuff I\'m doing, realtime would only be done for central, expressive themes on the main line so its no big deal. You gain some elements, lose others. And I haven\'t really finetuned some aspects of the compromising required within Sibelius.

I started this thread hoping somebody would do a workup on the Brahms so that others and I could learn more about the different approaches to our common problems. I did not originally intend on discussing sequencers, but of course that topic will quickly surface in any discussion about virtual orchestrations. There was a discussion on another thread about Sibelius vs. Finale, etc. a few weeks ago, but I just couldn\'t be bothered to pursue it, because, technically, everybody is right: Sibelius is a limp sequencer. I\'m just saying that it can be forced to be a great sequencer for specific applications, and that it must be made into a perfect sequencer in the future because then nothing in the world will ever catch up to it ever.

But we must find a way to stop wasting time and money on the bleeding-edge, and stop reading loser manuals about the gizmo of the month and dinosaur notation programs like Finale and start discussing some of the practical issues about working with samples. I feel that this whole approach (Sibelius with Giga) is currently the best hope for composers wishing to create orchestral simulations in an elegant and relatively fast environment. If its not working, it must be made to work because it\'s truly worth all the effort. These programs are the perfect synergetic interface, the project must not be dropped because too much is at stake. The problem, I feel, with programs like Sonar and Logic are that they aren\'t inherently musical in spirit, in fact, they are the antithesis of what real music is all about, which can never be better represented than the elegance of the standard syntax on aged virtual parchment which is almost tactile and organic in character. I believe the future holds Sibelius, a program like GigaStudio (or probably one of its me-too competitors) and a sample library such as the one Mel-Tron described with the result that virtual orchestration


If you and enough others, however, would like to share that knowledge, I wonder if there are enough people interested to justify a Sibelius/GS forum here - or somewhere else. <font size=\"2\" face=\"Verdana, Arial\">Yes, I\'ve thought of that. The Giga-Sibelius interface is being discussed on an ever increasing rate on the Sibelius site. What I was thinking is that it should be a membership site where people had access to uncommon midi files, some solo sample libraries (e.g I have friends that are willing to fill in some of the holes in currently available libraries), discussions on optimizing and developing the interface itself, discussions on the actual nature of composition as well (e.g. The Language of Music by Deryck Cooke, Leonard Meyer\'s Emotion and Meaning in Music and his theory about setting up expectations, etc., Swain\'s Musical Languages and so on). Compositions of members would be critiqued by others with similar setups and motivations, and so on. This is where it eventually has to go. It certainly will never happen on this site or at Sibelius where most of the discussions are about how to make a fermata look a little cuter.

Excellent summary of both some of the essentials and also very desirable nuances that must be developed, composer22.I totally agree. Because the concept of virtual orchestration is not going to go away. But was is the next step to drag it out of its infancy and take its first baby steps?

Cheers, Gungnir

[KingIdiot]

Well heres my thought on it.

You need to consider the fact that you are trying to recreate human espression. With MIDI you either have the option of creatiung you\'re own expression via multipe wavy(CC11 or GOS EXP, or Cfades, etc) or sticking with whats out of the box.

The problem is that whats out of the box willl not ALWAYS be what is right to EVERYONE\'s ears.

I\'lll say it AGAIN in another thread. MIDI orchestration/sequecing is an art unto itself.

It takes much tallent to get a computer to play samples \"realistically\"

Thats the thing. You are TELLIMG a computer to play something the way you want.

Computers are not intelligent enough to know when you want coffee. how are they EVER going to be intelligent enough to know what you want in a musical phrase?

and blah blah blah blah

images/icons/smile.gif

anyhow. I think the need for \"we want it NOW!\" is whats hurting us all. Its whats driving the \"generic\" dance music market (good electronic artists still tweak for weeks). Its also driving me insane. Pushing Music deadlines to \"we need it yesterday\". Thats plain stupid. Sample libraries that work well with ONE style of music is pushing these deadlines to smaller and smaller sizes.

ANYONE can let a library play itself for the musiv..... but writing music and making the library work is another task.

Its one that we all need to learn to get better at as MIDI musucians. Until we get the computer hooked to our brains that is!

who wants to work......indeed images/icons/smile.gif

I think I may be puttin my carreer to an early grave in sharing all I do. If all I have is a niche as it seems, then why should I.

You guys really need to work at this stuff. If you dont. The industry doesn\'t further in what we \"need\". but in what \"does *it* for us\". I\'m sorry, but if thats the way its going to be, it just looks like you\'re lazy. Or of course a Hobbyist....which is a term that I hate....but where does it put career profesionals? This is my career, its what I make my rent payments and insuatance and wahtever from. I share many of my findings with you guys and decelopers alike......
.....

but all I still see people not wanting to do ANY work for it, and wanting it done for them

If this is the case. I say write on paper and get real players to do the work.

I want to see results as much as the next person. But I personally dont want to see samples do everythign on their own (they will at some pont, and I hoope I win the lottery by then).

It may be that I\'m workign with 500k banks, or it may be the vodka....but man the Wants drive me nuts on this forum

lets get back to MIDI examples please!!! images/icons/smile.gif

I\'d be glad to take a look at a *CUSTOM MADE* MIDI file and MP3, and suggest some improvements if I can think of any.

and I do mean CUSTOM MADE!!!! I dont want to see simeple score to MIDI transcripts. I want people to try and use CC11 and or Mod Wheel. Its ecen better if they do Giga edits!!!No.....hey this MIDI should work on all libreary type stuff.

IMO thats like saying hey this vocal line will sound the same with different vocalists....Rock....pop...Operatic.......jeeez...

I need another drink......dammit wheres the absinthe!!!

[Hasen]

Originally posted by Kingidiot:

I\'d be glad to take a look at a *CUSTOM MADE* MIDI file and MP3, and suggest some improvements if I can think of any.
and I do mean CUSTOM MADE!!!! I dont want to see simeple score to MIDI transcripts. I want people to try and use CC11 and or Mod Wheel. Its ecen better if they do Giga edits!!!No.....hey this MIDI should work on all libreary type stuff.<font size=\"2\" face=\"Verdana, Arial\">That\'s pointless, no-one wants to share their own MIDI files.

[LifeForceExplorer]

[QUOTE]Originally posted by gungnir:

I feel that this whole approach (Sibelius with Giga) is currently the best hope for composers wishing to create orchestral simulations in an elegant and relatively fast environment.

...A membership site...(that includes) discussions on optimizing and developing the interface itself, discussions on the actual nature of composition as well (e.g. The Language of Music by Deryck Cooke, Leonard Meyer\'s Emotion and Meaning in Music and his theory about setting up expectations, etc., Swain\'s Musical Languages and so on). Compositions of members would be critiqued by others with similar setups and motivations, and so on. This is where it eventually has to go.

* * * * * * * *

Yep. Well said, gungnir. (Didn\'t we chat about all this once at noteheads on their mailing list? I think we were both evaluating \"Igor\" at about the same time.)

I think that most - but thankfully not all - serious classical composers have somewhat reasonably rejected MIDI and computer music technology in general as a place for making music. The fact that so much time and energy must be devoted to the technology itself is of course the big turn off. In reality ANY software tool that gets me to a realistic sounding performance is what I\'m going to jump on at any reasonable price that I can afford. I\'d like 90% of my creative energy (such as it is after the work week!) to go toward composing and harmony and 10% toward computer technology - which may not become a computer-based reality in our lifetimes! The fact that 80% of my energy has gone toward studying/learning technology for the past nine months has been a little insane! But the fact that Giga>Sibelius is so promising and that I found the worlds simplest yet fully compatible sequencer in PowerTracks7 (for preparing performances that I just cant seem to get from Sibelius) with excellent 960ppqn resolution and good digital audio capability now has me more optimistic!

I can actually see the day on the horizon when I\'ll be devoting 70 or maybe even 80 percent of my attention to composing again! So it was refreshing to see you mention composing which just about gets buried because we\'re in the midst of such huge technological changes. On the rock and roll / new age side of this whole computer-music industry, the skills of MUSIC and HARMONY are just about completely obliterated for the moment. Why learn music when someone\'s new whiz bang \"arranger\" plug-in will do \"music\" for you! In fact, you can just hold down a few keys for three minutes and someone\'s new software will do that complicated \"music stuff\" for you while you just have fun adding \"effects\" and changing around the \"instruments\" like a real DJ!

I went to a local community college and took my first MIDI class two years ago. I was a forty-eight year old, ogling the girls like Rodney Dangerfield, while back among the college students. It was amazing to see what went on. A very low level of musical knowledge among the students - all of whom couldnt WAIT to get into the MIDI lab and hold down some keys and just LISTEN to some \"effects\". The college had racks and racks of gear - Proteus synths, Akai samplers and Fatar keyboards for the most part - and 6 PCs running various software. Easily $50,000 worth of gear for this mainly kindergarten like playing around. Still, who am I to put down \"playing around\"? We do \"play\" music after all, even in the so called \"serious\" realm of classical music - a realm that gets way to effete and ego-centric with plenty of its own human issues sometimes. I\'m sure there were a few students with music in their blood who were trying to make the same long climb up through all the technology and somehow back into MUSIC again that I am trying to make. But this new generation is going to have a long climb - if they are going to reclaim their own natural musicality - because the technology and the fun of it is so overwhelming really.

You might also find \"Harmonic Experience\" by WA Mathieu to be a book worth considering for your library, gungnir.

[tomhartman]

Originally posted by KingIdiot:
[QB]Well heres my thought on it.

You need to consider the fact that you are trying to recreate human espression. With MIDI you either have the option of creatiung you\'re own expression via multipe wavy(CC11 or GOS EXP, or Cfades, etc) or sticking with whats out of the box.

The problem is that whats out of the box willl not ALWAYS be what is right to EVERYONE\'s ears.

I\'lll say it AGAIN in another thread. MIDI orchestration/sequecing is an art unto itself.
QB]<font size=\"2\" face=\"Verdana, Arial\">That\'s because what\'s come out of the box so far needs so much cajoling to make it realistic, or even near realistic, and that was my point.

The idea that \"no one wants to do any work\" sounds like a good argument, but in fact, anyone writing music and recording it is already doing a LOT of work...especially when it\'s due by Thursday.. The fact that I and others, like yourself KING, do this for a living makes it all the more imperative that we have something works without being a tweakhead or MIDI scientist. Time is way too short on these projects.

I appreciate the fact that with practice, premade templates, and other shortcuts, things improve significantly re the time factor, but 2 minute sequence with busy strings is still a bear with existing technology and sample libraries, unless you let a lot go by with the thought that \"They\'ll never hear the difference,\" which is what I\'m forced to do a lot of times.

I don\'t think that having a sample...or patch..or whatever you want to term it....that automatically shifts to say a short bow from a long bow if you hit the note quickly is a lot to ask, and even that would be a huge time saver. Perhaps when I get more hip to the Giga Editor I can make one of those. Who knows. When I have time ..... images/icons/wink.gif

Sending mp3s for helpful suggestions is a great idea. I for one would love to do a short 10 second passage or so, and get some feedback on what I should have done, etc, and I\'m all for it if you guys are.....think it would help all of us.

How do we proceed?

[Hasen]

Originally posted by KingIdiot:

Hasen
Its not pointless. I dont mean send me your own music. It can be another piece of music, but not some lame GM ... just the notes MIDI. There is no work involved in that IMO<font size=\"2\" face=\"Verdana, Arial\">So you mean to do a mockup of a classical or soundtrack piece but track everything yourself rather than just finding a MIDI file somewhere? I thought that was what this was about anyway...

I\'d like to do a film soundtrack mockup but I don\'t have access to the scores.

[tomhartman]

Originally posted by KingIdiot:


If you mean switching to shorter bows with harder vvelocity attacks. Its totally doable, but may take multiple ports depending on how the library is set up.

No, I mean switching to shorter bows regardless of velocity.

Another huge problem I come up against when combining different bowings to produce a simple phrase is that it often sticks out like a sore thumb. The samples were played differently a different times, and one phrase that goes from legato long bows to a few short staccato notes to something else can easily end up sounding patchy as hell.

I\"MO this is especially important of profesional composers/giga users. Its part of the job IMO. The more you do it, the quicker it gets. Soon enough you get to be like Thomas J who can do amazing 4 minute mockups in 5 hours! (well maybe no one will be like him...) He proves it can be done. And he is not an anomoly.\"

No doubt.

\"The fate of the art of Virtual orchestration is that it will disappear because no one considers it an art. The consider it a burden.\"

It IS a burden, unless you enjoy programming and editing. And yes it seems different than when we put the hours in to learn guitar or piano.

\"It is of my FIRM belief that if you dont want to do the work of MIDI orchestrating and only want to compose, that you should hire LIVE MUSICIANS!! and a Recordign engineer. They need the work! and it will always sound 100 times more realistic images/icons/wink.gif <font size=\"2\" face=\"Verdana, Arial\">Yes it will.

But I believe it\'s a bit of extreme black and white to say \"You either do it the hard way or hire real guys.\" This puts no onus at all on the sample and gear developers to search for ways to break new ground in getting closer to helping us produce music without hours of research so that we can \"do a great mockup in a few hours.\" Gary\'s Maestro is a great step in that direction, as are the release trigger samples in GIGA Drums.

I don\'t consider MIDI orchestration an art at all. . It\'s a skill. The art is in the music. JMHO.

Thanks as usual for your input KING...

[KingIdiot]

Obviously its more black and white in my posts. And I agree MIDI orchestration is more of a skill than an art (jsut using the thread topic for that)

Still. I encourage developers to make things better for us all....but the thing is...there are two ways of going about this. Giving us more control...and having it \"done for us\". Its when there is a combination of both that I\'ll be most pleased.

I\'ve been working on ways to do just this with some developers. Still the raw amterial is just NOT out there.

Also firmly believe that it will always be a little bit of work to get samples to sound \"realistic\". It can be MUCH easier, and involve less tweaking, but it will IMO, still require people to switch banks to different sounds when needed. However the time of MIDI tweaking can be cut down considerably and I hope it will be done in the future.

One thing I hate tho, is that MOST of the complaining always seems to come from people who haven\'t tried much in terms of \"tweaking\" OR AT ALL!! How is it that they are the ones who judge what is too difficult to do or not?

Sometimes I wish it was more black and white. I do feel sometimes that if you aren\'t willing to go the extra miles tweaking, you shouldn\'t be asking developers to do more work for you.

[LifeForceExplorer]

[QUOTE]Originally posted by KingIdiot:

Sometimes I wish it was more black and white. I do feel sometimes that if you aren\'t willing to go the extra miles tweaking, you shouldn\'t be asking developers to do more work for you.

* * * * * *

I can certainly understand that. As a musician who is pretty lo-tech I\'ve been pushing hard to learn the tweaking - asking questions, getting a little further into it slowly.

But the other side of the coin is this (and you already know these issues well, of course, King): Giga comes with virtually NO manual! And then you pick up a basic library like Prosonus and there\'s virtually NO documentation there either! So this field is essentially for the ADVENTURESOME right now who are willing to get in there and tweak and experiment and screw up and learn a little more gradually. And thank goodness for this forum to see that one is not alone in the woods on this. But just standing on the side lines and complaining that a beautiful string performance is hard to achieve, I agree, that\'s weak. Yet this is inevitable when the manufacturers are laying out nice glossy mass-market ads in keyboard magazines that make people feel like their orchestra is just $699 away. And then it becomes $2,699 really fast and then it turns out to be darned difficult at that.

So now if I\'m going to sit back with some good whiskey and visualize a virtual string performance ten years from now here\'s what it would be: A 3D animated violinist on my flat screen monitor who I can emulate with a dummy MIDI violin controller. He emulates my every move. (Of course I can change his dress style and hair style and gender and age and religion and political views or what fun would all this be, right?) As I attack a down stroke, he attacks down and my GigaStudio Year 2010 plays the according articulation perfectly. Then as I come back up, he comes back up and Giga 2010 switches seemlessly to that articulation. We could name him Hal. It would be a sort of Midi/3DStudioMax/Giga integration which would then just need to work seamlessly with Sibelius 2010 which would have become a fully capable sequencer too! Maybe I\'ll go to work on this with my son - who is a 3D computer graphic arts college student. Now if we could just get a 3 million dollar development grant. Oh well, Yamaha\'s probably hard at work on it right now with 6 million.

[Hasen]

[quote] Originally posted by LifeForceExplorer :

So now if I\'m going to sit back with some good whiskey and visualize a virtual string performance ten years from now here\'s what it would be: A 3D animated violinist on my flat screen monitor who I can emulate with a dummy MIDI violin controller. He emulates my every move. (Of course I can change his dress style and hair style and gender and age and religion and political views or what fun would all this be, right?) As I attack a down stroke, he attacks down and my GigaStudio Year 2010 plays the according articulation perfectly. Then as I come back up, he comes back up and Giga 2010 switches seemlessly to that articulation. We could name him Hal.

A MIDI violin? We\'ve already got a MIDI wind controller, a MIDI harp, of course a MIDI piano and now you want a violin and what\'s next, a brass controller? We\'re gonna have to be somewhat talented to even write anything anymore!

We\'ll have to hire real musicians to help with our mockups... images/icons/smile.gif

Hehehehe images/icons/wink.gif

[gungnir]

THE LAST LONG MONOLOGUE (promise!)

Because this thread is going nowhere and starting to splinter into peripheral topics like they always do (my philosophical dissertations on the state of the planet are exempted because I was waiting for someone in order to go somewhere and had time on my hands which therefore makes me not guilty).

It seems that only a few forum members have wanted to take the time to express their opinions regarding the issues involved in creating a virtual orchestra, so that you can finally stop visiting forums like this one and get on to spending your precious time creating music. So, I\'ll list a few of the concepts that I was hoping would have become apparent, had some people done a Brahms mockup, anyway:

QUESTION: What must be the next step in the development of having a full orchestra at your disposal in a fast, efficient, satisfying, purely musical environment that will yield a degree of realism of 95% for most applications, including smaller scale chamber orchestra pieces?

POSSIBLE ANSWER: A full orchestra must be commissioned where each and every commonly used articulation of solo and ensembles are recorded using the same hall and recording techniques in both stereo and mono to be driven in three or four GigaStations and receive midi (or new more compressed transmission protocol) from Sibelius, the only notation program (and possibly close to being the world\'s best sequencer) with a real future.

PRACTICAL POINTS FOR DISCUSSION:

ONE SAMPLED ORCHESTRA

Are the days of independent orchestral libraries coming to an end? Would the net cost to the end user of a large, very comprehensive orchestral library cost less than having to buy all the independent libraries individually, where you then have to go through all the frustration of calibrating them individually in order to unify their programming variables. Should the money be then spent on additional GigaStations instead?

Two Immediate Problems:

1. The first big problem with this I see is that it would push out smaller independent producers, who, (beyond the extremely important issue of it representing their livelihood), have so much to contribute in regards to their unique approaches to sampling, that in turn provides important performance nuances. With one, or even two, huge orchestral libraries, everybody would end up sounding the same, the very stuff of music, expressive nuances, could become standardized.

Possible Solution: The question of standardization amongst libraries in regards to programming must be addressed. The issues such the definition of dynamic signal levels (e.g. just what level should a ppp sample be), optimal reverb environments, midi response, etc., etc.

2. The polyphony required to provide the huge number of articulations.

Possible Solution: Mono samples, so easily captured at the point of recording, must be provided as well as stereo. Stereo, which is often lost in the panning, mixdown, etc. anyway, is too expensive in polyphony cost; ram is a non-issue, it\'s polyphony that has you purchasing cpu\'s and soundcards. Stereo would still be used for smaller ensemble applications, etc.

SOFTWARE

Forget midi. In the near future, people will smile when recalling how people used to make pathetic attempts to influence real world samples with midi parameters, much in the same way they now smile at the concept of looped samples. Midi may be good at sending fundamental triggers like note on/off, (debatable also) but only the creation of a large sample articulation palette will ever solve the problem of producing realism.

The money you are now spending on gearbage, the gizmo of the month built-in-obsolescence software with racing stripes, will no longer be necessary because it is used primarily in the negative function of manipulate (distorting?) the real world samples you spend so much for in the first place. Such programs are also used to handle midi, which has no future.

SEQUENCERS

Only a the best, fastest and most professional notation program should be used to trigger your GigaStations. No outside software (unless you\'re syncing to live audio, etc) should have to be used until mixdown (debatable also). Result: No proprietary Gatesified gizmos, with more options to learn than the space shuttle . . . that is, once you purchase a manual on how to use the manual (even though in doing so gets you a free propeller hat with the company logo on it).

The notation/sequencer interface is of the utmost importance to your composition process. A musical idea, often so transient in nature, must be quickly coming out your speakers the moment the phrase is entered on the page. I\'m not talking about all that after the fact post-production software. I\'m interested in creating a pleasurable and effective working environment that will allow people to create something worth recording in the first place.

The Case For Sibelius

Sibelius is close to being the ultimate sequencer, it just doesn\'t know it. Spawned for the Acorn computer platform, Sibelius seems foreign to both typical pc and apple programs. For those of you who use it, I think you\'ll agree that you seem to sense that Sibelius is a computer program that is a bit embarrassed being a computer program; it seems to aspire to something more. It is more an extremely clever digital musical friend that seems to enjoy creating music as much as you do. The original Sibelius slogan of \"Manuscript paper that can think!\" isn\'t far off the mark. Once you\'re accustomed to this compositional atmosphere, you just don\'t want to leave it. It\'s just too elegant, too professional. I want to tap in some sixths into the violas on aged virtual parchment manuscript paper, dab in the appropriate expressive nuances like a artist uses his brush, and immediately conjure a mp tremolo articulation within GigaStudio which then comes our my speakers. You first input the notes, then go back immediately and input the passion.

To a large degree, this can be done with Sibelius right now. But the user shouldn\'t have to go through the process, and absorb the minor setbacks that result from the process. However, the process also involves the worthwhile operation of calibrating all giga instruments to react to midi messages in more or less the exact same way. So, here at least, some of the battle with quickly creating a mockup and having a virtual orchestra setup that can immediately respond to any given page of sheet music has been addressed. Many I\'m sure have done this, but it can be done for Sibelius as well as dedicated sequencers.

And what we\'re talking about now is that other major expense we all have to deal with: time.

Even though there may be some practical programming issues inherent in Sibelius that makes it difficult to perform as a dedicated sequence (I\'m speculating only, have no idea), any amount of programming effort is worth the time and trouble because nothing can ever compare to Sibelius as a working environment for professional movie scores and classical composition. Forget the others.

THE SAMPLES

I believe it can be done. The articulations, in the expressive context that you require, 90% of the time. The first thing to do at the point of recording is to have two conductors, one classical and the other involved with movie scores, and then make sure than anybody who is even vaguely connected to the rock industry is kept silent and possibly tied up at all times (other than asking where the washrooms are located, etc.). You then have the conductors pick several representative prototypes of music from the various classical/movie genres, and have midi files of these compositions ready to go.
The scheduling is done so that after one section/ensemble is recorded, the programming is done (templates ready to go) and is then immediately used to sequence the representative compositions. The conductors will then have the opportunity to see what must be done to achieve realism in a purely musical context during the period of recording, not in retrospect. They are in charge. They conduct the virtual orchestra, not the technicians. This is what the project is all about, this stage is of the utmost, pivotal importance. So, a mp to mf flute vibrato swell in the context of Faure\'s Pavanne is verified to be expressively effective then and there, a fff horn ensemble attack in the context of the score for 13th Warrior is monitored before that phase of the project is deemed completed. If recording time is broken up and scheduled over a six month season, I don\'t think this should add to the cost that much. I also believe that you don\'t need virtuoso players, instruments or orchestras to do a great job in such a project.

I, for one, am prepared to pay $6000 on a library that will work for me forever so that I can finally stop playing with gearbage and loser manuals. I know lots of others who are willing to do so as well.

THE CURRENT SITUATION

More current practical applications (not all) are using symphonic orchestral samples like glorified synth patches (over powerful percussion tracks, etc.) than for true orchestral renderings. String attacks and sustains now sound better than they used to in the secondary roles in which they are used. This is how the majority of libraries are being used, on TV, etc. Big deal. Expressive deficiencies inherent in the samples themselves are often cleverly attenuated in the mix, while the positive attributes of particular samples are magnified and used like pre-recorded phrases, thus often making the user a slave to his samples rather than accurately rendering the musical idea in your head. So the sounds are rarely being used in the context of which they were created and advertised, and paid for: orchestral samples.

I wanted to try to get some idea of the state of the art by posting some mp3\'s to this site (thanks again Toyko Joe). I couldn\'t care less whether my workup was a good one or not, it\'s not a reflection about anything to do with me, but instead of the currently available libraries used within the context of a practical application. Any sequencer can do this type of workup, if discussions cropped up about the issue of time, I would suggest that Sibelius may have played an important role in that regard. But I don\'t even know that, because, unbelievably, on a world wide forum about purchasing and using Giga sample libraries, not one user asked for the jpeg I took the trouble to create (Brahms sheet music) so that we could have the opportunity to discuss the practical issues involved in doing a few bars of friggin string music. I even hinted that if you want some blood and guts brass and percussion, we could do that. But after five days of waiting, I\'ve decided, screw it, I\'m not going to be involved with anything like this cause (and especially for the next few months), I don\'t have the time for such discussion anymore.

But you gotta ask why. Why wouldn\'t a developer use this as an opportunity to show their library\'s potential before a huge group of Giga users? Why wouldn\'t the power users with gear stacked up their studio walls just do it as a workout for the later dissection by the bottom dwelling proletariat that feed on this forum, hence summoning copious amounts of adoration and general euphoric bliss. Is there a pecking order of elitism of this forum, are there those who feel above doing a workup on what they are usually blabbing about, the expressive potential of samples string instruments? (Yea, yea, I know, I\'ve been blabbing for paragraphs now but it\'s gonna end now).

This forum (in respect to practical library development, not compositions where you want your egos stroked) appears to be all mouth, no music. It\'s not a composition forum, it\'s a sample library forum! (There should be a composition site, we need one). Everyone goes gaga over discussing the latest new library, gizmos and spending their money but don\'t seem to ever want to test them. One would think that such exercises would be a regular function of a sample user forum for gods sake. There are no real trade-secrets (as LifeForceExplorer calls them) involved with samples, unless you created them yourself (which is not the case certainly with string samples). So all you\'re doing is holding up your own progress, and everybody else\'s. Your success in the musical world should depend on your compositional skills and performing abilities, not hindering the common tools that we all use, orchestral samples.

Gungnir

[KingIdiot]

Actually there is a MIDI violin out there already.

It works much like the Roland Guitar Synth pickup, where it changes Pitch and Volume/Velocity to MIDI data for each string.

When someone makes a Bow that responds to pressure we\'ll get closer.

And finally when Vibrato Control becomes really attainable we should be good images/icons/wink.gif

[gungnir]

Because this thread is going nowhere and starting to splinter into peripheral topics like they always do <font size=\"2\" face=\"Verdana, Arial\">I posted the above without checking the latest posts, and therefore must retract my opening remark above. Will have a chance later to read latest posts - Gungnir (quilty of firing before identifying if there is an enemy).

[Hasen]

Originally posted by gungnir:


SOFTWARE
Forget midi. In the near future, people will smile when recalling how people used to make pathetic attempts to influence real world samples with midi parameters, much in the same way they now smile at the concept of looped samples. Midi may be good at sending fundamental triggers like note on/off, (debatable also) but only the creation of a large sample articulation palette will ever solve the problem of producing realism.
The money you are now spending on gearbage, the gizmo of the month built-in-obsolescence software with racing stripes, will no longer be necessary because it is used primarily in the negative function of manipulate (distorting?) the real world samples you spend so much for in the first place. Such programs are also used to handle midi, which has no future.<font size=\"2\" face=\"Verdana, Arial\">Controlling samples in realtime is always necessary. Forget your articulations, I can find faults with a single legato sample. Are you saying you would just play the sustain straight with no volume changes throughout and no change to the initial attack? Why do you think LifeForceexplore is going on about this MIDI violin for? We want MORE control over samples if anything. No matter how many articulations you record they will never always be useable as is.



The Case For Sibelius
Sibelius is close to being the ultimate sequencer, it just doesn\'t know it. Spawned for the Acorn computer platform, Sibelius seems foreign to both typical pc and apple programs. For those of you who use it, I think you\'ll agree that you seem to sense that Sibelius is a computer program that is a bit embarrassed being a computer program; it seems to aspire to something more. It is more an extremely clever digital musical friend that seems to enjoy creating music as much as you do. The original Sibelius slogan of \"Manuscript paper that can think!\" isn\'t far off the mark. Once you\'re accustomed to this compositional atmosphere, you just don\'t want to leave it. It\'s just too elegant, too professional. I want to tap in some sixths into the violas on aged virtual parchment manuscript paper, dab in the appropriate expressive nuances like a artist uses his brush, and immediately conjure a mp tremolo articulation within GigaStudio which then comes our my speakers. You first input the notes, then go back immediately and input the passion.<font size=\"2\" face=\"Verdana, Arial\">And \'input the passion\'? Just how do you suppose I do that?



I wanted to try to get some idea of the state of the art by posting some mp3\'s to this site (thanks again Toyko Joe). I couldn\'t care less whether my workup was a good one or not, it\'s not a reflection about anything to do with me, but instead of the currently available libraries used within the context of a practical application. Any sequencer can do this type of workup, if discussions cropped up about the issue of time, I would suggest that Sibelius may have played an important role in that regard. But I don\'t even know that, because, unbelievably, on a world wide forum about purchasing and using Giga sample libraries, not one user asked for the jpeg I took the trouble to create (Brahms sheet music) so that we could have the opportunity to discuss the practical issues involved in doing a few bars of friggin string music. I even hinted that if you want some blood and guts brass and percussion, we could do that. But after five days of waiting, I\'ve decided, screw it, I\'m not going to be involved with anything like this cause (and especially for the next few months), I don\'t have the time for such discussion anymore.<font size=\"2\" face=\"Verdana, Arial\">I said I was willing to do a mockup, just not a classical one. I find this thread very interesting and am willing to discuss it more. It seems to me that you don\'t want to discuss it anymore \'cos not everyone is agreeing that Sibelius it the way to go for orchestral simulation.

[Hasen]

On the subject of tweaking, I have no problem with it. I\'m not saying I\'m great at it or anything but I have no problem with actually doing it. You just fiddle about with samples and stuff until you get what you want. If you have a crazy idea then you try it out.
I can\'t understand people who say they don\'t have time for all these tweaks; tweaking SAVES you time! images/icons/smile.gif

[LifeForceExplorer]

Well, yes, of course there\'s already a Midi Violin - but not interacting with different samples in GigaStudio triggered by the actions of a 3D graphics character. Look at the video games. The kids have their character punch one way - and the opponent reacts one way. And then your opponent reacts another way when you punch differently. Your race car approahes a brick wall and does one thing, it hits a soft haystack and does another. There\'s no reason why a virtual violinist couldnt be programmed to seek different channels or do \"key switches\" and so forth for you, if he was programmed this way. The Midi violin is just the input device and the virtual violinist\'s type of movements would actually trigger the changes.

[Thomas_J]

Ok I\'ve been reading up on this topic, and I have only one thing to say:

You guys who are whining that you can\'t do realistic mockups and that it takes a 100 years to build a realistic 5 second violins passage;
THIS FIELD IS OBVIOUSLY NOT FOR YOU.

I can\'t do much but laugh at these posts. They seem poorly reflected and dare I say immature.

Best of luck in building an automated instant realistic orchestral music production tool.

Oh and while you\'re at it why don\'t you throw in a conductor robot and some A.I musicians. Oh and I bet you guys want an automated composer too.

Your perfect gigasetup would undoubtly consist of a couch, a 53\" widescreen tv and an ON/OFF button that controls whether your giga rig is \"composing\" or \"not composing\".

I have two words for you LAZY A*SHOLES.

images/icons/smile.gif (I do hope you realize that I\'m exaggerating to make a point here) images/icons/smile.gif No offense meant to anyone, but it\'s with midi orchestration as it is with any other job. Work hard and you get results.

Don\'t expect to be able to pick up a violin and deliver a flawless performance of \"Vivaldi-4 seasons\" within one hour of practice. Or 1 month. Or 1 year. Or 20 years..... And (drawing a parallel with sample libraries) don\'t expect a $1000000 violin to sound better than your $1000 violin if you don\'t know how to play it.

So for now, stop whining and decide for yourself whether you can live with the limitations ,and the work that midi orchestration involves, because guys, there will always be limitations. Learn to deal with reality, and face your own limitations as well as those of technology.

Thomas

[Hasen]

Exactly.

[LifeForceExplorer]

Hehe...well, on the whole I agree with you, Thomas_J. Since I have nothing invested in this except my own dollars by my own choice and this is not my work for income. Yet it still suprises me how little supporting documentation there is in this field - GigaStudio being the worst. As I said before, supporting your point, this field (professional or not) is an adventure that obviously just takes a lot of hard work and experimentation. If it ends up being too much of a drain on what musicality I have, and I end up becoming too geeky and pointy headed, then once again, by personal choice, it\'s going to be back to the real manuscript paper, some well sharpened pencils, and a real piano. So there\'s no whining here - the GigaPC is sort of an early version of a new millenium organ, as I see it, and it apparently has a ways to go yet. That\'s all. So its natural to be speculating about how to make it better.

Good music to everyone.

[dwdonehoo]

It is no surprise Thomas got it right.
You work within the limitations as technology advances.

And:
\"Sibelius is close to being the ultimate sequencer.\" Pbbbt. Sure. That was a joke, right? Good one. The pen and paper approach is fast fading away, and Sibelius is just a bridge to the new technology. Almost every working pro composer is using midi and whatever controller that is most useful to them (sometimes several kinds) and tweaking the raw material into something creative using tools they are most comfortable with. I think that the Sibelius approach is becoming the backward way of doing things: I think the trend is to perform music with controllers and using tools (like even Sibelius) to tweak the composition and performance. But even so, there is no one true way of doing things.

[tomhartman]

Originally posted by LifeForceExplorer:
Hehe...well, on the whole I agree with you, Thomas_J. Since I have nothing invested in this except my own dollars by my own choice and this is not my work for income. Yet it still suprises me how little supporting documentation there is in this field - GigaStudio being the worst. As I said before, supporting your point, this field (professional or not) is an adventure that obviously just takes a lot of hard work and experimentation. If it ends up being too much of a drain on what musicality I have, and I end up becoming too geeky and pointy headed, then once again, by personal choice, it\'s going to be back to the real manuscript paper, some well sharpened pencils, and a real piano. So there\'s no whining here - the GigaPC is sort of an early version of a new millenium organ, as I see it, and it apparently has a ways to go yet. That\'s all. So its natural to be speculating about how to make it better.

Good music to everyone.<font size=\"2\" face=\"Verdana, Arial\">Agreed. However, saying that anyone who complains about how clumsy it currently is to put things together should get out of the business is a bit extreme, and sounds a bit defensive to these ears.

If everyone were satisfied to work within limitations, we\'d all still be flying on planes like the Wright brothers.

[KingIdiot]

I dont think anyone is satisfied with working within limits of techonology...

But there is a very distinct difference between actual limits of technology, and the limit of amount work one wants to do images/icons/smile.gif

Another thing is that last limit depends on the person.

Funny thing is, I think people are satisfied in working in that limit. They wont work more than they want to!

[tomhartman]

Originally posted by Thomas_J:
Ok I\'ve been reading up on this topic, and I have only one thing to say:

You guys who are whining that you can\'t do realistic mockups and that it takes a 100 years to build a realistic 5 second violins passage;
THIS FIELD IS OBVIOUSLY NOT FOR YOU.

I can\'t do much but laugh at these posts. They seem poorly reflected and dare I say immature.

Best of luck in building an automated instant realistic orchestral music production tool.

Oh and while you\'re at it why don\'t you throw in a conductor robot and some A.I musicians. Oh and I bet you guys want an automated composer too.

Your perfect gigasetup would undoubtly consist of a couch, a 53\" widescreen tv and an ON/OFF button that controls whether your giga rig is \"composing\" or \"not composing\".

I have two words for you LAZY A*SHOLES.



Thomas<font size=\"2\" face=\"Verdana, Arial\">Well, Thomas, if your posts were as classy as your music you would have something going there. But I resent this kind of reply, which is condescending and unproductive. You may pass it off as charmingly irreverant....I find it arrogant and insulting.

No one asked for an automated conductor or any of the other nonsense you cited. The current state of the art can be improved. That\'s all the point was. It does take an inordinate amount of time for me, and others I know, to mock up something realistic using the current tools. I do realize that some of this is my \"fault,\" if you\'d like to phrase it that way. I only got Gigastudio some months back, and have had Gary\'s strings for about six months. Long enough to make some inroads, not long enough to be an expert. So some of my frustrations can obviously be overcome with some more study. Obviously, if you are doing this for a living, and are constantly facing deadlines, the time available for such research, experimentation, and study, is directly related to how much you working, and I have been fortunate enough to be working a lot.

We cannot equate doing a MIDI mockup with learning the violin, because that would imply that we would all have had to have had Giga and Gary\'s strings since age 6 to get anywhere. A bit of a stretch, but necessary to imagine if you are serious about comparing the two.

Anyway, I enjoy your work very much. But this is not supposed to be a insult fest, and there\'s no need to assume that one who has issues with the current state of the art is any less talented, dedicated, or experienced than yourself.

[gungnir]

Yea, I know I promised no more long posts, but we got to end or start the sequencer aspect of this thing and half of this post are regarding comments from other users and the second half is trying to wrap this aspect up, from my perspective anyway. Also, as thread starter, I enjoy certain privileges that others don\'t. images/icons/wink.gif


But even with talent and experience, it takes WAY too long to do something really impressive....at least, it takes ME way too long. <font size=\"2\" face=\"Verdana, Arial\">All of this, I believe, in the context that you are discussing it, can be solved if you take the time to calibrate all the Giga libraries to your sequencer, in my case Sibelius. All the issues you are bringing up were directly involved in the Brahms mockup which nobody responded to.


Yep. Well said, gungnir. (Didn\'t we chat about all this once at noteheads on their mailing list? I think we were both evaluating \"Igor\" at about the same time.) <font size=\"2\" face=\"Verdana, Arial\">No, sorry, LifeForceExplorer, I\'m not familiar with noteheads but I\'m hearing interesting things about the Igor program


I can actually see the day on the horizon when I\'ll be devoting 70 or maybe even 80 percent of my attention to composing again! <font size=\"2\" face=\"Verdana, Arial\">I believe, with some limitations, you can easily do that right now. But I was trying to test a particular approach with the Brahms, to see what demands people want in terms of expression, midi control, time it takes, working environment, etc. and on a world wide web site about sample use I don\'t get one response! Unreal!


I don\'t think that having a sample...or patch..or whatever you want to term it....that automatically shifts to say a
short bow from a long bow if you hit the note quickly is a lot to ask, and even that would be a huge time saver <font size=\"2\" face=\"Verdana, Arial\">This would surely deteriorate the full expressive potential of the libraries you are using. But you can easily apply the exact articulations and dynamics required to fully enrich the phrase by direct application of the appropriate midi parameters right there on the score so that you write notes first, expression next. Done. (that is, within the capabilities of your libraries). But like King observes, you have to do some work to set up your studio properly, but it\'s the type of thing that should be done anyway


Sending mp3s for helpful suggestions is a great idea. I for one would love to do a short 10 second passage or so, and get some feedback on what I should have done, etc, and I\'m all for it if you guys are.....think it would help all of us. How do we proceed? <font size=\"2\" face=\"Verdana, Arial\">I clearly indicated I would send midi files, Sibelius files, mp3\'s and sheet music. I don\'t know what more I could have done.


The fate of the art of Virtual orchestration is that it will disappear because no one considers it an art. The consider it a burden. <font size=\"2\" face=\"Verdana, Arial\">Exactly! The concept has been around for years but is stalled when it should be surging ahead. But it can be done right now, with the tools we have.


I\'d like to do a film soundtrack mockup but I don\'t have access to the scores. <font size=\"2\" face=\"Verdana, Arial\">John Williams\' scores are available at: http://www.alexuniv.com/scoring.html (\"http://www.alexuniv.com/scoring.html\")


Gary\'s Maestro is a great step in that direction <font size=\"2\" face=\"Verdana, Arial\">Actually that is the brainchild of Jeff Hurchalla, whose Maestro Tool contribution is nothing short of brilliant in its practical contribution to the world of orchestral emulations. It is so important in fact, that I personally would pay hundreds of dollars to it alone. It is an absolutely essential component in virtual orchestrations.


So for now, stop whining and decide for yourself whether you can live with the limitations ,and the work that midi orchestration involves, because guys, there will always be limitations. Learn to deal with reality, and face your own
limitations as well as those of technology. <font size=\"2\" face=\"Verdana, Arial\">This is true as long as they are real limitations and not just perceived ones. I don\'t think technology has any limitations (unfortunately), and its our ability to approach things differently that removes limitations and hence shapes our reality. Most of the comments on this thread are perceived limitations, not actual ones.


If everyone were satisfied to work within limitations, we\'d all still be flying on planes like the Wright brothers<font size=\"2\" face=\"Verdana, Arial\">Right on.

CONCLUSION:

I\'m not trying to lead anybody on here. No bs. I sequence mainly for composition and so my friends and I can play along with some orchestral pieces. I have limited experience with other sequencers and setups because I almost broke my mouse hitting the yes prompt so hard on the \"Do you really want to remove this program?\" of most of them. The thought of working outside of Sibelius was unthinkable. So, like some others, I messed around with things and ended up, for the type of stuff I do, sequencing by direct input (mainly) large orchestral works, not needing other programs to any great extent at all (need the editor and wave editing programs). The amount of precise fine-tuning in some areas of midi manipulation may be less refined, but this, in my opinion, is compensated by extreme advantages in other areas.

But, because I rarely use other programs, I\'m at a massive disadvantage in regards to knowing the big selling points of other sophisticated studio setups. So that I would not end up making any false statements, or mislead anyone in any way, I thought a string mockup involving many of the practical issues related to virtual orchestration, would give a clear idea of how people are currently doing things on a step by step level. Then, without jerking around with people\'s time, or expectations, or degree of professionalism, I could assess if the Sibelius-Giga approach had in fact enough practical advantages to outweigh some of the disadvantages, and be therefore worth serious further investigation.

There are many different levels of proficiency on this forum. For example, I personally find the GigaStudio Editor extremely easy to use, I\'m not bragging, anyone can do it if they can find the time, but things like this, like King and Thomas J point out, are necessary if you are serious about setting up things right. But once done, things begin to get real easy fast. The various points that I am making above are all legitimate, within the context that users are presenting them on this particular thread. I may be overlooking some critical issue, but I have been sequencing sections of major orchestral works for some time now, and so far haven\'t run into any particular problems. But, by and large, so far, based on the information I have, many of you are just not approaching things correctly. Again, I may be wrong, but even if it is proven that I have overlooked something, I am quite convinced that this approach, or something quite similar, is the way to go in the future. What I mean here is staying in your composition environment to edit midi, etc.

I am willing to share some (not all) insights concerning this interface with others who wish to establish a dialog on how they are currently sequencing large works. But this will only be truly productive if several people do a workup of at least some of the Brahms and tell me exactly how they did it. I\'m not trying to do a power thing here where you are being jerked around by some wingnut. I just want to know how everybody\'s doing it and how long it takes. I couldn\'t care less how they sound, only the exact procedure used. Or describe the methods for your sequencer in general, step by step, keeping your \"trade secrets\" to yourself. I\'m obviously concerned with not being aware of some element that hasn\'t been important enough in my setup to yet come to my attention and I don\'t want to make any serious blunders. I\'m simply attempting to see why others are not approaching things this way, so that it can be determined if the approach should be more fully explored (possibly with Sibelius themselves). They may say that developing this method can\'t be done. I don\'t know. Maybe someone else has tried it. But it sure works very well for me.

So I\'m up front. That\'s why I started this thread. So, how can you tell whether such an interface is worth some of your time? I will now give this disclaimer that you can use to decide if you want to do a mockup and send it to me, or post it on the forum, or whatever you want to do: The Brahms mockup, that you can hear very clearly on the mp3, was done in about four hours (I could do it faster now, maybe two and a half) in Sibelius Version 1 software, driving GigaStudio, using GigaStudio reverb and recorded to the harddrive on one of the cpu\'s using my wave editor to capture the wave file with no other software involved. Use those facts to base your decision on. Make sure you aren\'t using the sounds as criteria, they have little to do with what we\'re talking about. What you hear, the sequencing aspect, is what you get. (I could send one more mp3 to Tokyo Joe providing the inner parts and the elegant phrasing nuances). But what you should also be hearing is the term about four hours (once you are used to working with it maybe faster), in the world\'s best notation program environment (my opinion), never having to leave it and enjoying it because of its instant monitoring of phrasing articulations.

(It may take several hours for me to respond to posts as I have now run out of free time)

Gungnir

[Chris Beck]

I want some of what Gungnir\'s been smoking.

[composer22]

I\'m sorry, Gungnir. I got distracted with this 4 Movement Symphony for 2 orchestra\'s I\'ve been writing since you started posting on this subject. Almost done. Could you repeat that? images/icons/grin.gif images/icons/tongue.gif

[KingIdiot]

Hasen

Its not pointless. I dont mean send me your own music. It can be another piece of music, but not some lame GM ... just the notes MIDI. There is no work involved in that IMO

Tom, the idea of a giga patch that switches to a short bow when you play quick is impossible on its own. With a maestro tools like app, maybe, but you have to decide what is \"Quick\"

Possible making a keyswitched instrument with long and short bows is possible, but you start to eat up dimensions for some more \"useful\" things like layering.

If you mean switching to shorter bows with harder vvelocity attacks. Its totally doable, but may take multiple ports depending on how the library is set up. IMO its a WAY WAY WAY WAY WAY WAY simple edit, that in fact can already be done with GOS and port layering a short bow and the SUS EXP. It can be enhanced to add velocity controlled attack envelopes on the short bow.

This is something that takes LITERALLY less than a minute to make the edit (a little more to actually save the file because of how GSEdit works on BIG Gig Instruments)

And now another point. I personally wouldn\'t work with this type of patch set up all the time. I find it a pain on my Roland Orch Board that has that type of set up. Developers dont know what ALL people are going to be writing and wanting. Which is why, people need to learn to make the simple edits on their own.

IMO this is especially important of profesional composers/giga users. Its part of the job IMO. The more you do it, the quicker it gets. Soon enough you get to be like Thomas J who can do amazing 4 minute mockups in 5 hours! (well maybe no one will be like him...) He proves it can be done. And he is not an anomoly.

The fate of the art of Virtual orchestration is that it will disappear because no one considers it an art. The consider it a burden.

It is of my FIRM belief that if you dont want to do the work of MIDI orchestrating and only want to compose, that you should hire LIVE MUSICIANS!! and a Recordign engineer. They need the work! and it will always sound 100 times more realistic images/icons/wink.gif

[Thomas_J]

Tom, I\'m sorry if I offended you (I tend to offend people no matter what I write if I write it with a sarcastic and a bit over the top \"tone\") - Let me just say that I don\'t mean to insult anyone, and I don\'t think I was being arrogant at all, I was just talking about the facts of midi orchestration, as I see them. I\'m struggling every day with these limitations, but yet I feel soooo FAR from mastering what I\'m doing. I was simply making a (very exaggerated, to make a point) statement to people who expect stuff out of the box, coz anyone who have been in this game a while knows that it\'s simply not the case.

Some tools have made it easier (like the Maestro Tools and keyswitching etc) and more efficient to use giga samples, but it\'s still a lot of work to make it sound real, and I don\'t think you can take any shortcuts here.

This is just how I feel. Of course I have nothing against people who do this on a hobby basis and want more for their money than they are getting, or whatever the case would be.

Just don\'t forget that orchestral music is perhaps the most complex form of music out there!
It requires a lot of knowledge just to write for a real orchestra, and it requires a lot more to write for a midi orchestra, if it\'s going to sound as good. I think many people on this forum have good grips on this, but they are the ones who love to work with it and tweak to get it right.

Personally I\'m not really that concerned with REALISM. At least not down to the point where I\'d spent 1 hour on a 5 second violins passage. I use this technology to learn more about orchestration, and help realize my musical ideas.

So Tom, just accept my post as a \"humorous\" attempt to enlighten those who are new to this and think it should be easier than it is!

I don\'t want to pick on anyone, I\'m just sharing my experiences, like everybody else.

Sorry again if I offended you. I really didn\'t mean to, but seriously though: Sometimes people need to wake up, realize and accept the hard truth. And someone has to be the big bad wolf and tell them. I\'m that wolf in that last post. I didn\'t want to diminish their hopes, but maybe it is inevitable? (That violin comment was more a humorous remark than a hard fact of life) images/icons/smile.gif

One fact of life is for sure though. Things aren\'t getting any WORSE in this field. Everything is improving, all the time, but the same fact lives on:

Be prepared to WORK for results.

Thomas

[Hasen]

Originally posted by gungnir:


John Williams\' scores are available at: http://www.alexuniv.com/scoring.html (\"http://www.alexuniv.com/scoring.html\")
<font size=\"2\" face=\"Verdana, Arial\">I know but with all the money I\'m spending on samples I can\'t afford to buy them right now. I plan to - just not yet.

[csduke]

Originally posted by gungnir:
Forget midi. In the near future, people will smile when recalling how people used to make pathetic attempts to influence real world samples with midi parameters, much in the same way they now smile at the concept of looped samples. Midi may be good at sending fundamental triggers like note on/off, (debatable also) but only the creation of a large sample articulation palette will ever solve the problem of producing realism.
<font size=\"2\" face=\"Verdana, Arial\">I am wondering what your vision is for a new computer protocol that will replace MIDI? Is that part of what you are proposing? What is absent in MIDI that does not allow you to \"influence real world parameters?\" \"Inputting passion\" be it changing a velocity value to 42 or placing an mp on your Sibelius score both result in a MIDI message. The mp saves a little time and looks more traditional but trades off the subtlety of control (less passion?). You don\'t get something for nothing - usually.

Inevitably, you must deal with the details of the protocol you are using to drive your samples (sample player) if you want subtle musical control. This is, of course, because musical notation is very inexact. That, of course, is one reason why different musician interpret music so very differently - because the notation is not precise. MIDI/sequencers realizes the choices that the musician makes while playing. You have to \"program\" those choices.

Inevitably, you must deal with that \"large sample articulation palette\" by not only triggering it but controlling the other musical attributes associated with it - via MIDI or some other computer protocol.

The two aspects of MIDI or its replacement (downward compatible) that will probably change in the future are a faster bit rate and the specific details of its control words e.g. a larger range for velocity. But, I don\'t see how MIDI get replace with anything that is not similar to MIDI.

I am misunderstanding your comments?

[tomhartman]

Originally posted by Thomas_J:
I was just talking about the facts of midi orchestration, as I see them. I\'m struggling every day with these limitations, but yet I feel soooo FAR from mastering what I\'m doing. I was simply making a (very exaggerated, to make a point) statement to people who expect stuff out of the box, coz anyone who have been in this game a while knows that it\'s simply not the case.\"

Thomas<font size=\"2\" face=\"Verdana, Arial\">Thanks Thomas. I get exasperated at times with all of this and I was probably being oversensitive. I certainly agree with you about how much work it takes to do all of this. I started with MIDI on a Mac Plus in the 80s and have been using multiple MIDI tracks and audio tracks to create convincing passages for many years now, and now with advent of wonderful sets like Gary\'s, it becomes even more of a challenge.

Funny, but after posting last night, I took about a half an hour or so to make up a new Performance in Giga with Gary\'s strings that will speed things up for me. If I could discipline myself to spend even a half an hour a day trying out different things my gripes about it taking so long would probably diminish somewhat. So I realize some of this is on me.

I don\'t really mind digging in for the long haul on a difficult orchestral piece, trying to get it right with MIDI. But as you know, when a client calls and says,
\"We have a 8 minute video. We\'d like it to sound like a cross between The Matrix and John Williams, and we need it in 10 days, can you do it?\" the prime enemy becomes time. On a job like that, you already know you are going to be screwing around for hours finding oddball synth sounds and great drum sounds, so when you get to the strings and brass you just don\'t have hours to spend working on a 20 second sequence. I usually get by OK, but most of you fellow pros would hear a million things \"wrong\" with the brass and string parts, simply because there wasn\'t enough time. As I said to KING, templates and the like can help this out, but since every piece is different, even having premade set ups only gets you so far. And of course, our imaginary client doesn\'t have 15,000 for the project, he has 5000, so its MIDI or nothing. The good news is, we keep the 5000 and it\'s pure profit if we pull it off;)

I too believe it is an excitng time for all of us. Giga opened up the doors, and Gary and the other fine developers up here ARE listening. Learning right along with us. So my hopes are high. Yes, it would be great to be able to play some things right out of the box. And actually, thanks to Maestrotools, you can play a legato string line right out of the box!

Sorry if I seemed oversensitive. I do admire your work, and we are all in this together, hopefully pushing for the same end result...realism with less programming and more time to compose, record and mix.

Thanks again Thomas.

[Plundrik]

...rushing in.. post nr 100! images/icons/cool.gif

Thomas, by the pieces of music I\'ve heard from you, it seems that you have mastered midi orchestrations almost perfectly, the music is good enough to score any movie in my opinion.. images/icons/smile.gif

[Thomas_J]

Hehe, Tom! I\'m sure it\'s frustrating when clients approach you like that. Obviously you\'d just have to do your best and hope that they\'re satisfied enough with your job to choose you in a future project. It seems we agree with each other in the end images/icons/smile.gif So there\'s not much left to say, except good luck on your \"Matrix\" piece (if that was an actual job you had) images/icons/smile.gif And thanks for your comments on my music!

Thanks to you too Plundrik images/icons/smile.gif (Where\'s that nick from anyway? images/icons/smile.gif )

Thomas

[KingIdiot]

Thomas,

You have no idea how many times someone will ask for \"the matrix\"

Its quite frustrating. Everyone wants that Spybreak sound.

[gungnir]

I think there is little possibility that a practical, hands-on dialog about what\'s involved in the optimal setup for virtual orchestration, using the tools currently available, will happen on this user forum.

So, from my perspective, I have concluded what I suspected when initiating the topic on Virtual Orchestration, the fate of art. The fate of the art will never be determined by a user forum as inactive and fragmented as this one.

Outta here,

Happy arpeggios, Gungnir

Endicott, Endicott, you still out there in cyberspace, man? You were right all along? Endicott, f o r g i v e m e, Endicott, I doubted you, f o r g i v e m e . . . . . . . . . .

[Plundrik]

Originally posted by Thomas_J:
Thanks to you too Plundrik images/icons/smile.gif (Where\'s that nick from anyway? images/icons/smile.gif )

Thomas<font size=\"2\" face=\"Verdana, Arial\">Hi, Thomas. Plundrik is what I was called as a child when I did something stupid (name is Henrik). I\'m from Molde by the way, so the nick is sort of from here. hehe

[peter269]

The late Herb Spencer, John Williams\' orchestrator for many years, once told me that if you want to be successful in Hollywood, you have to know about 1000 orchestral devices (instrumental combinations).

Clark Spangler, one of the first programmers of the DX7, once told me that to be a successful programmer, you have to know both what a sound sounds like, AND what it doesn\'t sound like.

Herein are the core secrets of effective MIDI \"orchestration\". You have to know your orchestral devices, what they sound like, and what they don\'t sound like to make judicious musical choices.

This means that annually you have to build several hundred hours of listening time to learn these devices along with compositional techniques. You can \"speed up\" the process by focusing on selected pieces and knowing them inside out. Ravel\'s Mother Boose Suite is a treasure trove of devices as is Pictures at an Exhibition. For Vaughn Williams, consider his London Symphony Second Movement, Variations on a Theme by Thomas Tallis, Greensleeves and Symphony Antartica. For Debussy, Prelude to the Afternoon of a Faun, La Mer; Georges Bizet, Carmen. Borodin, On the Steppes of Central Asia; Stravinsky, Rite of Spring and the Firebird.

Another \"secret\" is to practice your keyboard performance skills with both technical exercises and playing pieces so that your musical expression is up to par.

Then, you have to learn to master your sequencing program, whatever you choose. This is a four-fold consideration since it requires knowledge of the devices you want to test record, keyboard skills, mastery of the sequencing program, and samples that will do the job.

Next comes recording and listening to how you mixed and blended the orchestration. Was it authentic? Did the reverb you selected put the score into the best soundng room that gave your digital samples more life?

Thus, the final element, the \"trade secrets\" part, are simply the answers you found based on what samples you have in your library today and the resultant devices you discovered they\'ll sucessfully produce after you recorded and mixed them.

Cordially,

Peter Alexander

author, The Revised Rimsky-Korsakov\'s Principles of Orchestration
http://www.alexuniv.com/orch_princ.html (\"http://www.alexuniv.com/orch_princ.html\")

Writing For Strings & Logic Made Easy
http://alexuniv.com/AU/classlist.html (\"http://alexuniv.com/AU/classlist.html\")

[peter269]

[dash]

Read the answers here and iy seems its a great idea to have a central source for audio and midi files with a write up of each to show what has been done and how for each arrangement and what library was used and why.
I\'ve got an upload section on my site at the mo. but it is only for midi. You can zip up files and upload them aswell, so this might be useful to ya all!! (please don\'t upload silly sized files or my server will go bang and it will be a loss for everyone.
Have a look at the section (file uploads) and see if its of any use or how it can be improved.
IF we all can ultimately make realistic orchestral works from samples then more power to our elbows!

[dash]

My site, www.studio-33.com, (\"http://www.studio-33.com,\") web page (\"http://www.studio-33.com\") has an upload centre which is only for midi files, but you can upload audio, midi and a decription of how, what and why you put your arrangements together if you zip them up. Please don\'t upload silly sized files or my server will go bang!!
Please contact me if you have any suggestions on improvement to this facility.
If the section is used regulary for peeps to exchange ideas and tips on this area, then i\'ll expand the facility and webspace.
Please email me with any suggestions at webmaster@studio-33.com

[dash]

oops, sorry, posted twice by accident, thought the first post didn\'t work!! <<<<< muppet!!

[dash]

oops, sorry, posted twice by accident, thought the first post didn\'t work!! <<<<< muppet!!

[dwdonehoo]

Kinda late on this but I have been thinking about this a lot for the past week or so:

\"The late Herb Spencer, John Williams\' orchestrator for many years, once told me that if you want to be successful in Hollywood, you have to know about 1000 orchestral devices (instrumental combinations).\"

Boy, I really think this is the truth. I keep a notebook...