Chris Hein Horns (vs Garritan JABB) - Fully polyphonic?

[Ouch that hurts]

I've just discovered the limitation of Garritan's Jazz and Big Band collection, that I bought some time ago but hadn't got around to using - that the horns are all monophonic. Apparently there will be an update allowing them to be used polyphonically, but it won't allow for all the MIDI control of sound quality subtleties (growls, air flow etc)

The significance of this for me is that I was hoping to use them in my live rig as well as in the studio. To do this, I need to be able to play chords. I currently use QLB in Giga, but was hoping to get better brass sounds and more expressive range. But I need to be able to play four or five note passages and have them realized in real time.

Question: Does Chris Hein Horns have the same monophonic limitation? And if you CAN use it polyphonically, can you do so with full access to all the MIDI control of breath, growls etc? And if you CAN do this, how did Chris manage to pull it off while Gary says it's impossible while maintaining the advanced sound control?

I don't mind stumping up the money for Chris's set if it will do the job as I want.

Thanks.

[Tom Hopkins]

I've just discovered the limitation of Garritan's Jazz and Big Band collection, that I bought some time ago but hadn't got around to using - that the horns are all monophonic. Apparently there will be an update allowing them to be used polyphonically, but it won't allow for all the MIDI control of sound quality subtleties (growls, air flow etc)Actually, the update allowing polyphonic use has been out since January but you are correct that this is limited to the "lite" instruments with reduced features. If you haven't updated, please do so because the important changes are not limited to the addition of the "lite" instruments. It can be found here: http://www.northernsounds.com/forum/showthread.php?t=41107


Question: Does Chris Hein Horns have the same monophonic limitation? And if you CAN use it polyphonically, can you do so with full access to all the MIDI control of breath, growls etc? And if you CAN do this, how did Chris manage to pull it off while Gary says it's impossible while maintaining the advanced sound control?I'm sure Chris can chime in with an answer about the features of his fine instruments but, in our case, it was not a matter of "impossible," it was a matter of a critical feature (tongue/slur capability as designed in our programming) that would need to be eliminated to accomplish the polyphonic behavior - and that was unacceptable, in my opinion. So, polyphonic behavior was supplied in the form of 'lite" instruments by programming in the same way as GPO so that the tongue/slur feature was not sacrificed (but some of the fancy features like air flow noise and key clicks were necessarily removed - since the GPO approach to programming won't support that.) Every programmer weighs the choices afforded by the platform and makes choices based on what he/she/they consider to be most important. As a consumer you must examine the choices and decide if they fit your needs. I've stated before, on other threads, that I don't consider live performance to be a particularly practical use of JABB, in its present form, because of the multitude of controllers for sculpting various aspects of the sound - something not easily done in real-time. Future scripting developments may make both choice of polyphony and simplification of complex controller use possible for the fully-featured instruments in live performance, but that's a ways off.

Tom

[Alan Russell]

Hi,

I don't know the answer to the Chris Hein inquiry but here's my take for other content within your post.

The fact that JABB is Monophonic is a benefit and that it really forces you in to writing ensemble parts the way they are as a Pro-Arranger. Each part can articulate a tad different from the other sectional part. This gives the section a unique sound and not a repetetive block sound in velocity and articulation. Certainly it is more work for the midi arranger but in my view it delivers a more realistic outcome.


Alan Russell

[Hans Scheffler]

i wish there was a script for kontakt that would automatically split the notes in a chord and send each voice to a different instrument for realtime polycontrol of a hornsection.
Would this be possible?

Hans

[kotori]

i wish there was a script for kontakt that would automatically split the notes in a chord and send each voice to a different instrument for realtime polycontrol of a hornsection.
Would this be possible?

Hans
Hi Hans,
unfortunately scripts in K2 are always tied to a specific instrument so it's not possible to have a script receive notes and send them to different instruments. There's a possibility this will change with Kontakt 2.1.

Nils

[Chris Hein]

Question: Does Chris Hein Horns have the same monophonic limitation?
Chris Hein-Horns is completly polyphonic.

I bought JABB recently and like it very much together with my Horns.
When opening JABB instruments in kontakt, you can make them polyphonic
by setting the voices-groups to off in the source page.


i wish there was a script for kontakt that would automatically split the notes in a chord and send each voice to a different instrument for realtime polycontrol of a hornsection.
Would this be possible?
Thats possible with kontakt scripting.
Not with different instruments, but with different keygroups.
So, it needs special designed instruments for that function.

Chris Hein

[Ouch that hurts]

Chris Hein-Horns is completly polyphonic.

I bought JABB recently and like it very much together with my Horns.
When opening JABB instruments in kontakt, you can make them polyphonic
by setting the voices-groups to off in the source page.

Thanks Chris. Just to clarify, when you do this do you still sacrifice some of the advanced programming as Tom said?

[Ouch that hurts]

Hi,

I don't know the answer to the Chris Hein inquiry but here's my take for other content within your post.

The fact that JABB is Monophonic is a benefit and that it really forces you in to writing ensemble parts the way they are as a Pro-Arranger. Each part can articulate a tad different from the other sectional part. This gives the section a unique sound and not a repetetive block sound in velocity and articulation. Certainly it is more work for the midi arranger but in my view it delivers a more realistic outcome.

I'm not denying any of that from an arranging point of view, but maybe you missed the point of my post - that I need something for live use.

[thesoundsmith]

Ouch, that was my primary disappointment with JABB as well, i was hoping they would work live, but even playing poly witht he lite horns - they sound best in a fully controlled environment, any time you go polyphonic you lose the individual control, there is no way around it (except I suppose it would be possible to script something that would allow only the top note of a chord to respond to the controller data, but if you look at a midi file of live perfornace, I'll bet your accuracy in hitting a chord is not going to be clean enough to let the script perform its task once per chord - any slop at all, even a small amount, and you'll be getting some very un-authentic sounding results (multiple attacks, sloppy slurs with odd enveloping, etc. as the script tries to guess your intentions - or is delyed enough for the fingers to catch up, but then the attack/slur is late as well - you can't beat Mother Nature.)

[Chris Hein]

Thanks Chris. Just to clarify, when you do this do you still sacrifice some of the advanced programming as Tom said?
The complete programming is based on multi-samples.
Air Flow, Vibrato, Growl, Sax-Keys, Spit, Top-Kicks, Shakes, etc.
everything is just done by blending samples.
Even these controller-samples have up to 10 velocities.
Just the growl-effect on the trumpet consists of 102 different samples.
There are no synthetic controllers controllers in Chris Hein-Horns.

I found no reason to do anything monophonic.
You know what you play, so you can avoid overlapping notes if that is the reason.

Chris Hein

[Ouch that hurts]

The complete programming is based on multi-samples.
Air Flow, Vibrato, Growl, Sax-Keys, Spit, Top-Kicks, Shakes, etc.
everything is just done by blending samples.
Even these controller-samples have up to 10 velocities.
Just the growl-effect on the trumpet consists of 102 different samples.
There are no synthetic controllers controllers in Chris Hein-Horns.

Thanks again Chris. I think this may well be my answer. I really wish you'd release it for Gigastudio though.

Just out of interest, how do you handle things like blending non-vibrato samples into vibrato? Isn't there a problem with the effect sounding like two instruments, as you get the two different sounds overlapping, each at different points in their legato envelope?


I found no reason to do anything monophonic.
You know what you play, so you can avoid overlapping notes if that is the reason.

Yes, I agree with that. I've never found playing monophonic instrument parts with a polyphonic sound a problem from that point of view.

I've also now had a bit of a play with JABB, and on first hearing I really don't like the tongue/slur programming. The removal of the attacks to make the slurred notes sounds really fake to me. In fact there's some wierd thing that happens at low velocities even without the pedal down that sounds fake like this, like the way they make quasi-legato on cheap synth keyboards.

I also find some of the growls and other things very thin, sterile and fake-sounding. Currently I use QLB for my live brass. It's what, six or seven years old now? And while it has massive flaws and limited flexibility, the sound quality is definitely better than JABB.

In theory I'm all for the design concept of JABB, massively reducing the sample content and size, and using programming instead for the flexibility. But you can only sacrifice basic sound quality so far - I don't see sample libraries having anything to worry about any time soon.

No disrespect to Gary, Tom or anyone else - these are just my reactions.

[Journeyman]

But one thing comes to mind: You might be making the mistake in thinking that JABB's purpose is to fool the listener into thinking that what they're hearing is the real thing. Granted, that's the purpose of most sample libraries out there, but I don't think that's really the case with JABB. JABB is more for those that want to make a midi mockup of their arrangements, but not really trying to replace a real big band. There's just no way that JABB comes that close; nor was it expected to. So holding it to the high expectations that you hold it to is unrealistic. It's just not meant to compete on that level. Gary has said as much.

[Tom Hopkins]

First of all, journeyman is correct. JABB is not designed for the kinds of things you are apparently trying to do. You are criticizing the apple for not being an orange. Both Gary and I have repeatedly stated the aims of the library as a composer's tool which has very little to do with live performance. To reiterate, here are the goals of the library (and a similar statement is also included with the product when offered for magazine reviews):

“1. JABB is designed to give composers, educators, and students useful information during the writing process so that they can better judge the success of a chart before placing it in front of real musicians. As a writer's and educational tool, it is capable of supplying much better feedback on timbral combinations, section interplay, and articulation choices than any previous sound source for this kind of material.

2. The library can be used (with a commensurate expenditure of time and energy during the editing process) to create reasonably convincing mockups which can be used by a composer for promotional purposes.

Unlike the goals of most sample libraries, JABB makes no attempt to replace real instruments, especially in a solo context. We consider this to be a futile goal in a genre which contains such a wealth of subtle detail in playing technique, inflection, and unique stylistic characteristics.”

The success of the second point above I believe is amply illustrated by the existing big band demos. These are more than sufficiently convincing to effectively communicate the intentions of the composer and to help a prospective purchaser of a chart get a good idea of what it would sound like when given to real players. That accomplishes our initial goals for the library.

I've never found playing monophonic instrument parts with a polyphonic sound a problem from that point of view.No doubt, in live playing - but that's not one of the purposes of the JABB library. In mockup production I find it extremely valuable to use monophonic behavior to simulate instruments that are intrinsically monophonic in reality and, as I've said, the particular programming in JABB depends on this behavior to function precisely as intended.


I've also now had a bit of a play with JABB, and on first hearing I really don't like the tongue/slur programming. The removal of the attacks to make the slurred notes sounds really fake to me. In fact there's some wierd thing that happens at low velocities even without the pedal down that sounds fake like this, like the way they make quasi-legato on cheap synth keyboards.Now, here we have a significant disagreement concerning this issue. In creating the big band demos (including the recently posted scream trumpet demo) I spent a great deal of time with this programming and the tongue/slur feature played a major role in the success of the demos. One of the essential aspects of horn playing, and one usually missing from simulation attempts, is the use of the tongue. I found that applying intricate tonguing and slurring articulations using cc64 in JABB often made the difference between an obviously synthetic rendition and one that started to sound convincing. It was emphatically important. It must be used properly though to get good results. You must know exactly where (and how) the tongue get used in a passage and correctly place the cc64 data to make it work convincingly.

As to the “weird” low velocity behavior of the tongue layer: In JABB, velocity controls attack strength. Low velocities give gentle attacks that have a number of purposes. Horn players can create legato lines in two ways: By slurring, where the tongue does not interrupt the air flow; and by very gently tonguing, where the tongue lightly “feathers” the air stream. Both can be simulated in JABB. Furthermore, the lowest velocities can be used in special situations. An example is the case of slurred octave skips that I encountered during the redo of the Paul Read flugelhorn feature. As a flugelhorn player myself, I knew what was required in a skip like this. I was able to use a combination of low velocities and carefully-drawn pitchbend data to effectively create these skips. There are other situations that benefit from the adjustment of velocity to get precisely the effect required. The "shakes" layer adds even more advanced options for articulation simulation. Between the slur feature, adjustable velocity, and the "shakes" layer, there is usually a combination that will achieve a given articulation. Notice that my description is one of careful editing procedure – not live playing. I believe, therein lies one of the problems here – the library is designed to be used one way and you are trying to use it another.


In theory I'm all for the design concept of JABB, massively reducing the sample content and size, and using programming instead for the flexibility. But you can only sacrifice basic sound quality so far - I don't see sample libraries having anything to worry about any time soon.From what you have said so far I suspect that you will find Chris's programming approach more to your liking. Now, you are, of course, perfectly free to use the JABB library in any way you desire but don't confuse the results (of your personal approach) with the library's inherent sound quality. Used as designed, it is capable of high quality mockups. In fact, certain things cannot be accomplished with any other existing library of which I'm aware (like the scream trumpets in the latest demo.) I've said it before: JABB is just another collection of samples until you apply the many controllers as intended. The “sound quality” happens only after the controllers have been applied in detailed, careful, and knowledgeable ways. The user must know precisely how and where to apply the controllers to get convincing results. That will remain the case until advanced scripting becomes available.

Tom

[Raindog]

That will remain the case until advanced scripting becomes available.

Tom

Just out of curiosity....WILL advanced scripting be available for JABB? I´m a very passionate K2 user and I find the possibilities of scripting unbelievable. What about a future update (which will be not for free) for Kontakt 2? In the meantime I could try some of the available scripts (such as Big Bob´s excellent legato and vibrato script). But think about it...PLEEEEASE
regards
Raindog

[Tom Hopkins]

WILL advanced scripting be available for JABB?We've said from the release of the product that we plan to use scripting with JABB. In fact, some of the programming (like the shakes layer) is specifically tailored for future scripting. So, yes.

Tom

[Houston Haynes]

i wish there was a script for kontakt that would automatically split the notes in a chord and send each voice to a different instrument for realtime polycontrol of a hornsection.
Would this be possible?

Hans

You might be able to get what you're looking for with Plogue Bidule, but you'd be opening an instance of GPO with multiple instruments loaded into it, and then setting up some MIDI parsing algorithims to analyze the incoming data and split the notes up to each instrument input - not easy but it most certainly can be done. All of this would run inside of Bidule and you would then use Bidule as a VSTi wrapper or as a ReWire slave.

You might also get the logical editor in Cubase/Nuendo to do that for you, but I'm not entirely sure. There are logical editor presets that you can design and then call up in a MIDI channel insert (and route the input to another channel) so perhaps there's a chance to "cascade" through to another channel if the logic is right.

Then again, why? It's not like players play that way or arrangers arrange in that fashion - and you can end up with some unmusical hocketing where one chord is assigned according to the order in which you play the chord. To me that's just too far away from useful to be worth the effort.

I'm reminded of a story told to me by an organ instructor/professor. When he was young his parents signed him up for school band and on the first day he picked up the clarinet and asked how he was supposed to play chords. When the band director told him it was impossible he put it down, left and never walked into the band room ever again. ;)

[Ouch that hurts]

Tom, you're undoubtedly right that I'm trying to use JABB in a way that is quite different from its design concept, so it's by no means the library's fault that it isn't working the way I want. I should point out that I did buy it largely for studio use - just with one eye on the fact that it might be useful live as well - so I don't feel disappointed or anything in this regard.


Now, here we have a significant disagreement concerning this issue. In creating the big band demos (including the recently posted scream trumpet demo) I spent a great deal of time with this programming and the tongue/slur feature played a major role in the success of the demos. One of the essential aspects of horn playing, and one usually missing from simulation attempts, is the use of the tongue. I found that applying intricate tonguing and slurring articulations using cc64 in JABB often made the difference between an obviously synthetic rendition and one that started to sound convincing. It was emphatically important. It must be used properly though to get good results. You must know exactly where (and how) the tongue get used in a passage and correctly place the cc64 data to make it work convincingly.

Actually with a bit more work behind me, I am starting to get used to this.


As to the “weird” low velocity behavior of the tongue layer: In JABB, velocity controls attack strength. Low velocities give gentle attacks that have a number of purposes. Horn players can create legato lines in two ways: By slurring, where the tongue does not interrupt the air flow; and by very gently tonguing, where the tongue lightly “feathers” the air stream. Both can be simulated in JABB. Furthermore, the lowest velocities can be used in special situations. An example is the case of slurred octave skips that I encountered during the redo of the Paul Read flugelhorn feature. As a flugelhorn player myself, I knew what was required in a skip like this. I was able to use a combination of low velocities and carefully-drawn pitchbend data to effectively create these skips. There are other situations that benefit from the adjustment of velocity to get precisely the effect required.

OK, I suppose it's just a question of getting used to the particular way the lowest velocities are designed and using them appropriately (or more inportantly, not using them inappropriately). I suppose I'm just used to the "normal" way of doing things, where velocity determines a combination of volume and timbre, but not specifically attack strength to this degree, so you can play fluently right down to the lowest velocities and it sounds natural.

I find in JABB those lowest velocities sound completely unnatural in most situations. But then that's probably just a case of training my playing so that I don't use them in most situations, but program them in for the specific needs that you mention. I wonder though, if they need to be used so sparingly and deliberately, whether velocity is the most logical way to achieve such an effect. I've always found that the advantage of linking things to velocity is that realising them becomes a much more flowing and easy part of the playing process than keyswitching or slider manipulation. For example I have all sorts of things programmed into the highest velocities of the brass patches in my live setup so I don't have to remember which bit of machinery to reach for and what to do with it, I just hit the keys harder. Other techniques on the other hand work better by being programmed in with MIDI CCs or whatever, so you can get more precision. The kind of unusual articulations you're referring to would seem to come more under that category. Just a thought...


From what you have said so far I suspect that you will find Chris's programming approach more to your liking. Now, you are, of course, perfectly free to use the JABB library in any way you desire but don't confuse the results (of your personal approach) with the library's inherent sound quality. Used as designed, it is capable of high quality mockups. In fact, certain things cannot be accomplished with any other existing library of which I'm aware (like the scream trumpets in the latest demo.) I've said it before: JABB is just another collection of samples until you apply the many controllers as intended. The “sound quality” happens only after the controllers have been applied in detailed, careful, and knowledgeable ways. The user must know precisely how and where to apply the controllers to get convincing results. That will remain the case until advanced scripting becomes available.

Thanks again for your input. I'm not sure about Chris's library. I don't really need such a vast array of variety for the kinds of live lines I'm doing (mostly funk/soul riffs and stabs), and I'm not 100% sold on the sound quality from the demos. I'm sure it's probably superior in many ways to QLB, but the problem is I'd have to load it in Kontakt to adapt it to my needs, and then load Kontakt on my live rig which currently runs Giga. I rather think the improvement in sound quality would not be radical enough to justify the cost (both of the library and an extra Kontakt licence) and hassle.

I'll probably just carry on tweaking my custom gigs using QLB and some other stuff, and see whether the next round of development in this area either brings things back to the Giga platform, or moves them on enough to convince me to switch. In the meantime I've got JABB to play with in the studio, and I'll try that polyphonic upgrade...

[Ouch that hurts]

i wish there was a script for kontakt that would automatically split the notes in a chord and send each voice to a different instrument for realtime polycontrol of a hornsection.
Would this be possible?

Hans

I asked about that a while ago and there were some ideas raised, but nothing very definitive. I'm now looking at the possibility of using the Giga Editor's key mapping to pre-program horn parts in this way. I'm sure the same could be done using the key mapping in Kontakt.

Most of the songs I'm playing are pretty simple and the horn parts are pretty limited, say a two- or four-bar riff repeated a few times in the verse, followed by a couple of harmonised stab chords in the chorus, stuff like that. So what I plan to do is arrange the gig files I'm triggering so that I play a single melodic line at all times, but it's sent to all the instruments in either unison or harmony according to how I want it.

I figure I can do this by using different octaves of the keyboard. For example I play the unison bits high on the keyboard and Giga transposes them an octave down for the trumpet and alto sax, and two octaves down for the bone and tenor sax, giving me octaves. Then I play the harmonised bits in the middle of the keyboard, and the instruments are actually remapped in Giga so that when I play a C, say, the trumpet plays a C, the alto sax the G below it, the tenor sax the E below that and the bone the Bb below that, for a C7 chord. I'll literally arrange all the voice-leading in detail and map the notes of the melody accordingly.

The limitation of this is that I'll only be able to play as many different voice-leadings as I have octaves on the keyboard. For me this is not a huge problem as the songs are pretty simple, and also I can also just switch to a new preset on the keyboard, sending prog changes to Giga, to access a whole new set.

Also, it means everything will be arranged, not improvised. On the other hand, that's most how horn players tend to do this stuff anyway, notwithstanding solos which are a different matters.

One big advantage on the other hand is that I'll be able to play everything with one hand, leaving the other free for slider and knob manipulation (oo-er!), adding to the expressivity and realism.

It does take it a little too close to just playing backing tracks for my liking, but you can't have everything. It does mean I'll be able to really "play" the arrangements in time with the band, rather than killing the feel with click tracks etc, and adapt the sound manipulation with regard to how the band is playing, so it won't be quite so bad as that. I can also imagine adding aspects of flexibility to it, for example by playing the voice-led chords via the melody in my right hand and adding bari sax lines on their own in the left hand.

At the moment I just use a general brass and sax sound spread over the whole keyboard and it really just ain't right - you either have brass only or sax only or EVERYTHING doubled, all the time. I've played around a bit with arranging things this way and they already sound a million times better. Lot of programming ahead though...

[Chris Hein]

I'm not sure about Chris's library. I don't really need such a vast array of variety for the kinds of live lines I'm doing (mostly funk/soul riffs and stabs)
All the single articulations are available as single instruments.
And all these instruments are available in 3 different versions with different amount of velocities, samples and ram usage.


I'd have to load it in Kontakt to adapt it to my needs, and then load Kontakt on my live rig which currently runs Giga.
You don't need kontakt.
You can load up to 8 articulations like stabs, falls, sustains, crescendo in to one kontakt player.

I am planning to do an update on the single instruments with keyswitches.
So, an instrument with a single articulation only playes, when you hit the corresponding keyswitch.
With theese instruments you could fill up your player with 8 articulations
of your choice and play them on one midichannel with keyswitches.
You could even use an Alto-Sax crescendo and a Trumpet stab with a Trombone slide and Tenor-Sax and Trumpet playing together the sustian....

So, there are a lot of choices to put you section together with different
performance needs and combinations of instruments.

Chrs Hein