• Register
  • Help
Page 1 of 2 12 LastLast
Results 1 to 10 of 20

Topic: Do accumulated solo strings really work?

Share/Bookmark
  1. #1

    Do accumulated solo strings really work?

    The problem is the fellowing: I can sum up e.g. 12 solo violins and the result is always that they sound more like a 12-times-doubled solo violin than like a string body. The don’t merge into one sound body as in real life. What are the reasons for that?

    1. solo strings are recorded close mic. They don’t have the room reverb like a recorded string body. Recording a string body from a forced certain distance behaves quite different.

    2. the single string instruments don’t make the neighbour instrument oscillate like in real life

    3. generally still the lack of 100% natural reverbs although the impulse reverbs are very close to that

    4. the psychological componet: real string players hear and influence each other and contribute instinctlively their component to the whole sound mix


    Any other ideas?

  2. #2

    Re: Do accumulated solo strings really work?

    2 and 4 would be tough to do anything about, though it\'s hard to know exactly what impact they have on the final sound.

    I think for GPO, all the instruments are miked in the same fashion and from the same perspective (anyone know for sure for the solo instruments?). That would mean number 1 doesn\'t apply here. If I\'m wrong, at least that would be easy to create for some future sample library.

    I don\'t agree that number 3 is a problem, at least not one specific to section vs. solo strings.

    One thing I would add is that usually when a solo violin is sampled, they are playing like a solo instrument, not a section instrument. That usually means a little more vibrato, at the very least. A violinist plays a passage very differently if they are doing it as a solo, as opposed to in a section. Of course, this can also be remedied in the recording process, by asking the player to do otherwise. That suggests that it may be useful to have two different versions of each sampled string instrument: one played as an actual solo, and one played as if they were in sections. I wonder if players could make that psychological leap, and whether the microphone could capture it.

    In any case, GPO doesn\'t do all that bad, does it? I can\'t wait to see where GOS 2 might take us. I\'m guessing we\'ll all be blown away.

  3. #3

    Re: Do accumulated solo strings really work?

    Ever since I purchased GPO, I have always been confused when everyone talks about building string sections, from what....... the solo samples? They all have too much vibrato to be used collectively,,,,, do string players play with that much vibrato when they are playing as part of a section?
    Also, there is no replacement for high quality digital reverberation. I use Altiverb, and although there is not a ton of adjustments as other great reverbs provide, you just choose the IR (impulse response) that best suits the sound/mood. I am always very pleased with Altiverb.

  4. #4

    Re: Do accumulated solo strings really work?

    Now I just can\'t believe that so less people take part at this interesting discussion. Try to see this higher-ranking...no one is trying to critisize your holy GPO! The problem that I described also appears e.g. if you want to build a choir out of several layers of your own voice. They simply won\'t merge into that typical choir body sound. It is also the reason why effect plug-ins like \"clone ensemble\" do not really work .

    So where are the physicists, musicians and sound engineers? Any declarations besides my humble effort to find some?

    PS: someone censored my thread headline. This is really very poor.

  5. #5

    Re: Do accumulated solo strings really work?

    I think a combination of micing and player ensemble technique (your 1 and 4 points I think) is the answer. I don\'t think induced vibrations from one instrument to another would be a significant point, as it is in a piano, for example, where all the strings share the bridge.

    I supose that removing some high and lo freqs to rudely push the strings far away could help in melting the ensemble sound. But take into account that you usually need a minimum of 4 string real players to build up an \"ensemble\" feeling.

    If you listen to \"The sad composer\" by Francesco Marcetti, I don\'t think it sounds so bad. I was posittively surprised when I heard this one, actually. The ensemble sound he reaches form solo instruments is astounishing IMHO. I think the key here is to make each line from each track sound convincing by itself, AND, make them independently. If you keep on the grid, and make rude expression editing, it will just not work. Little natural tempo sweeps between lines and expression independence is the key. But this is applicable to the rest of the orchestra as well.

    Anton

  6. #6

    Re: Do accumulated solo strings really work?

    Yes....GPO is a good tool and even I use it ...I enjoy it and I have made already some quite fine tunes with it.
    Yes, it sounds not bad. Sure. But if you are honest that certain thing does not happen. If you sum up these solo strings they sound just like the sum and not more...they do not begin to get that certain silky oscillating body sound. Also quite obvious if you try to unite the woodwinds to one body btw.

    And I am riddling why this is so. Generally. Another explaination could be that an instrument ensemble gives another reverb impulse to the room than added single instruments. Don\'t know. Just thinking.

  7. #7

    Re: Do accumulated solo strings really work?

    [ QUOTE ]
    The problem that I described also appears e.g. if you want to build a choir out of several layers of your own voice. They simply won\'t merge into that typical choir body sound.

    [/ QUOTE ]

    Theres a big difference between a bunch of your own voices and a bunch of different voices that Im sure you can understand with a tiny bit of thinking. To apply the same reasoning to a bunch of solo violins would be illogical.

  8. #8

    Re: Do accumulated solo strings really work?

    [ QUOTE ]
    The problem is the fellowing: I can sum up e.g. 12 solo violins and the result is always that they sound more like a 12-times-doubled solo violin than like a string body. The don’t merge into one sound body as in real life. What are the reasons for that?

    1. solo strings are recorded close mic. They don’t have the room reverb like a recorded string body. Recording a string body from a forced certain distance behaves quite different.

    2. the single string instruments don’t make the neighbour instrument oscillate like in real life

    3. generally still the lack of 100% natural reverbs although the impulse reverbs are very close to that

    4. the psychological componet: real string players hear and influence each other and contribute instinctlively their component to the whole sound mix


    Any other ideas?

    [/ QUOTE ]

    1. Add your own reverb ambience. I\'m doing a surround mix using Voxengo Pristine Space and an impulse from GOS. The trick is routing things for proper positioning and then carefully downmixing to two channels. If your impulses are adjusted correctly, you get a good 3-D impression from even in a stereo mix. There are also other more processor-intensive answers to this problem, but I will leave that to others to describe.

    2. This will never happen with sampling. The best hope is that 64-bit systems and physical modeling environments will allow a \"sympathy\" coeficient such that this effect can be recreated. I know that Scarbee Imperial Drums does a lot to sample a variety of positions on each drum with the entire kit present to also capture some of the sympathetic resonance, as you will hear the same with a real kit. The size of library that generates is impressive, to say the least, and I\'m sure is out of scope for GPO. Besides, sampling resonance for a struck instrument is much simpler than capturing the infinite varieties of resonance from a driven oscillation like a violin. I also think that this effect is not nearly as important as room positioning in the acoustic space.

    3. This touches number 1 - convolution and surround processing in real-time is the next big thing. Just like sampling, it sounds like the real thing until you hear the real thing. (apologies to Jeff Rona) Garritan has mentioned in a few threads that he has amassed a huge catalog of concert hall impulses. It will be interesting to see what he does with those. I\'m sure that he will break ground with that project the same way he redifined orchestral sampling with GOS.

    4. Could be partly touching number 2 - but I think this is more of a factor that you CANNOT just copy MIDI parts verbatum and play them on separate ensemble instruments and have it sound like an ensemble of different players. There has to be some factor of individual expression in each part that the MIDI arranger has to imply as coming from a separate source. Then the ensemble begins to \"amass\" an ensemble quality. Listen to Marchetti\'s \"Till the End\" and you will get the picture of how this works. I remember back in the early 90\'s when a Japanese artist gave a talk on how he used the Kurzweil K250 to play individual violin lines onto a multi-track in real time. He mentioned that not only do you NEED independent vibrato on each line, but you also need to vary the pitch and timbre of each line to get a realistic result. This is not only because you hear the differences, but also because of the \"averaging effect\" that is caused by the complex evolution of phase and pitch relationships that is perceived by the human ear (or more accurately, the brain) and translated as one body of sound. I believe that Francesco does this in spades, to fantastic result - PARTICULARLY coming from a $249 library.

    It\'s the end result that counts. For my money, I prefer to use sectional sample sets with a solo instrument or two to mix in and \"speak out\" during soli passeges. That provides a good effect without an overinvestment of time. I also have a background in mixing and mastering, which I understand that not everyone has, but at this point I prefer to \"own\" certain areas of the process while letting others slide. Everyone makes those compromises - and it can be easily heard in their work.

    I\'ll be producing a few tutorials for GPO and Nuendo, perhaps even later today - and I\'ll be including some tidbits on surround mixing and convolution as well. Just FYI.

  9. #9

    Re: Do accumulated solo strings really work?

    Hey cool... A tutorial...awesome. I\'ve never heard of that reverb plugin you mentioned. Sounds very intriguing...

    My comment about combining the solo strings....

    I personally think that this is absolutely the best way to come up with great sounding string ensembles. If you do it all the way...the hard way...they will sound amazing..precisely because of all the vibrato and also because if you do it right..because of all the variation that you will introduce for each \"player\". Of course...its takes 10-20 times as much time as whipping it together with some great ensemble samples...but personally..I think the results are finer and more detailed...with the possibility of going off into truer divisi sections, etc... doing it this way, you are directly controlling each and every string part with a fine degree of control....more work yes...perhaps more subtle detail also...

    If you take the time to play all the parts seperate, make use of the random pitch control in GPO, get a good reverb and use it wisely..I think you will come up with some very convincing string parts.

    But its definitely true that it takes a lot of time to do it. For me..its a good thing...because i\'m still learning about arranging and orchestration and it forces me to put on a magnifying glass and think about every part....nay...every PLAYER! I suppose in the long run...if I were on a clock..I\'d be looking at GOS or something to augment..and to provide quicker and easier string sections.... But that isn\'t to say that GPO can\'t deliver the goods. It just requires a bit of work to do it.

  10. #10

    Re: Do accumulated solo strings really work?

    Check out Voxengo Pristine Space. Alexy Vaneev has also promised to update his impulse designer software that bookends PS to also support generation of multi-channel impulse files. That should be interesting, especially as surround impulses start showing up in greater numbers on the market. Right now, I use a stereo impulse routed to each of three pairs of inputs, with a slight delay on the surround (about 35 milliseconds). I\'m also downmixing to stereo using the SRS Circle Surround Encoder, so things are a bit crowded and a bit more wet than I like, but it seems to transfer very well to other locales. It will be interested to see how the mp3 fares.

Go Back to forum

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •