• Register
  • Help
Page 1 of 2 12 LastLast
Results 1 to 10 of 18

Topic: A Request for Orchestral Lib Developers NON-NOISE REDUCED VERSIONS

  1. #1

    A Request for Orchestral Lib Developers NON-NOISE REDUCED VERSIONS

    I've have heard a few demos from composers who have recorded custom libraries recently. There is no question in my mind that the air and natural hiss gives their work life that I have not heard in QLSO, Sonic Implants, GOS, or any of the other major orch libraries. Vienna is obviously a different concept and falls out of this particular thread.edit: SAM libraries, to my ears do NOT use noise reduction. I don't mean this as a dig against these developers. I use and appreciate these libraries every day.

    I realize there are a number of concerns around this topic. I realize the noise itself is cumulative. At the very least, I would be thrilled if major developers would consider releasing non-noise reduction versions of Violins and Trumpets. These seem to be the most effected by noise-reduction.

    One could take this one step further and release versions that have not been pitch corrected. Again trumpets and high violins would benefit the most from these inperfections IMO. Thats another worthwhile discussion to have.


    Colin O'Malley

  2. #2

    Re: A Request for Orchestral Lib Developers NON-NOISE REDUCED VERSIONS

    I've got an orchestral noise library in the works that will be going for a very reasonable price. It will be available in Gigastudio and Kontakt formats with both 16 and 24 bit samples. Included will be a special Gigapulse noise convolution of the noise from several major performance halls. Crossgrades will be free of charge.

    Please forgive me. I couldn't resist :-)


  3. #3

    Re: A Request for Orchestral Lib Developers NON-NOISE REDUCED VERSIONS

    I think that would work great for solo instruments but only in a simply orchestrated piece . So you would have a "noise channel" playing constantly even if the instrument ( eg violin ) changes articulations . If you have a note off though , you would hear the noise dropping down .

    On the other hand if a string section had noise in each note, a big fat chord would mean 5 to 7 times that building up which is quite restrictive.

    Since recordings are so well done lately, even a non-noise reduced library would not have noise ( correct me if i'm wrong ) .
    Also, the composers which you say recorded custom samples, maybe did
    not do it with the extent of technology that's why they got so much noise.
    Theo Krueger - Composer


    Kontakt 2 Scripts

  4. #4
    Senior Member Bruce A. Richardson's Avatar
    Join Date
    Sep 1999
    Dallas, Texas

    Re: A Request for Orchestral Lib Developers NON-NOISE REDUCED VERSIONS

    Quote Originally Posted by Lee Blaske
    I agree. Having the noise recorded with the sample simply would not work because it would be unpredictable. You'd usually have too much or not enough. The only way to do this is to add it in after the fact. Also, real orchestra noise is not at all random. You'll often hear wind players inhaling before playing a phrase. You'll hear string players breathing along as they phrase. There's no way you can incorporate that into samples.

    Actually, if you really wanted to, you could go into your studio, sit down in a chair, and "perform" some breathing that would by synced to the score.

    Lee Blaske
    Dead on.

    Noise in the samples is the absolute wrong way to go. If you are interested in designing a virtual orchestra performance, then these extramusical elements must be specifically designed. It's a sound design gig, the sound design of an orchestra playing a particular work in its particular physicality. It is not the combination of recording imperfections that will produce this result, but the study and designed reproduction of a set of physical motions combined with an overall ambient feel (whether that includes noise or not). It has nothing to do with the notes being played, therefore tying it to note values is a design mistake.

    Now sure, you may serendipitously get some element of this via "noisy samples." But as Lee says, its unpredictability is death to its useful function...self canceling. It's not even analogous to the musical/physical behavior it is being credited with aiding. Room noise does not reflect the number of voices playing within the room. It is what it is, a single-state condition. To attach it on the note level makes it anything but a single state condition. Likewise with tuning anomalies. Hence the word anomaly. To place something in incorrect tuning with the intent of invoking realism must be approached on a controlled, not random, basis. Frankly, every place I have ever done this, I've eventually lived to regret. I thought it would be great to have one horn out of a section of four execute a scoop lazily, and behind the beat. So I did it, in a show that I deal with every year. Every time I hear it now, it drives me insane. An imperfection for perfection's sake doesn't make sense. Face it, in a recording environment, the ideal is to produce the performance one dreams of having, not the one that we ACTUALLY have. This is why we have take two and three and so on. We are looking for an idealistic performance, not a realistic performance. Realistic performances are for people to hear once, on a good belly full of food and wine. They are to be forgotten.

    Hell, no orchestra leaves a bunch of sonic or musical boogers on their albums. They go in and clean that crap up. There is no particular nobility in hiss or botched tuning.

    Ditto timbral inconsistencies. For the purposes of sampled virtual orchestra, you must have enough consistency to realize any line you may want to play. You make the most music in the least amount of time with libraries which offer predictible behavior across the board. These are the ones in which you can conceive any line in the range of the instrument, and be able to coax a performance out of the samples. That "high personality" note which is charmingly scoopy may work in one setting, but will be disastrous in one thousand. And to create that condition is bad library design. It makes the tool difficult to use, and it narrows the creative options. One must realize that the analogy for an anomalous response is an anomalous technique--an exception, not the rule. A deliberate act, not one which comes off the shelf.

  5. #5

    Re: A Request for Orchestral Lib Developers NON-NOISE REDUCED VERSIONS

    I have spent hours cleaning up chair creaks and such like, so for me there is no place on a sample for these noises. However, I would like somehow to be able to recreate that wonderful studio "air" that happens at the start, end and silent sections of a recording, as the lack of this is what often makes virtual orchestras sound rather lifeless to my ears.


  6. #6

    Re: A Request for Orchestral Lib Developers NON-NOISE REDUCED VERSIONS

    My request has NOTHING to do with chair squeaks and clicks etc. If you have a chance, listen to the demos of Craig Sharmat, Maarten Sprujit (A Child Rises- True Strike demo) and Simon Ravn at V.I. Control. There is a shimmer and air to the samples that is missing in most libraries.

    Sonic Implants Brass for example, is a great library. Obviously great players, room and well performed samples with a lot of life. For me there is a noticable blanket on the trumpets in particular. I believe the culprit is the Waves Restoration X-Noise, X Hum plug-in they used. I understand there are situations in which Noise Reduction is necessary, but I've never used it on any score I have recorded with an orchestra. Particularly for forte and tutti passages I think it kills some of the life of things. For soaring trumpets I feel strongly the samples should be left alone. I'm sure its a bit of a balancing act for developers, but I don't believe processing every sample is the way to go.

    Perhaps Maarten can correct me here, but I believe the SAM libraries do not use noise reduction. This RAW sound is one of the reasons these libraries work so well.


  7. #7

    Re: A Request for Orchestral Lib Developers NON-NOISE REDUCED VERSIONS

    I have to agree with this to a large extent. It wasn't only me that preferred the sound of the PMI Emperor prior to noise reduction. The early demos had a uniquely realistic sound to my ears, mainly due to the ambient noise in the samples. Nevertheless I bought it because I believe it's still one of the finest Bosendorfer multisamples in existence.

    The big question is what is "noise" and what is "instrument"... It would be nice if that were left to the customer to decide if he or she so wishes...

  8. #8

    Re: A Request for Orchestral Lib Developers NON-NOISE REDUCED VERSIONS

    I'm not trying to say using noise reduction is wrong. It would be great to have other options though, i.e. an 18 Violins RAW patch, or 3 trumpets RAW. I think these patches would be incredibly useful.

    Regarding the room, I understand there are different room characteristics. Maybe that is some of what I am hearing. However, I love the SI rooom. At the end of the day this is a pretty subjective thing, everyone has their preference.

    It would be very interesting to hear an A/B of a few samples noise reduced and not. Our ears are pretty receptive in the higher registers and its worth considering the role noise reduction may be playing. Until we hear that its all informed speculation.


  9. #9
    Senior Member
    Join Date
    Jul 2004
    Calgary, Alberta, Canada

    Smile Re: A Request for Orchestral Lib Developers NON-NOISE REDUCED VERSIONS

    I have some percussion and bass libraries that use fret and hand scrapes on some not off samples. This is something you can turn off or on at will as it is a seperate sample. If done well I too feel that in certain tracks this could add a level of authenticity. I agree that solo or chamber music is where this would be the most usefull, while in film/TV/game music it would be horribly destracting. Musicians make alot of extraneous noise and it would be interesting to capture some of it (and have the option to turn it off).

    Of course, one of the things that makes recordings so great (IMHO) is the lack of this noise. If I wanted noise I would go to a concert where it seems more acceptable. Part of this is that there is an electricity at a live concert (mentioned in a previous post) that comes from knowing you will never hear this piece played the same way ever again. You see the musicians (an underated part of concert going) really getting into the music and there is a hushed concentration (in a classical concert) or a raucos chorus (at a rock or pop concert). It is also a unique combination of a shared moment and a deeply private moment as everyone listens to the same unique performance but each has there own experience.

    In my experience even a recording of a live event doesn't really capture the whole of it. Perhaps samples can't create this feel because they rely on repeated recorded events, where as a live performance has an infinite variety of possibilities. I do however sympathize with what you are saying and would probably go for some small tastefull breathing samples.

    Imagine a Keith Jarret or a Glenn Gould patch.



  10. #10

    Re: A Request for Orchestral Lib Developers NON-NOISE REDUCED VERSIONS

    FWIW, I totally agree with Colin. I too would love to have the option of 2 versions, raw and clean.

Go Back to forum


Posting Permissions

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