• Register
  • Help
Results 1 to 3 of 3

Topic: Newbie problem with samples

  1. #1

    Newbie problem with samples

    Hello all,

    I'm sure this is user error, but I seem to be having a problem with phase cancellation using some of the JABB samples. I'm using 6 instruments (alto, tenor, bari, tpt 1&2 and bone). I've tried panning with no success. Is there some additional setup required when loading the samples? My hardware/software setup is:

    MacBook 2Ghz (2GB RAM)
    Pro Tools 7.1.3
    Kontakt 2.2 (full version)

    I'm aware of the compatibilty issue with the embedded player and MacTels but is there also an issue with the actual samples? Any help would be appreciated..

  2. #2

    Re: Newbie problem with samples

    I'm going to assume that you are referring to "phasing" between instruments playing unison lines. This will occur when two (or more) instruments are playing the same pitch and there are consistent and unchanging, slight differences in intonation and volume between them. This can also happen with real acoustic instruments but it is rarely distracting because real players can't maintain pitch and amplitude with that kind of consistency. And therein lies the clue to mitigating the problem. Here are the things I did in the big band demos when unison lines exhibited phasing:

    1. Positioning - I made sure that each instrument had a different position on the stereo stage - panning. I also treated each section (saxes, trumpets, trombones, rhythm) with a different reverb treatment using convolution to reflect the differing distances of my "players" to the microphones - front to back.

    2. Pitch - on sustained unison notes I used pitchbend data to draw in small, rapid fluctuations in pitch accuracy - different data for each unison instrument. On notes of short duration (rapid passages like scales) I used VAR1 and VAR2 to add random variability in both pitch and timbre. Variability data was only drawn into the locations requiring it.

    3. Amplitude - on sustained unison notes I used mod wheel data (cc1) to not only shape the sustains but also add rapid fluctuations in volume, similar to the fluctuations in pitch mentioned in the previous point, at the locations of any phasing problems - again, different data for each instrument.

    4. Timing - players never execute lines in perfect synchronization. I played in each part separately to give timing discrepancies to the parts. This also serves to cause unison waveforms to overlay with differences from note to note. Always approach each part as a separate player.

    5. Instrument choices - use different instruments for different parts; i.e. don't use Trumpet 1 for all 5 trumpet parts. Each trumpet has a different tonal character and instruments with different characters combine with fewer "phasing" problems than identical instruments in the same situations.

    6. Don't make the mistake of (on average) playing at the loudest end of the mod wheel travel. You should mostly have your players working at mid values except for the loudest passages. Consistently using high values in your horns will emphasize the upper overtones which is where phasing is most audible. It will also make your players too generally aggressive in most cases. With a few notable exceptions, most horn players do not consistently blow at maximum capacity - they use dynamics and pace themselves. Also, lead players tend to blow harder than players on lower parts. This can also be used as a tool for differentiating between players.

    As you can see from the above suggestions, the point here is to avoid consistency between players - especially consistency of amplitude and pitch. Real players aren't near as consistent as unmodified samples and you must take steps to consciously add imperfections with controller data. The imperfections can reduce or eliminate the source of phasing, which is static pitch and amplitude relationships between players holding the same notes. Strictly speaking, since this is a mathematical phenomenon, it can also occur between any notes that have a relationship of multiples but it is most obvious with unisons. Anyway, those are the main things I did in the demos to reduce the audibility of phasing on unison lines. But don't forget: It's not necessary to completely eliminate this phenomenon because it occurs in nature too. I was surprised to note when I used some excellent big band recordings as a reference that live recordings had it sometimes too; especially in the trumpets. Interestingly, the better the players and the better rehearsed the bands, the more frequently it appeared.

    Finally, you can examine the midi data of the posted demos for detailed information on what I did. This long answer assumes that I interpreted your question about "phase cancellation" correctly. If not, . . . I'm sure you'll let me know!


  3. #3

    Re: Newbie problem with samples

    Thanks for the very detailed reply. I'll try your suggestions this weekend...

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