• Register
  • Help
Results 1 to 10 of 10

Topic: Stupid GPO Question 134, Solo Strings

Share/Bookmark
  1. #1
    Senior Member
    Join Date
    Mar 2004
    Location
    Wilton, NH
    Posts
    2,450

    Stupid GPO Question 134, Solo Strings

    I’ve been experimenting with GPO a little but I’ve only been able to squeeze in about an hour a day to play with it. [img]images/icons/frown.gif[/img] Sometimes not even that. (2 weeks ago I didn\'t play/make any music on my computer) Anyway, Since I haven\'t had time to find out for myself I have a few questions about building sections.

    I know you shouldn’t use the solo instrument with the ensembles built with that instrument because of phasing problems. Is this only if it is exactly doubling the line (i.e., copied the midi data to another track)? If I play the parts in manually without quantizing, different velocities, different settings for VAR1 and VAR2 (yielding slightly different pitches and filter setting), etc would I still have phasing problems? Would it maybe be worse? I know this is in the manual, but the wording was such that it seemed to imply it was only a problem if the parts were exactly the same. Maybe I’m taking this “exactness” too literally.

    On a similar subject I’m wondering about the use of the exact same instrument. I know I can use violin1-ensamble1 for both my first and second violins (since the notes will be different) but how about doubling that instrument on the first violins? It seems like it should cause phasing problems, but does it? (I am, of course, talking different instances of the same sound.)

    I have other questions, but I’ll save them for a rainy day. Actually, with the weather we’ve had on the East Coast, I’ll save it for a sunny day – we should have one by August, I’m sure.

  2. #2

    Re: Stupid GPO Question 134, Solo Strings

    Originally posted by trentpmcd:
    ...a few questions about building sections.
    ... phasing problems.
    <font size=\"2\" face=\"Verdana, Arial\">Trent,

    I\'ve had similar questions about using the same instrument. As best I can hear, if the same instrument is used to play two or more notes in unison you get a phasing problem.

    The phasing problem (as best I can hear) is that you lose much of the characteristic timbre of the instrument and it sounds unnaturally loud.

    But I personally think that it\'s fine if you use the exact same instrument in multiple simultaneous non-unison parts -- like a piano plays normally.

    It would be interesting to hear from someone on this forum who has a physics definition of a phasing problem. I\'m just going by ear.

  3. #3

    Re: Stupid GPO Question 134, Solo Strings

    I don\'t have a physics explanation or anything, but Tom Hopkins confirmed over the phone for me that its okay to use the same instrument for different parts as long as they\'re not playing the same notes.

    That is correct Tom, isn\'t it? It was like 2am my time, I hope I remembered correctly...

    Anyway, a good friend of mine is a physics major at the University of Illinois, I\'ll direct him to this thread...

  4. #4

    Re: Stupid GPO Question 134, Solo Strings

    phasing problems ONLY happen when you are playing exactly the same sample with exactly the same pitch (or very close pitch) at exactly the same time. Well, actually..what happens is that when two are layered, they may NOT be exactly starting at the very same instant (down to sample accurate time)..and that\'s where the phasing problems come in.

    In fact if they were starting exactly at the same time there would be no point...it would just be the same sound only too loud. But if they are offset a wee little bit..the phasing problems can create all kinds of wierd EQ artifacts from the two identical waveforms being combined together. EQ artifacts and the sound will seem to jump around being too loud and too soft.

    You might be able to get away with combining the same sound twice in unison if you have them both detuned a littl bit...but you might also get some phasing artifacts. Perhaps combining time and pitch variation will give you an acceptable sound..perhaps not. Mileage will vary. But bottom line, its just easier to avoid the problem by not using the same sample twice in a unison line (or even a shared note by two voices that otherwise are seperate).


    Phase problems suck. Avoid them. Don\'t use the same sample twice. And for GPO that means you need to understand which solo samples are used in the ensemble patches (explained in the manual).

  5. #5

    Re: Stupid GPO Question 134, Solo Strings

    Alan,

    Your barely-awake memory is correct. Phasing is only an issue when two identical instruments are playing the same note(s). There are advantages (beyond avoiding phasing problems) to using different instruments, even when there are no unison notes, but “same instrument – different notes” should work fine.

    Tom

  6. #6
    Senior Member
    Join Date
    Mar 2004
    Location
    Wilton, NH
    Posts
    2,450

    Re: Stupid GPO Question 134, Solo Strings

    Thanks. I guess the lesson is playing the same part, even if entered separately, is to be avoided. Same, I’m assuming, when using a solo instrument along with ensemble instruments based on it.

    To a small degree I understand the physics behind the phase problems. If the waves are identical in some instances the two sounds will cancel each other and other places they will reinforce each other. Identical waves 180 degrees out of phase will completely cancel each other out – when I first learned this in high school physics I wanted to make noise canceling headphones –I wish I had because they are now very common and I would have been a rich man.

    I was wondering if the samples are complex enough that off-setting the parts by more than just a few cycles and maybe slightly different pitches plus vibrato and other subtle variations of performance would be enough to cancel out the effect. Sounds like a \"no\".

    Most of the time I don’t think I’ll run into a problem, however I’m currently playing with Vivaldi’s “4 Seasons” and, as written, in several of the movements the solo violin, when not actually soloing, doubles the first violin part. If I play as written I remove choices for my first violin part, at least in movements where this occurs. I could do it as a modern concerto and have the soloist do rests during the tutti parts. Of course, when I have my first violin section complete I can try putting in an ensemble instrument based on the soloist and see if problems occur, but most likely only as an experiment.

  7. #7
    Senior Member
    Join Date
    Jan 2002
    Location
    Dorset, UK
    Posts
    470

    Re: Stupid GPO Question 134, Solo Strings

    I don\'t have GPO, but another way to do it is to record one of the tracks 1 semitone down and use an audio editor to pitch shift the rendered track back up, so that diffent samples are used for coinciding notes (assuming chromatic sampling). Over one semitone the formant change is not too noticable, even less so if you use an editor which preserves the formant, (like Melodyne).

  8. #8

    Re: Stupid GPO Question 134, Solo Strings

    <font color=\"orange\"> Subtitle: Stupid GPO Question 135, Real-Live String Sections. </font>

    Why don\'t the string sections of large orchestas produce the phasing problems discussed above?

    Don\'t the better musicians seek to match one another\'s playing: pitch, timbre, attacks, decays, bowings, etc.? It seems like the better they worked as an ensemble the worse they\'d sound. But because I\'ve never heard an orchestra with a phasing problem I assume I\'m overlooking something <font color=\"black\">obvious </font>!

  9. #9

    Re: Stupid GPO Question 134, Solo Strings

    [ QUOTE ]
    <font color=\"orange\"> Subtitle: Stupid GPO Question 135, Real-Live String Sections. </font>

    Why don\'t the string sections of large orchestas produce the phasing problems discussed above?

    [/ QUOTE ]

    Phasing problems in samples come from combining samples which are mathematically identical, or very similar. Real musicians who listen to each other to unify styles, timbre, dynamics, timing, etc. still have enough subtle differences. They could play different brands of instruments, different strings, different rosins, different bow tensions, etc. Even if they played with robotic similarity, those subtle variations make a difference.

    --Eric

  10. #10

    Re: Stupid GPO Question 134, Solo Strings

    Adam,

    Don\'t underestimate the interacting complexities here. No matter how hard even the best musicians try to match one another they never actually get there. There are too many variables from note to note, not to mention acoustic variations because no two musicians are occupying exactly the same position on the stage. Try performing and recording the identical note with as much precision as possible (same attack, tone quality, volume level through the sustain, vibrato, etc.) multiple times and then examine and compare the waveforms for each. You will find that even ones that sound very similar are not an exact waveform match - probably not even close. On the other hand, with samples, the same note is literally exactly the same (if it uses the same sample) and small offsets in start time result in phasing issues because of the matching waveforms.

    Tom

Go Back to forum

Tags for this Thread

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
  •