Loading ...
Sorry, an error occurred while loading the content.

Polyphony issues?

Expand Messages
  • Jon(athan)
    So the K2500R has 48 notes of polyphony, right? (for such an amazingly powerful unit it certainly dates itself with its polyphony. oh well) I have a setup
    Message 1 of 12 , Jun 30, 2008
    • 0 Attachment
      So the K2500R has 48 notes of polyphony, right? (for such an amazingly
      powerful unit it certainly dates itself with its polyphony. oh well)
      I have a setup layering "Synbrass Sect" (a 2-layer program) and
      "Analog Brazz" (a 3-layer program), with 2 copies of each for doubled
      octaves. Actually I have a few such setups, but in a few of them the
      synth brass layers act monophonically and I don't know why. Even if a
      3-layer program takes of 3 notes of polyphony with each key I play (I
      don't remember if that's the case or not) I should only be using 10
      notes for each key.

      It appears that this only happens in setups using KB3 programs; the
      synth brass polyphony appears to work fine in other setups. The KB3
      setups also have, split elsewhere, other sounds (one of them uses "Att
      Ctl Fast Str" -- 1 layer == and another uses "Dyn Hi Brass B" -- 3
      layers -- which, ironically, is my favorite trombone section sound),
      and these other sounds are fine polyphonywise. BUT if I'm holding any
      number of notes in those sounds (strings or brass), and then I play a
      synth brass key, the first sound is cut off. If I hold the synth brass
      note first, I can add any number of string/brass notes.

      I have a similar problem in another setup using a KB3 program with
      Braz Sect and Synth Brass layered (both 3 layers each for 6 poly notes
      per note), where each brass sound will only play 3 notes at a time
      when I solo them, and when layered together, well, it's pretty hard to
      tell since one of the sounds has a pretty fast decay.

      Anyway, is there something I'm missing? I'm not too happy about this,
      is there a way to fix it?
    • James Cason
      ... Short answer is to make setups that do not involve using KB3. The KB3 mode CONTINUOUSLY uses the number of voices that are allocated to KB3 (I don t
      Message 2 of 12 , Jul 1, 2008
      • 0 Attachment
        --- "Jon(athan)" <tuzmusic@...> wrote:

        > So the K2500R has 48 notes of polyphony, right? (for
        > such an amazingly
        > powerful unit it certainly dates itself with its
        > polyphony. oh well)
        > I have a setup layering "Synbrass Sect" (a 2-layer
        > program) and
        > "Analog Brazz" (a 3-layer program), with 2 copies of
        > each for doubled
        > octaves. Actually I have a few such setups, but in a
        > few of them the
        > synth brass layers act monophonically and I don't
        > know why. Even if a
        > 3-layer program takes of 3 notes of polyphony with
        > each key I play (I
        > don't remember if that's the case or not) I should
        > only be using 10
        > notes for each key.
        >
        > It appears that this only happens in setups using
        > KB3 programs; the
        > synth brass polyphony appears to work fine in other
        > setups. The KB3
        > setups also have, split elsewhere, other sounds (one
        > of them uses "Att
        > Ctl Fast Str" -- 1 layer == and another uses "Dyn Hi
        > Brass B" -- 3
        > layers -- which, ironically, is my favorite trombone
        > section sound),
        > and these other sounds are fine polyphonywise. BUT
        > if I'm holding any
        > number of notes in those sounds (strings or brass),
        > and then I play a
        > synth brass key, the first sound is cut off. If I
        > hold the synth brass
        > note first, I can add any number of string/brass
        > notes.
        >
        > I have a similar problem in another setup using a
        > KB3 program with
        > Braz Sect and Synth Brass layered (both 3 layers
        > each for 6 poly notes
        > per note), where each brass sound will only play 3
        > notes at a time
        > when I solo them, and when layered together, well,
        > it's pretty hard to
        > tell since one of the sounds has a pretty fast
        > decay.
        >
        > Anyway, is there something I'm missing? I'm not too
        > happy about this,
        > is there a way to fix it?
        >
        >

        Short answer is to make setups that do not involve
        using KB3. The KB3 mode CONTINUOUSLY uses the number
        of voices that are allocated to KB3 (I don't remember
        the default number, but it takes up most of the
        available voices). These voices are allocated to keep
        the virtual tonewheels spinning whether or not any
        notes are depressed.

        You will only be able to use very simple non-KB3
        programs along with KB3 in a setup (something like a
        bass guitar note, maybe a few notes).

        Of course, the new PC3X starts with 192 note
        polyphony, so there are a lot more notes left in mixes
        with the PC3 KB3 module and non-KB3 sounds.


        Jim Cason
      • Jon Stubbs
        Both of these questions were answered, but I ll give you some basic information and some details. First, you are gonna have to go ahead and put that big manual
        Message 3 of 12 , Jul 1, 2008
        • 0 Attachment
          Both of these questions were answered, but I'll give you some
          basic information and some details.

          First, you are gonna have to go ahead and put that big manual
          in your bathroom, and eat lots of pizza. ;B

          Yes. If you had asked us about pro's & con's of the K2xxx series,
          you would have heard that polyphony is a weak point. Still,
          we would have had no problem recommending the K2xxx anyway.
          You would just need to understand a few things to work around it.

          The K2xxx has a voice stealing system that should make
          the loss of polyphony a bit less nasty than on other rigs.
          But you will have to be more careful in programming if you
          are making sequences with lots of tracks.

          If you are working with live stuff, you can easily make
          your tracks & programs & setups more efficient without
          feeling to hemmed in.

          The KB3 mode was a free gift-- a special mode that takes over
          most of the oscillators to use them as virtual tone wheels.
          You can think of the KB3 as some special alternate way of using
          your K2500. You can use it by itself, and when you are in KB3 mode,
          but don't think of doing anything else.

          The folks at Kurzweil could have said, "we better not give
          this gift to them, they might not appreciate it, since it
          sort of shuts the rest of the rig down." But they gave it to
          us anyway. You just have to understand that it is a mode
          that takes over the rig-- if you need it.

          If you want to stack organ sounds with other stuff, use the
          non-KB3 organs, there are some built-in, there are hundreds more
          available for free. Just watch for the KB3 thing and you can
          avoid using it in songs or setups where you need other sounds.

          It seems you have dove right in with the SETUP mode. Which
          is fine. And there are folks on this list who will tell you
          that is the way to go for setting up live sounds.

          But you should realize that the SETUP mode is where you
          mix and match programs. It is working at a macro kind of
          level.

          The program mode is where all the finer controls are for
          stacking and keymapping, etc. etc. Its the inner powerhouse
          of VAST!

          If you are wanting to stack oscillators in octaves, you can
          get some extra oscillators using vast that won't kill polyphony.
          In program mode you can edit and select VAST algorithms that have
          SAW's, or SQUARES that you can transpose and detune.

          In my opinion, you really can't unlock the power of the K2xxx series
          without completely understanding the archetecture and the way
          objects relate. The manual is a beast, but it's well written
          and will clarify most of it.

          I have posted an overview of the objects and structure on this list
          several times. If you are interested, let me know, and I will dig up
          one of those emails, and send it to you. It might make your early days
          a bit more efficient.

          Jon
        • Peter Schouten
          Jon Stubbs wrote Tuesday, July 01, 2008 8:27 AM ... Many of the Kurzweil onboard analog sounds use both sampled SAW and SQUARE as well as internal DSP variants
          Message 4 of 12 , Jul 1, 2008
          • 0 Attachment
            Jon Stubbs wrote Tuesday, July 01, 2008 8:27 AM

            > If you are wanting to stack oscillators in octaves, you can
            > get some extra oscillators using vast that won't kill polyphony.
            > In program mode you can edit and select VAST algorithms that have
            > SAW's, or SQUARES that you can transpose and detune.

            Many of the Kurzweil onboard analog sounds use both sampled SAW and SQUARE
            as well as internal DSP variants in the same patch. What I'd do is
            eliminate the sampled ones and see if you can work with the remaining sound.

            Hit Edit, go to Keymap and set to "None." If it doesn't not make too much
            difference to the sound, I would do that with all layers. If it helps, maybe
            leave one of the layers in the original program alone.

            Jonathon wrote (Monday, June 30, 2008 10:28 PM)

            > I have a setup layering "Synbrass Sect" (a 2-layer
            > program) and "Analog Brazz" (a 3-layer program),
            > with 2 copies of each for doubled octaves.
            > Actually I have a few such setups.

            A 5 layer stack like this would not be a polyphony issue if you were using
            only DSP wave forms.

            The stack will in fact use very little polyphony if the basic program does
            not use samples, as per my suggestion. So I would go ahead and modify both
            "Synbrass Sect" and "Analog Brazz" and see how they sound.

            If the basic programs are strictly sample-based and don't use DSP waveforms,
            I'd go ahead and make me some that do. I have programmed lots of big,
            punchy synbrass sounds using just the DSP waveforms

            ~Peter Schouten
            Pyramid Sound Productions
            http://www.pyramid-sound.com
          • Jon Stubbs
            Unfortunately, Peter s suggestion does not work. A layer uses at least one voice of polyphony per note, or two if it is a stereo keymap (or in stereo-mode).
            Message 5 of 12 , Jul 1, 2008
            • 0 Attachment
              Unfortunately, Peter's suggestion does not work.

              A layer uses at least one voice of polyphony per note,
              or two if it is a stereo keymap (or in stereo-mode).

              Turning off the keymap (i.e., setting the keymap to zero or
              not found) will not prevent this. The "voice" is the combination
              of sample-playback and the VAST DSP chain. If you are wanting to
              stack, detune or make intervals out of oscillators, then keep
              that keymap and mix it with the DSP-based oscillators.

              It is true that using a stereo keymap (with the keymap mode
              in stereo) uses an extra voice, that you might want for other
              purposes.





              >Many of the Kurzweil onboard analog sounds use both sampled SAW and SQUARE
              >as well as internal DSP variants in the same patch. What I'd do is
              >eliminate the sampled ones and see if you can work with the remaining sound.
              >
              >Hit Edit, go to Keymap and set to "None." If it doesn't not make too much
              >difference to the sound, I would do that with all layers. If it helps, maybe
              >leave one of the layers in the original program alone.
            • poser_p
              ... A few weeks ago I rewrote the K2000 entry on Wikipedia, adding a bit of an explanation of VAST: http://en.wikipedia.org/wiki/Kurzweil_K2000 My main goal in
              Message 6 of 12 , Jul 1, 2008
              • 0 Attachment
                --- In KurzList@yahoogroups.com, Jon Stubbs <jonstubbs@...> wrote:
                > I have posted an overview of the objects and structure on this list
                > several times. If you are interested, let me know, and I will dig up
                > one of those emails, and send it to you. It might make your early days
                > a bit more efficient.
                >
                > Jon
                >

                A few weeks ago I rewrote the K2000 entry on Wikipedia, adding a bit
                of an explanation of VAST:

                http://en.wikipedia.org/wiki/Kurzweil_K2000

                My main goal in writing was to give an overview of VAST and describe
                some of its major features. It would be good if someone else could
                read over it and correct errors or fix wording that they find unclear
                (or rewrite the whole thing).

                Of course similar explanations could be added for setups, the
                sequencer, sampling and sample processing, and so forth.

                -Andrew-
              • Marsiglio, Clifford C
                Why did you list the 2500 - the PC3 on the wiki site under K2000 Models? Just curious :-) Because they ain t! clif ... A few weeks ago I rewrote the K2000
                Message 7 of 12 , Jul 1, 2008
                • 0 Attachment
                  Why did you list the 2500 - the PC3 on the wiki site under K2000 Models?

                  Just curious :-) Because they ain't!

                  clif


                  On 7/1/08 6:33 PM, "poser_p" <poserp@...> wrote:

                  --- In KurzList@yahoogroups.com, Jon Stubbs <jonstubbs@...> wrote:
                  > I have posted an overview of the objects and structure on this list
                  > several times. If you are interested, let me know, and I will dig up
                  > one of those emails, and send it to you. It might make your early days
                  > a bit more efficient.
                  >
                  > Jon
                  >

                  A few weeks ago I rewrote the K2000 entry on Wikipedia, adding a bit
                  of an explanation of VAST:

                  http://en.wikipedia.org/wiki/Kurzweil_K2000

                  My main goal in writing was to give an overview of VAST and describe
                  some of its major features. It would be good if someone else could
                  read over it and correct errors or fix wording that they find unclear
                  (or rewrite the whole thing).

                  Of course similar explanations could be added for setups, the
                  sequencer, sampling and sample processing, and so forth.

                  -Andrew-



                  ------------------------------------

                  To QUIT KurzList, send a *blank* message to kurzlist-unsubscribe@yahoogroups.com
                  It will take from 24 to 48 hours for Yahoo!Groups to cancel you.Yahoo! Groups Links





                  Clifford C. Marsiglio
                  Manager of Development and Outreach Activities
                  The Testing Center
                  Indiana University-Purdue University Indianapolis
                  620 Union Dr. G003
                  Indianapolis, IN 46202
                  phone: (317) 274-2897; fax: (317) 274-3400
                  email: ccmarsig@...
                  website: http://tc.iupui.edu/
                • Peter Schouten
                  Hi Jon, You wrote...... ... I believe it would be more accurate to say that it s one voice of polyphony per DSP layer. As for stereo playback, what is the
                  Message 8 of 12 , Jul 1, 2008
                  • 0 Attachment
                    Hi Jon,

                    You wrote......

                    > A layer uses at least one voice of polyphony per note,
                    > or two if it is a stereo keymap (or in stereo-mode).

                    I believe it would be more accurate to say that it's one voice of polyphony
                    per DSP layer. As for stereo playback, what is the sound of NONE in stereo
                    mode? Is there any reason to believe the Stereo On/Off switch on the Keymap
                    page will have any effect on the sound..... ;-)


                    > The "voice" is the combination of sample-playback and the VAST DSP chain.

                    There is no sample playback when the voice is based only on DSP waveforms as
                    your sound source. That was the point I was making when I suggested
                    eliminating all or most sample playback for the synbrass sounds.

                    A 5-layer setup with one voice of polyphony per DSP layer (no
                    sample-playback) would add up to 5 voices. That would actually leave some
                    voices free.

                    Hope that helps,

                    ~Peter Schouten
                    Pyramid Sound Productions
                    http://www.pyramid-sound.com
                  • Jon Stubbs
                    Hi Peter! Your last email made it sound like getting rid of the samples by selecting none keymap would have some effect on polyphony; which it will not. ...
                    Message 9 of 12 , Jul 1, 2008
                    • 0 Attachment
                      Hi Peter!

                      Your last email made it sound like getting rid of the
                      samples by selecting "none" keymap would have some effect
                      on polyphony; which it will not.

                      >A 5-layer setup with one voice of polyphony per DSP layer (no
                      >sample-playback) would add up to 5 voices. That would actually leave some
                      >voices free.

                      Any combination of programs adding up to 5-layers will use 5 voices
                      per note no matter whether samples are playing or not.

                      >> A layer uses at least one voice of polyphony per note,
                      >> or two if it is a stereo keymap (or in stereo-mode).
                      >
                      >I believe it would be more accurate to say that it's one voice of polyphony
                      >per DSP layer. As for stereo playback, what is the sound of NONE in stereo
                      >mode? Is there any reason to believe the Stereo On/Off switch on the Keymap
                      >page will have any effect on the sound..... ;-)

                      It WILL have an effect on the polyphony. If the keymap is in Stereo Mode,
                      then even if "None" or "Not Found" is chosen for the both of the stereo
                      keymaps, it will still use 2 voices of polyphony.

                      So it is more accurate to say a layer can use one or two voices of
                      polyphony per note depending on the keymap mode being mono or stereo.
                    • Peter Schouten
                      Hello again Jon, You wrote .... ... Jon, just curious, what was your rationale when you suggested this: You can edit and select VAST algorithms that have
                      Message 10 of 12 , Jul 1, 2008
                      • 0 Attachment
                        Hello again Jon,

                        You wrote ....

                        > Your last email made it sound like getting rid of
                        > the samples by selecting "none" keymap would
                        > have some effect on polyphony; which it will not.

                        Jon, just curious, what was your rationale when you suggested this: "You can
                        edit and select VAST algorithms that have Saw's or SQUARES that you can
                        transpose and detune." (Tuesday, July 01, 2008 8:27 AM). Your suggestion
                        would seem to favor adding DSP Saws or Squares to programs that have samples
                        in the Keymaps -- which you'd keep there because deselecting them presumably
                        won't help. Isn't that kinda like bringing the player who is already
                        approaching poly limits to his knees even sooner? ;-)


                        > (The Stereo On/Off switch on the Keymap page) WILL have an
                        > effect on the polyphony. If the keymap is in Stereo Mode, then
                        > even if "None" is chosen for the both of the stereo keymaps,
                        > it will still use 2 voices of polyphony.

                        Again, Jon, why would anyone want stereo playback for mono DSP waveforms? I
                        don't understand where you're going with this.

                        There are various other ways of dealing with the polyphony issues that can
                        be used in combination:
                        http://www.sonivoxmi.com/SupportKurzweilVast.htm

                        The above suggestions relate to drum sounds, but they apply to most any
                        polyphonic situation.

                        ~Peter Schouten
                        Pyramid Sound Productions
                        http://www.pyramid-sound.com


                        .
                      • Jon Stubbs
                        ... Thanks, Peter. I should have specified to use +saw and +square , etc. As those blocks with the plus sign will mix the waveform with the sample. It
                        Message 11 of 12 , Jul 2, 2008
                        • 0 Attachment
                          >Jon, just curious, what was your rationale when you suggested this: "You can
                          >edit and select VAST algorithms that have Saw's or SQUARES that you can
                          >transpose and detune."

                          Thanks, Peter. I should have specified to use " +saw" and " +square",
                          etc. As those blocks with the "plus" sign will mix the waveform
                          with the sample.

                          It is a fairly simple matter to have the keymap samples mixed with
                          two other independently tunable oscillators (in DSP blocks) and
                          a fixed resonance filter. So an easy 3 oscillators with filter
                          taking only one voice per note.

                          >again, Jon, why would anyone want stereo playback for mono DSP waveforms? I
                          >don't understand where you're going with this.

                          I wasn't recommending stereo. I was just being accurate & thorough.
                          The number of voices used by a layer is one-voice-per-note, unless
                          stereo is enabled on the keymap page, then it is two-voices-per-note.

                          KB3 mode, like the vocoder mode, are exceptions in that they
                          are using the available processing power & polyphony in a different way.
                          So they don't count in the above rule.

                          Jon
                        • poser_p
                          Just for reference to show the evolution of the product line. I think a bit of the K2000 as a staring point in a long chain of similar products (to my mind
                          Message 12 of 12 , Jul 2, 2008
                          • 0 Attachment
                            Just for reference to show the evolution of the product line. I think
                            a bit of the K2000 as a "staring point" in a long chain of similar
                            products (to my mind the K2000 is radically different from the K1000,
                            while successive KXXX and PC series synths all use VAST and other
                            things that started w/the K2000). Now that I think of it, though,
                            there should be a note w/the PC3/X that it doesn't do sampling.
                            Otherwise it appears to build on/expand the VAST concept.

                            -Andrew Plewe-

                            --- In KurzList@yahoogroups.com, "Marsiglio, Clifford C"
                            <ccmarsig@...> wrote:
                            >
                            > Why did you list the 2500 - the PC3 on the wiki site under K2000 Models?
                            >
                            > Just curious :-) Because they ain't!
                            >
                            > clif
                            >
                            >
                            > On 7/1/08 6:33 PM, "poser_p" <poserp@...> wrote:
                            >
                            > --- In KurzList@yahoogroups.com, Jon Stubbs <jonstubbs@> wrote:
                            > > I have posted an overview of the objects and structure on this list
                            > > several times. If you are interested, let me know, and I will dig up
                            > > one of those emails, and send it to you. It might make your early days
                            > > a bit more efficient.
                            > >
                            > > Jon
                            > >
                            >
                            > A few weeks ago I rewrote the K2000 entry on Wikipedia, adding a bit
                            > of an explanation of VAST:
                            >
                            > http://en.wikipedia.org/wiki/Kurzweil_K2000
                            >
                            > My main goal in writing was to give an overview of VAST and describe
                            > some of its major features. It would be good if someone else could
                            > read over it and correct errors or fix wording that they find unclear
                            > (or rewrite the whole thing).
                            >
                            > Of course similar explanations could be added for setups, the
                            > sequencer, sampling and sample processing, and so forth.
                            >
                            > -Andrew-
                            >
                            >
                            >
                            > ------------------------------------
                            >
                            > To QUIT KurzList, send a *blank* message to
                            kurzlist-unsubscribe@yahoogroups.com
                            > It will take from 24 to 48 hours for Yahoo!Groups to cancel
                            you.Yahoo! Groups Links
                            >
                            >
                            >
                            >
                            >
                            > Clifford C. Marsiglio
                            > Manager of Development and Outreach Activities
                            > The Testing Center
                            > Indiana University-Purdue University Indianapolis
                            > 620 Union Dr. G003
                            > Indianapolis, IN 46202
                            > phone: (317) 274-2897; fax: (317) 274-3400
                            > email: ccmarsig@...
                            > website: http://tc.iupui.edu/
                            >
                          Your message has been successfully submitted and would be delivered to recipients shortly.