The Official Sequential/DSI Forum

Differences in filter tracking

Differences in filter tracking
« on: June 01, 2021, 01:12:50 AM »
Does anyone else notice this?

My Rev 1/2 filter seems to track all the way up the keyboard but the Rev 3 is off on the top 19 keys.

Re: Differences in filter tracking
« Reply #1 on: June 01, 2021, 06:58:55 AM »
Maybe depends on how much the filter is open to begin with ? 

Cheers !
1976 MiniKORG700s // 1978 Prophet-5 rev.2 // 1981 KORG CX-3 // 1984 DX7 // 2020 Prophet-10 rev.4 // MoPho Box // 2 Creamware MiniMax // Creamware Pro-12 // 2 EMU-Proteus 2000 // EMU-Vintage Keys  // Casio VZ-10M // Roland VK-8M // Fatar SL 880 //

Re: Differences in filter tracking
« Reply #2 on: June 01, 2021, 11:31:02 AM »
Maybe depends on how much the filter is open to begin with ? 

But then that would have to also happen when the 1/2 filters are engaged.

@Quatschmacher - Let me check mine - Any particular patch that pronounces this best ?

Re: Differences in filter tracking
« Reply #3 on: June 01, 2021, 12:54:54 PM »
No patch in particular. I simply turn resonance to maximum, put key tracking on full and tune the filter to either a key or to the reference pitch. If you mix on a little oscillator volume, you can hear when the tracking goes out.

I get that it might be to do with the difference between V/octave and equal temperament tuning. Just curious why one filter is dead on and the other not.

Re: Differences in filter tracking
« Reply #4 on: June 01, 2021, 01:17:52 PM »
On the other hand, if there was no difference the 1/2 and 3 filter options
would be rather pointless .......
They mimic the behavior of Rev.1 / 2  vs  Rev.3 according to the man who
made them in the first place, so I think they do what they supposed to do ....

If Rev.3 filter does not do what you want it to do, stick to Rev 1 / 2 filters    ;)


Cheers !
1976 MiniKORG700s // 1978 Prophet-5 rev.2 // 1981 KORG CX-3 // 1984 DX7 // 2020 Prophet-10 rev.4 // MoPho Box // 2 Creamware MiniMax // Creamware Pro-12 // 2 EMU-Proteus 2000 // EMU-Vintage Keys  // Casio VZ-10M // Roland VK-8M // Fatar SL 880 //

Re: Differences in filter tracking
« Reply #5 on: June 02, 2021, 01:06:15 AM »
No patch in particular. I simply turn resonance to maximum, put key tracking on full and tune the filter to either a key or to the reference pitch. If you mix on a little oscillator volume, you can hear when the tracking goes out.

I get that it might be to do with the difference between V/octave and equal temperament tuning. Just curious why one filter is dead on and the other not.
Well you stumped me on this one. I for some reason can not recreate this issue. Or I am not knowing what to do/listen for. LOL But it seems kinda self explanatory and I followed your instructions to a tee. The 1/2 Filter is a little more slow but more pronounced than the 3 filter. But neither seems to go dead at the top 19.

By tracking, you mean the keyboard tracking thing right ? Where the filter opens up more and more as you climb the keys ? In turn kinda gets louder as a byproduct on the settings you said (max res), cutoff tuned to A ref.

Re: Differences in filter tracking
« Reply #6 on: June 02, 2021, 03:34:32 AM »
No patch in particular. I simply turn resonance to maximum, put key tracking on full and tune the filter to either a key or to the reference pitch. If you mix on a little oscillator volume, you can hear when the tracking goes out.

I get that it might be to do with the difference between V/octave and equal temperament tuning. Just curious why one filter is dead on and the other not.
Well you stumped me on this one. I for some reason can not recreate this issue. Or I am not knowing what to do/listen for. LOL But it seems kinda self explanatory and I followed your instructions to a tee. The 1/2 Filter is a little more slow but more pronounced than the 3 filter. But neither seems to go dead at the top 19.

By tracking, you mean the keyboard tracking thing right ? Where the filter opens up more and more as you climb the keys ? In turn kinda gets louder as a byproduct on the settings you said (max res), cutoff tuned to A ref.

The issue is that on the top 19 keys, the tuned filter is out of tune with the oscillators on one filter revision but not the other.

Iím simply trying to ascertain whether this is expected behaviour or not. I know the envelope response between the two filters differs but not sure that scaling ought to differ.

I guess Iíll likely email support about it to get a definitive answer.

Re: Differences in filter tracking
« Reply #7 on: June 02, 2021, 03:38:03 AM »
On the other hand, if there was no difference the 1/2 and 3 filter options
would be rather pointless .......
They mimic the behavior of Rev.1 / 2  vs  Rev.3 according to the man who
made them in the first place, so I think they do what they supposed to do ....

If Rev.3 filter does not do what you want it to do, stick to Rev 1 / 2 filters    ;)


Cheers !

I certainly know that their envelope responses differ by design; I donít recall reading anything about differences in their ability to track the keyboard across the five-octave span so wanted to check if other users notice similar behaviour. Does your unit do this?

Re: Differences in filter tracking
« Reply #8 on: June 02, 2021, 04:53:09 AM »
Tried a little, and switching from rev 1/2 to rev3 went out of tune, but after adjusting
cutoff knob it worked well over the whole keybed .......
On some sounds it went off a little in the high register, but not on all sounds ......
(Perhaps my P-10 was not warmed up enough  :o   )

Might depend on how much "detuned" OSC 1 and 2 is, or if there is any "Pulse With" applied ???
Or "Vintage" knob position .....    Have no clue, and itīs not that important to me.
If I ever need some sound like that Iīll use Rev. 1/2 mode    :)

Cheers !
1976 MiniKORG700s // 1978 Prophet-5 rev.2 // 1981 KORG CX-3 // 1984 DX7 // 2020 Prophet-10 rev.4 // MoPho Box // 2 Creamware MiniMax // Creamware Pro-12 // 2 EMU-Proteus 2000 // EMU-Vintage Keys  // Casio VZ-10M // Roland VK-8M // Fatar SL 880 //

Re: Differences in filter tracking
« Reply #9 on: June 02, 2021, 05:18:34 AM »
Tried a little, and switching from rev 1/2 to rev3 went out of tune, but after adjusting
cutoff knob it worked well over the whole keybed .......
On some sounds it went off a little in the high register, but not on all sounds ......
(Perhaps my P-10 was not warmed up enough  :o   )

Might depend on how much "detuned" OSC 1 and 2 is, or if there is any "Pulse With" applied ???
Or "Vintage" knob position .....    Have no clue, and itīs not that important to me.
If I ever need some sound like that Iīll use Rev. 1/2 mode    :)

Cheers !

Oscillators were tuned precisely, Vintage knob set to 4, sawtooth only (so pulse width not an issue). Iím using this to create some organ patches and it works fine on 1/2 but not on 3.

Re: Differences in filter tracking
« Reply #10 on: June 02, 2021, 01:14:12 PM »
The issue is that on the top 19 keys, the tuned filter is out of tune with the oscillators on one filter revision but not the other.
Oh.. out of tune - I swear you said "dead" (sorry I somehow misread that part, as you said "dead on" - my bad man, I been working a lot on projects around my house and when I log on and even mess around on my synths lately, I am just not 100%)

- I was listening for them to disappear. At same time, when something is out of tune I should hear that too. BUT I did what HockeBocke said he did, but I started out doing that (just by default). I used the cutoff filter to tune the filter resonance sound to A ref. I thought that is what you meant.

Maybe, just maybe me doing that caused me to not hear the out of tune thing as HockeBocke said using cutoff filter fixed the issue for him ??? Just maybe ??

IDK, now that I understand more, I am going to test again later when I got time (painting my game room now/soon)

Re: Differences in filter tracking
« Reply #11 on: June 03, 2021, 12:51:49 AM »
Ok, I have verified this myself. With all settings remaining the same. There is a definitely tracking issue with the upper register in the rev 3 filter vs the 1/2 filter. Very weird. This is beyond just difference in characteristics. You can hear that too within the keys that sound good between them both, but once you hit the high register, all hell breaks loose on the rev3 filter option and it sounds out of tune, ugly, etc..

I'd say this is official, unless they did this on purpose. I never owned a vintage rev3, so I can not say if thats its normal character. That said, this issue only rears its head up when the settings are as Quatschmacher said - Res maxed is the main thing.

LPF83

  • ***
  • 526
Re: Differences in filter tracking
« Reply #12 on: June 03, 2021, 04:51:46 PM »
Ok, I have verified this myself. With all settings remaining the same. There is a definitely tracking issue with the upper register in the rev 3 filter vs the 1/2 filter. Very weird. This is beyond just difference in characteristics. You can hear that too within the keys that sound good between them both, but once you hit the high register, all hell breaks loose on the rev3 filter option and it sounds out of tune, ugly, etc..

I'd say this is official, unless they did this on purpose. I never owned a vintage rev3, so I can not say if thats its normal character. That said, this issue only rears its head up when the settings are as Quatschmacher said - Res maxed is the main thing.

I just did some tests on mine,  I felt it tracked as expected across the range of keys on both filter types.  It can get a bit metallic and unappealing sounding on the far upper keys, but I didn't feel the filter tune was off.

Does it behave the same whether you have it in P5 voice allocation vs RR?   I tried it on the Prophet6 and was surprised at the havoc slight voice tuning differences make in this particular test (my P10 is in P5 voice mode currently, how I normally play it).
Prophet 10, Prophet 6, OB-6, Prophet 12m, Prophet Rev2-16, Moog SlimPhatty, Hydrasynth desktop, Korg Minilogue XDm, Roland SPD-30, Roland SPD-SX Special Edition, Roland KT-10, Maschine, Focusrite Scarlett 18i20 3rd Gen + Octopre, Strymon Pedals, Cubase Pro 11.

Re: Differences in filter tracking
« Reply #13 on: June 03, 2021, 06:09:05 PM »
I just did some tests on mine,  I felt it tracked as expected across the range of keys on both filter types.  It can get a bit metallic and unappealing sounding on the far upper keys, but I didn't feel the filter tune was off.

Does it behave the same whether you have it in P5 voice allocation vs RR?   I tried it on the Prophet6 and was surprised at the havoc slight voice tuning differences make in this particular test (my P10 is in P5 voice mode currently, how I normally play it).
You might have discovered it does not do it when in 5voice mode. I had mine in 10voice mode. Here is what it sounds like if dialed in correctly to hear it and also in 10voice mode.

https://youtu.be/WFvLBotudhs

The video does not explain much as I try to keep it short, but it is dialed in like this, Filter Res: Max, Envelope minimal, Filter Cutoff dialed in to make "A" key be in the key of "A" (approx 10o'clock'ish). Osc B get a tone up, make in tune to filter tune and loud enough to just begin to hear it enough so that if the layering filter tune goes out of pitch, you'll hear it.

Then I just toggle between filters 1/2 and 3. And that is the difference. IMO, that's not normal. And if it is, it shouldn't be. lol.

LPF83

  • ***
  • 526
Re: Differences in filter tracking
« Reply #14 on: June 03, 2021, 07:39:13 PM »
I just did some tests on mine,  I felt it tracked as expected across the range of keys on both filter types.  It can get a bit metallic and unappealing sounding on the far upper keys, but I didn't feel the filter tune was off.

Does it behave the same whether you have it in P5 voice allocation vs RR?   I tried it on the Prophet6 and was surprised at the havoc slight voice tuning differences make in this particular test (my P10 is in P5 voice mode currently, how I normally play it).
You might have discovered it does not do it when in 5voice mode. I had mine in 10voice mode. Here is what it sounds like if dialed in correctly to hear it and also in 10voice mode.

https://youtu.be/WFvLBotudhs

The video does not explain much as I try to keep it short, but it is dialed in like this, Filter Res: Max, Envelope minimal, Filter Cutoff dialed in to make "A" key be in the key of "A" (approx 10o'clock'ish). Osc B get a tone up, make in tune to filter tune and loud enough to just begin to hear it enough so that if the layering filter tune goes out of pitch, you'll hear it.

Then I just toggle between filters 1/2 and 3. And that is the difference. IMO, that's not normal. And if it is, it shouldn't be. lol.

I might try more tests tomorrow and try to mirror your settings.  Wanted to mention though, I wasn't referring to 5 voice versus 10 voice mode, I'm referring to P5 mode versus round robin voice allocation.   Round robin is like the P6, tapping the same key plays a different voice with each successive tap, versus classic P5 where tapping the same key plays the same voice.
Prophet 10, Prophet 6, OB-6, Prophet 12m, Prophet Rev2-16, Moog SlimPhatty, Hydrasynth desktop, Korg Minilogue XDm, Roland SPD-30, Roland SPD-SX Special Edition, Roland KT-10, Maschine, Focusrite Scarlett 18i20 3rd Gen + Octopre, Strymon Pedals, Cubase Pro 11.

Re: Differences in filter tracking
« Reply #15 on: June 03, 2021, 09:00:30 PM »
Wanted to mention though, I wasn't referring to 5 voice versus 10 voice mode, I'm referring to P5 mode versus round robin voice allocation.   Round robin is like the P6, tapping the same key plays a different voice with each successive tap, versus classic P5 where tapping the same key plays the same voice.
Dang, my bad brother.. You are right. lol I have no clue why I concluded that sentence wrong. Yes I am familiar with the RR. My Oberheim OB-Xa and Matrix 12 does that.

But like you, I had mine in the P5 Voice Allocation as well. Also, 10 voice. I will try 5 voice, and also RR and combinations thereof to see if any make a difference.

Thanks for clearing that up btw.

Re: Differences in filter tracking
« Reply #16 on: June 04, 2021, 12:19:15 AM »
Round robin causes noticeable differences - some voices are in tune with the filter others are not.

I switched between revisions on a held note and the difference at the upper end was that the rev3 filter can do a semitone fiat. I need to do this test on each repeated note when playing round robin.

LPF83

  • ***
  • 526
Re: Differences in filter tracking
« Reply #17 on: June 04, 2021, 03:10:09 PM »
Wanted to mention though, I wasn't referring to 5 voice versus 10 voice mode, I'm referring to P5 mode versus round robin voice allocation.   Round robin is like the P6, tapping the same key plays a different voice with each successive tap, versus classic P5 where tapping the same key plays the same voice.
Dang, my bad brother.. You are right. lol I have no clue why I concluded that sentence wrong. Yes I am familiar with the RR. My Oberheim OB-Xa and Matrix 12 does that.

But like you, I had mine in the P5 Voice Allocation as well. Also, 10 voice. I will try 5 voice, and also RR and combinations thereof to see if any make a difference.

Thanks for clearing that up btw.

I just tried the settings you've described and could not reproduce the sound in the video.  Tuning stays consistent across the key range assuming envelope amount is at 0 (I think that's what you meant by minimal, but once its anything but zero then naturally the filter starts to go a little nuts as that's going to affect the quantization).

Speaking of which, as I was doing these tests, the whole reason Dave likes to quantize the filter really hit home with me.  Even though the stepping is a tradeoff, the tones coming out of that filter are so sweet it was hard to stay focused on the test and not start writing musical hooks with it.  I think that has always been Dave's thing and why the Prophet 5 found itself into so much music, the sheer versatility of what could be done with so few controls.  The way that filter can sound  is amazing.

If your envelope amount is zero in that video, I'd recommend contacting support to see what they say.  The symptoms sound a little like keyboard tracking isn't really on full (though I assume it is in your test).
« Last Edit: June 04, 2021, 03:25:46 PM by LPF83 »
Prophet 10, Prophet 6, OB-6, Prophet 12m, Prophet Rev2-16, Moog SlimPhatty, Hydrasynth desktop, Korg Minilogue XDm, Roland SPD-30, Roland SPD-SX Special Edition, Roland KT-10, Maschine, Focusrite Scarlett 18i20 3rd Gen + Octopre, Strymon Pedals, Cubase Pro 11.

Re: Differences in filter tracking
« Reply #18 on: June 04, 2021, 04:34:03 PM »
Watch these. I got more technical -Killed the Osc, so Filter is solo'ed, Used a tuner, also tested in all settings 5voice, 10 voice, P5&RR allocation - all equals same thing. Subtle differences of course, but all equals same result = Rev.3 filters do not track good in the upper registers.

https://youtu.be/ZfnGGDwDQPU
https://youtu.be/vHqelFZE49A
https://youtu.be/Ob8Zb0Bb3Wk

Re: Differences in filter tracking
« Reply #19 on: June 05, 2021, 07:16:10 AM »
Yes, envelope amount was set to zero and filter tracking on full. I have a support ticket open but seem to being told itís normal, though I have now sent links to Infraredís videos as further demonstration.