| Author |
Message |
Burp
Joined: Dec 18, 2006
Posts: 71
Location: NL
G2 patch files: 7
|
|
|
Back to top
|
|
 |
iPassenger
Joined: Jan 27, 2007
Posts: 1068
Location: Sheffield, UK
Audio files: 5
G2 patch files: 78
|
 Posted: Thu Sep 17, 2009 6:06 am Post subject:
|
  |
|
I've not looked at your patch (I'm at work), I'm not sure I would understand it if I did but could it be related to the higher sampling freq of audio signals? Doesn't any signal sent through a bus get converted to an audio signal. Not sure how that could create distortion but thought it may be related.
Do signals sent through a bus get DC filtered too? Like the main outputs do. Could this not also introduce some variance? Where's Tim?
EDIT:
| Burp wrote: |
When sending the ctrl signal directly to a OSC (pitch) it all seems ok
|
This would suggest I am wrong tho.
As a test to establish if the problem is the BUS/FX connection. Could you not make a syncable OSC in both the poly area and the FX area. Have both of them sync from the Status module's patch active. Have an lfo in the poly area and route the ctrl signal into the FX area, route the audio from the poly area also into the fx area and invert it. In theory if you mix the two osc signals they should cancel and any errors you here would be due to either calculation order (should be a constant signal) or the BUS/FX connection. - Just a thought.
_________________
iP (Ross)
- http://ipassenger.bandcamp.com
- http://soundcloud.com/ipassenger |
|
|
Back to top
|
|
|
Burp
Joined: Dec 18, 2006
Posts: 71
Location: NL
G2 patch files: 7
|
| Posted: Thu Sep 17, 2009 8:13 am Post subject:
|
|
|
Hi iPassenger, thanks for your suggestions.
I'm not near my G2 too at this moment, but your test idea is a good one. I'll try it when i've got the time. (hopefully somewhere this weekend).
_________________
if you don't know where you're going, you will allways arrive |
|
|
Back to top
|
|
|
Burp
Joined: Dec 18, 2006
Posts: 71
Location: NL
G2 patch files: 7
|
|
|
Back to top
|
|
|
Tim Kleinert
Joined: Mar 12, 2004
Posts: 1148
Location: Zürich, Switzerland
Audio files: 7
G2 patch files: 236
|
| Posted: Mon Sep 28, 2009 1:14 pm Post subject:
|
|
|
| iPassenger wrote: | | Where's Tim? |
Here. 
I haven't got the time right now to debug the patches in depth.
But what I can already say is that it is problematic to drive a delay readout pointer with a controlrate source to generate sample-pitching, as this only clocks at 24kHz and will cause audible signal degradation.
The reason why the version with the filtered-out carrier signal works, is that the controlrate signal is heavily lowpass-filtered below 24kHz and converted to audiorate. In this way, the 24kHz controlrate signal (which intermodulates with the 96kHz audiorate and thus causes artefacts) is interpolated (smoothed out) at audiorate, and sounds much better. The problem here is that this interpolation will also smooth out the sharp flank of the sawtooth LFO, which will cause a glitch when used for sample-pitching. Audiorate DIY sawtooth readout drivers is the only solution.
Another thing I noticed is that the 4-tap delay has been replaced by a 2-tap delay. Don't do this. Only the 4-tap delay has a "direct" output that is not modulatable. The G2 delays suck, as all the modulatable tap outputs increasingly degrade the signal (probably through buggy interpolation) when used for looping. Also, I cannot guarantee that my sampling circuit will not drift over time when hooked up to a modulatable delay tap, even when it's at maximum I've never tested this, but I don't trust the delay modules at all, as they are rather shitty. That's why I've spent so much time with them 
cheers,
t |
|
|
Back to top
|
|
|
Burp
Joined: Dec 18, 2006
Posts: 71
Location: NL
G2 patch files: 7
|
| Posted: Tue Sep 29, 2009 1:58 am Post subject:
|
|
|
Hi Tim,
Thanks for your input.
Mmm the delays are indeed more shitty than i thought  thanks for pointing out some problems. I'll replace the 2-tap delay with the 4-tap and have to re-think my patch.
I noticed indeed the glitch on the sharp LFO flanks you mentioned, but only when the filtering of the control signal is done at relatively low frequencies, that's why i used such a steep filter. It's not very nice, just a pragmatic workaround
Conclusion:
- There is no bug in sending low frequencies control signals through a bus.
- The delays are more shitty than i thought: see Tim's comments above.
_________________
if you don't know where you're going, you will allways arrive |
|
|
Back to top
|
|
|
|
Forum index » Clavia Nord Modular » Bugs
FAQ
Calendar
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Chat Room
