Author |
Message |
Tim Kleinert
Joined: Mar 12, 2004 Posts: 1148 Location: Zürich, Switzerland
Audio files: 7
G2 patch files: 236
|
Posted: Wed Mar 22, 2006 10:06 am Post subject:
Please verify this bug Subject description: notorious delay line weirdness possibly detected! |
 |
|
IMPORTANT UPDATE: I found a robust workaround for this patch described lower in this thread.
Direct link:
http://www.electro-music.com/forum/post-409316.html#409316
As some of you might know, the delay lines have been bothering me since I bought my G2s two years ago.
There was always something strange going on with them when using them for high accuracy stuff like physical modelling.
For a long time I believed that the 16 bit dynamic depth was the culprit, but somehow that didn't scan right. The artefacts I was repeatedly confronted with sounded very different from the kind of typical harmonic distortions you get from low dynamic resolution. And anyhow, 16 bits would still be good enough. Proof is the CombfFilter module, which is based on the same 16 bit memory, but doesn't produce those artefacts and sounds much much better for physical modelling applications.
But using the comb filter module all the time is restrictive in its own right, as it uses considerable more DSP and only offers logarithmic delay length response (very convenient for some purposes, less so for others).
I'm currently working on a patch similar to AAS string studio. It was initially based on comb filters, but I started breaking it down to delays (for DSP/polyphony reasons) and ran into those same old gremlins again. But I think I got them this time.
Please try out the attached patch. If you get what I got, then you should hear a strange cyclic rustling coming out of delay line 2. But please notice that no audio is running into delay line 2! There is, however, audio running into delay line 1 (whose output isn't going anywhere). Turn off audio into delay 1, and delay 2 becomes silent too.
Or, turn off constant modulation value going into the time mod input of delay 2, and it also becomes silent.
I suspect that something funky is going on with the readout pointers, causing delay 2 to sometimes fetch a few samples of what is in delay 1.
Could you please verify this.
thanks,
tim
Description: |
notorious delay line weirdness possibly detected! |
|
 Download (listen) |
Filename: |
Bugreport 220506.pch2 |
Filesize: |
1.35 KB |
Downloaded: |
2156 Time(s) |
Last edited by Tim Kleinert on Sat Mar 28, 2015 10:17 am; edited 2 times in total |
|
Back to top
|
|
 |
mosc
Site Admin

Joined: Jan 31, 2003 Posts: 18235 Location: Durham, NC
Audio files: 222
G2 patch files: 60
|
Posted: Wed Mar 22, 2006 10:12 am Post subject:
|
 |
|
Yep, I think you have certainly identified a problem. My G2X V1.4 does exactly what you say it does.
I moved this to the Bugs forum...  _________________ --Howard
my music and other stuff |
|
Back to top
|
|
 |
Tim Kleinert
Joined: Mar 12, 2004 Posts: 1148 Location: Zürich, Switzerland
Audio files: 7
G2 patch files: 236
|
Posted: Wed Mar 22, 2006 10:16 am Post subject:
|
 |
|
Yeah, even the demo version does it.
So, good that you moved it to the bugs section, where it belongs now.
Ha! This is a personal victory for me. ! FINALLY, I found what was going on all the time! Ah, the relief.
But -will Clavia ever fix it?  |
|
Back to top
|
|
 |
cebec

Joined: Apr 19, 2004 Posts: 1098 Location: Virginia
Audio files: 3
G2 patch files: 31
|
Posted: Wed Mar 22, 2006 10:17 am Post subject:
|
 |
|
i can confirm it on the Demo, too...
i would e-mail Clavia, and link them to this thread, perhaps... |
|
Back to top
|
|
 |
Tim Kleinert
Joined: Mar 12, 2004 Posts: 1148 Location: Zürich, Switzerland
Audio files: 7
G2 patch files: 236
|
Posted: Wed Mar 22, 2006 10:54 am Post subject:
|
 |
|
Well, at least I can go on with the patching now, knowing that it's not my fault.
I'll have to live with the artefacts though, until Clavia comes up with a fix.  |
|
Back to top
|
|
 |
Chet

Joined: Nov 19, 2004 Posts: 231 Location: Lititz,PA,USA
Audio files: 7
G2 patch files: 35
|
Posted: Wed Mar 22, 2006 11:06 am Post subject:
|
 |
|
That's excellent detective work, Tim. Clavia should be able to track it down pretty easily now. |
|
Back to top
|
|
 |
v-un-v
Janitor


Joined: May 16, 2005 Posts: 8933 Location: Birmingham, England, UK
Audio files: 11
G2 patch files: 1
|
Posted: Wed Mar 22, 2006 11:37 am Post subject:
|
 |
|
Getting the same sound on the demo version here too.
Actually I quite like this sound- it reminds me of the noises/ feeling that you get when you stand under major electricity grid overhead powerlines!!
There are many quirks in hardware modulars and some would even argue that these quirks add character.
This has always been my complaint with the digital stuff- that it's too perfect.
Oh well- well spotted anyway  _________________ ACHTUNG!
ALLES TURISTEN UND NONTEKNISCHEN LOOKENPEEPERS!
DAS KOMPUTERMASCHINE IST NICHT FÜR DER GEFINGERPOKEN UND MITTENGRABEN! ODERWISE IST EASY TO SCHNAPPEN DER SPRINGENWERK, BLOWENFUSEN UND POPPENCORKEN MIT SPITZENSPARKSEN.
IST NICHT FÜR GEWERKEN BEI DUMMKOPFEN. DER RUBBERNECKEN SIGHTSEEREN KEEPEN DAS COTTONPICKEN HÄNDER IN DAS POCKETS MUSS.
ZO RELAXEN UND WATSCHEN DER BLINKENLICHTEN. |
|
Back to top
|
|
 |
mosc
Site Admin

Joined: Jan 31, 2003 Posts: 18235 Location: Durham, NC
Audio files: 222
G2 patch files: 60
|
Posted: Wed Mar 22, 2006 11:40 am Post subject:
|
 |
|
v-un-v wrote: |
This has always been my complaint with the digital stuff- that it's too perfect. |
Yes, this is perfect. Perfectly awful. _________________ --Howard
my music and other stuff |
|
Back to top
|
|
 |
3phase

Joined: Jul 27, 2004 Posts: 1189 Location: Berlin
Audio files: 13
G2 patch files: 141
|
Posted: Wed Mar 22, 2006 12:34 pm Post subject:
|
 |
|
Chet wrote: | That's excellent detective work, Tim. Clavia should be able to track it down pretty easily now. |
In case somebody informs them... You cant expect them to scan the forum regulary.. |
|
Back to top
|
|
 |
3phase

Joined: Jul 27, 2004 Posts: 1189 Location: Berlin
Audio files: 13
G2 patch files: 141
|
Posted: Wed Mar 22, 2006 12:37 pm Post subject:
|
 |
|
v-un-v wrote: | .
This has always been my complaint with the digital stuff- that it's too perfect.
|
Its not perfect..its just simple... |
|
Back to top
|
|
 |
Tim Kleinert
Joined: Mar 12, 2004 Posts: 1148 Location: Zürich, Switzerland
Audio files: 7
G2 patch files: 236
|
Posted: Wed Mar 22, 2006 12:45 pm Post subject:
|
 |
|
I don't need the sound of "electricity grid overhead powerlines" in my waveguides, thank you.  |
|
Back to top
|
|
 |
v-un-v
Janitor


Joined: May 16, 2005 Posts: 8933 Location: Birmingham, England, UK
Audio files: 11
G2 patch files: 1
|
Posted: Wed Mar 22, 2006 1:41 pm Post subject:
|
 |
|
 _________________ ACHTUNG!
ALLES TURISTEN UND NONTEKNISCHEN LOOKENPEEPERS!
DAS KOMPUTERMASCHINE IST NICHT FÜR DER GEFINGERPOKEN UND MITTENGRABEN! ODERWISE IST EASY TO SCHNAPPEN DER SPRINGENWERK, BLOWENFUSEN UND POPPENCORKEN MIT SPITZENSPARKSEN.
IST NICHT FÜR GEWERKEN BEI DUMMKOPFEN. DER RUBBERNECKEN SIGHTSEEREN KEEPEN DAS COTTONPICKEN HÄNDER IN DAS POCKETS MUSS.
ZO RELAXEN UND WATSCHEN DER BLINKENLICHTEN. |
|
Back to top
|
|
 |
blue hell
Site Admin

Joined: Apr 03, 2004 Posts: 24389 Location: The Netherlands, Enschede
Audio files: 296
G2 patch files: 320
|
Posted: Wed Mar 22, 2006 2:29 pm Post subject:
Re: Please verify this bug Subject description: notorious delay line weirdness possibly detected! |
 |
|
tim wrote: |
I suspect that something funky is going on with the readout pointers, causing delay 2 to sometimes fetch a few samples of what is in delay 1.
|
Something got a bit over-optimized I guess ... when you throw in another unconnected delay the sound will stop (well, just to be sure, on the modified version that I have now) throwing in other modules or removing ones changes the character of the sound sometimes.
There have in the past been examples of this effect involving other (non delay) modules (that got solved).
Good work of you to pinpoint this with such a simple patch ! _________________ Jan
also .. could someone please turn down the thermostat a bit.
 |
|
Back to top
|
|
 |
dasz

Joined: Oct 16, 2004 Posts: 1644 Location: victoria, canada
Audio files: 29
G2 patch files: 56
|
Posted: Wed Mar 22, 2006 2:41 pm Post subject:
|
 |
|
Nice find. Could also be some memory which is not cleared fully
/Dasz |
|
Back to top
|
|
 |
Antimon
Joined: Jan 18, 2005 Posts: 4145 Location: Sweden
Audio files: 371
G2 patch files: 100
|
Posted: Thu Mar 23, 2006 5:24 am Post subject:
|
 |
|
Ah! Good work spotting that. I've run into this before as well, with a feedback loop over a delay that started living all by itself without any input when the delay length attenuator was modified by a MIDI send module.
/Stefan |
|
Back to top
|
|
 |
dorremifasol

Joined: Sep 28, 2006 Posts: 823 Location: Barcelona, Spain
Audio files: 7
G2 patch files: 49
|
|
Back to top
|
|
 |
Axiom Crux
Joined: Sep 16, 2008 Posts: 8 Location: Detroit Florida
|
Posted: Sun Oct 18, 2009 3:45 pm Post subject:
|
 |
|
So can someone at clavia fix this already!!!!!!!!!!!!!!!!!!!!!! |
|
Back to top
|
|
 |
drapdap

Joined: Oct 11, 2004 Posts: 204 Location: London
Audio files: 1
G2 patch files: 1
|
Posted: Thu Oct 22, 2009 5:49 am Post subject:
|
 |
|
Axiom Crux wrote: | So can someone at clavia fix this already!!!!!!!!!!!!!!!!!!!!!! |
i don't think so, since they went a bit of church organ, you know...
they miseralby failed with the second run of expansion boards too.
seems to me we are boiling in our soup and Nord couldn't care less.
it's not that it bugfixes didn't happen, nor the dsp board, but the way they
don't communicate at all, at least they'd mail or tell somebody like Jan that
sorry folks it's not going to happen, or something, just say something...
For me it's not the G2 who died but nord keyboards.
Flames off...
róbert |
|
Back to top
|
|
 |
Chrono

Joined: May 18, 2008 Posts: 229 Location: NL
Audio files: 37
G2 patch files: 38
|
Posted: Fri Oct 23, 2009 4:59 am Post subject:
|
 |
|
that bug has been bothering me as well. it took me some time in the beginning before i really noticed it. But it seems to produce unwanted noise and sounds low quality. and this should still be fixed as i think clavia (nordkeyboards) is a high quality product range..
shall we just collect Signatures from everyone and send the mail to the nords chief. and just bother them a bit untill they cant go around it. |
|
Back to top
|
|
 |
Tim Kleinert
Joined: Mar 12, 2004 Posts: 1148 Location: Zürich, Switzerland
Audio files: 7
G2 patch files: 236
|
Posted: Fri Oct 23, 2009 1:04 pm Post subject:
|
 |
|
Chrono wrote: | that bug has been bothering me as well. it took me some time in the beginning before i really noticed it. But it seems to produce unwanted noise and sounds low quality. and this should still be fixed as i think clavia (nordkeyboards) is a high quality product range..
|
This bug compromises the chorus module (makes it unusable in a professional recording session if mixed upfront), as well as DIY chorus/delay stuff.
Interestingly though, I've not run into the bug when using the delay module for elastic audio stuff (which I'm very relieved about).
But in terms of audio fidelity, it's the worst bug on the G2 IMO.
It's definetely got to do with the delay module memory allocation and/or the readout pointers. Probably just some wrong scaling or offsetting. In other words: just sloppy coding. According to Rob, the delay modules are still in alpha state, and never got finalised at all. A high quality product.
All in all, it's obvious that Clavia overtaxed themselves when developing the G2. Because, when you look at it, the G2 system was a herculean task. And when they saw the low sales figures rolling in, and the resulting lack of ROI, they probably had to put a lid on it immediately. I can understand that. It's a business after all.
I still believe the G2 could have been a top seller if Clavia had done the marketing differently and invested more time on factory algorithm voicing. The omission of comprehensive preset documentation especially was the downfall of the G2 IMHO. Not everybody can read a signal flow, and without understanding what's going on, a preset isn't really useful. Hell, even I never understood how most of the presets operated or what the assigned parameters (often not properly named) actually did. Last edited by Tim Kleinert on Sat Mar 28, 2015 10:50 am; edited 3 times in total |
|
Back to top
|
|
 |
rutgerv
Joined: Nov 22, 2010 Posts: 26 Location: Elst, Netherlands
G2 patch files: 3
|
Posted: Wed Dec 11, 2013 12:10 pm Post subject:
|
 |
|
I just want to confirm that I've identified the bug too, after spending more time with my G2. It's even in some of the preset sounds that Clavia delivers with the G2. Check for instance the 'G3' organ emulation patch. This has similar noise generated in the leslie emulation section. Strangely enough, only audible on the left channel. |
|
Back to top
|
|
 |
Tim Kleinert
Joined: Mar 12, 2004 Posts: 1148 Location: Zürich, Switzerland
Audio files: 7
G2 patch files: 236
|
Posted: Sat Mar 28, 2015 9:48 am Post subject:
|
 |
|
As described in the elaborations of my newest design...
http://electro-music.com/forum/topic-64110.html
...there's a robust (albeit a bit expensive) workaround for this bug.
I described it pretty well there, so I'll just copy/paste the relevant paragraph here for reference:
Quote: | I found a way to work around that old infamous "delay line bug" which causes crackles and glitches. In preparation of this algorithm I had to do lengthy experiments with the modulated delay lines (for 24bit precision addressing of individual samples), where I had the hunch that the artifacts were caused by the fractional sample interpolation window not "wrapping around" within the memory array allocated to that module, grabbing samples from another one and causing glitches. By that reasoning, placing a "dummy" delay module of equivalent size and filled with identical audio material right before the module being used should solve the problem. This is a bold assumption to make, and I might be completely wrong, but... IT WORKS! The crackles vanish. Of course, this is a rather wasteful solution, but I had to implement it in this algorithm to preserve sound quality. So if you see seemingly nonsensical delay modules with unused outputs somewhere in the patch, they are there for a reason. (Other imperfections in the mod delay modules remain, but they are negligible for most purposes.) |
|
|
Back to top
|
|
 |
|