delay module timing

roy thinnes · Post by **roy thinnes** » Thu Nov 02, 2006 1:30 pm

have you ever checked it? In a recent patch, I was confused about the fact that something with the timing of a tempo delay module was wrong. So I exchanged it for a standard delay module, and everything was ok again. Damn! I'm using the Modular since many years, but it took me that long to come across this.
What I found out is that there are just two delay modules with correct timing:
Delay and pitchShiftDelay.
all others have incorrect timing:
longDelay, TempoDelay, SSBDelay and (!) Flexors PitcherEngineFB.
If you want a 1/8 delay in a 120Bpm track, you'll use 250ms. With the "bad" delay modules, you have to set the delay time to 275ms to have tight timing.
Can somone confirm this queerness?

pling · Post by **pling** » Thu Nov 02, 2006 4:02 pm

maybe samplerate bug? i have timing problems with some modules on 48khz. don't know exactly what modules, but for sure flexor ramp-module.

roy thinnes · Post by **roy thinnes** » Fri Nov 03, 2006 1:56 am

na, I don't think it's associated with sample rate. You can try the patch below at any sample rate with identical results.

Ben Walker · Post by **Ben Walker** » Fri Nov 03, 2006 4:39 am

I wonder if it's related to your ULLI settings. SpaceF's Echo comes in different versions to suit your ULLI setting, so he must have come across something similar.

roy thinnes · Post by **roy thinnes** » Fri Nov 03, 2006 5:12 am

Hi Ben! No, I've tried it with various ULLI settings without avail. Have you checked the patch? I'm the only one...?

Lima · Post by **Lima** » Fri Nov 03, 2006 9:58 am

Very interesting!

I'll make the test soon

djmicron · Post by **djmicron** » Fri Nov 03, 2006 11:25 am

the same here

pling · Post by **pling** » Fri Nov 03, 2006 2:10 pm

ok that sounds like something different than samplerate. simply calculating comes to: 250ms at 44.1 = 272,1ms at 48khz. that is not exactly 275ms...

i'll test the patch soon.

btw: the flexor modules are probably fixed with the forthcoming version!?

roy thinnes · Post by **roy thinnes** » Fri Nov 03, 2006 5:22 pm

well, would be nice if someone could test the patch (just load it, hit the start button and switch on the on/off switches). If all delay modules produce the same sound / have the same timing, then it's my setup. If the mentioned modules (longDelay/TempoDelay...) have different timings, it's not.
ps. in the former case: one of my cards is still a Pulsar I (4dsp), which I bought back in '99. I hope she's not the one to blame..

wayne · Post by **wayne** » Fri Nov 03, 2006 8:20 pm

All is as you say, Roy.

I had noticed this every now and then over the years, and sorta scratched my head and assumed it was my lack of electro-savvy causing it - i never thought of loading all deley modules and comparing. Well done.

Creamware?

roy thinnes · Post by **roy thinnes** » Sat Nov 04, 2006 4:39 am

thank you for testing wayne.
SO this is something for the module wishlist.

dawman · Post by **dawman** » Sun Nov 05, 2006 7:07 am

I never thought I would live to see the day when analog symptoms came out of something digital.

roy thinnes · Post by **roy thinnes** » Sun Nov 05, 2006 4:54 pm

yes, Creamware always got a new surprise in store for us...
just curious, I tested the other CW delays (via insert module) and - guess what...half of them 's got wrong timing too:
Delay LM and LS
Delay LCR LS
Delay LS
Dual delay LS
Ducking delay M and S

the good ones:
Delay M and Delay S
Delay LCR S
Delay M and Delay S
Dual delay S
Multitap Delay M and S
Pattern Delay
(in short: all delays capable of long delay times and also DuckingDel have wrong timing)

dawman · Post by **dawman** » Sun Nov 05, 2006 6:46 pm

The same is actually correct on older hardware delays. Lexicon always paid attention to detail, but their PCM 70 was inacurate in that respect also. But it was actually sold as a reverb unit. They got it right with the PCM80, and 90 though. The PCM80 is the best multi effect box in that medium price range. You always know when a company is going to milk it's designs down like we recieve server technology down from Intel and AMD, is when they start to go into the consumer market with the same design chips, but less I/O's and cheaper DA / AD's etc. Their NuVerb that I used with ProTools was a real disappointment. Everyone back in 1993 thought it would be the bomb. We ended up using thje Lexicon PCM70, and believe it or not the excellant sounding Ensoniq DP4, as it's chorus and four effect matrix was way ahead of it's time then. I'm glad that you have found a chink in Creamware's armor. I shall now demand the Body Of Ralf.

Post by **hubird** » Sun Nov 05, 2006 7:35 pm

as I couldn't remember ever having faced timing problems with the stock delays, I tested the LS and the LCR LS, and encountered no problems.
I entered the right BPM to fit the loaded 4/4 kick loop, zero'd the feedback for clean listening, and choose 1/2, 1/8 and 1/16 delay settings, with cubase as sinc master, and everything was fine, even when then feedbacking to heaven...
I'm on mac but why would that make any difference?
So at least it's not an universal bug, there's something specificly wrong here

roy thinnes · Post by **roy thinnes** » Mon Nov 06, 2006 1:07 am

happy hub (happy Mac-Users?)
my LCR LS Bd (120Bpm - 1/4) sounds like this. Thats a little bit too shaky for me.

roy thinnes · Post by **roy thinnes** » Mon Nov 06, 2006 1:42 am

gosh! I just tried it with a a different Ulli setting (the slowest ) and everything is tight....exactly as BW suggested...before I only tested the faster ULLI settings... Shame on me

.
Conclusion1: if you run into problems with delay timing => use slower ULLI settings
Conclusion2: Scope is and will ever be the best audio card ever created here on this planet earth!

wayne · Post by **wayne** » Mon Nov 06, 2006 4:19 am

Thank you so much, Roy

wayne · Post by **wayne** » Mon Nov 06, 2006 4:44 am

And Ben

Shroomz~> · Post by **Shroomz~>** » Mon Nov 06, 2006 5:23 am

Ah, that's good news. ;¬'