Pops and clicks

[marc]

Originally Posted by Tiphares

I do exactly have the same problems, but even sometimes when i use single mode. I also have my TI for many years now and the problems still exist even with the newest TIOS's.
I got in contact with the support team, but they explained me that the patches I use are ressource heavy and the problems are normal. I don't think that it's normal to max out the polyphony with only one patch and three notes at the same time. So I also would like to know if anyone else has the same problems.

For example, when I use ROM B 108 SuperWrmMS, three notes at the same time in the sequencer, i sometimes get nasty clicks, that can even be heard in the delay of the synth. It must be an internal problem. The don't occur very often, but only one such click can destroy your export or dump.

If anyone's interested in the cubase project, i will upload it.

What's also interesting is that the DSP load indicator on the TI is full (5 dots), even when I don't play a note with this patch. That means that any note can max out the dsp?

Thanks for the help and cheers

Tiphares

the DSP load indicator shows how expensive a patch is. it is not an indicator for the current load. send me a cubase project and i'll have a look if it can replicate your findings.

best, marc
access music

[motorik]

Add me to the "still getting this after several years list." Subjectively, it doesn't sound like something caused by "a total dsp work asked of the Virus vs. it's total dsp capacity issue" ... it seems more like certain configurations / modulation setups cause unexpectedly large spikes in the required calculations.

I did some tests, and was unable to reproduce the behavior with high polyphony loads of a simple 2 osc test voice. I'm guessing it is a DSP overrun, but not a simple "making it do too much" thing. The best analogy I can come up with is that it's like a really strong person that can carry twenty cinderblocks, but can only juggle three of them.

I get this sometimes from a single, but mostly it's multitibral that causes it. It's gotten worse with the new OS. I believe it also got worse with the first version of the last big update, but improved again with subsequent builds.

It would be nice to have system total dsp display someplace.

This all relates to using Virus Control / Logic 8, btw. From what I remember from the old email list, people using the analog outs don't have this problem.

[marc]

Originally Posted by motorik

This all relates to using Virus Control / Logic 8, btw. From what I remember from the old email list, people using the analog outs don't have this problem.

anybody who discovers pops and clicks ONE with the USB outs and not with the analogue outs has a USB transmission problem. you're right in the respect that in the early days there were issues in virus control which could have resulted in those artefacts. i'm sure those don't exist anymore.

marc

[Ingo]

Originally Posted by marc

anybody who discovers pops and clicks ONE with the USB outs and not with the analogue outs has a USB transmission problem. you're right in the respect that in the early days there were issues in virus control which could have resulted in those artefacts. i'm sure those don't exist anymore.

marc

Hello, i am new to the forum but i own my Virus TI Desktop since around 2,5 years. I encountered problems from the first day until now. With OS 3.0 i still get pops and clicks in multimode (e.g. 3 parts: 2 "DSP-heavy" pads and a lead). Using Virus Control via USB gives less or no problems. As a result i see more problems when using the TI in Standalone mode.
I use a Virus TI Desktop with OS 3.0.0 pb. Computer: Windows Vista SP1 on Core2Quad Q6600 CPU (Desktop, 4 x 2,4 GHz) or Core2Duo (Notebook, 2 x 2,53 GHz) with Ableton Live 7.0.14. As a soundcard i use a M-Audio Ozonic firewire controller with built-in audio interface.

[Tiphares]

Originally Posted by marc

anybody who discovers pops and clicks ONE with the USB outs and not with the analogue outs has a USB transmission problem. you're right in the respect that in the early days there were issues in virus control which could have resulted in those artefacts. i'm sure those don't exist anymore.

marc

Thanks for the help.

I still have this artefacts even without USB-Connection, only with a midi cable coming from my midikeyboard and headphones plugged in the TI.

And if you set at the posted project the outs to analog, you also can hear the clicks. So it can't be the transmission...

Cheers,
Tiphares

[suzzymackenzie]

Originally Posted by marc

the DSP load indicator shows how expensive a patch is. it is not an indicator for the current load

Really? I'm convinced I've seen that load indicator fluctuating for a given patch, in real time.

[Tiphares]

Thanks for the help!

The clicks also occur with a single patch, but if I want to be sure that they occur, I have to load more than one patch. It seems to me as if those problems mostly occur when the ti allocates the dsp power for a patch. After this "allocation" there are fewer problems. The clicks can be heard mostly at the beginning of a sequence of a "new" patch.

Here is a cubase project with an mp3-export.
We can clearly hear the clicks.

MP3:
http://home.zhaw.ch/~projebal/ti/problem.mp3

Project:
http://home.zhaw.ch/~projebal/ti/problem.cpr

I know that I'm kind of overloading the TI, but it's the only way to be sure that the problems occur.

Thanks in advance
Tiphares

[marc]

Originally Posted by suzzymackenzie

Really? I'm convinced I've seen that load indicator fluctuating for a given patch, in real time.

i know this effect when it hops from say 4 bars to 3 bars and back again. that's more like an inaccuracy in the measurement.
it updates in realtime when you change the sound but it doesn't take the amount of voices played into account.

marc

[suzzymackenzie]

Originally Posted by marc

i know this effect when it hops from say 4 bars to 3 bars and back again. that's more like an inaccuracy in the measurement.
it updates in realtime when you change the sound but it doesn't take the amount of voices played into account.

marc

Aha, I see. Thanks, Marc

[DigitAl56K]

I've had this problem from the moment I first turned on my TI through the OS 2.75 upgrade and now through the 3.0 beta upgrade, and I play it standalone. The problem seems to be related to the events that take place when a voice is triggered, this seems to be when pops/crackles occur, somewhat randomly. Personally I believe that what's happening is some kind of bug in the way the virus locks resources between its two DSPs so that one DSP can think a voice is released while another modifies its components, for example. If that happens while several voices are playing due to thread scheduling it may be a couple of milliseconds before the first DSP actually completes modifying original voice, so you hear a pop (depending on the patch properties and voice state). The pop then gets blasted through the effects unit and so with some patches it sounds worse than others (e.g. if there is a reverb). Anyway, that's my suspicion and I urge marc & team to double check their locking code and ensure that all voice-related components are stopped before locks are released, particularly in the case of stealing. Personally I can't believe that access can't fix this, it's the most widely reported problem with the TI ever since it was first released and made me worried about buying one in the first place. When I first turned on the TI I was sad to experience this issue almost immediately. Fortunately I'm glad to say that even with this issue I do love the virus, but that's no reason that access shouldn't fix the problem.