DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Arceo · Post by **Arceo** » Mon Mar 05, 2012 2:06 pm

Sorry to report it, but I've been confirmed by Martin Saleteg (Software Developer of Vienna Symphonic Library) that:

I can confirm that DP does not latency compensate properly. The audiounit does however report the latency changes properly, and all other hosts with ADC pickup on this and adjust the latency accordingly. It might be an issue with DP only expecting a static latency for the plugin - not considering any changes in the latency value during runtime. This is necessary for VEPro, since the latency depends on (a) soundcard buffer size, (b) VEPro numbuffers setting.

This is the topic at the VSL forum where you find the info.
I must admit that I'm quite baffled that such an important feature (which has been a selling point for me) has not been tested by the VSL beta testers with DP.
I mean, only after VEPRO5 has been commercialized it turned out that DP "Bounce to Disk" feature doesn't work with their new "Event plugin" (which is a clever workaround to circumvent the 16 MIDI channels limit of AU's specs). And they apologized for not including this info in the manual. Now it turns out that DP VEPRO tracks aren't properly latency compensated: it looks like DP at VSL is not exactly popular...

This would be the perfect moment for Magic Dave to chime in and help us with the technical implication of Martin statement.

Cheers
Arceo

Marc7777 · Post by **Marc7777** » Mon Mar 05, 2012 4:57 pm

Ah HA! So i'm NOT losing my mind!!! pheeww..

When you do those tests that you speak of, with the kick drum, are they WAY off??

I'm asking because when I have a buffer of 256 in DP and a VEP "Plugin buffer" at 2, it's not That bad..

As in maybe like 20 samples or so off.. noticeable if it's a sharp percussion, but hardly noticeable with any other instrument..

And if I put the plugin buffer down to 0, it's really really close.

I was thinking that the samples themselves might not be exactly on the grid.. again, i'm talking 5-20 samples..

PS - i didn't like reading that it's a DP problem.. that means it'll never get fixed..

What do you think!

~Marc

doodles · Post by **doodles** » Mon Mar 05, 2012 6:02 pm

That's a nightmare. If this is the 2nd (or 3rd?) problem with the new VE Pro (which I haven't downloaded yet) and DP, that doesn't make for good reading.

I hate it when people blame DP. Sure, it might be something unique to DP, but if there are now 2 or 3 problems with VE Pro and DP, I'm not sure they should advertise it as playing nicely with DP.

The Event/bounce to disk problem is a deal breaker for me.

mhurwitz · Post by **mhurwitz** » Mon Mar 05, 2012 7:41 pm

Yes, I can confirm that VEP 5 does not work for me in DP. Not only does it not latency compensate, it is more sluggish in performance.

Arceo · Post by **Arceo** » Tue Mar 06, 2012 5:33 am

As I suspected , unfortunately, this is an issue with VEPRO4 and VEPRO5. It's been confirmed once again by Martin:
Latency compensation also doesn't work with VEP4 in DP. As I wrote earlier, it looks to be a DP issue

Seems like the right time for MOTU and VSL software teams to get in contact and try to sort this out...

Anyway, since Martin wrote (three times in a row!) that this is a DP issue I'd like to hear a MOTU official word about it, so I'm going to mail them all these info and see what they have got to say. I'll keep you updated...

mhurwitz · Post by **mhurwitz** » Tue Mar 06, 2012 5:40 am

The engineers at VSL are now saying that ALL versions of VEP dobn't latency compensate in DP. Can anyone confirm this? I don't recall having this problem with VEP 4. This is terrible news because it makes my slave PC pretty much useless.

mhurwitz · Post by **mhurwitz** » Tue Mar 06, 2012 5:54 am

I also just emailed MOTU tech support. This is horrible.

Arceo · Post by **Arceo** » Tue Mar 06, 2012 6:59 am

MOTU TechLink posted!
Now I can start tapping my foot on the floor!!!

cbergm7210 · Post by **cbergm7210** » Tue Mar 06, 2012 8:52 am

I too, have posted a tech link. Thank you for the work in confirming this.

labman · Post by **labman** » Tue Mar 06, 2012 10:30 am

We have also posted to motu techlink ...

"Hi Folks, There is a dreadful problem with DP not working with Vienna Ensemble Pro 4 or 5 with the delay compensation. Can you please look into it and work with them Vienna asap to get it resolved? With all the great new video features you are about to release with DP8 it is an absolute must for all us film folks to have the link between DP and VEPro fixed. Delay compensation with VPro is working fine on some of the other platforms."

malditoyanki · Post by **malditoyanki** » Tue Mar 06, 2012 10:33 am

Wow...bummer. I actually hadn't noticed any issues, but I have my buffer set to "0" in the VEpro plug. I set DP to 512. Maybe my brain latency is equal to that.

labman · Post by **labman** » Tue Mar 06, 2012 10:46 am

malditoyanki wrote:Wow...bummer. I actually hadn't noticed any issues, but I have my buffer set to "0" in the VEpro plug. I set DP to 512. Maybe my brain latency is equal to that.

Hah. Brain Latency. We need a 'neuron time shift plug'.

Buffer at "0" is one possible workaround. But that will bring our host DP CPU to its knees with all the slaves and instantiations we need.

Do you get glitches using "0"? How many trks? Instantiations?

cbergm7210 · Post by **cbergm7210** » Tue Mar 06, 2012 11:00 am

labman wrote:
malditoyanki wrote:Wow...bummer. I actually hadn't noticed any issues, but I have my buffer set to "0" in the VEpro plug. I set DP to 512. Maybe my brain latency is equal to that.
Hah. Brain Latency. We need a 'neuron time shift plug'.

Buffer at "0" is one possible workaround. But that will bring our host DP CPU to its knees with all the slaves and instantiations we need.

Do you get glitches using "0"? How many trks? Instantiations?

Can someone point me to where the VEP buffer setting is located? I cannot find it.

labman · Post by **labman** » Tue Mar 06, 2012 11:10 am

VEP buffer is called 'latency' directly inside their plug. You pull it down to the desired number. The number you select is multiplied times DP's buffer. If you pull down '2' that will be an add'l 2x DPs buffer for it to make its roundtrip.

mhurwitz · Post by **mhurwitz** » Tue Mar 06, 2012 11:31 am

I'd just like to add that it does not work in Protools 9+ either... though I think it's being chalked up to some sort of "MIDI playback latency bug" in Protools that people have known about for years.

I see people running all kinds of sophisticated tests for this... there is a very simple test you can do:

- Program quarter note snare hits that are quantized perfectly to the quarter note and send it to your VEP server.
- Record the audio that comes back into the computer.
- Compare the waveform with the MIDI note.

You will easily see how far off the recorded audio is from the MIDI note.

MOTUnation.com

DP/VEPRO - VSL confirms ADC doesn't work properly! :(

DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(

Re: DP/VEPRO - VSL confirms ADC doesn't work properly! :(