Comments

Arnar wrote on 9/24/2003, 3:36 AM
It seems that when you disable plugins then it sticks this time and doesnt turn them on when you push play BUT the CPU munching is still there.

That was my biggest concern as well as the Pultec render problem.
The cpu munching is still there but i havent tried the Pultec Bug.
PipelineAudio wrote on 9/24/2003, 12:50 PM
Been eager for this release. DX CPU munching is still there. Dreamverb is OK but nothing special, nowhere near enough controls to make a decent gate. Still the PC hangs when I shut it off. Still cant put effects in certain orders without the " buffer size mismatch" bug.

good job UA!

bgc wrote on 9/24/2003, 3:36 PM
I recently got the UAD-1 (last week). I had some buggy/hanging behaviour with it and was able to get a lot better performance by PCI slot hopping, so it must have been some IRQ issues. Try that approach to see if the card performs better.
I don't know if I'd call the buffer mismatch message a bug. If there are plug-ins before the UAD-1 in the FX chain that do not have sample consistency (i.e. it doesn't output the same number of samples that are input) then the UAD-1 can't handle that and complains. These plug-ins cause trouble with input monitoring in Vegas as well.
BGC
Arnar wrote on 9/24/2003, 7:14 PM
The buffer mismatch bug happens even though you only use UA plugs.

Cambridge on a track insert with a send to the Dreamverb causes the buffer mismatch.
I think it only happens when i use either Realverb or Dreamverb.
PipelineAudio wrote on 9/24/2003, 10:24 PM
la-2a and pultec can cause it too in some circumstances
bgc wrote on 9/25/2003, 11:02 AM
I haven't tried or seen that. I got some buffer errors when my ASIO buffer size was set to 96ms. By lowering it to 48ms I got a lot of buffer issues to go away. You guys might be doing monitoring and have your ASIO buffers set really low though.
PipelineAudio wrote on 9/25/2003, 12:45 PM
no monitoring, not messing with the buffers

I put LA-2A on a track insert, then put realverb ( HAHAHAHA) on a assignable fx buss. Send the LA-2A track to the realverb and buffer size mismatch error
bgc wrote on 9/25/2003, 12:57 PM
I'll try this test later today at the studio and report my findings.
B.
bgc wrote on 9/25/2003, 8:09 PM
Pipeline,
I tried this LA-2A / RealVerb test and had no buffer mismatch errors at all.
B.
PipelineAudio wrote on 9/26/2003, 6:25 AM
damn billybk had the same results as you

I wonder what the deal is
fishtank wrote on 9/26/2003, 10:05 AM
Did they fix ANY of the Direct-X problems in the 3.2 release? They have known about the CPU munching problem before 3.1 and it is still there after 3.2. The scary thing is that I read that the 3.3 release will address OSX support, so I am not optimistic we will see any benefit - will these DX bugs ever get resolved?

I beginning to think that the only way to get the UAD-1 working properly is to switch to a VST application.
bgc wrote on 9/26/2003, 10:45 AM
I'm not seeing any DirectX problems. Maybe using the VST versions of the plugins with a VST wrapper might help. Just a thought.
fishtank wrote on 9/26/2003, 11:32 AM
You are not seeing the CPU munching?? AFIAK - this is a Direct-X problem as I have not heard about it with VST apps.

I have seen a fair amount of people complain about the buffer mismatch error with Direct-X. This has been showing up for me from time to time.
bgc wrote on 9/26/2003, 12:52 PM
I'm not seeing it, but then I'm not loading up lots and lots of instantiations of the plug-ins (which i think is when the munching shows up).
I'm selectively using the compressors and reverbs on select tracks.
PipelineAudio wrote on 9/26/2003, 3:07 PM
1 plugin will do it

Something is going weird with your stuff bgc, if you arent seeing any of these problems, or you have something very right that noone else can figure out. The dx CPU munching happens across the board, regardless of how many plugs you have going
bgc wrote on 9/26/2003, 3:46 PM
Hmmm, I'm definitely not seeing any CPU munching from my end then, if 1 plugin can do it.
Let me know if any system configuration information would help.

Again, the only weird thing that I saw was that the card was hanging my machine. After playing "hop the card" in different PCI slots it cleared up.

B.
billybk wrote on 9/27/2003, 1:24 PM
After installing tha latest 3.2 update, I decided to dig a little deeper, to see if some of the troubles I have been experiencing could be of my own doing.
Read from my humbling experience:

http://www.chrismilne.com/uadforums/topic.asp?TOPIC_ID=3781

It reminds me once again, that as in life, one should never take anything totally for granted.

Billy Buck

P.S. Hey Peter H, the rendering problem with the UAD-1 and Vegas 4 / ACID Pro 4, that we discussed, was an IRQ sharing configuration issue, after all.
bgc wrote on 9/27/2003, 1:29 PM
That's great that it helped Billy! I'm always conscious of IRQ issues with audio hardware after having a horrible time with an Event Layla setup in another studio. It would hang and crashe and exhibit all kinds of nasty behaviour. After doing some IRQ investigations and slot hopping it finally fixed it and the system worked beautifully from then on.
B.