"kind of defeats the object... powered plugins using CPU! .... is it like a lot? i.e like running similar plugs natively? "
Yes, it is like the native CPU is mirroring the UAD-1's DSP usage. Watching the XP Windows Task Manager, there seems to be a severe spike in the native CPU usage
that does not go away when you first press play in a project that uses a
lot of UAD-1 plugins. It is only really evident when you use a lot of powered plugins (7 or more). The remedy, is to disable and then enable the PPI's, using
the global desktop switches (on the Performance Meter) while the project is
playing. This releases the native CPU load back to normal. This happens
in ACID 4b and Vegas 4a. This does not affect the mixdown render. During playback, I simply leave a couple measures lead time so I can quickly do the disable/enable procedure to get the native CPU usage back to normal. Only thing is, if you stop and then start your project, you need to repeat the same off/on procedure with the PPI's. This should not be that difficult to fix. Hopefully, this little quirk will be addressed and fixed soon, in an upcoming update. This is really the only problem I have encountered using the latest 3.01 UAD-1 update. Otherwise, the PPI's are just like using any other native DX real-time automatable effects. Except, I have yet to find any native DX/VST effects that can touch the UAD-1's vintage compressor/eq's (Teletronix LA-2A, UA 1176LN, & Pultec EQP-1). Also, the new Cambridge EQ (think Sony Oxford) has raised the bar even higher.
"Does tha Uad allow usinf the new sony plugs , as in their new compressor ?"
No, DX/VST plugins need to be coded specifically for the UAD-1. As of right now, UA has no plans to add third party development for the UAD-1. Once the Cambridge EQ and the DreamVerb are finally released later this month, the UAD-1 will have 21 DX/VST plugins available for the card. Even more are in development, but no official news on those yet. For my processing needs, the UAD-1 covers most of the bases for me already. That UA has made a committment to development additional PPI's is just like adding gravy on my meat and potatos.
"So what about the latency...?
One could probably live with it on an insert by moving a track or whatever (sucks but could work)
but as an aux fx its impossibel to deal with latency´s.
So how does it cope with this?"
The automatic delay compensation in Vegas takes care of any timing issues. The UAD-1's 3.01 software fully supports Vegas's DX implementation and the automatic delay compensation keeps everything(track inserts, aux buses, main out),in sample accurate sync. Current VST based apps(Cubase, Logic, Nuendo 1.61) do not have this luxury though and require you to manually compensate for inherent latencies on anything other than track inserts. DX based apps like Vegas 4, ACID Pro 4, SONAR and Samplitude 7 compensate automatically on everything. So it is really nothing you need to be concerned about, unless of course you are using a VST based app.
The dropping out problems seem gone, there is still some funnyness from time to time but Im not sure it's UA's fault. The cpu munching is still around, but we'll see what happens. All in all working as good as any software plug would at the moment
ONE problem for sure, they wont work in DX mode after any sort of lookahead plugin, messes with the buffer settings.
My bitches on the card now have more to do with UA's odd behaviours and claims, aside from that the LA-2A and 1176's are OK, the pultec may or may not be your cup of tea as it suffers a latency that cant be compensated for and bleeds off a lot of hi end. The reverb is sort of OK but a better one is supposed to be coming. You may or may not like the nigel, as it has a very weird buck rodgers interface and a SEVERE lack of hi end. The channel strip compressor and eq are sort of ok, tho the eq is pre compressor and the compressor has a magic numbers auto gain compensation that cant be turned off, and the eq has the silliest buck rodgers interface Ive ever seen.
<<The channel strip compressor and eq are sort of ok, tho the eq is pre compressor and the compressor has a magic numbers auto gain compensation that cant be turned off, and the eq has the silliest buck rodgers interface Ive ever seen. >>
Seems like you're fairly underwhelmed with the UA DX. Personally I'm mostly interested in what the reports will be for the Dreamverb plug. Unless its a real killer, rivaling one of the PCM model Lexicons (I have a PCM 80), then I doubt I'll be getting one.
so in other words, I might as well stick with the waves renaissance series of plugins because they're better quality, low CPU overhead, and have killer interfaces?
Of course there's 8 zillion happy Nuendo/UAD-1 users out there that think Im an idiot, so take my opinion for what its worth, and I think Billy has a rosier picture as well
put an LA-2A in the insert of a channel
Now put Realverb pro as an assignable FX
now send some of the channel that has LA-2A on it into the realverb pro's send
Hey Pipeline,
I routed a LA2A track insert to a RealVerb on an assignable FX bus. Is there something I should be hearing/seeing that is out of the norm? It seems to be working fine. The only problem I am still encountering in both ACID 4b and Vegas 4b, is the inexplicable native CPU usage when using a lot of UAD-1 plugs (DX or VST) in a project. Try using (7) or more UAD-1 plugs in a project and watch the Windows Task Manager for native CPU usage. Now while the project is playing, disable and then enable the UAD-1 plugs from the Performance Meter. Do you see a drop in CPU usage?
If you stop and re-start your project, the native CPU usage goes way back up again.
This is the one and only problem I have with the aforementioned SOFO apps and the UAD-1 3.01 plugins. In SONAR, the one and only problem I have is inability to properly loop a portion of a project for more than a couple of times before I lose sync, but the native CPU usage does not increase like it does in my SOFO apps when using the UAD-1 plugs.
nuts
I get a " buffer mismatch error" when I do that. COuld you try it on a 24 bit 48k project? What type of soundcard are you using? ASIO or MME ?
Are you using the DX UAD plugs or the VST UAD plugs in a wrapper?
Can you try with : LA-2A then pultec in the channel insert and Realverb in the assignable fx?
As for the CPU use, yeah this thing MUNCHES cpu! Just a few plugs and vegas is already "echoing" if I hit save while playing, or open an insert fx button. By echoing I mean that error where it repeats a certain part a bunch of times till it catches up. I tried turning it off then on with the global disable but didnt seem to work.
"I get a " buffer mismatch error" when I do that. COuld you try it on a 24 bit 48k
project?"
Are you sure the buffer sizes of your audio card, the UAD-1 and the wrapper (if you are using one with the VST plugs) are set to the same exact number? I routinely use 24/44 in my projects.
"What type of soundcard are you using? ASIO or MME ?"
Delta 66, both ASIO and WDM modes work fine. MME does not seem to work well with the UAD-1. Low latency drivers are best I suspect.
"Are you using the DX UAD plugs or the VST UAD plugs in a wrapper?"
I have used both with good success. Although I prefer the DX, as it auto configures
to any host latency changes, whereas you have to manually set the FXpansion wrapper each time. Also, it is one less thing (when using the VST-DX wrapper) you need to match, buffer size wise, with the host and UAD-1.
"can you try with : LA-2A then pultec in the channel insert and Realverb in the assignable fx?"
No problems here, as one would expect when routinely inserting/routing real-time effects plugins (native or PPI) into the audio chain.
" I tried turning it off then on with the global disable but didnt seem to work."
It should work, keep trying to disable and then enable the UAD-1 plugs from the Performance Meter. If you can't, then it may be an indication of a source of your problems. With v3.01, I am able to easily disable and enable the PPI's in a project, whether I am using the DX or VST plugs. You should see the DSP usage indicator go to zero, once you disable the PPI's during playback.
"ok if I stop the transport before I add the effect I dont get the error, however, if I mute a track thats sending then unmute it I still get the error"
I am not getting any errors at all. The only thing that happens if I try to add a UAD-1 effect during playback, is a brief hesitation, a slight gapping if you will, but never any error messages, or other playback problems. I am using all the latest updates:
Vegas 4b
UAD-1 3.01, using the DX versions exclusively since 3.01 came out in February
I am also using a Delta 66 in ASIO mode (would not work with MME, it needs low latency drivers I guess) with a host/driver buffer size of 256 samples (5msec)when using the Nigel in realtime and mixing small projects. When mixing larger, more CPU intensive projects, I may boost the host/driver buffer size to 1024 samples (23.2msec), but not any higher than that. Normally, I just keep it at 256 samples (5msec). The less latency, the better, IMO.
Other than the inexplicable native CPU usage (which can be remedied, by disabling and then enabling the PPI's from the Performance Meter, during playback) I don't have anything negative to say about the UAD-1 from a operational standpoint. BTW, I always keep the UAD-1's Performance Meter on my desktop, so that I can easily monitor the UAD-1's DSP usage and glean any operational info as needed.
Will you be getting the Cambridge EQ when it becomes available at the end of the month? I will, even though I have to pay extra for it. The demo impressed me enough to want to part with some cash ;)