Subject:Post your "usable" lowest ASIO latency here (Acid 4, etc)
Posted by: Cowboy
Date:9/13/2002 7:30:34 PM
I have an Echo Gina/20, I'm running Windows XP. In the Acid 4.0 "Advanced Audio Preferences" I can set it as low as 512 samples (11ms latency) before it starts to crack up. This is with a simple 6-track project, with an average of one DX effect per track. Just running Native Instruments' B4 or Pro52 VSTi as stand-alone apps, I can get them working fine at 128 samples (3ms latency) I want to get an idea of what other people can set their soundcards to, what is the lowest USABLE latency you can set it to, how does it work? I want an idea of what the best soundcard is for Acid 4.0 + ASIO + VSTi. Post your "usable" lowest ASIO latency here! - Cowboy |
Subject:RE: Post your
Reply by: groovewerx
Date:9/13/2002 7:33:48 PM
a3 + xp + wamirack = 0ms latency @ 64 sample |
Subject:RE: Post your
Reply by: Cowboy
Date:9/13/2002 7:52:49 PM
Have you tried Acid 4, with VSTi's? |
Subject:RE: Post your
Reply by: nlamartina
Date:9/13/2002 8:10:54 PM
21.66 ms, rock-solid. With the following: 1 GHz Athlon Win2K - SP3 9 VSTi's running at once (inc. multiple instances of Triangle 1, Triangle 2, Quadra 64, and Absynth) Beefy projects (15 - 40 tracks, different mixes of loops and MIDI) 16 bit, 48 kHz SBlaster Live! Platinum using kX-Project drivers (ASIO mode) Using just one or two synths, I can turn it down to its minimum, 2.66 ms. The number of audio tracks doesn't seem to affect performance as much in terms of static and such, relative to VSTi's, so I could get away with 2.66 there too if I wanted. But to save myself the trouble of constantly changing the buffer, I leave it at 21.66, where I know I can get away with just about anything. - Nick |
Subject:RE: Post your
Reply by: TeeCee
Date:9/13/2002 8:45:04 PM
Everyone should open up Hybrid.acd and test with that. It use three DLS tracks and some audio. I'll post mine soon. TeeCee |
Subject:RE: Post your
Reply by: Cowboy
Date:9/13/2002 9:06:19 PM
Hybrid.acd uses very little effects and CPU. I can set my Gina/20 ASIO to 128 samples (3ms) and it works almost perfectly (occasional glitching). The thing is, this file is not even remotely representative of one of my projects. - Cowboy |
Subject:RE: Post your
Reply by: groovewerx
Date:9/13/2002 9:09:25 PM
not interrested in a4 but i often use 4-5 virtual instruments and several other demanding apps simulaneously @ 0ms without a glitch. |
Subject:RE: Post your
Reply by: Rockitglider
Date:9/13/2002 9:37:07 PM
Hello, 384 samples, 8Ms Delta DIO 2496 and Audiophile 2496, both synced with Delta control panel, External Oxygen-8 MIDI controller, oodles of VSTi's to choose from, But mostly I use Native Instruments Brand and LinPlug, and Big Tick See ya, Rockitglider |
Subject:RE: Post your
Reply by: TeeCee
Date:9/13/2002 9:48:29 PM
It doesn't have to be representative of a typical project (let's hope it's not). It's just a benchmark for us to compare drivers and systems (not necessarily mine vs yours, just any and all). TeeCee |
Subject:RE: Post your
Reply by: Spirit
Date:9/14/2002 2:48:19 AM
Creamware Pulsar-II with a P4 1.8MHz Northwood: 5ms This is very smooth, no crackles, running Creamware synths and effects plus Kontakt and Absynth as VSTi |
Subject:RE: Post your
Reply by: Cowboy
Date:9/14/2002 8:54:16 AM
Well, it kinda does have to be representative. For example, if SOFO advertised that I'd get 3ms latency using my Gina with VSTi and Acid 4, because they tested it out with this tiny project with next to no plugins or tracks.. and then I found out the reality which was "there's no way you're getting CLOSE to that in your projects" I would be rather disappointed. Either way, what I'm interested in is "realistic" latency values for "non-sofo-optimized" projects! - Cowboy |
Subject:RE: Post your
Reply by: Cowboy
Date:9/15/2002 5:45:09 PM
Rockitglider, is this wih Acid 4? Do you get 8ms ASIO working reliably (and fast) while you have many tracks playing, with effects, while recording a VSTi track? - Cowboy |
Subject:Latency? Infinite...
Reply by: victors
Date:9/15/2002 6:55:12 PM
AMD Ahtlon XP Pro Delta Audiophile (new beta drivers) 1.4 ghz 1 gig ram Big fat, fast HDs Waves plugins Fruity Loops w/Soundfont loader VSampler I get out-of-control crackling no matter what I set the latency to, even without any VSTi (leave alone when I add just one). It crackles on playback on every clip start/stop, and sounds like it's doing my sound system harm when I move my mouse. The application is regularly (but not repeatably) crashing, usually on a 'render' or leaving the program and the program seems to using about 30% more memory than Acid 3 (including some apparent leaks). (There are painting gliches and several new usability annoyances but these are fluff compared to the overall experience.) Basically Acid 4 is unusable to me. I understand very well the pressures to release software as early as possible in a competitive environment, but I get the impression that M-Audio cards are relatively popular -- it's therefore all the more confusing that sofo would ship (and charge) such an obviously beta codebase and call it a 'release' and yet not work with such popular hw/sw as M-Audio and Fruity Loops (a whole other set of interop disasters). Speaking of charge, there are rumors around that some people paid $99 for the upgrade. Is that true? I was charged $157.95. As I recall, Acid 2 was a open (i.e. $0), beta and as a result the most stable music program I have ever worked with. Acid 4 meanwhile (in my experience, of course) is at least as bad as the worst Cakewalk release that I can remember -- which is maybe the worst thing I can thing of saying. A snake will take the shape of its prey for a while after swallowing. Going head-to-head with Cakewalk/Cubase et. al. it appears sofo has forgoed their usual standards of quality (the thing that kept many of us loyal *despite* having a fraction of the features those full-blown sequencers have, even after they included ACID-looping caps) in order to compete on features. Hopefully the snake will resume it's original shape after a period of digesting and therefore I await 4.0a and M-Audio's final realease of upgraded Delta drivers. A piece of unsolicited advice to (still) my favorite software company: to all the clever marketers (who rushed to product out to compete) and the naive developers (who rushed to the product out because they were eager and proud and it worked on their machines): do the math (!) -- the cost of lost sales from people who never up-sell from an open, free beta will be less than ruining the whole company because your flagship product did too good job at copying the competition. Sigh, VS |