Using VST or Audio Sample

  • Thread starter Thread starter edward0328
  • Start date Start date
E

edward0328

New member
Hi i came across a situation where my computer cant handle all of the instrument plugins that i use for my pro tools session. I have a standard desktop computer with Potools mp 9. Sometimes when i use too many VSTs,especially drums, my program stsrts to say errors saying to use less VST plugins. So maybe i thought that instead of using VSTs for drums,just take audio samples and use audio tracks insteadof instrument->VSTs to reduce CPU usage.. Is it true, that Audio samples use less CPU than VST instrument plugins?? Or am i just makig this up..please help!! Thanks!!
 
PT should have a method for "Freezing" your VSTi tracks. Do that and release the vsti and you'll have more cpu resources available.
 
Is it true, that Audio samples use less CPU than VST instrument plugins?? Or am i just makig this up..please help!! Thanks!!

Don't worry, it's true. It's called 'bouncing' and sometimes 'freezing'. In the days of analogue it had to be done I suppose at nearly every stage when working with limited tracks but not so with modern computer DAWs. But when the computer is clearly over-loaded, then it's time to bounce. Or it might be time to increase memory or get a new computer. But yes, overall, bouncing to audio is a perfectly respectable way of managing resources in a DAW.

In theory, a VST would be using more CPU resources than an audio stream because it's actually a mini program but it would depend on the format you save in. As well, there might be an option for you to leave the sample on disk instead of constantly loaded into memory.
 
Just to elaborate on the concept of "freezing" a track, in case it might help:

When the DAW is playing an audio track directly, it's doing very little more than just passing the data along, from the way it was recorded to the way it's played back. It might be doing minor things like changing the volume or panning the signal to one side, but those kinds of things are very simple in terms of the actual processing that has to be done to the data. I'm oversimplifying here for the sake of brevity and clarity, but the main point is there isn't a lot of "figuring" that the computer needs to do.

But when you're using a VST, or a more complex kind of effect like a chorus or reverb, there is a LOT of math that the computer has to do to generate the effect. It has to take the original signal data, which is a stream of numbers, and run them through an intense and complex bunch of mathematical calculations, and generate a new stream of numbers - the output of the effect. And if you're using the VST or effect in real-time during playback, it has to do ALL those calculations for EVERY number, tens of thousands of times every second. When you string a lot of VSTs together, the computer has to do all that math processing for everything, and all of it has to be done fast enough to keep the stream flowing, without slowing things down. When you reach the limit of what your computer's processor can handle, it can't keep up: it can't do all the math fast enough. And that's when you see errors, drop outs, or other problems with the playback.

What "freezing" a track does is this: it lets you apply some of the VSTs ahead of time, and generate the finished stream of numbers (i.e. the finished audio track), and stick that into the DAW's track. So now THAT particular track has already had all its VSTs and effects applied, and none of the heavy-duty math has to be done in real-time during playback for that track. That takes a lot of the strain off the processor and lets it handle the remaining VSTs and effects more easily.

Now, a track that is "frozen" can't easily have its effects tweaked as it's playing, because they've already been applied, and they're not being processed "live" during playback. You generally pick a track (or two, or three, or more, depending on the song and how many tracks and effects you're using) that you're fairly comfortable with, and freeze it - generate a finished audio track with all the effects applied - and then leave another track (or two or whatever) with the VSTs and/or effects still being processed in real-time, so you can easily tweak them and listen to how they change as they're played back. When you're fairly well satisfied with the results, you can freeze THOSE tracks. The DAW will also let you go back and "unfreeze" a track so you can adjust the effects; then you can refreeze it. You can do that as often as you like without losing your original audio track.

Again I'll stress that this is a quick, very simplified overview of what's happening. But I hope it gets the picture across so you can see how freezing can be used, and why it's sometimes needed.
 
Back
Top