I don't know if someone requested such a feature, I didnt find anything related.
I thought it could be nice to be able to by passe the vst in automation... I have very heavy modules with several plugins using a lot of CPU. it could have been interesting to bypass them when they're not in use in the track.
Let me know what you think about that
I think most plugins should be easy on the CPU when they're silen and/or have no input. Then again, some probably aren't.
This sounds like an feature only available through automation which is still a dream for OMPT
Hmm... As far as I know, the VST spec actually dictates that a plugin should always be active, processing null midi events or silence. OpenMPT doesn't actually follow this, because plugs are just "stopped" when you hit the "stop" button - that's why you hear the tails when you restart playback. In fully-compliant systems, when you hit stop, everything will fall away nicely. So I guess that if we want to stay as close to the spec as possible, we shouldn't implement this feature.
The other problem is that plugs don't always respond to audio data or midi events - sometimes, they generate their own. For example, you might use the sequencer capabilities of a plug, or use something like the "tb_field" plug. It would, of course, be possible to check whether the plug's producing any sound, but I think that polling regularly enough to pick up the moment a sound starts would probably use more processing power than the plugs would.
I see what you mean about using automation to bypass the plugs (so that you stop the plug when you want to), and the only thing that I can suggest is that you use something like xlutop Chainer, which allows you to bind the "wet" value of plugs to specific CCs. :/ Not exactly what you're after, but it's getting there.
Quote from: "bvanoudtshoorn"
... the only thing that I can suggest is that you use something like xlutop Chainer, which allows you to bind the "wet" value of plugs to specific CCs. :/ Not exactly what you're after, but it's getting there.
binding the wet value won't stop the cpu use of the plugin... it will just mute the silence! :s
Would a new modeffect for setting VST bypass on/off sound like a good idea?
My first thought was: just use the wet/dry to put it off. But indeed that wouldn't stop the processing.
Then again, wouldn't it be better if the plugin would be bypassed automatically if the wet/dry would be set to 0% anyway?
Quote from: "LPChip"Then again, wouldn't it be better if the plugin would be bypassed automatically if the wet/dry would be set to 0% anyway?
In comparison to sample based instruments, wouldn't that be somewhat similar to freezing sample play position, envelopeprocessing etc. in case putting channel volume to zero?
Hmm, I see your point. In that case, it could be unwanted, and you might want to put it on 0% on purpose as gate effect...
Yeah, then it would be better if you can do Z7f = on or Z00 = off as bypass option using a new macro.
Quote from: "Relabsoluness"Would a new modeffect for setting VST bypass on/off sound like a good idea?
Quote from: "LPChip"
Yeah, then it would be better if you can do Z7f = on or Z00 = off as bypass option using a new macro.
indeed the idea sounds really nice... :)
Boy if we could add that feature, it would actually expand our musical capabilities. My CPU is only about 100 under emulation, so i can only use one very low-CPU VSTi in one song. If i could "Ignore VSTi process" with some kind of note FX or channel command, i'd be able to use more VSTi's and really put together a nice opus....
Quote from: "bvanoudtshoorn"Hmm... As far as I know, the VST spec actually dictates that a plugin should always be active
having such a "always stay active" mode for VST plugins
and audio devices would be a good idea. This would proably also solve one of my problems (http://forum.openmpt.org/index.php?topic=1820.0) with the audio buffer, too...
I'm also having issues with my first songs that used VSTs which could be solved this way: If there's first my global "mda Dynamics" plugin and
then a global "JS Equalizer", there's some strange noise in the beginning. I think it doesn't occur if I reverse the VST order or if I use Electri-Q as an equalizer.
this sounds like a continuation of this (http://forum.openmpt.org/index.php?topic=935.0) feature request...
the bypass-macro idea sounds good to me
Quote from: "BooT-SectoR-ViruZ"this sounds like a continuation of this (http://forum.openmpt.org/index.php?topic=935.0) feature request...
the bypass-macro idea sounds good to me
yes it sounds like that thread, sorry i didnt find it, when i wrote the post...
by the way the clik/pop issues is no big deal with a leveler fx in the output!
clicks are always a problem if they exist...
Quote from: "le_parasite"Quote from: "BooT-SectoR-ViruZ"this sounds like a continuation of this (http://forum.openmpt.org/index.php?topic=935.0) feature request...
the bypass-macro idea sounds good to me
yes it sounds like that thread, sorry i didnt find it, when i wrote the post...
by the way the clik/pop issues is no big deal with a leveler fx in the output!
no need to apologize... that thread was 2 years old and this one puts up some nice ideas concerning that problem