I don’t see much reason for me to do either test. (Unless there’s a bug in Pigments that makes it run much more inefficiently as a Plugin. But already knowing i have room there makes this less interesting to me at the moment. I suppose more tests could be to compare Pigment head to head with another software synth also running standalone, like UVI Falcon, or Steinberg Halion 6 – comparing apples to apples.Īnother possible test would be to run Pigments as a VST over top of a DAW like Live or something. My system has lots of performance headroom over top of what i’ve seen already.
(You can think of it like an audio compressor, which means you have further to go before you get to hard clipping.) The processor just works harder to compensate to the increased demand. Given that the CPU clock speed increases with demand means that the correspondence here will not be linear. I do have a fairly good GPU, with its OpenGL, which may be part of the explanation for this. From this i’ve concluded that very little of the CPU usage on my system is due to display routines. I do see a dramatic difference between making both engines analog versus making them both wavetable. (I’ve got a i7 3.4 GHz, with 4 cores and max memory. And this is an extreme, in typical use, demand is at less than half this. I doubt this is even enough to be pushing the CPU to speed up it’s clock, it’s barely above idle. (Too bad there’s not a numeric percentage displayed there.) But this only shows a 9.3% usage for the Pigment application with the Windows System Manager. I got the meter on the front of Pigments all the way up to almost 25%. Definitely could push up CPU demand by making both engines wavetable, with plenty of effects and tons of notes.
0 kommentar(er)
