Does Satin's auto-mute save CPU?

Official support for: u-he.com
Post Reply New Topic
RELATED
PRODUCTS

Post

I am keen on using Satin, and am wondering if the auto-mute function saves CPU usage when no audio is being fed into Satin?

Post

auto-mute only affects hiss/noise, therfore: no.
Sascha Eversmeier [formerly digitalfishphones]
TOURAGE DSP
croquesolid drum processor- mix real drums fast & focused

Post

sascha wrote:auto-mute only affects hiss/noise, therfore: no.
Yes, but I thought that vst3 spec allowed the plugin to use 0% cpu load if no audio is passed through it? In the case of satin, I would have thought it would still use full cpu if it is generating hiss/noise, therefore surely vst3 satin + auto-mute SHOULD in theory = yes?

Post

well, all VST3 (as well as AU since 2001) does is tell the plugin that it's processing silence.

That alone however isn't enough to put a plugin to sleep. For anything but the most simple plugins it often involves a lot of work to determine whether or not a plugin can save CPU. Most obviously with delays and reverbs that still sound even if the input is silent. But in practice this accounts for anything containing filters, envelopes or any other feedback driven device.

In case of Satin I'm not sure if there's a way at all to determine the point of "ok, let's go in standby". We would probably have to add another "auto standby" mode to it, adding more complexity and possibly more CPU when not in standby.

Post

Urs wrote:well, all VST3 (as well as AU since 2001) does is tell the plugin that it's processing silence.

That alone however isn't enough to put a plugin to sleep. For anything but the most simple plugins it often involves a lot of work to determine whether or not a plugin can save CPU. Most obviously with delays and reverbs that still sound even if the input is silent. But in practice this accounts for anything containing filters, envelopes or any other feedback driven device.

In case of Satin I'm not sure if there's a way at all to determine the point of "ok, let's go in standby". We would probably have to add another "auto standby" mode to it, adding more complexity and possibly more CPU when not in standby.
Thanks and sorry for the late reply, Urs, I lost my laptops hard disk last week and had to do some maintenance :/ Just bought my licence now, it's absolutely incredible, and thanks for taking the time to explain the situation with vst3. If it COULD be implemented whereby Satin could detect audio going in, and do a dynamic standby based on that without taxing the cpu in the detection system, I think you could be onto a perfect winner with Satin. At the moment, envelope bypasses are the best I can come up with to save on cpu juice, if Satin could do this thinking for me, it would be the finishing touch I think, especially with the vst3 version having then something to stand-out from vst2 from.

Once again, thank you for Satin, I often think about how much work and talent it went into creating this, Satin actually makes me think and feel like I'm working with analog, and my possibilities seem expanded so much further than they ever could be without it!!

Post

Urs wrote:well, all VST3 (as well as AU since 2001) does is tell the plugin that it's processing silence.

That alone however isn't enough to put a plugin to sleep. For anything but the most simple plugins it often involves a lot of work to determine whether or not a plugin can save CPU. Most obviously with delays and reverbs that still sound even if the input is silent. But in practice this accounts for anything containing filters, envelopes or any other feedback driven device.

In case of Satin I'm not sure if there's a way at all to determine the point of "ok, let's go in standby". We would probably have to add another "auto standby" mode to it, adding more complexity and possibly more CPU when not in standby.
Thanks and sorry for the late reply, Urs, I lost my laptops hard disk last week and had to do some maintenance :/ Just bought my licence now, it's absolutely incredible, and thanks for taking the time to explain the situation with vst3. If it COULD be implemented whereby Satin could detect audio going in, and do a dynamic standby based on that without taxing the cpu in the detection system, I think you could be onto a perfect winner with Satin. At the moment, envelope bypasses are the best I can come up with to save on cpu juice, if Satin could do this thinking for me, it would be the finishing touch I think, especially with the vst3 version having then something to stand-out from vst2 from.

Once again, thank you for Satin, I often think about how much work and talent it went into creating this, Satin actually makes me think and feel like I'm working with analog, and my possibilities seem expanded so much further than they ever could be without it!!

Post Reply

Return to “u-he”