Stereo Tool

those like 0.0000000x

or in compressor upspeed/downspeed (BTW, is this technically correct name?), where 2 digits after decimal should be OK as well.

Regards,
SB

Speed is measured in meters per second or m/s , so Speed in audio is measured in DB per second or dB/s ?! What is your wonder about that ?
I think that this must be the unit for the Multibands too.

Upspeed/downspeed term, not the units dB/s per se



Where we read, that dB/s or dB/ms is for attack/release rate. There's no such technical term as compressor's upspeed/downspeed

Does not happen in beta5.

_________________
control point
control point2

BS412 is apparently still using some very old compressors, which I made before I knew (yes, really) what Attack and Release was.

Chances are that I'm going to redesign that part soon anyway.

Ok, I'm making BETA005 an "almost-release-candidate", tomorrow I will update the presets and release it.

So, crash with the buffer size will not be fixed ?

Also , AGC will be changed ? That means no more Up Speed , down speed and what else ?

I think Slawomir B. means also the many digiots behind the comma on some values, including the AGC/BS.412-Compressor up/downspeeds...
There I would agree, 2 Digits are enough I think...
Regards,
Robert

Yes (AGC etc. less digits) - but not now.

Buffer crash is most likely not new, it only happens when you're adjusting the buffer size (so not when Stereo Tool is playing unattended), and it happened once when Bojcha was constantly testing a buffer/hiccup bug that I introduced in 7.80. Which makes it less important than other issues. It could potentially even be a sound card driver issue (a while ago, adjusting the buffer size could cause BSOD's on some systems, and a 'dead' sound card on others, which turned out to be a driver issue. I ran a test which constantly opens and closes the sound card, and that test showed that the driver basically 'dies' on average once every 1000-2000 times. (That was the ESI Juli@ driver). To solve that, I put in some anti-hysterisis code that only closes and opens the sound card a few times per second instead of on every key press. I haven't heard about this problem since - but it's not unlikely that this is that same issue.

By lowering the number of times the sound card gets opened/closed from on every key press (basically the key repeat speed of the keyboard, usually 18-50 times per second) to about 3 times per second. Which means that now you need to hold the key for about 5 minutes to get a single crash, before that was in worst case (50 times per second) 20 seconds.

Since this is a driver problem, it's not something that I can fix - I can only attempt to work around it. I don't know if this driver crash is caused by the same thing, but since we're planning a complete sound card handling rewrite anyway, I'm not going to look into it now.

Yes, exactly this. Thank you.