Stereo Tool

So, this means that creating a high CPU load in a SEPARATE ST instance still causes clicks in the first... That would mean that offloading the processing to a separate process won't help either.
And it gets weirder: It's also not caused by the drawing. So what else can I be doing that is so different from what any other application does?!

Brian: For what you're doing, it would actually make sense to have a lowpass filter at the end of the processing instead of (or on top of) at the start. I should warn you though that doing so will cause BIG spikes in the output, so you'll have to reduce the output level a lot, probably by halt of so.

About the GUI update thread: I'm now moving it to priority "BELOW_NORMAL" instead of "IDLE", please let me know if that works better for you (will be in the next BETA), otherwise I'll move it back to NORMAL and see if I can keep it on a separate thread or not.

Actually, this is very interesting. I'm now building a 601-BETA030B version where I've not only turned off sending the drawn bitmaps to the video card, but also drawing them in memory.
- If that solves it, I'll try what happens if I add a command to hand control back to the Windows scheduler after every single bitmap that I'm drawing. That would make the 'burden' of the GUI thread even lower than currently at full screen mode.
- If that doesn't solve it, it means I'm looking in a completely wrong direction, and that I need to check the code that handles Windows events instead, or maybe even how Windows itself updates its display before control is handled to Stereo Tool...

Here it is:
http://www.stereotool.com/download/radi ... TA030B.exe

Tested BETA030B when set to normal or realtime both times still some clicks with buffer of 2.8 ms. And moving the task manager screen around or the stereotool screen. Will do some more tests later.

Busy weekend...

First test..
64samples-NormalOut-48kHz
- Beta27, 1.7ms minimal buffer size, High Priority, still clicks
- beta30A, 2.0ms(?) minimal buffer size, Less clicks is if i move ST window, but it's still there.
So problem is not only ST window...

Let's see b30b...
EDIT:
- beta30B, 2.0ms(?)minimal buffer size, High Priority, much same as beta30A
Same is with MultiCore On or Off. Click free is at ~10ms, but i guess on slower PCs will again be clicks.

I think you meant highpass and not lowpass, but at any rate, I'm mostly ok with where it's at now with the preamp adjustment. The only downside here is there is becoming little use for the AGC. That might not even be a "downside". I just don't know. I'm muddling my way through this on some basic knowledge from the last time I was involved in stuff like this, which was 20 years ago. You're younger, so you know what they say about young minds and fresh ideas...

I guess though my issue is that since this is all software and not any actual circuitry doing the magic, why it is that filters after a bandpass do not know about the bandpass variables? In the Java world, there'd be your various "getters and setters". So, if I set a highpass of 50, why, for example, in Bass Boost, would you attempt to boost the signal at 20 if I already said that I wanted to be attenuating that area?

Of course, it's possible that I don't understand things...and am making a silly argument.
Like I've said before, you have to go with what works for the bulk of your audience. I'm kinda hanging out here as a hobbyist...

I managed to get ASIO working on my Juli@ card again - but so far I'm unable to reproduce any clicking behavior, even at latency 1 ms... Maybe that's because a have a quad core system...

I did notice that something is very wrong with the beep tones in the FM calibration settings - they cause horrible clicks at latencies below something like 40 ms or so. I'll check what's wrong there.

LESS clicks in 30A - that is good news... If you have none in 30B I think I have a solution...


Edit: Oh, eldoradofm already reported that there are still clicks

btw..
BA ASIO have 2 processes, but when "real time audio" is On in software. When is OFF, there is One process, but there are still no clicks in sound.. even if i set both or one process to Low (from task manager) and even if i set Affinity to ONE core.. still no clicks... blah

That then leads to a version 601-BETA030C, where I'm not doing ANYTHING anymore when something happens in the display - note that you won't be able to configure anything in this version, and need to kill it to close it...

http://www.stereotool.com/download/radi ... TA030C.exe

If this still causes clicks it must be the Windows window handling itself...


Edit: Upload to server was stalling... File is there now.