Stereo Tool

Using and testing ND at stream..
Without ND cpu usage around 60%, with ND 75-80%.
CPU: Itel E5700 not OCed. - 3GHz.

BETA725-002B has me back in the game with Composite Clip Strictness at 5 (down from 19 w/out ND), Quality at 75% & refresh at 40%. The previous BETA would not allow ND function no matter how much strictness/quality was sacrificed. Hoping for further CPU efficiency gains. Using a very open preset, the loudness audibly increases when ND is engaged. Intel(R) Core(TM)2 Quad CPU, Q6700 @ 2.66ghz.

Yes, but if you properly tuned AGC and MB amount of loudness should be same + better/more dynamic sound

Appreciate it Bojcha...just got it running...lots of tweaking ahead

I had planned to do some work tonight but there's a strong thunderstorm over my head and I don't want to risk my laptop on it.

@brian: if you want to get a "difference" recording for your creative card, here's how:
Connect output to input
Recorx the same thing twice, once with and once without effects
Open both recordings in a wave editor, take one channel of each and creatd a new waveform where you align the 2 manually.

This alignment probably won't be perfect, use fhe AZIMUTH correction filter in Stereo Tool to fix that.
Then subtract the 2 channels in a wavr editor again.

Damn! My roof is leaking at multiple places! this is probably the heaviest rain in years!

Batten down the hatches!!

BETA725-003:
Windows Winamp: http://www.stereotool.com/download/dsp_ ... 25-003.exe
Windows stand alone: http://www.stereotool.com/download/ster ... 25-003.exe
Windows VST: http://www.stereotool.com/download/vst_ ... 25-003.dll

I have changed the first detection step (what is done BEFORE Punch Protection). At first it sounded completely different from the original filter, I've played a bit with settings and more or less replicated the same sound with the new algorithm. Not sure if that's good or not yet, but I think there are a few advantages here.


What I changed:
[*] Old method calculated a 'boost' value and took the square root of it.
Result: Softer sounds will get a bigger boost. I made this before I noticed that I really need to boost 'punchy' sounds that usually aren't soft.

[*] New method calculates (boost * NMul) - NSub. So you can increase NSub to stop softer sounds (mainly voices etc.) from being boosted at all, and you can increase or decrease NMul to boost sounds more or less. The resulting value is limited below NMax. (NMax=1 = no boost at all).

- I have also made some small changes to the behavior at big volume changes.
- And smoothing to the right (volume dropping back to normal after boost) can now be set at much lower speeds, defaults are also much lower. This may help reduce effects when the volume goes up and down really fast which can happen during choruses.

I've just added the stand alone version to the post above; VST will follow tonight.

that natural dynamics, is a nice feature or how to call it

now i'll hold listeners on my stream!