Stereo Tool

BETA620-027:
I performed another full build. Also made one very minor change: In some cases, extremely loud highs could be reduced too much. That should be fixed now.

WOOPS. THE CORRECTION I MADE CONTAINS A BUG. DO NOT USE THIS VERSION!

Winamp DSP plugin: http://www.stereotool.com/download/dsp_ ... 20-027.exe
Stand alone version: http://www.stereotool.com/download/ster ... 20-027.exe
VST version: http://www.stereotool.com/download/vst_ ... 20-027.dll
VST version (No SSE2): http://www.stereotool.com/download/vst_ ... 20-027.dll
Command line version: http://www.stereotool.com/download/ster ... 20-027.exe
Linux command line version: http://www.stereotool.com/download/ster ... ETA620-027 NOT AVAILABLE
Linux GUI version: http://www.stereotool.com/download/ster ... ETA620-027 NOT AVAILABLE

Question to Bojcha and others: Do you hear any difference if you increase the 'subtract' slider a bit (say to 0.05 or 0.10)? Because I can't really say that I hear anything - but it does reduce the clipping quite a bit and hence should give cleaner highs.

Also, in the recent versions some extra filtering has been added due to which even a multiplier of 0% sounds _way_ better than before this new filter was added (on top of everything else, in every processing step the highs are clipped between -32767/+32767).

I'd like to reduce the effect of this new filter as much as possible, because in the end it causes (some, mild) distortion in the high frequencies.

So I would prefer:
- High value for Vib - 1.0 is high enough there!
- High value for Subtract
- Low value for Multiply
- Low value for Maximum


Fixed versions:
Winamp DSP plugin: http://www.stereotool.com/download/dsp_ ... 0-027A.exe
Stand alone version: http://www.stereotool.com/download/ster ... 0-027A.exe


Edit: Just setting 'subtract' to 0.05 already has a big impact on the amount of clipping that occurs.


Edit #2: Subtract to 0.08 and Multiply to 400% gives roughly the same result as the current settings, but with less clipping if the highs don't have that much volume effect.
Edit #3: Same with Subtract at 0.12, Multiply at 500% (new version will allow for this).
Edit #4: But the volume level of the lows IS affected by this. So keeping the default setting is probably better in the end.

Here's another track for testing S sounds and distortion: Christina Milian - When You Look At Me

After doing a lot of tests I'm going for:
- Subtract: 0.05
- Multiply: 360

For loud highs this behaves the same as Bojcha's settings, for softer highs or less effect on the lows the clipping is slightly reduced.

Found another track which is causing problems with highs.
The beginning of Britney Spears - Till The World Ends (Single Version)
http://processing.jongerenwebsite.net/a ... s_test.wav

Edit: But new highs protection doesn't seem to fix it. Even with extreme settings.
The only way i can get rid of the S sound distortion is lowering the De-esser quite a bit. But then i loose too much highs.

Edit: Tried this track with breakaway and i have same problem.

This filter is so much better then old De-Esser...
Old De-Esser works like ... De-Esser. This new filter is not lowering Highs almost at all, at level domain.
New default settings are good (beta27A) Not sure with this "substract" I have feeling that it's bit better at 0.01 - 0.0.5, but at 0.00 is good and more safe 'i think'.

I just tryed to tweak more mid-highs and highs in B-Town preset (comparing to O.11), seem new filter handles more highs comming from MB.

Bojcha,

In your preset, why did you set 'Limiting strictness' to 0.0%? That does help the audio quality, but causes spikes in the end result which need to be lowered using a slower responding filter - hence causing volume effects.

oops.. not overviewed.. will be fixed, i just checking "version4"

Ok... Because with that setting, Hard Limit has to take over...

Setting it too high won't sound good either, but SMALL values have very little effect on the audio quality. Say 5-10%.

Ah, ok. I'll take a break then. I was just preparing the processed files for uploading...