Stereo Tool

I checked version 6.10 there is also the same problem. I then tried version 6.00 there the problem is gone.

BETA039:
Stricter declipper behavior: In many cases the analysis algorithm can now determine for samples above the threshold that they are not clipped at all. So far less samples should be rejected now. This can be controlled using a new checkbox; if you notice that in some cases the declipper is less effective, please let me know, that would indicate a problem with the detection mechanism. (I don't expect this to happen, but you never know).

Edit: There still is a problem here - in certain songs the audio isn't restored properly with this filter on (the Metallica track among others). For MOST tracks though it works fine. Will have to check what's wrong.

I still want to make this filter smarter, but not for version 6.20.

From the previous beta (still valid):
The CPU load is also pretty high because I'm calculating 32 different tilt levels to find the best match (so automatic tilt detection works, but it needs serious optimization).

New slider: "Declipping margin" - if you set this to 75%, everything above 75% of the maximum level is considered to be potentially clipped.

Changed slider: "Never consider samples to be clipped if they are lower than:" - rule out soft sounds, whatever they look like, they are not clipped.


Winamp DSP plugin: http://www.stereotool.com/download/dsp_ ... 20-039.exe
Stand alone version: http://www.stereotool.com/download/ster ... 20-039.exe
VST version: http://www.stereotool.com/download/vst_ ... 20-039.dll
VST version (No SSE2): http://www.stereotool.com/download/vst_ ... 20-039.dll
Command line version: http://www.stereotool.com/download/ster ... 20-039.exe
Linux command line version: http://www.stereotool.com/download/ster ... ETA620-039 NOT AVAILABLE
Linux GUI version: http://www.stereotool.com/download/ster ... ETA620-039 NOT AVAILABLE

TODO:
- Fix loading changed multiband frequencies
- Add buffer and filter for SCA output (SCA1 ok, SCA2 ok)
- Finish AGC improvement - make mono value configurable (replace checkbox by slider) 1 hour --> NO, not needed - anything else needed? -> NO
- Fix Punch
- Check what to do with new filters (such as bass AGC) - keep them, remove them, change them? --> KEEP
- Save new BASS_AGC setting in VST mode
- Loudness: Annoying cracking sound in bass. Slightly present in 5.00, worse in 6.00, maybe even worse in 6.10. Only when bass is too loud. Much worse than in Final Limiter (at same input level!) - so this clearly indicates a bug. Most likely cause: The filter that was added to remove bass artifacts....... - No, it's the louder bass. But it can be fixed by changing some settings. Default settings updated, and behavior for 'not Very strict' improved. Also Deep bass boost and Very deep bass protection are enabled for latency 512 now.
- Fix crash at program close
- Fix VST plugin version (does not run)
- Dynamically drop 'Allow louder highs, even if it causes vibrations' to 0 when bass filter suspects noticeable voice vibrations. 1-2 hours
- Reduce Loudness CPU load days?
- Check and remove static variables
- Finish new de-essing filter (check what to do with the settings, remove at least some!) 1 day
- Convert Multiband input to MONO, then use arrays [2][4096] --> should give speedup. - FAILED
- Finish declipping filter (clipping level detection + level reduction in dB). 2 days
- Update presets? (BASS_AGC etc.) 1 day
- Finish blind interface
- Add AGC start level
- "Test Right Channel" in "FM Transmitter Calibration" does not work since v6.10 (standalone) (bojcha)
- Channels L/R swap in stand alone version when changing filtering/quality (?) (eldoradofm)
- Change version number 1 hour
- Release 1 hour
- Add lowpass filter for stereo signal (will cause a lot of extra latency!) - it might be possible to avoid this latency using a Hilbert transform
- Add smarter clipping detection. Maybe something much simpler suffices: Current clipping detection with threshold + flat line detection

I'll have to disagree here. I discovered a few years ago that I can easily hear the difference of switching the polarity of asymmetrical bass waveforms. On headphones and on speakers, even in not so great acoustics.

So it makes sense that a processor that alters the symmetry of bass waveforms should be audible, at least to some subset of subjects. Subtle? Yes. But I wouldn't say inaudible.

Hans, I'm going to start working on a preset in the next week or two.

Interesting... At the NAB someone came to the Omnia 9 booth and said the same thing (well, he didn't mention that they have to be asymmetrical!). Leif then quickly built in a phase polarity filter and both of us could not hear any difference - but I don't think the tracks we used for testing had asymmetrical bass waveforms.
Even worse: The phase rotation filter moves sounds in time. If I move them in the opposite direction the effect is really noticeable (for pumping bass the highs are played before the lows, leading to a 'tjoonk' sound (no idea how to better describe this). So yes, it does lead to differences. However, I don't understand how that would improve clipped audio (unless... the ears notice pieces of 'flat' waveform - a temporary absense of all frequencies. The flat parts will not be flat anymore because phase rotation moves different frequencies to different times, so the flat part gets filled with other data). But clipping distortion frequencies are not removed.

What MIGHT happen is that for example for 'S' sounds that are temporarily gone (they disappear on bass frequency peaks) could be less noticeable after phase rotation. Will have to test that...
Cool

Multiband still needs a lot of work, as you'll probably notice. And I have a 'natural dynamics' filter (somewhat similar, but built using a completely different algorithm based on what I think I know about what the Omnia 9 does) that I wrote and tested but still need to add. Maybe I should add it for version 6.20 - it does have a big effect on the audio...

With Phase rotator ON, difference can be heard on lower freqs. In ST when you set 'start freq' lower then default, difference is even bigger. but in most cases sounds worse.
For ST? Nice. Really want to hear that
Last days i am trying to make something like "Zenith" but seems not possible in ST.
good! D:

Good luck !
Jesse said that he is going to work on a preset.
But maybe it is a private preset ?

Passive Aggressor Stereo Tool Edition ?

Afaik bbp, that's really not possible.

BETA040:
Input waveform display displays which samples are considered to be clipped.
Made the detection more greedy (but it's currently TOO greedy, however I think I know how to fix that).

Note: For the Metallica track, set the 'Declipping margin' slider to 67%! (75 is TOO HIGH to fix everything!)

From the previous beta (still valid):
The CPU load is also pretty high because I'm calculating 32 different tilt levels to find the best match (so automatic tilt detection works, but it needs serious optimization).

New slider: "Declipping margin" - if you set this to 75%, everything above 75% of the maximum level is considered to be potentially clipped.

Changed slider: "Never consider samples to be clipped if they are lower than:" - rule out soft sounds, whatever they look like, they are not clipped.


Winamp DSP plugin: http://www.stereotool.com/download/dsp_ ... 20-040.exe
Stand alone version: http://www.stereotool.com/download/ster ... 20-040.exe
VST version: http://www.stereotool.com/download/vst_ ... 20-040.dll
VST version (No SSE2): http://www.stereotool.com/download/vst_ ... 20-040.dll
Command line version: http://www.stereotool.com/download/ster ... 20-040.exe
Linux command line version: http://www.stereotool.com/download/ster ... ETA620-040 NOT AVAILABLE
Linux GUI version: http://www.stereotool.com/download/ster ... ETA620-040 NOT AVAILABLE

TODO:
- Fix loading changed multiband frequencies
- Add buffer and filter for SCA output (SCA1 ok, SCA2 ok)
- Finish AGC improvement - make mono value configurable (replace checkbox by slider) 1 hour --> NO, not needed - anything else needed? -> NO
- Fix Punch
- Check what to do with new filters (such as bass AGC) - keep them, remove them, change them? --> KEEP
- Save new BASS_AGC setting in VST mode
- Loudness: Annoying cracking sound in bass. Slightly present in 5.00, worse in 6.00, maybe even worse in 6.10. Only when bass is too loud. Much worse than in Final Limiter (at same input level!) - so this clearly indicates a bug. Most likely cause: The filter that was added to remove bass artifacts....... - No, it's the louder bass. But it can be fixed by changing some settings. Default settings updated, and behavior for 'not Very strict' improved. Also Deep bass boost and Very deep bass protection are enabled for latency 512 now.
- Fix crash at program close
- Fix VST plugin version (does not run)
- Dynamically drop 'Allow louder highs, even if it causes vibrations' to 0 when bass filter suspects noticeable voice vibrations. 1-2 hours
- Reduce Loudness CPU load days?
- Check and remove static variables
- Finish new de-essing filter (check what to do with the settings, remove at least some!) 1 day
- Convert Multiband input to MONO, then use arrays [2][4096] --> should give speedup. - FAILED
- Finish declipping filter (clipping level detection + level reduction in dB). 2 days
- Update presets? (BASS_AGC etc.) 1 day
- Finish blind interface
- Add AGC start level
- "Test Right Channel" in "FM Transmitter Calibration" does not work since v6.10 (standalone) (bojcha)
- Channels L/R swap in stand alone version when changing filtering/quality (?) (eldoradofm)
- Change version number 1 hour
- Release 1 hour
- Add lowpass filter for stereo signal (will cause a lot of extra latency!) - it might be possible to avoid this latency using a Hilbert transform
- Add smarter clipping detection. Maybe something much simpler suffices: Current clipping detection with threshold + flat line detection

( Input waveform display displays which samples are considered to be clipped. )


yea I was just looking at that...looks cool to. makes life much easer now you can see what's been de-clipped.

It wouldn't be possible for two reasons. Even by using the altered crossover points to simulate the crossover poitns used for that in Breakaway/O.9, there's no way to properly run the multibands anywhere near as slow as I am, and still catch the sudden transients and volume increases the way that Breakaway/O.9's AGC does... thanks to the dual wideband compressors relational to the AGC to catch stuff like that. Secondly, a lot of the magic of how that's balancing the spectrum and loudness so perfectly is the density/loudness effecting the ratio. Not deliberately either, but just a combination of the slow timings and the "Infinite above 0dB" ability of any given band (selectable on each band individually).

When Breakaway/O.9 moves the psycho-acoustic density/dynamics detection stuff into the multiband, to allow the ratio to be biased based on that... WOW. Then I'll be able to have that effect but much better and more accurate... and I'll be able to have it on presets that use much faster timings. It's really going to make it possible to make some really off the wall but somehow controlled and "right" sounding presets.

---

With Stereo Tool, I'm going to go for a more technical sound. Taking advantage of all 10 bands, and yet trying to make it sound open. Probably something with high ratio multiband AGC, and very little multiband limiting except to control bass transients before they hit the bass clippers. But we'll see. It's been since v5 since I've played with it.