Stereo Tool

CPU load of this will be lower when I'm finished, I expect it will be roughly equal to that of Multiband.

Right now I'm still playing with the shape of the boost behavior (what does it do at each input level), that might be expensive.

BETA043:
Natural Dynamics filter: Highs and lows now behave equally. And if you lower the slider, before the volume dropped again (that was wrong!), that's now fixed. Unfortunately this results in a new issue: In some (rare) cases, loud sounds get a big boost, making them far too loud. So I think I'll need to limit the boost if the level is already high (but controlled, not due to a bug like before. And probably not for bass sounds.)

Apart from this, I hear some bass artifacts that need to be fixed.

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_ ... 20-043.exe
Stand alone version: http://www.stereotool.com/download/ster ... 20-043.exe
VST version: http://www.stereotool.com/download/vst_ ... 20-043.dll
VST version (No SSE2): http://www.stereotool.com/download/vst_ ... 20-043.dll
Command line version: http://www.stereotool.com/download/ster ... 20-043.exe
Linux command line version: http://www.stereotool.com/download/ster ... ETA620-043 NOT AVAILABLE
Linux GUI version: http://www.stereotool.com/download/ster ... ETA620-043 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
- RDS issue reported here: viewtopic.php?f=15&t=3703&p=11524#p11524
- Always oversample clipping (configurable)
- Finish declipping filter (clipping level detection + level reduction in dB). 1. Figure out why removal of unwanted frequencies causes flat lines at high quality setting with small overlap. This causes distortion, with this fixed repairs are MUCH better. 2. Implement MP3 correction, automatically scale down when this deteriorates the sound.
- 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)
- Move pre-emphasis to end of processing 1 day
- 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

Note to self:
New declipper ideas:
- In case of a very sharp clipping detection at a single level, ONLY mark samples at that level
- If samples above a certain threshold are marked as clipped, samples in between that are higher than those that are NOT marked as clipped might also need to be marked.

I do know this, I think it's one of the first things Hans told me in a PM about a month ago... Hopefully I can give a few ideas to improve on to Hans after I sink my teeth into it some.

Basically... assigning high-frequency bands you don't want to use into the maximum frequency.

Michi95,

I know English is not your first language, so I'll be kind, sort of.

Hans did not indicate that there was supposed to be this kind of dramatic increase with the Natural Dynamics filter. The note about CPU usage was about the Declipping.

Just a few versions ago, if we had held to your same belief, that we should all just go out and buy new computers for very small improvements in audio processing, then there would likely not have been the reduction in CPU usage that we saw in the software just a few releases ago.

In short, unless you have something constructive to say, please add me to your ignore list.

Alternatively, I will be happy to provide you with a way to send me money via Western Union so that I can waste your money on unneeded computer upgrades because you like excess bloat in your software

Thanks... I felt like a doubling of CPU load was more than what should be expected.

Beta043 links do not work...'

Edit: They work now...

If this misrepresentation of the facts is your definition of kindness, then it's clear why your circumstances are the way they are.
I allow you to ignore me too.

Links are fixed.

About the CPU load: You are both right.

Brian, your pc is really, really, really old. A P4, that's at least 10 years ago I think. I'm using some FFT libraries for better performance, and they stopped supporting anything below a dual core over a year ago - well it still works, but they stopped including very strongly optimized code for it. That's one of the reasons why I'm still using an older version of that library (there's no performance gain if I use the newer version, on newer systems). About the performance increase of the dynamics boost filter: I hadn't even noticed it. Just measured it on my own pc, and it does increase the CPU load by about 5% (in absolute numbers). I checked what it will do after optimizing it, and it looks like it will drop to about 2%. Now my pc is already 3 years old, and the lifetime of pc's is usually considered to be 3-5 years, so it's approaching the end already.

Having said all that, of course, if I CAN make it work on a P4 I should. Because it allows people to use very cheap pc's, and because it reduces the cpu load on newer ones (which also helps to keep the latency low etc.).

But there is of course a reason that for example the Omnia 9 will contain an i7 chip - which is about 20 times faster than a P4. Even my own 3-year-old pc is over 10 times faster than a P4. More processing power means that more can be done - if you choose (or are forced) to use an older pc, you might not be able to use everything. Just keep in mind (assuming that the average pc out there is about 3 years old) that your system is about 10 times slower than the average system out there...

By the way, on the Onmia 9, BEFORE I did my optimizations, the declipper used about 2-3% of the CPU. And there wasn't even an i7 in it yet - the test system had an older, slower cpu.

I made mention earlier that I didn't much care for being doubted. You're seeing that come back out, because, well, you're doubting. I COMPLETELY UNDERSTAND that there will be a need for SOME increase in CPU utilization. Where you and I will differ is if you think it's great to have a 100% increase for a maximum of 5% increase in value of a product. That is catering to a niche market, pure and simple.