Stereo Tool
https://forums.stereotool.com/

Low latency secondary input (microphone)
https://forums.stereotool.com/viewtopic.php?t=853		 Page 42 of 75
Author:  hvz [ Thu Feb 25, 2010 1:52 am ]
Post subject:  Re: Low latency secondary input (microphone)
NOT USABLE FOR PRODUCTION. THIS VERSION HAS AN INSANELY HIGH BASS LEVEL WHICH CANNOT BE CONTROLLED (YET).

Links:
Winamp DSP plugin: http://www.stereotool.com/download/dsp_ ... BETA3A.exe
Stand alone version: http://www.stereotool.com/download/ster ... BETA3A.exe
VST version: http://www.stereotool.com/download/vst_ ... BETA3A.dll

DO NOT USE FOR PRODUCTION.

Bojcha, can you check if the bass level is high enough with this version (it's probably way too high), and if the quality is 'tolerable'? The bass level might be far too high (if it is, I'ld like to know that too...)
Author:  hvz [ Fri Feb 26, 2010 12:16 am ]
Post subject:  Re: Low latency secondary input (microphone)
In an hour I'll upload a new version which should be a bit better. Instead of using 2 steps (way too much) and performing the bass boost before multiband, I'm now using a single step, and performing it after multiband.

In this version I'm only using bass boost on the left channel, to enable easier comparisons.
Author:  hvz [ Fri Feb 26, 2010 1:58 am ]
Post subject:  Re: Low latency secondary input (microphone)
Available (Left channel only!!). Tomorrow I'm going to check if there's a better way of doing this.
Author:  Bojcha [ Fri Feb 26, 2010 7:33 pm ]
Post subject:  Re: Low latency secondary input (microphone)
hi Hans ..
this is actually near same as EQ after Multiband .. except there is a little soft bass clip with this now which is good but again not controlled good.

usefull:
http://www.cgould.com/index.php?option= ... &Itemid=37
http://www.vorsis.com/whitepapers/White ... S_Rev2.pdf
Author:  hvz [ Sat Feb 27, 2010 12:40 am ]
Post subject:  Re: Low latency secondary input (microphone)
Hi Bojcha,

Apparently there are 2 different things that are both referred to as bass boost... Which doesn't really clarify things.

The 2 things are:
1. *ALWAYS* adding higher frequencies to improve bass sounds on cheap speakers, or to make the bass sound 'warmer'
2. *CLIPPING*, then lowpass filtering the bass to allow higher peak levels without introducing (too much) distortion.

The first one changes (deforms) any bass shape that goes in.
The second one only changes loud bass sounds. The louder the bass gets, the more the peaks are 'flattened'. The overall bass level can be increased in this case, because the peaks are less high. For low bass levels there's no change in the sound, for high bass levels some harmonics are generated.

What's needed here is #2.


(follow-up coming soon)
Author:  hvz [ Sat Feb 27, 2010 2:47 am ]
Post subject:  Re: Low latency secondary input (microphone)
Follow-up: I'm starting to think that in order to do this I may have to combine the first 3 bands of the multiband compressor into a single band. That would make it possible to determine the maximum volume of the bass. In fact, I wouldn't have to do anything special afterwards anymore, because I could then use the existing multiband clipper for the combined bands to clip the bass. The combination of increasing the equalizer and lowering the clipping value would then correspond to bass boost...

Before I take rigorous actions like that, I'll post a new test version. This time the 'expected' maximum volume is based on the Multiband settings.

(Edit: Yet another possibility would be to change the clipping frequencies for the first 3 bands in the Multiband filter. For each band, I know exactly what the "normal" maximum and the peak level is. As an extra bonus, different frequencies would not so easily get in each others way, which may make it possible to clip more aggressively without getting annoying effects.)

Before I assumed that the amplitude of the bass after Multiband would range between -100% and +100% of the total waveform. But as it turns out, for _most_ sounds, if the audio behaves 'neatly' as configured in the first 3 bands of the Multiband compressor, the bass stays roughly between -50% and +50% (based on FM Loud Bass preset). Which explains why there was almost no change in the shapes of the bass waveforms - most of the time that stayed below half of the range, only sudden spikes got much higher.

In the new version the expected peak level is no longer set at 100%, but determined based on the multiband settings of the first 3 bands. This should lead to much bigger changes in the shapes of the bass waves.

(In some cases, far too much clipping would occur, which in fact reduces the outgoing bass level. If that happens, the remaining part of the original bass sound is merged back into the end result, so the resulting bass level is ALWAYS higher than the original).


Note: Single tones in Stereo Tool will _never_ reach high volumes (at least not while I'm using 3 bands in the bass range), and hence should not be used to test this! Please use your ears on real audio instead... I could fix that, but I expect that the sound will be better if I don't. If you don't agree with this statement, PLEASE let me know.

Update is being built now, will be here in an hour.
Author:  hvz [ Sat Feb 27, 2010 3:24 am ]
Post subject:  Re: Low latency secondary input (microphone)
Update is available.

Left channel: Bass boost
Right channel: Normal
Author:  Bojcha [ Sat Feb 27, 2010 5:07 am ]
Post subject:  Re: Low latency secondary input (microphone)
Actually this is good.
if i measure right, Bass beat-sound has same (sometimes a bit higher) peek level but more RMS power !?
So this bass boosting can be reduced a bit. In same time this clipped bass, loved by many, but unnatural, will be less unnatural.
Image
Quote:
Note: Single tones in Stereo Tool will _never_ reach high volumes (at least not while I'm using 3 bands in the bass range), and hence should not be used to test this! Please use your ears on real audio instead... I could fix that, but I expect that the sound will be better if I don't. If you don't agree with this statement, PLEASE let me know.
No need if this works good. i think it will be enough.
Author:  hvz [ Sat Feb 27, 2010 5:19 am ]
Post subject:  Re: Low latency secondary input (microphone)
Great :D

I'll still make it configurable, so you can set it between no bass boost and this (the maximum) level.
Quote:
if i measure right, Bass beat-sound has same (sometimes a bit higher) peek level but more RMS power !?
Well, that's the point of changing the shape of the waveform :-)


Apart from lowering the Bass Boost strength, are there any other changes needed?

By the way, I have compared the filter that I just created with what some other processors do, and my version is a bit different. I expect/hope that it works better (more natural, less sharp change between normal processing and clipping, probably less harmonics).


Oh, about unnatural: I'll also make the clipping level configurable. So you can choose to set it a bit higher to get higher peaks with less clipping. And change the frequency range etc. (currently hard coded at 100% at 0-100Hz, going down to 0% at 300 Hz).
Author:  Bojcha [ Sat Feb 27, 2010 5:35 am ]
Post subject:  Re: Low latency secondary input (microphone)
Quote:
By the way, I have compared the filter that I just created with what some other processors do, and my version is a bit different. I expect/hope that it works better (more natural, less sharp change between normal processing and clipping, probably less harmonics).
fast check.... Actually yes :mrgreen: Cleaner ;)

But is still think bass boost can be reduced a bit.
Quote:
Apart from lowering the Bass Boost strength, are there any other changes needed?
If Pre limiter and loudness can handle this enhanced Bass Peek level then OK.


hmm .. how many bands has Final Limiter? 2? :oops:
Page 42 of 75 All times are UTC+02:00
Powered by phpBB® Forum Software © phpBB Limited
https://www.phpbb.com/