Stereo Tool

@ everyone:

1. I hope version 6.20 will be ready within a few weeks (I took 8 weeks of holiday to work on it - should be able to finish it in that time! Not sure though if the Natural Dynamics filter will be part of it; I might release something as soon as I'm happy with the declipper.)

2. The clipper is in the Loudness filter. That's actually a combined clipper and volume boost - which might be odd (grown this way historically, but maybe I should change that in the future).

3. What Leif has is a composite clipper, I want to make one of those but I don't have it yet. His clipper maintains transients and punch better than mine, which is probably at least partially explained by the fact that it's a composite clipper. By the way, 140% modulation doesn't mean that you can reach that level constantly - it only works for highs and stereo sounds, and not always, so in effect you gain less than that (but still a lot more than using my own 'FM Overdrive' which should be changed to a composite clipper). 140% is approximately the peak level that can be reached when the pilot is on one side (9%), and the signal and stereo signal are pointing in different directions.

Actually, I might even get away with keeping my current clipper, removing HARD LIMIT and only replacing the FM Overdrive section with a composite one... But completely composite clipping will give better results than that.


Oh, and when I *have* composite clipping, stuff like SSB etc. is really simple.

2. And that's why is aksed. Do we talk about L/R Clipper or Composite. What do you compared is certainly not composite "clipper". So, it is L/R clipping. But again BBP has much better L/R clipper

3. First L/R - then composite.


De-CLipper: What is can hear only problem is soft-highs-on-loud-lows get smashed and distorted.
Maybe naive idea, but since de-clipper is anyway not good for restoring higher frequencies, maybe soulution is to just remove higher frequencies from de-clipping.. at least part of it.

Personally I would love a composite clipper in stereotool.
I think it has some major advantages in all fields. Louder sound while still maintaining the same bandwidth is the biggest ofcourse.
No software, except for Leifs, has done this yet. I actually wonder how his thing works with a mono input signal like voice.

2 - Leif said that he prefers mine over what's in BBP, because the voices are clearer (at least in the one track that we compared). His is better for transients, which is something I really need to look into (I think I know what's causing it though). The Omnia 9 clipper is a completely different story - it has clean voices and transients, and he's going to put that one (not composite though) in BBP soon.

Actually I think I need to start with my clipper from scratch, using all the things I've learned in the past. The current code has gotten so complex that I'm having more and more troubles even understanding what's happening. And making changes is even worse.

3 - Declipper: The declipper is actually good at restoring highs, but not at restoring unpredictable highs (hiss etc.). If I don't restore the highs at all, the higher harmonics will still be there. But I have found a new method to (probably) much more reliably detect clipping - working on it right now. This might fix the whole issue, and I hope it will even allow me to (somewhat) distinguish hard clipping from 'filtered' clipping like Leif and I are using.

Yes, sometimes it's better too just start from scratch.
I guess with a mono input you can't get higher then 100% audio peaks.
Probably only when you have the stereo encoder enabled you can get higher peaks.

Well, if both stereo and RDS are switched off there's just nothing composite in the signal

For mono sound with stereo pilot + RDS you might still be able to get upto 113.5% modulation (assuming 9% pilot + 4.5% RDS level).

Yes instead of "hard limit" a clipper composite a 75 khz

7 Khz pilot
3 Khz RDS
and 65 Khz Audio

it is very nice if it is possible !

Because actualy they are little overshot up to 75 Khz...

I have tested more sound card, the best actually is Asus Xonar Essence STX at 192 Khz

That's problem.. What is Clipped there !? if there is any clip inside 'loudness' that's about 5% and 95% you can more call it "limit".
When high Punched track hit loudness filter, it's doing limit .. and when you same try in BBP it's unchanged and clearly goes into clip. (depend on preset settings)
Proof: Just try "Rustonium" preset with some bass kick track, like "Jack is Back" (no FM mode)
m? First what we talking about here ?? We need to talk about De-Modulated Output, which is upto 140% in Omnia.9 and in Stereo tool about 92%, without composite 'limiter' and safely can reach no more then 100% with composite 'limiter'

Again..
In Omnia.11 demo you can clearly see and hear brand new clipper how nicely punching at whole band, and again. De-Modulated level is not more then 100%.. meaning.. L/R clipper is more important and first thing to be properly done.

Edit:
Just checked: With Stereo OFF and RDS ON, Audio Level is 96% (96.2%), with Stereo ON Audio level is 90% (~90.2%)

What ? ... instead of "hard limit" ????
Nooo , no way .
Hard limit is doin his job very well . And another thing .. For what the hell is this over 100% modulation . New laws in FM modulation we make ... ? Again NOOO . Laws are laws , don`t forget the receiver . For transmitter there is no any problem to make 1000% deviation , But receiver is made for +-75kHz deviation , which means 100% of deviation .... Above is dead ! Guys who want more than 100% , don`t forget it !

And let me tell ya , I test ST for a long time ago and no predeviation yet .no hicups , no clicks , no anything ... just what I ever dream . Come and see other stations if You don`t believe me.Use expensive Stereo generators with all kind of sound processing .....and...... ST beat it all.

We are talking about 75 kHz deviation / 100% composite modulation.
Only demodulated and stereo decoded audio with higher levels. Try the omnia 9 demo in mpx tool to see what we mean.