Stereo Tool

If you are a professional , commercial radio station tech guy then you probably know that even with GPS syncronized fm exciters (plls ) digital or analogue... you HAVE PROBLEMS when broadcasting from 2 mountains at the same frequency...

a new kind of "trick " technology came and it is compatible with old fm radio recievers!! ( it's just a little bit of DSP code!! )

My question :

Can you ( hvz) make some extra steretool code.. that makes this ?
SSBSC Single Sideband Suppressed Carrier
http://omniaaudio.com/downloads/white-p ... -Paper.pdf

http://www.youtube.com/watch?v=KlxeI2uqCOw

Yes. But I need to have a composite clipper first.

If you don't need a loud sound (so this is not for you, lol): In the FM Calibration settings you can actually already enable SSB mode (just enable "Similar, but only using 23-38 kHz"). But you'll get huge spikes in the MPX signal, so you'll need to lower the volume - a lot.

huh, no

Ow, that needs some improvements indeed

The sound should be ok though.

yes, but ofc one channel only..

* moved to 'wish list' section

In some earlier betas it was indeed possible this way with this checked to have SSBSC.
But that was a bug and Hans fixed that. So now that's not possible anymore.
So indeed the solution is a new composite clippper.

O lol I just forgot that it's one channel only.. Yeah that would make it kinda unusable.

I don't know yet when I'll make the composite clipper, but I should at least warn you that it will require a lot of Processing power (reason: I need to clip at 4 times the MPX sample rate!!!). The Omnia.9 has an i7 cpu, my composite clipper will probably require something similar.

I have taken a quick peek at what I would need to do to create a composite clipper.

Conclusions:
- Converting the current L/R clipper to a composite clipper is very simple. Would probably take 1-2 days. All the protections that are built in would still work after the conversion (until yesterday I though that that wasn't possible, but I just figured out a very simple solution for this).

There is another problem though.
- The current code contains a lot of annoying stuff to split the processing over multiple cores, and to combine processed data. Rewriting this code is difficult, and will take much more time.

An alternative would be to place the composite clipper *after* this code, but that would double the latency.


At least I know now what I need, and whenever I need to make changes to the code I'll make sure it moves towards the needed setup. I've actually already moved several parts of the code today to make a future move to composite clipping easier.

You need to replace current "composite Limiter" (?)
Problem is only if that is also part of loudness section.
Alsom, you need ~1.3kHz and up. under that keep to loudness peek level, i guess you know that already

The composite limiter will probably have to stay as well (reason: Clipping is not 100% perfect, so something is needed to remove the small peaks that remain, similar to HARD LIMIT after Loudness).

I know that I won't get bigger spikes on all frequencies, but for the algorithm that doesn't matter - it will be *exactly* the same algorithm that I use now, except that it operates on an MPX signal (and hence on different frequency ranges) instead of the normal signal. (And of course that the stereo pilot and RDS signal may not be changed by it. But that's the easy part - it works already for the composite limiter).