Stereo Tool

Hello Hans,
Could you please take a closer look into the BS.412- Limiter again?
What I want is neither to compress the Audio (via fast Speeds in the compressor) nor to let the BS.412- Limiter to perform any frequency changes.
That means I'd like to behave it like a slow leveling - not more. And all I tried in this point lets me get into trouble.
There are some main reasons for this:

At first the compressor:
The BS.412 compressor doesn't respond really to the mpx power. When there is a song with less amount of Bass the BS.412 compressor's gain reduction is too big compared to Songs with much (deep) bass.
The compressor should respond to the very short integrated mpx power (e.g. 200ms) to give a better pre- control.

Secondly the Slow response limiter:
The base Integration time of this limiter is very long. Could you make it adjustable? Only a few seconds would be great.
The thing is that the limiter now creates much "over- reduction" especially after Level- jumps. Let's assume you have the headroom set to 0.5dB, a jump comes, the compressor reduces and also the limiter. For one Minute the MPX- Power is at e.g. 0.2dB under the limit, then for a very long time (20-30s) the slow response limiter's gain reduction grows up until the the power falls down again.
But at this Point the gain reduction is too high, so that the power then falls largely under the power Limit.
It is reduced too far when you see the 60s power.

Do you know what I mean?
The reduction should occur more asymptotically.
Following example:
You have a dense song with nearly the same structure after the Impact for longer than one minute.
The effect should be this way, that - assumed the headroom is set to 0.5dB - after the start of the Impact (slow response limiter starts to act), the power grows up to -0.1dBr, then falls down to -0.5dBr and then stay at this Level until the Input changes (new song etc...).

I am NOT able to get an effect like this with the BS.412 controller like it is now.

I would be happy if you could look into this.
Kind Regards,
Robert

Additionally i would wish to make a Switch, that puts the compressor in an other mode:
Like the MPX Guard in the well known IDT processors.
Many stations especially in austria and switzerland do their processing at the Studio site and put an IDT DEP@FM unit at the Transmitter site.
The IDT DEP@FM (Digital Encoder Plus) is a Stereo Encoder with additional functionality like:
-AGC (Level- based)
-Deviation limiter
-Stereo with limiter
-Stereo Boost (optional)
-RDS Encoder (optional fixed or dynamic via RS232- Data port)
-Streaming functions (optional stream Encoder and Decoder)
-MPX Guard (optional)

And this MPX guard is the reason for so many sold products.
The algorithm of this is very simple:
As base for the control they take a 3.2s Integration time.
And now Comes a simple if/else relation:
If the 3.2s integrated power is bigger than the target, the gain reduction grows up with 0.75dB/s.
If it is under the target, the reduction gets down with 0.5dB/s
The speeds are fixed for power level above 1dBr below the target. If the power Level (3.2s) falls below this, a gating function kicks in, that means the upspeed is slowed down.
This occurs via a linear function, so at -1dBr there are the full 0.5dB/s until it stops at a level of approx. -8dBr.

Also a Simulation with matlab or octave brings you the desired "flat line" at the Output.

It would be great if the compressor would be able to work this way (like your new compressor is able to work, but based on mpx power):
-fixed up and down Speeds (like your example with the door, where you push but it doesn't close any faster)
-an adjustable base Integration time (e.g. to be able to set it to 3.2s)
-an adjustable gate threshold and slope

Kind Regards,
Robert

Good point! I'm also looking for something like this in the BS.412-section...

No response here?
I could imagine that we (radiofreak and me) are not the only one who would like to see another behavior in this...
Kind Regards,
Robert

Hi,

You're not, and I'm going to look into this later this year.

Hi,

it also sometimes (but fortunately rarely) happens, that the gain reduction does very steep and very quick dips, e.g. if a song with very heavy bright mastering and deep sub-bass is played between two average songs. Would be great if such fast movements could be smoothed out, as they sound like an unwanted ducker-effect.

Plus, it would be great if you can set the MPX-power-limit to a "soft" or sloppy-mode by choice, so that small overshoots of 0,1 or 0,2 dBr are allowed in extreme cases (as described above), because in many countries, the regulation for MPX-power is also soft and allowes a margin of +- 0,1 ... 0,2 dBr.

Setting a bigger headroom will fix the steep drops. The limiter soms at staying at that headroom level, so you could slightly increase the target dB level to get the occasional overshoots.

Hello radiofreak,
In Addition to the Response from Hans i have also some Facts that maybe interesting for you.
-Try to let the BS.412 compressor and its bass- part to do the most of the work, the slow Response limiter shouldn't do more than 1dB reduction.
-The naming of the debasser- slider "pumping vs. bass reduction" is reversed, so 1.00 means wideband, 0.00 means bass reduction only, I prefer 1.00 to minimize the Sound Change of the BS.412 section.
-The time constants of the debasser are based on the BS.412- compressor speeds.
-If you see the steep gain reduction Spikes in deat you have to increase the headroom, I suggest not to go under 0.25dB.
Regards,
Robert

Thanks Hans and Robert, i've fixed it by increasing the headroom (was at 0.05 dB only ). Soft limit is also working by increasing the MPX Max Power Limit by the number of headroom

There was a bug in the current BS412 encoder: It didn't work properly in some cases for non-MPX output. I fixed it in 784-BETA002. If any of you is using it without generating MPX, can you check if this version works better? (It should!!)