Limiters - Finally understood

[sine143]

Explain your reasoning byacey. If 1 ch can supply the neccessary voltage to be limited, how would using 2 channels be any different.

and escape. I still maintain you need to be limiting via peak voltage, not rms. what reasoning do you have for converting rms?

[Grant Bunter]

Explain your reasoning byacey. If 1 ch can supply the neccessary voltage to be limited, how would using 2 channels be any different.

For my thinking:
The absence of the other channel ie not a summed signal so 3dB down? Is that it?

Of course, this is about to become moot as a DCX is on the way. So L+R inputs can then be used, still run in mono but sub input summed. Since all outputs on the DCX are assignable, you could end up with 3 sub outputs and three mid high outputs.
Daisy chain 2 amps on every output for 2 cabs every channel and you'd have a rather large system, but I digress...

[byacey]

It's not a matter of limiting, or 1 or 2 channels, but of reproducing the sub levels as they were recorded.

If you run stereo, but only drive the subs from one channel instead of L+R summed, you lose 3db of level.

Here's some info on RMS / Log conversion for sidechain control:
http://www.google.ca/url?sa=t&rct=j&q=& ... GU&cad=rja
http://en.wikipedia.org/wiki/Dynamic_range_compression

[sine143]

you might lose. 3 db on the input level, this can be made up for on the output

If he is running subs 1 channel, tops on the other, I hope that he is running L/R summed. while we might be concerned about "losing 3 db) in the bass, I think losing half the song in the mid/highs would have a much greater impact.

[byacey]

I'm not sure, but I got the impression he was running mains and subs stereo, but was only driving the sub amp from the left sub crossover output. If this is the case, the L&R sub outputs of the crossover could be summed to mono with just with a simple Y cable. Even if the level is leaning (panned) one way or the other in the subwoofer passband, it won't hurt the output stages of the crossover summed in this manner.

Of course the approximate 3db loss can be made up with the crossover sub output gain. The whole description sounds more complicated than what it needs to be.