I got a short vocal test clip i use, it peak at 0 db ( no clipping ) on a audio track ( new project )

I add the PC4K channel comp in the Prochannel :
lets say i set threshold to 0 db, and output remains at 0 db, then i play the track
all is good no peaking or change in the volume level.

Ratio Issue : first test, slow attack
if i raise the ratio , the output volume raises too it will peak at +1.8db
that should not happen, because it is just the amount a compressor should compress, not raise the level.

when i turn the threshold to atleast -15db, it start to compress and reduce the output ( ratio is set to 4:1 )
but i need to go all the way down to -18 db to have peak at about - 0.1db 

Adjusting the Ratio shall not affect the volume at all.

bug or inaccurate measurements and programming ?

I tested the Sonitus Compressor as insert on the track, it peak at -0.2 db
regardless of Ratio / Threshold or other, so that works as a compressor should no strange behavior.

If i use pro-channel and insert the FX chain, then add the Sonitus Compressor it peak at - 0.7db
a bit lower volume at a difference of -0.5db
same result both post and pre, non activated EQ.

Test two :

I have both Peak and RMS  level meter in the BUSS preview.
From that i see the vocal has a RMS peak at -9db and start about at-14
guess average would be around -12 db

test RMS level ( threshold ) - 12db  / Release 120ms  / ratio 4:1 / attack "around" 14ms
( working area of the compressor in this test )

I use the Sonitus compressor in the FX chain and set it to ratio 4:1 , threshold -12 db
and then it compress the vocal about 6db at the peak compress.

When i use the  PC4K channel comp at ratio 4:1 / rel 0.120s / fast attack off
i need to set the threshold all the way down to -26 db for the  "same " amount of gain reduction
(6db) on peak compress.

With fast attack setting it compress to peak 6db reduction, at about threshold -14 db

Now tell me how come this big difference, its certain not the same adjusted level compressors work with !

i Guess PC4K channel comp, seriously need an adjustment, cause its not accurate
compared to the Sonitius compressor

And yes it is repeatable, everyone can check this out....

Jan

Do some tests ,and see what you guy's find out
You can also use a third party compressor to verify, what is "most" correct

Jan

I'm pretty sure this was discussed before, and properly models the behavior of an SSL 4K compressor, with the exception that the gain reduction metering is off because of the difference between the analog and digital domain (digital stops at 0dbfs and analog has headroom up to 20dbfs or something). Adjusting the ratio on a compressor like that also adjusts the knee point, which may cause what you're hearing.

See this thread, and the post/thread it links too:
 
http://forum.cakewalk.com/PC4K-Bus-Compressor-Ratio-backwards-m2947867.aspx
 
 

Thanks for the link's