- M-var-Multisonics VST
- FreezeRMS Group
- RMSBuddy VST
- HNM Filter VST
- Dither ITU Group
- The Spreadsheet
- the layout.
I included everything in the package that you can dowload here : CCGLCR Package
Put the layout in the layout folder of Plogue Bidule and place the CenterCutGui
on the desktop. There are two audio players, one for the fronts and one for the
rears. The one for the fronts gets the original stereo and the player for the
rears gets the output from Center Cut Gui.
Because Center Cut Gui can give you artefacts, it's good practice to use it
this way :
- change the 16 bits of the original stereo to 32 bits (NOT FLOAT). CCG doesn't
- work with 32 bits float.
- after that upsample to 96 Khz.
- Save the changed stereo .
Your directory should contain now stereo 16/44.1 and stereo 32/96 wavs.
The easiest way to handle this program is to copy it from the destop into the
directory of the stereo !
Make sure you have the 32/96 file (I always add this to the name of the stereo)
and load it in Center Cut Gui with "Add".
Make sure the settings of CenterCutGui are the same as this picture :
Now press run and after a while the following files are saved in the directory :
stereo3296-S.wav and stereo3296-C.wav
Actually you don't need the C in CCGLCR in this layout.
Now listen to the stereo3296-S file to hear if it need some improvement.
For instance a bit of BBE or Yamaha Final Master to remove some of the artefacts.
After improving the sound, downsample the original stereo3296 and the
stereo3296-S to 32/44.1 . If you wonder why you need to upsample the same
original stereo and later downsample it again and not use the original
unprocessed stereo, it's because you avoid small (very small) tonal differences
when you do it this way.
REMINDER : if your original stereo sounds bad, do some preprocessing before
starting the whole proces.
Now you're done with preprocessing the files, here is the layout :
Put the 32/44.1 original stereo in the fronts player and the stereo3244-S in
the rear player and start the process for the first time.
Use the gains to adjust to a good soundfield with the settings we always pro-
moted here : Front L and Front R equal RMS, Center 0.5 Db louder RMS than fronts
LFE 4 Db softer than fronts, Rears 1 Db softer than fronts.
Use the spreadsheet !!
That's actually all there is, however you might have sometimes after the first
run bigger differences than in other methods. An example would be :
fronts -0.1 Peak - 18 RMS
rears -0.1 Peak -25 RMS
Now you could lower the fronts to -24 but then the peak will be -6.1 Db, which
makes the mix rather dull. When you ever have a situation like this : leave it!
Separation and the sense of separation comes from the peaks, not from the RMS.
The RMS is more responsible for the soundfield.
In this method you will notice that in the described way to leave it as it is
with extreme values (except of course the fronts, center and lfe adjustment),
you still get an excellent separation. It differs in so far from other methods
that some parts are almost quiet in the rears and sometimes louder than the
fronts. And that's where that great sense of separation comes from here !
Example : a singer start a song with a piano only for one minute and both are
steered to the fronts. The rears will keep silent. But after that minute the
band falls in and sound is steered to the rears too. In most other methods some
of the piano is also steered to the rears which makes it easy to get them 1 Db
softer than the fronts, but with this method it actually leaves it more where it
belongs : in the fronts. Why is that ?
That's because we don't do anything on the center, but use M-var-M which is not
a vocal remover, but more a tool to let stereo hear over 3 instead of 2 speakers!
So you have the full stereo in three speakers in the fronts and accents from the
stereo steered to the back.
Happy Encoding !!