speaker cal pink noise

thothsong

Member
Thread Starter
Joined
Jun 29, 2021
Posts
423
From the REW help:

"The Speaker Cal option applies 2nd order (40 dB/decade, 12 dB/octave) filters at 500 Hz and 2 kHz, producing a signal with its energy centred on 1 kHz. Sub Cal applies filters at 30 Hz and 80 Hz. Both are broadly in line with the THX test signal recommendations."

I'm looking for best-practice advice on the type of pink noise to use when balancing a mixed set of non-sub speakers, where the speakers have varying capability, especially below 200 Hz. Is the REW Speaker Cal option preferred, or something else?
 
The speaker cal is a good test signal for level matching, though the roll-offs are quite slow so there is significant energy outside the band. You can use the Custom option to apply faster roll-offs with pink random noise or use pink periodic noise with brick wall filtering to isolate that band.
 
The speaker cal is a good test signal for level matching, though the roll-offs are quite slow so there is significant energy outside the band. You can use the Custom option to apply faster roll-offs with pink random noise or use pink periodic noise with brick wall filtering to isolate that band.
Thanks! I'm a bit confused, though, further down the Help also says "The Sub cal and Spkr cal signals are brick wall filtered." So are they BU2 or Brickwall? Comparing previews with Custom 500-to-2000 Brickwall, it looks like Speaker Cal is actually Brickwall?
 
Oh, I see, Periodic Speaker Cal is Brickwall, and Random Speaker Cal is not.
 
If I use Periodic Sub Cal for the sub, and Periodic Speaker Cal for the other speakers, what dBC offset should I be targeting for the sub vs the other speakers (factoring in the usual +10 dB LFE offset)?
 
Zero if you want a level response, otherwise however hot you want the sub.
I want to check if the following is the correct understanding of your answer:

If Speaker Cal with pink periodic noise for each speaker (apart from the subwoofer) is X dB, the subwoofer(s) (sub cal, pink periodic noise, LFE channel) should be X + 10 dB, for a standard calibration?
e.g.
Speaker Cal, pink periodic noise, centre channel = 75 dB
Sub Cal, pink periodic noise, LFE channel should be 85 dB

I'm asking because Dolby say the SPL difference should be something else when using an SPL meter, but first let's confirm if the above is how you think calibration should be done for a standard calibration.
 
Last edited:
REW's cal signals are adjusted to have the correct rms level for their bandwidth. AV processors apply 10 dB gain to signals on the LFE channel. A better way to calibrate is to measure the system response, rather than relying on averages over the bandwidth of a test signal.
 
Back
Top