Voxengo - Sound with a Soul
Professional VST PlugIns and
AU PlugIns

I can successfully capture spectrums in both Cubase 5 and Wavelab 7.

The resulting equalizations are grossly exaggerated.

In other words: If the source signal is bass heavy, the new equalization from the capture is Very bass heavy.  It does not match up with the source signal's spectrum.

I can supply a screen capture if that would be helpful.

What am I doing wrong?

Spectrum capture does not take dynamics into consideration.  So, if the source spectrum you are trying to match to a target spectrum contains infrequent bass notes, they may be boosted a lot if that target spectrum contains some constant bass content.

*nods*

I know Curve EQ looks at the "average" energy across the spectrum.

This is new behavior and I don't know if I'm doing something wrong or if the latest updates to C5 and W7 and causing the behavior.

May I send you a picture of what I'm seeing?

I found a spot to upload it to.

http://4.bp.blogspot.com/_j97n2DE1WTc/TTR3x3c2ajI/AAAAAAAAAE4/CQwswRK6cc0/s1600/Captured%2BSpectrum%2BOverEmphasized%2B.jpg

This screenshot does not tell much, really.  It only tells that the average power of the lower frequencies of the target spectrum was that much higher than those of the source spectrum.  You could try using GlissEQ on two tracks and use its interchannel spectrum monitoring to see how track spectrums differ in dynamics.

This is the display just prior to ending capture and then matching the EQ curve.

http://2.bp.blogspot.com/_j97n2DE1WTc/TTSzRAb7XpI/AAAAAAAAAFA/7gUo_wDY-No/s1600/Spectrum%2BDisplay%2BBefore%2BCapture%2B.jpg

Here is the resulting EQ curve.

http://4.bp.blogspot.com/_j97n2DE1WTc/TTSzWwka3HI/AAAAAAAAAFI/mGSVcYjdlPo/s1600/Captured%2BSpectrum%2BEQ%2B.jpg

As you can see, the EQ curve does not match up with the overall curve of the source spectrum.  It's is over exaggerated.

How do I fix that?

The spectrum display has a different scale to the EQ curve display.

I input Pink noise (which is essentially "flat" to the human ear) and captured the spectrum then I applied that captured spectrum to the same Pink noise.

I monitored the output using SPAN.  Here are the results:

During capture (Curve EQ)
http://3.bp.blogspot.com/_j97n2DE1WTc/TTXDudhWWGI/AAAAAAAAAFQ/--x-1gGzSIg/s1600/Pink%2BNoise%2BSpectrum%2BCapture%2B-%2BCurve%2BEQ%2B.jpg

During capture (SPAN)
http://4.bp.blogspot.com/_j97n2DE1WTc/TTXDui2CwlI/AAAAAAAAAFY/J9xbKb-c-RA/s1600/Pink%2BNoise%2BSpectrum%2BCapture%2B-%2BSPAN%2B.jpg

The captured EQ (Curve EQ)
http://3.bp.blogspot.com/_j97n2DE1WTc/TTXDu5qg3WI/AAAAAAAAAFg/rJR4FYBswqs/s1600/Captured%2BSpectrum%2B-%2BPink%2BNoise%2BInput%2B-%2BCurve%2BEQ%2B.jpg (The blue line is the input.  The red line is the output.  The "dots" are the EQ curve.)

The resulting output (SPAN)
http://3.bp.blogspot.com/_j97n2DE1WTc/TTXDvmA1NlI/AAAAAAAAAFo/5nXtb63KXvQ/s1600/Captured%2BSpectrum%2B-%2BPink%2BNoise%2BInput%2B-%2BSPAN%2B.jpg

How do I set up Curve EQ so that the captured EQ matches the source signal's spectral balance?

Please tell me the steps you are doing.  When you are matching pink noise to pink noise you should get a flat EQ curve which makes me wonder what you are using as target spectrum.

I reset Curve EQ to 60 bands and tell it to capture the source.

I turn on the generator and let it run for many seconds until Curve EQ's display stabilizes.

I turn off the generator so Curve EQ is hearing silence and tell Curve EQ to stop capturing the source.

I then press M and tell it to match 60 bands and then S to stop and show the EQ curve of the source.