i have an issue using the frequency response combiner (frc) speadsheet.
when trying to determine the q of the extracted curve with frc, using that value, say 0.5, results in a too narrow bandwidth ie incorrect eq, when the original audio file is re-eq'd according to the curve generated on frc to match the difference extracted beforehand.
there is a discrepancy between the q definitions of frc and typical vst eq's i have tried, ie vst eq of q=0.5 result in a narrow bandwidth whereas frc shows this as a much bigger bandwidth.....
when trying to determine the q of the extracted curve with frc, using that value, say 0.5, results in a too narrow bandwidth ie incorrect eq, when the original audio file is re-eq'd according to the curve generated on frc to match the difference extracted beforehand.
there is a discrepancy between the q definitions of frc and typical vst eq's i have tried, ie vst eq of q=0.5 result in a narrow bandwidth whereas frc shows this as a much bigger bandwidth.....
Last edited:
an update on the use of "frequency response combiner" (frc)
i extracted an eq peak filter curve of 60hz, +6db and bandwidth of 3 octaves.
this equates to a q of 0.4
i checked it manually by center, and -3db upper/lower frequencies.
when it was loaded into frc, the centre freq and amplitude were a good match, but the "best fit" for q was 0.6
unless i have made an error, frc's q matching is a bit off...
i extracted an eq peak filter curve of 60hz, +6db and bandwidth of 3 octaves.
this equates to a q of 0.4
i checked it manually by center, and -3db upper/lower frequencies.
when it was loaded into frc, the centre freq and amplitude were a good match, but the "best fit" for q was 0.6
unless i have made an error, frc's q matching is a bit off...
Hmm... how's the match when gain is set to say +3dB.
When I build the filter graph display for my EqualizerAPO GUI software I found out that the bandwidth didn't match with the drawn curve when gain was above 3dB ... (I used filter formulas from EQ-cookbook (released by BJ)) ... I asked about this issue on some forums (dsprealated.com/KVR) but the answer was "cookbook formulas are OK").
When I build the filter graph display for my EqualizerAPO GUI software I found out that the bandwidth didn't match with the drawn curve when gain was above 3dB ... (I used filter formulas from EQ-cookbook (released by BJ)) ... I asked about this issue on some forums (dsprealated.com/KVR) but the answer was "cookbook formulas are OK").
thats not an area that i can help you with or comment on.
in using "frequency response combiner" what i can say is it will give me a estimate of Q, which allows me to make a good Q match after 2 or 3 attempts.
if there is anything out there that will import a .frd type curve and has a "curve generator" function, so that its freq/ampl/Q can be resolved would be useful
- Status
- Not open for further replies.