I have been told that to find an offending frequancy, use a notch scan and move back and forth until you find it. I can't seem to get this to work. I think it is because the problem is in the both primary and overtones. Is there a more effective method? I have a fiddle player whose fiddle at about f3 has a very stridant sound and I'd like to make it less noticable.
Thanks,
Bax
Finding a frequency
-
mflorio
sweep
There really is no way to sweep across frequencies like in a conventional eq. It's a bit harder to identify those strident harmonics in Har-Bal, in my experience. Best way I've found is to even out the curve according to general harbalization procedures. Eventually you will pinpoint the offending frequency.
-
Har/Bal
finding a frequency
Stridency shows up most prominently in the peak spectrum trace. You should be looking for peaks that stand head and shoulders above the rest and cutting them down is size a bit.
If the fiddle sounds strident in the recording but there are no obvious peaks that stick out in the spectrum it is very likely that any EQ you will apply to fix it is going to be less than optimal. This is because, given that there aren't any peaks out of control in the mix it indicates that the fiddle is not the only instrument occupying that part of the spectrum so anything you do to control the fiddle will colour whatever else is in the same band.
If this is the case my suggestion to you would be to go back to the multitrack source and just work on the fiddle part. If it isn't obvious in that spectrum then you are probably going to have difficulty fixing the sound with whatever processing you do.
Oh, and by the way, since HarBal doesn't yet support mono files(New beta does) you'll need to convert the fiddle part into stereo format before attempting to process it. Mono support will be in the next release sometime next month.
Regards,
Paavo.
If the fiddle sounds strident in the recording but there are no obvious peaks that stick out in the spectrum it is very likely that any EQ you will apply to fix it is going to be less than optimal. This is because, given that there aren't any peaks out of control in the mix it indicates that the fiddle is not the only instrument occupying that part of the spectrum so anything you do to control the fiddle will colour whatever else is in the same band.
If this is the case my suggestion to you would be to go back to the multitrack source and just work on the fiddle part. If it isn't obvious in that spectrum then you are probably going to have difficulty fixing the sound with whatever processing you do.
Oh, and by the way, since HarBal doesn't yet support mono files(New beta does) you'll need to convert the fiddle part into stereo format before attempting to process it. Mono support will be in the next release sometime next month.
Regards,
Paavo.
-
SpiderM69
Re: finding a frequancy
bax3:
It sounds like what you're referring to is what you do to find an offending frequency while mixing, as opposed to HarBal-ing.
First, make sure the gain on the band you're using is cut by quite a bit: 8-10dB is a good range to start, then sweep. If you're having a hard time zeroing in on it, then try the same process, only boost by 8-10dB and listen for where it sounds the most offensive; then cut to taste. You may want to lower the fader when you boost, however .
I hope this helps.
Regards,
Steve
It sounds like what you're referring to is what you do to find an offending frequency while mixing, as opposed to HarBal-ing.
First, make sure the gain on the band you're using is cut by quite a bit: 8-10dB is a good range to start, then sweep. If you're having a hard time zeroing in on it, then try the same process, only boost by 8-10dB and listen for where it sounds the most offensive; then cut to taste. You may want to lower the fader when you boost, however .
I hope this helps.
Regards,
Steve