Those are reasonable starting points, but it really depends entirely on the sample. There is no single EQ formula that works equally well for any instrument, just as there is none for any microphone or voice.
For the orchestral shootout, I did a rendition with Miroslav Philharmonik. I had just acquired the library at the time we began the project, so I was still feeling my way around in it. The first problem I encountered was the violins had an annoying screechiness to them that was particularly troublesome in this piece. I treated it using the standard method for identifying the most audible resonances in a sound source, the swept high-Q filter.
What I did was loop on the most annoying section of the song, soloing the violin. I inserted an instance of the Sonitus EQ, enabled one bandpass filter, set its Q to 0.5 and the gain as high as could go. I started about where my ear told me the screech was centered and started moving the filter's midpoint up and down until I had achieved the absolute worst possible, painful-sounding violin. I slowly widened the filter until further widening stopped making it sound worse. Then I inverted the filter and made it a -4db notch.
There were two frequencies I had to notch out, one at 450Hz and another at 3KHz. The 450Hz notch I kept narrow, the 3KHz notch was wider. But I might not use those exact settings for any old violin patch. It's specific to that particular sample, the formants and resonances of the sampled instrument, and to some extent this particular melody.
All acoustical instruments have resonances.
Here is a graph showing the typical resonances in both the air and in the wood for a violin. Note the wood resonance at 440Hz. This may correspond to the 450Hz notch I had to put into the Miroslav violin sample, I don't really know.
If you want to get an idea just how complex a violin's sound is (and any other acoustical string instrument for that matter), here's an
introduction to the subject. Note that in graph at the bottom of the page, the peak appears to be more like 300Hz rather than the 200Hz cited in your reference, and there appears to actually be a
dip at 200Hz, so if your violin matches that example, notching 200Hz could actually be harmful.
post edited by bitflipper - 2010/08/17 19:22:28