Spectrum Analyzer

F

FunkyShredder

Member
I was playing around with tone match today and after I'd finished I compared the source and the tone match audio in a spectrum analyzer. In the attached image the original is on the left, the tone match on the right, does anyone know why the tone matched one appears way more jagged?Tonematch.png
 
Interesting. The first trough at ~1K is 8dB lower, yet the lowest frequencies match amplitude, and other areas differ by 6dB or less.
 
I always thought that "Tone Match" is the compensation needed to change an input to match the provided source or target. In order to make the adjustment, it boosts and attenuates various frequencies to achieve a match.
As a result, the only true comparison is the output of the Tone Match against the source and both taken with the same degree of smoothing and number of samples.
 
That's really interesting. I don't know if really is the way Tone Matching works, but looking through the graphs seems it's like it has several narrow bands eqs (high "Q"). So, the eq corrections that tone matching applies is not perfect to the desired frequency only, it has side effects to neighbors frequencies, so it has to eq again to correct the side effect, that causes other side effects and goes on and on.
Well... ends up that it's not perfect seeing close, but seems pretty close on average.
 
Vertical scale of the two graphs is slightly different. Might be making them get resampled differently in the graph display. Since they both peak at the top but are at different levels, it looks like the graph scale is getting normalized. See if there's an option to turn that off and force it to use the same scale.
 
Yes, if the audio being matched is of a cleaner or dirtier amp than the amp in the preset, maybe this produces the variance. For instance, if your source is clean, and your tone match block is after a heavily driven amp, the purpose of the tone match block is not then to perfectly reproduce the 'signal type' of the source audio, i.e. to clean it up... (It's my understanding that the user needs to do a good job of factoring in what a similar signal chain is to get the desired result, and use this IN COMBO with the tone match block).

For example a synth sound can be smooth, or have a lot of distinct, amplitude-varying harmonics. If I were to try to match that harmonically-laden synth, the resulting tone match block would in theory provide allowance to let thru that sort of harmonic environment, but if there is no way for an Axe FX preset to generate a very accurate version of this quality (and ordinarily there really wouldn't be) then of course the two graphs would again be very different.

Er, I rest on my guitar case..
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom