[Project] Hardware independent modelling

[Flow Of Time]

Hi guys, I came up with an idea today. First of all: I don't have too much knowledge about this matter, so everything I'm writing right now could be utter bullshit.

Everyone knows about the idea of match eq/tone matching. And most people agree that it's accurate enough. But the thing with those matched tones is, that they're pretty hard to share, due to different guitars; pickups; etc. (leaving the kemper aside right now).

The idea to avoid this now would be to create some kind of database (I'd set up a website or something), where people can download a sine wave. They'd reamp that sinewave through there setup of choice, and submit this reamped sinewave (and maybe the ampsettings he used) to the database. Now everyone could match the two sinewaves through his modeller of choice and voilá. The only thing you'd need to use those files would be a matching eq.

Do you guys think, that would be a good idea, and would anyone be in to take part in this project? (Maybe Clark Kent, I know you have a thing for match eqing)

 

[onqel]

I think personally white/pink noise would be a better choice for matchEQ'ing

 

[deLuther]

I think personally white/pink noise would be a better choice for matchEQ'ing

Or amp send on high gain. I`m made comparison of different approaches to cab impulse retrieval, among them was white noise. White noise requires long averaging in time to get better result.

Final results:

 

[Flow Of Time]

Matching white noise sounds like a good idea, that's also what I prefer in the comparison you posted. If I'm understanding correctly, this is what you did with NoiseIR2, right?

 

[onqel]

DeLuther: I'm not referring to deconvolution, i then would use a swept sine.. But a noise signal for matchEQ'ing would make it faster for a fft analysis to find the magnitude of the entire spectrum.

Flow of time: i have no relations to that software

 

[Flow Of Time]

I was referring to the comparison deLuther posted. He had a clip in there called NoiseIR2 Could someone who is experienced in this topic (maybe deLuther) post a white noise clip, with "perfect length", so we can determine this for the project?

 

[onqel]

Ah okay, sorry

 

[deLuther]

I`m used just 5 seconds of noise, to much peaks still persists, from some expeiments better to have 40 seconds or even more, maybe just cycle till sufficient smoothness. I`m used my own tools, currently working on my own matcher (C++ this time)

For cab impulses just match between line signal and miced signal give sufficient result in very easy way, just record additional line signal and you can obtain very good impulse.

 

[Flow Of Time]

I think wee'd have to keep it as simple as possible for this, to be not too confusing, so a predetermined white noise clip would be the best. I can post it myself later. At which level do you have your white noise?

 

[deLuther]

Usually I`m prefer level in the tange -1/-6 dbFS, better for converters and output buffers, because in the range higher than -1 dbFS harmonic distortion can be higher.

 

[Flow Of Time]

Allright, I'll upload the clip later!

 

[Flow Of Time]

Here you go:
https://copy.com/k7IVohumGddw/white_noise.wav?download=1

It would be awesome if some people could reamp this! If you do, settings, tube screamer yes/no, etc. would be helpful