Acoustic fingerprint tool?

Tue Mar 31, 2020 9:55 pm

Is anyone aware of a tool (software?) that takes an acoustic fingerprint of your guitar?

For example I have 2 guitars with the same exact electronics and they sound very different plugged in or not. So factors besides the pickups are contributing. This is even true with heavy distortion.

It would be a lot easier to buy a guitar or sell one if you knew something about it’s acoustic sound. Or if you knew you liked the sound of a certain type of guitar you could find other guitars that are similar.

Is there anything like this?

Re: Acoustic fingerprint tool?

Tue Mar 31, 2020 10:10 pm

I've been thinking of taking Impulse Responses of differing blocks of wood.
Are you familiar with IR's or FFT's? There is some free software out there.
Voxengo is one program I have that seems to work well. I'm still a newb.

Driving guitar bodies with a (loud) frequency sweep would be interesting. :think:

Re: Acoustic fingerprint tool?

Wed Apr 01, 2020 8:24 am

Maybe this idea is not quite the same as yours- But as obsessive as guitar players can be, I’m pretty alarmed it never became a standard to compare frequency response graphs for guitar output waveforms, similar to the ones often published in specs with speakers. It seems like whenever they are used or brought up online, someone mentions they would be useless because ‘not all visual fluctuations in the graph would be picked up by the human ear’ - never mind that audible differences still *would* be portrayed visually. A bit of science and study would probably aid in interpretation. It seems like a lot of guitar players and makers are unwilling to trust technology, or would not want to sacrifice their empirical ‘seat of the pants’ knowledge to trust the machine.

Re: Acoustic fingerprint tool?

Wed Apr 01, 2020 11:32 am

Danley wrote:Maybe this idea is not quite the same as yours- But as obsessive as guitar players can be, I’m pretty alarmed it never became a standard to compare frequency response graphs for guitar output waveforms, similar to the ones often published in specs with speakers. It seems like whenever they are used or brought up online, someone mentions they would be useless because ‘not all visual fluctuations in the graph would be picked up by the human ear’ - never mind that audible differences still *would* be portrayed visually. A bit of science and study would probably aid in interpretation. It seems like a lot of guitar players and makers are unwilling to trust technology, or would not want to sacrifice their empirical ‘seat of the pants’ knowledge to trust the machine.


totally agree.

the visual fluctuations argument makes no sense to me. you can filter away half of the frequencies of a song and still very much know the difference between hip hop and country. you don't need all the information. you just need the critical information.

no one would think about buying a mic if it didn't tell you some basic stuff. pickup pattern, freq response, sensitivity, S/N. i'm guessing these specs were not there when commercial mics first came out. now they are standard because they help you narrow down the huge number of options.

for this tool i was thinking of taking the pickups out of the equation since they are relatively easy to swap. also modern pickups, i hope, have more manufacturing consistency then organic materials like wood. so, in theory, the wood is the big variable. really it's the whole acoustic system (body, neck, bridge, nut).

cause what's happened to me in the past: i bought a dog and changing pickups was not fixing that. : )

it would be cool if pickups had something like a waterfall chart. input is something like a sine sweep. output is freq response over time. same type of thing i use to measure my listening rooms character.

at first people would be confused as i was when measuring my room. but as you get better at reading the data it's really helpful.

a person might have a preference for a certain type of neck radius, thickness, shape. this is the same thing for the guitars inherent acoustic characteristics.

but i don't think there is much motivation for manufactures to make this tool. lots of purchases are based on emotions and marketing. also, if i'm right, they would have to measure every guitar. especially the cheap stuff where quality is less consistent. and it might work against some of the larger companies that are selling more on name than quality.

i did notice that semour duncan has some tools on it's site for comparing pups. a step in the right direction. most of my time shopping is spent cutting through the marketing BS.

last thing. : ) i had a really hard time getting my acoustic guitar set up. i took it to some of the well known shops in my area (San Francisco) and told them to set it up to taylor factory specs. the guitar came set to those specs and played great when i first got it in 2004. no point messing with that! i had one tech play it for 20 seconds and say it was in specs. it was not. had another tech tell me it's about feel and not specs. "how does it feel?" i'm like "it felt better when it was set at taylor specs for 10 years". measuring should be standard for these guys. how do you know what you've changed on a guitar? if a customer likes a certain relief why not write that down for next time? in my experience the basic necks numbers tell a lot of the story. there is one shop which takes before and after measurements here. i only go there now.

Re: Acoustic fingerprint tool?

Wed Apr 01, 2020 11:57 am

Elwood wrote:I've been thinking of taking Impulse Responses of differing blocks of wood.
Are you familiar with IR's or FFT's? There is some free software out there.
Voxengo is one program I have that seems to work well. I'm still a newb.

Driving guitar bodies with a (loud) frequency sweep would be interesting. :think:


it would be.

i'm familiar with room measurement software. sending a sweep sine way out and measuring a waterfall. with a guitar the waterfall would be much shorter i'd guess.

what i was thinking is let the strings vibrate the guitar. put a mic close to the body. make the test super simple. gather data. dial it in later. you'd want it to be easy enough for anyone to do. more data.

if you drove a guitar with a sweep... not sure how that would work.

the system i'm thinking of is like this:
strings are strummed - vibration transferred across bridge/nut - resonance of body/neck/trem - effect on strings (sustain, resonance) - effect on what pickups "hear"

at least that's how i'm thinking about it.