Frequency spectrum analyzer to evaluate guitar tone?

H-evolve

Member
Apr 21, 2014
499
41
28
Montreal, Canada
Hi guys,

I am trying to find a "visual" way of knowing what a good high gain guitar is, in a qualitative way. It might be a completely stupid, but bare with me one second.

You see, I keep reading everywhere that you must have a lot of mids in your tone, to make it sound good and alive in the mix. First problem I face, as a noob, is that some amps are stronger than others, I think, on the "mid" range. For instance, mids at 10 o'clock on a Dual Rec might be a lot more mids than 12 o'clock on an ENGL. So I can't really rely on "where should I put the mid knob".

And then there is a question of how it sounds "not in a mix". I realized that I have never heard a good "recording" tone on one single guitar. I mean, I do like the tone on the album Godless Endeavor from Nevermore, but have no idea how a single guitar sounds like on it. For example, I feel that more than 9 o'clock on the Dual Rec starts to sound not so nice, at least in my opinion. Too crunchy, not enough of heavy "chugs". However, from what I read here from Dual Rec users, everybody seems to go from 10 o'clock to 12 o'clock in their mix.

So here I am on a quest to characterize qualitatively a good modern high gain tone.

What do you guys think? Stupid idea or ?

Keep in mind that I am not a sound engineer... I am actually a "mechanical engineer"! :loco: So there are surely some things I miss that would perhaps make this idea unrealistic.
 
visually the best that you can do is get a wysiwyg eq like equality and and find some of the spikes in the upper range like around 4k and flatten them out a bit, but really you cant do what you are talking about just due to the medium you are working in. finding your tone takes a while because as you said what you hear in the final product is not necessarily the same as the individual elements that comprise it. I like to use a supersaw synth as an example. one saw sounds boring and dated whereas 8 of them panned out and detuned a bit sound like this huge massive thing. my advice is to put every knob in the tone stack and power section at noon, set there usually you have achieved factory default. then go left to right setting each where it sounds the best to you. after that you start in the center of the cone with the mic/cab sim and move it away by increments etc., to sort of preview you finished tone maybe put a 10ms delay on a second track panned opposite to get a "faux" double. every tone is set up differently esp in relationship with the bass guitar and the eq on that. one usually fills the gaps of the other i.e. middy super high passed guitars might get a scooped bass tone or a scooped guitar sound would get a middy less low end centric bass tone.

sorry for the novel...
 
What they said, and I would add that the having plenty of midrange thing is really dependent on what tuning your playing in as well as what the mix around it dictates.
 
I definitely think you are overthinking the process too much.

Just experiment, get to know the gear you're using inside out, and learn to trust your ears

Well maybe I didn't explain the objective correctly. It's not really that I absolutely need this to mix, I was just trying to get some tool created, out of curiosity and perhaps practicality, as long as it would be feasible. Just experimenting you know.

But for this specific idea, I wasn't so sure, so I asked.
 
I think the biggest challenge is that every "good" guitar tone is good for THAT mix, but might be complete garbage somewhere else.
 
I spent so much time worrying over guitar tones when I first started mixing and sculpted them to death. Realized after a few months that basically as long as you get a decent-sounding guitar tone with solid gear and reasonable settings (i.e., not deviating hugely from noon on most knobs on most amps), a lot of different stuff will work and frequency spectrum analyzers aren't worth much. HPF @ 85 hz, lpf @ 11000, tiny boost at 1200, 2-3db cut @ 300, and you're done. Anything else I do is to allow the kick/snare/vocals to breathe or to suck out 1 or 2 resonant nodes and that's it. To the extent you spend time worrying about the mids or whatever, just use your ears and compare it to albums during parts where the guitar is isolated. Stems from the Rock Band/Guitar Hero games can help too, and so can videos like this:



But also bear in mind that a lot of guitar tracks sound like crap when isolated. It's whatever works in the mix. tl;dr don't overthink it; guitars are by far the easiest instrument to mix in my experience.
 
Last edited by a moderator:
I spent so much time worrying over guitar tones when I first started mixing and sculpted them to death. Realized after a few months that basically as long as you get a decent-sounding guitar tone with solid gear and reasonable settings (i.e., not deviating hugely from noon on most knobs on most amps), a lot of different stuff will work and frequency spectrum analyzers aren't worth much. HPF @ 85 hz, lpf @ 11000, tiny boost at 1200, 2-3db cut @ 300, and you're done. Anything else I do is to allow the kick/snare/vocals to breathe or to suck out 1 or 2 resonant nodes and that's it. To the extent you spend time worrying about the mids or whatever, just use your ears and compare it to albums during parts where the guitar is isolated. Stems from the Rock Band/Guitar Hero games can help too, and so can videos like this:



But also bear in mind that a lot of guitar tracks sound like crap when isolated. It's whatever works in the mix. tl;dr don't overthink it; guitars are by far the easiest instrument to mix in my experience.


Thanks, that is a very interesting video.

But ya, like most people said here, I think the idea of using frequency analyzer isn't that good after all!

Well, too bad... Haha! The idea came from the fact that in my line of work, we tend to "standardize" everything, or to figure out what is common in everything that we do, so that we can come out with a certain recipe, easy to replicate.

I guess what you listed in your message is that recipe, and nothing more than that is required.

Thanks a lot
 
Last edited by a moderator:
I actually learned a lot from using frequency analysers with raws, not only with guitar but particularly with bass. I don't do this every time I mix but it was very useful to use when practicing and isolating why a particular mix sounds the way it does.

When I do this I used the Waves PAZ and occasionally Ozone.
 
If you need a spectrum analyser VST for whatever reason, this one is pretty killer :
http://www.voxengo.com/product/span/
span-large1024.jpg


And it's free.
 
No, SPAN has always been a freeware and there are no limitations.
They recently released a commercial version called SPAN Plus with a few additional features, but the freebie is already great IMHO.
 
If you want to do an effective comparison of your guitar tone to an other, just find a spot in a song that has 1 single guitar playing and render that to wav. Record yourself playing the same segment and you'll get a good idea of how they differ.

I started building a guitar tone reference folder with isolated segments like that. Funny how some tones sound weak when there's no bass present.


As for the analyser, pretty useless IMO, except if something is incredibly flawed in the tone, which you should be able to hear anyway.
 
Find some guitar stems from great sounding mixes. Look at them through a spectrum analyzer to get a general feel for how much bass, mids, and highs are there. That's as far as a spectrum analyzer will help you. The point of this is to orient you, if you have no mixing skills yet.

Then the rest only your ears can tell. Your ears have evolved to detect tone, character, and meaning of speech -- they can distinguish one sonic fingerprint from another where an analyzer might show both fingerprints as looking basically the same (because it goes beyond just bass, mids, highs -- it gets into harmonics, formants, the numerous little peaks and valleys and where they are located).

So then do your best to get your guitar tone in the ballpark of the stems. You'll learn from this process what amps, settings, mics, mic position, etc. do the job. Also I advise against using one amp/cab/mic combo to try and replicate a completely different setup, especially trying to EQ your way towards that. If you want an exact sound, then match your equipment accordingly. If you have selected correctly, then you won't need to do much of any post-EQing.

Lastly, play around with EQ and/or compression prior to the guitar hitting the amp sim if you're going DI. That affects the gain structure, tightness, looseness, crunchiness, etc.
 
This all kind of reminds me of the Match EQ craze a little while ago. While I don't really recommend relying on it, it is a great learning tool to get an idea of mixed guitar tone sounds like or what problems there are in your tone.

By matching visually it sounds like you are basically trying to achieve the same goal but manually.
 
Yea, as others have said; keep in mind that what makes a guitar sound good may or may not change depending on the other instruments. I find cymbal size or how dark or bright they are can effect what tone sounds good for example. Experimenting and getting to know your gear is good because it will let you adjust to different situations better.
 
You get them from tones you like in the music you like. (or not, might be a crap song with cool tone) If you're lucky you'll catch some of those "isolated segments" dude is talking about. ;)