On the subject of fizzy guitars...

[Nebulous]

The recent threads about guitar fizziness and harsh highends has had me thiking. I remembered hearing that amps/ speakers don't actually reproduce anything above 5kHz (which is incorrect, they do, but not much).
Point in case:
http://professional.celestion.com/guitar/products/classic/detail.asp?ID=4
http://professional.celestion.com/guitar/products/classic/detail.asp?ID=5

So if the speakers are voiced so that the fizz is so much lower than the "body" of the sound, where the hell do we get the problems?
Is distorting a signal doing all this to the sound, increasing the high/ harsh frequencies so much that they even seem to "reverse" the speakers character?

 

[Ermz]

Look at a fizzy guitar signal through a spectrum analyzer and all your questions will be answered.

Whilst the fizz is usually much lower than the body of the sound, there are still certain nodes of high frequencies which contribute to the harshness we hear. It seems that that high frequency content can get so harsh with distorted guitar signals that we hear it as an unpleasant sensation, even though it's below the body of the sound. This is one of the reasons I like tracking guitars sounding dark, and allowing for a linear EQ boost afterwards in mixing, to attempt to minimize these irritating high frequency nodes.

 

[Nebulous]

Hm, good call. Is the harshness entirely due to the distortion then?
While i understand what you're saying about the nodes creating a "perceived" harshness, there is infact ALOT of energy in the 5kHz + region, and even beyond 10kHz (at least compared to what you would expect based on those graphs). How are the speakers creating this? I mean, very often the harshness that I get is in 6-8kHz and beyond 11-12kHz. Could the nodes really be creating a 20dB peak that starts to catch up to the body of the sound?
Is it the electric signal which would create/ contain the nodes, or do they occur at the speakers?

BTW, I'm in no way dismissing what you've said, I'm just trying to have a discussion about it and maybe get some peoples takes on why we have these problems with respect to what the manufacturer claims of it's products.

 

[Mutant]

Hm, good call. Is the harshness entirely due to the distortion then?

Easy to measure.
Put a white noise through a clean channel and use an analyzer to check if those harsh frequencies are present.
If yes then its your speaker and mic, if not then it's caused by distortion.

 

[Ermz]

Yeah, as Mutant is saying, you may need to bust out the technical measures and check it out for yourself. It's the best way to be sure.

 

[Nebulous]

Interesting....

 

[Soundlurker]

Very narrow cuts pretty much fix the nasty fizziness. If you have Izotope Ozone 3 or something similar you can do this: choose a refresh rate of at least 1 sec, preferably 3 sec, playback some great sounding guitar (i.e. a part of say an Arch Enemy song where there's only guitars) and take a snapshot. Then, after roughly matching the level of the guitars you've recorded, see how they differ from the snapshot and where you have spikes of high frequencies peaking significantly over the snapshot make narrow cuts; where your guitars are below the snapshot try a slight wide boost. Don't know how those nasty spikes should be called as I am not completely sure with why they emerge ...standing waves?
Anyway, hope this little trick helps you fight the fizziness.


Edit: of course first you should do all you can to minimize such spikes by chianging mic position, but IMO you shouldn't take away too much highs doing so.

 

[NickL]

this is only sort of related but do any of you guys ever run into a really nasty cab resonance every once in a while? It's not often but every once in a while there'll be a palm mute that has the sound of vibrating wood under it

 

[Nebulous]

Very narrow cuts pretty much fix the nasty fizziness. If you have Izotope Ozone 3 or something similar you can do this: choose a refresh rate of at least 1 sec, preferably 3 sec, playback some great sounding guitar (i.e. a part of say an Arch Enemy song where there's only guitars) and take a snapshot. Then, after roughly matching the level of the guitars you've recorded, see how they differ from the snapshot and where you have spikes of high frequencies peaking significantly over the snapshot make narrow cuts; where your guitars are below the snapshot try a slight wide boost. Don't know how those nasty spikes should be called as I am not completely sure with why they emerge ...standing waves?
Anyway, hope this little trick helps you fight the fizziness.


Edit: of course first you should do all you can to minimize such spikes by chianging mic position, but IMO you shouldn't take away too much highs doing so.

I guess my point was sort of lost. I know how to fix it, it's just a matter of discussing where it comes from, ie: understandnig where it comes from, and why it even exists when speakers shouldn't be reproducing those frequencies based on the manufacurer specs.

 

[Nebulous]

Ok so after some thought I think I've come up with an answer to my simplified question: "Why do we have to deal with harsh frequencies when the speakers barely produce them relative to the body of the sound?"

The answer lies wtihin:
http://www.shure.com/stellent/group...ts/web_resource/site_img_us_rc_sm57_large.gif

Summed up, the speakers we use have a (~)-20dB reproduction of the freq's above 5kHz, but the mics we use to record have a relative (~)+12dB reproduction of those freq's, especially 6 and 8kHz where we often have the fizz problems.
I'm sure the distortion plays a part in the higher freq's being so prevalent too.

One thing that I have left to be answered after all of this, which I doubt I will find an answer for, is how manufacturers can even justify using such graphs to represent their products when the sound of a speaker is so different from one place to another on the cone, one centimeter away?

/end rant