Comparison of real amp, AXE-FX and KEMPER preamps!!

ML SOUND LAB

ML SOUND LAB

Cab Pack Wizard
Vendor
Heyy!! So I got some preamp sound of the Kemper to study and I found out something quite weird. I must say that the tone that comes out of the Kemper is just fine and with the cab it sounds like this: https://dl.dropbox.com/u/6455606/Ros...nas 5150.mp3 but if you know how a preamp signal should sound/look like in meters then you'll find this somewhat weird. This is the preamp signal: https://dl.dropbox.com/u/6455606/Ros...amp mono.wav

Now here's a comparison image for you guys:
REAL_AXE_KEMPER_preamps.jpg


So just like Cliff told us.. the Kemper low end is located at 125hz and something weird happening there in the middle.

The reason I wanted to make this test is because I feel like the Axe-Fx (both gen1 and gen2) preamp sound cuts out at 17khz. Now those are extremely high frequencies to begin with but I do feel that some mojo is still up there. I might be deaf above 10khz for all I know. :D

But what's happening in the Kemper high end? Ice picks at 7khz, 10khz and 12khz and it cuts out at 14khz already but after that it comes back extremely high. Even as a hobbyist audio engineer I'd be worried with that high end.

What do you guys think?

I've made some EQ tone matches for my Axe-Fx to get some of that uber high end back to my Ultra. But... I just wish the preamp could go further than that. Is it possible?
 
It's high gain riffage but different riff and notes being played with every unit so in the low end it's not good to make assumptions based on that. The differences should be seen with wider bandwidth.
 
Hey Clark, you didn't have the cabs on for these sweeps did you?

If you've ever looked at a 10" - 12" guitar speaker's response you'll see that they roll-off steeply, and that by 7K they're generally around 20db+- down, so what happens in the guitar amp's high freqs is relatively inaudible in the real world through guitar speakers.

Are you saying all the KPA profiles exhibit this trait, or just this single one you picked (some amp designers put a little peak up top to add "air", i.e. tuned negative feedback in the power section etc)?

BTW, thanks for your gen 1 tone matching video!
 
Interesting test... It would be interesting to conduct this same experiment using the II with the latest firmware to see if it does the same thing or gets even closer to the curve of the real pre.
 
Nice shots. The curves don't look that different somehow? Do these curves show the differences in sound?
I guess frequency response curves are nice for linear transducers like cabs (that are not cranked up to the point of distortion) and eqs and alike. I wonder if response curves tell enough about nonlinear systems, do they?

The AxeFX and the KPA have different vibes somehow, their own voices, that I couldn't dial away, no matter which eqs I'd use for it. Maybe a tonematch/profiling could?
 
Clark Kent said:
It's high gain riffage but different riff and notes being played with every unit so in the low end it's not good to make assumptions based on that. The differences should be seen with wider bandwidth.
Click to expand...

i'd prefer to see a pink noise sent through all three, so the comparison is more accurate....you can use the synth block in the axe to generate it
 
simeon said:
i'd prefer to see a pink noise sent through all three, so the comparison is more accurate....you can use the synth block in the axe to generate it
Click to expand...

That's what I was wondering too.

Also, the gain matching that Cliff shows in the amp match tutorial. That should make the harmonic content closer too?
 
this is great! I agree with the others about using the same source for more accurate measuring.

It would also be cool to like "graph out" the different firmwares which are switchable in the axe-fx II. maybe it does not show up so well in eq, but is more of a feel thing.

great project Clark Kent, thank you very much for doing these highly interesting things! :)
 
DJD100 said:
Hey Clark, you didn't have the cabs on for these sweeps did you?
Click to expand...

It's the preamp signal so NO CAB or IR in there. A cab would cut out around 4-8khz.

Longhair said:
Is there a noticable difference to the untrained ear?
Click to expand...

Well ofcourse in Kemper's case there's the convolution process afterwards so it'll probably cut that top end out. The problem will be in the dynamics after that though. However if that's the preamp signal that goes to the power amp then you might have issues. I'm thinking the high end thing is there to compensate for some dynamics maybe?

Sixstring said:
Interesting test... It would be interesting to conduct this same experiment using the II with the latest firmware to see if it does the same thing or gets even closer to the curve of the real pre.
Click to expand...

The curve on the Axe-II was almost identical to the Ultra curve.

simeon said:
i'd prefer to see a pink noise sent through all three, so the comparison is more accurate....you can use the synth block in the axe to generate it
Click to expand...

That's true but ofcourse amp sim also matters.

Longhair said:
Is there a noticable difference to the untrained ear?
Click to expand...

An average human can hear up to 20khz so yes.
 
The hz limit has to do with the perceived fidelity. F.ex. I know lots of people refuse to listen to MP3s that are 128kbps quality because they "sound like sh*t". The reason for that is that frequencies above 16khz are being cut away. So... although those frequencies are almost inaudible you will most likely feel like the real preamp signal was higher fidelity after comparing for a while.

I don't know if you read my thread: http://forum.fractalaudio.com/ultra...ast-real-amp-only-using-ultra-guess-what.html

Anyways what that tone match did was create an EQ curve to compensate for that high end being cut out. It seems to work just fine.

EDIT: That being said, Soundcloud is 128kbps which many of these "purists" don't know so. Hahahahah...

EDIT: Also something good to know: Not every amps have these high frequencies.. Seems like the expensive ones do. F.ex. Mesa Dual Rectifier doesn't go that far but Mesa Road King does. Orange Rockerverb 100 MK II does... etc.
 
Last edited:
Clark Kent said:
F.ex. I know lots of people refuse to listen to MP3s that are 128kbps quality because they "sound like sh*t". The reason for that is that frequencies above 16khz are being cut away
Click to expand...

It's not really that. It's first of all the psychoacoustic encoding. There are all kinds of tricks and whatnot to be calculated when reducing the file size, and some are worse than others 8) In addition, you'll often have files encoded more than once....that's a gurantee for disaster. E.g. when people upload mp3's to Soundcloud, they'll go through another encoding process (So, always upload wav's to Soundcloud)
 
Last edited:
antcarrier said:
Not the average human (can hear up to 20 KHz) - young people.
Click to expand...
This. In fact, the average teenager can hear frequencies almost to 20 KHz. And they only hear those very high frequencies at a low perceived volume. After adulthood, the high frequencies get progrssively shaved off. Sometime in my 20s I lost the ability to hear the 15 KHz whine of a TV's flyback transformer, and I subjected my ears to a lot less abuse than most people I know.

If there's any electric-guitar mojo above 15 KHz, the average adult can't hear it. And even if they could, it's being cut to almost nothing because, as Clark noted, guitar cabs cut out above 4 to 8 KHz.
 
You must log in or register to reply here.
Back
Top Bottom