Okay... another comparison....

Drew_TNBD

So... which is Kemper and which is the amp?

https://dl.dropboxusercontent.com/u/630473/GearDemos/PreampComparison/1.wav
https://dl.dropboxusercontent.com/u/630473/GearDemos/PreampComparison/2.wav

jonnyburgo

Hmm, B had more hiss and was less smooth sounding, so Id be inclined to say that b is the real amp. I'm wrong arent I?

ICBM

1 is the real amp.

If it's not, it still sounds miles better!

guitarfishbay

I preferred 1, it sounded more balanced. 2 was duller on top and had less punch. Are they level matched? Not on my DAW comp, 1 felt louder to me but not sure if that is just the low end.

steamabacus

Well, 1 sounds better than 2 for me. Same sort of reasons as last time - fuller, more depth, more open, more '3-d'. Also more intrusive background hiss on 2.

So, based on the assumption that the real amp must sound better (?) I'd say the first one is the real amp. Although all I can really say for certain is that I prefer the first one.

equalsql

Reckon 1 is the Kemper and 2 is the amp. There seemed more natural bite when you were digging in around the 25 sec mark.

guitarfishbay

2 might be the Kemper, but only based on hearing the youtube video I sent to Drew that was using the Kemper powerhead.  I've not really heard any valve preamp vs Kemper DI tests so I'm not sure what to expect.  From this video the Kemper Powerhead DI profile isn't as thick sounding and the mids are placed differently.

Comparison goes amp first, Kemper Powerhead second.  I think the difference on the clean tones is massive, and the EVH gain sounds thicker and ballsier on the real amp IMO.  The Engl seems closer... annoyingly (as discussed by PM) there is no real explanation on how the DI profiles were made - so it is impossible to know how the Kemper profile was created and whether or not it is just the preamp signal, or included the poweramp signal (e.g. taken using something like a Palmer PDI).

guitarfishbay

I just re-listened to the comparison on the MDR-V55 Sony headphones that I leave at work.  When I last listened I used my Audio Technica ATH M50s at home.  Just once again shows how different listening environments/equipment make a difference -  I didn't really like the Kemper EVH gain tone at home on the ATH headhones but on the Sony headphones it sounds good.

Drew_TNBD

Interesting comments.

Well 1 was Kemper, and 2 was Axe FX.

So first of all; this was not really a fair test, because I don't have a proper methodology for comparing yet. In fact I don't think it's really possible to compare them in a scientific manner, because the mechanisms for separating amps from cabs work very differently in both units. Running the Axe into the Kemper for cabs wouldn't make sense, because you cant completely turn off the aspects of the cab in the Kemper. So running the Kemper through the Axe's cab block wouldn't make sense either. I just had to make do with more of a subjective comparison. I tried to get as close as I realistically could.

So this was comparing a 5150 rig from the Kempers profile exchange that was recorded with a Mesa Boogie 2x12 cab with v30's... versus the Axe FX's 5153 Red model using a Redwirez Mesa Boogie 4x12 impulse response.

So yeah, totally not fair, and not exactly comparable. But I was really just trying to illustrate something I am noticing in a rather general sense when comparing, and bear in mind this is just my opinion:

The Kemper tends to sound more bassy and full than the Axe FX, with a more pleasing resonant character. To me it feels closer to the experience of having a cab mic'd up in a live room, where you're monitoring it from the control room. The Axe FX has this rather abrasive nastiness on palm-mutes that is really difficult to dial out, and with my setup is most noticable on open notes and a low power-chord at the 10th fret. It doesn't quite resemble the control room monitoring experience as well as the Axe. It doesn't sound bad, but... just different, and from a subjective experience, I prefer the Kemper.

By default the Axe models generally seem treble heavy and kind of distant, even when using impulse responses taken where the microphone was only an inch away from the speaker cone. So it takes a fair amount of work to dial them in. Some of the default presence and resonance settings are not very good, and (having had this wish before I got the Kemper) it often makes me wish I could save default setups for each amp model, so that any time I switch to it I get the sound that I want.

Now... I've said before that I got the Axe FX to sound almost indistinguishable to my real Diezel amp through a real guitar cab. This is still true. If you take just an amp block, route the Axe into the FX return of the Diezel and then dial the EQ, presence, and depth (resonance) in... you get something very close to a real amp experience.

If you do this with the Kemper, it sounds like a modeller but louder. It doesn't really sound like a real tube amp... you lose something when you disable the Kempers cabinet section. Even when you use direct profiles (I got some of the rig exchange).

So ... all of this leads me to a few things: I strongly suspect I don't bond well with impulse responses. In actual fact, I think I really dislike them. They make every amp sound distant and unrefined. Even the IR's I've taken of my own cabs suffer from this. The stock Axe FX cabs especially suffer from this. The only IR's I've used that I liked were the GuitarHack ones... but they too suffer from it.

The Kemper is doing something different when it comes to the cabinet modelling. I don't know what exactly, but it sounds much much more realistic to me. If you have a friend play some low tuned power chord palm muted chugs through a real amp and you then move your ear over the speaker cone... you'll hear a massive world of tones available to you, and you'll hear the characteristic low-end thump that the Kemper is producing quite nicely actually. The Axe FX does not do this.

The Axe has really good amplifier models; preamp and poweramp. But I do not like the cabs. Not when compared to the real thing and not when compared to the Kemper.

The Kemper has really nice amplifier profiling capabilities, and whilst it isn't always 100% you can get to 90% of the original tone quite easily. The cab modelling just sounds more realistic to me. It has that trouser flap quality that the Axe is lacking.

So here is where I stand at the moment:

Overall I prefer a real valve amp into a real cab. That is (and probably always will be) the benchmark.
The Kemper I prefer for direct recording.
The Axe FX II I prefer for running into a poweramp and cab, and using as an amp head in its own right.


This could all be honeymoon period stuff of course. My intention is to keep all three solutions for the foreseeable future. We are most likely moving house before the end of the year, and I will not be able to crank a valve amp in our new flat. So Axe FX II and Kemper are going to be lifesavers in that respect.

I do want to figure out if I can get closer to the real amp with the Kemper. That's kind of the next thing here, but right now it's more than usable. It's very enjoyable to play, and requires almost no tweaking at all.

PVO_Dave

Drew, have you upgraded to the Beta 2.6 F/W on the Kemper? It's supposed to further improve the profiling process.

Drew_TNBD

PVO_Dave said:

Drew, have you upgraded to the Beta 2.6 F/W on the Kemper? It's supposed to further improve the profiling process.

Yes, I jumped straight to it. Didn't bother with 2.5 or 1.8.

I have to say, my whole experience with the Kemper this time round has been a completely different one from when I owned one a few years back.

PVO_Dave

Drew_fx said:

PVO_Dave said:

Drew, have you upgraded to the Beta 2.6 F/W on the Kemper? It's supposed to further improve the profiling process.

Yes, I jumped straight to it. Didn't bother with 2.5 or 1.8.

I have to say, my whole experience with the Kemper this time round has been a completely different one from when I owned one a few years back.

Great

Absolutely love mine, not profiled anything myself, best purchases I've made so far have been from TopJimi, worth checking out, although not sure he does heavier amps (as yet)

guitarfishbay

Interesting.

A lot of talk on the Kemper vs IR's focusses on speaker movement, IR's are technically static and any 'movement' is simulated (e.g. through the speaker tab on the Axe).  Regardless of this result (and so far I'm surprised there is this big a difference) I still say the Axe is way ahead of the software plugs I've used and I am quite enjoying using it through a cab.  I don't think these tests would sway me to look at the Kemper, but that is mostly a workflow thing plus I don't want to have to own another amp, even if I'd like to.

IIRC, Periphery recorded '2' using the Axe through an EVH power section and mic'd up.  The name escapes me but there's a rock producer who mics up the Axe for his tones too.  If that is the way of getting the best tones with it then maybe the Kemper is the best option for you for recording.  I don't know of anyone micing up a Kemper profile, but why would you if you can get the direct profile to sound good already.

steamabacus

Haha. Interesting to hear them both. That's the first time I've heard a direct comparison between Kemper/AxeFX and I'm actually quite surprised the Kemper sounds that much better. And I do mean 'better quality' rather than 'different in a way I like more'. That's just one test, though.

Maybe it would be useful to do a comparison where, rather than trying to match tones to sound the same, you recorded a short section of music - one real amp, one Kemper, one Axe FX - and just mixed each to sound 'right' in the context of the track? One may be much quicker and easier to get what you want? Or one may sound absolutely superb but take ages to get it to a point that works for you? Listening back after six months, all 3 may sound quite acceptable - but a great performance will always stand out. Whichever method makes it easier to capture an inspired take is surely the best one to go with

It kind of reinforces my view that, whatever method you use to record - real amp, digital modeller, analogue DI'd preamp, whatever - production and recording skills are really important. I've no doubt that digital modellers can sound great on a recording. I've also no doubt that a real amp can sound poor (if badly recorded or in a crappy sounding room).

guitarfishbay

steamabacus said:

It kind of reinforces my view that, whatever method you use to record - real amp, digital modeller, analogue DI'd preamp, whatever - production and recording skills are really important. I've no doubt that digital modellers can sound great on a recording. I've also no doubt that a real amp can sound poor (if badly recorded or in a crappy sounding room).

Exactly how I feel.

The old saying 'only as strong as your weakest link' holds true.  The nice thing is that these days you have a lot of choices for workflow.

ICBM

Drew_fx said:

The Kemper is doing something different when it comes to the cabinet modelling. I don't know what exactly, but it sounds much much more realistic to me. If you have a friend play some low tuned power chord palm muted chugs through a real amp and you then move your ear over the speaker cone... you'll hear a massive world of tones available to you, and you'll hear the characteristic low-end thump that the Kemper is producing quite nicely actually. The Axe FX does not do this.

What power amp are you using with the Axe FX?

I mentioned in the other thread that one of the major giveaways (as it turned out!) was that the Kemper seemed to be more heavily damped than the real amp. But if the Kemper is also *less* damped than the power amp you're using with the Axe FX, that might explain the lack of realistic speaker response from the Axe FX. If the power amp is too damped, it tightly controls the movement of the speaker cone and does not allow it to resonate as it does with a valve amp.

I'm coming very much to the conclusion - I'm not the first of course - that the amp-speaker interaction, and particularly the damping factor, is one of the most critical parts of the 'valve amp sound'.

Drew_TNBD

It's just a feel+intuition+experience+vibe thing. Someone else may think I'm completely wrong and that the Kemper sounds like a Pod! Who knows.... it's all subjective I suppose.

But considering you bring up software plugs... actually, the Axe has the same characteristics a lot of the time (like when I compared the Soldano model in Amplitube 3, and there were minutae differences...) and what are we using there as well? Impulse responses.

I have a feeling that impulse responses may be the bane of all my woes. As you say... they're static and movement is simulated. Essentially the Axe FX is faking speaker movement and resonance by changing the outgoing EQ curve and poweramp dynamics... the IR is then just acting as a filter. I think the Kemper is actually changing it's built in cabinet model. It will have some internal filtering parameters and dynamic parameters, and they will be set and the min-max curves will be changed based on the information being picked up by the microphone. Okay... that's totally a guess, but seems logical. So what you end up with is a set of parameters given values that correspond to the real world observations the Kemper is making, and then the rest of it is picked up by the dynamic speaker models built within the box. Again.... guessing.

Drew_TNBD

ICBM said:

Drew_fx said:

The Kemper is doing something different when it comes to the cabinet modelling. I don't know what exactly, but it sounds much much more realistic to me. If you have a friend play some low tuned power chord palm muted chugs through a real amp and you then move your ear over the speaker cone... you'll hear a massive world of tones available to you, and you'll hear the characteristic low-end thump that the Kemper is producing quite nicely actually. The Axe FX does not do this.

What power amp are you using with the Axe FX?

The FX return of the Diezel D-Moll. I did have a Valvestate too, but it didn't perform as well for me as it did for Clarky.

I mentioned in the other thread that one of the major giveaways (as it turned out!) was that the Kemper seemed to be more heavily damped than the real amp. But if the Kemper is also *less* damped than the power amp you're using with the Axe FX, that might explain the lack of realistic speaker response from the Axe FX. If the power amp is too damped, it tightly controls the movement of the speaker cone and does not allow it to resonate as it does with a valve amp.

I think part of that 'giveaway' was more to do with the fact that I cranked the presence a little higher on the Kemper and other EQ differences. It doesn't feel any more damped to play through than the amp. But I understand what you're saying.

I'm coming very much to the conclusion - I'm not the first of course - that the amp-speaker interaction, and particularly the damping factor, is one of the most critical parts of the 'valve amp sound'.

I 100% agree. With the advances in modelling tech, once this nut is cracked... I expect digital will take over. *gulp*

ICBM

Drew_fx said:

ICBM said:

What power amp are you using with the Axe FX?

The FX return of the Diezel D-Moll.

Well that blows my theory out of the water!

Drew_fx said:

With the advances in modelling tech, once this nut is cracked... I expect digital will take over. *gulp*

I agree, and in my opinion it will be no bad thing, as long as digital amps are made to the same sort of quality as valve ones.

Valve amps have a lot of maintenance issues, especially these days with poor quality valves. I don't think they will ever disappear entirely, but I could see them becoming much more a niche thing - almost exclusively simple single-channel designs intended for no-compromise recording studio use or home enthusiasts, rather than working musicians.

guitarfishbay

Drew_fx said:

I have a feeling that impulse responses may be the bane of all my woes. As you say... they're static and movement is simulated. Essentially the Axe FX is faking speaker movement and resonance by changing the outgoing EQ curve and poweramp dynamics... the IR is then just acting as a filter. I think the Kemper is actually changing it's built in cabinet model. It will have some internal filtering parameters and dynamic parameters, and they will be set and the min-max curves will be changed based on the information being picked up by the microphone. Okay... that's totally a guess, but seems logical. So what you end up with is a set of parameters given values that correspond to the real world observations the Kemper is making, and then the rest of it is picked up by the dynamic speaker models built within the box. Again.... guessing.

I think you're right.  But the Axe does let you tweak the response in a more detailed fashion (frequency, Q and amount) vs a plugin (at least I'm not aware of any that let you do that yet).

I was under the impression that the Kemper was doing something like you've described.  On paper if it works as described it should make it the better sounding solution for DI recording.

Drew_fx said:

I'm coming very much to the conclusion - I'm not the first of course - that the amp-speaker interaction, and particularly the damping factor, is one of the most critical parts of the 'valve amp sound'.

I 100% agree. With the advances in modelling tech, once this nut is cracked... I expect digital will take over. *gulp*

Yep.  IIRC Cliff did allude to having worked something out relating to this for use with a poweramp and cab (I won't be able to find the post, sorry, but IIRC it was in a massive thread on poweramps and speaker impedances) but he said it wouldn't be viable until the next generation product.  Can't source a quote for this...

It is a shame nobody has posted 'correct' data for commonly available standard cabs for tweaking the speaker tab during use with SS poweramps.

Drew_TNBD

guitarfishbay said:

It is a shame nobody has posted 'correct' data for commonly available standard cabs for tweaking the speaker tab during use with SS poweramps.

I was kind of under the impression that the data needs to change from amp to amp. IE: Using a range of valve amps with the same cab will result in the cab seeing a range of different impendances, frequency responses, and thus different resonance Q's and amounts... or have I got that wrong?