This. Previous FW versions lacked the ability to compare 2 profiles, determine their differences, and save the results in a cab.
What I believe is that the Profiler was initially aimed at profiling whole rigs (or rigs once-and-for-all, so to speak). Then popular demand made swapping cabs - and the faithfulness while excluding cabs and using a real guitar cab - a higher priority, so they implemented further profiling functions.
Because now you can make profile in 2 steps:
1. profile with cab+mic
2.profile with DI box (between amp&cab).
So KPA will have AMP+CAB+MIC and AMP DI profiles.
Yes, I got that now, I think. 
Thank you.
Ingolf's question stands: KPA has relied on magic since day one, but it'd be interesting to understand how the magic has changed, how generating tones and analyzing those tones to create a profile has suddenly evolved to where the Kemper can now distinguish which aspects of those sampled frequencies can be attributed to specific components of the profiled combo. No one's complaining but it's cool; like, who figured it out? And where do we send the flowers?
Because you now make two profiles, at different points in the signal chain and combine them.
a. A profile of a DI signal taken post power amp (which is used to fill the amp block).
b. A standard profile via mic (who's cab is used to fill the cabinet block).
The DI profile is taken via an attenuator that's basically run on the speaker cable from your amp, so this means that the amp is still reacting to the speaker. So all that's missing from this picture is the speaker (and mic) itself, it's basically already split by you at this point, not algorithm involved, all that's missing is the speaker & mic component.
When you merge the cab from the second profile with the first (which has no cab) then it just tries to get the result to match the sound of (b), what's required in order for that to happen is the separated sum total of the cab & mic. Therefore perfect separation.
The trick is the evaluation of what parts contribute to that final sound, to my ears it seems like it's just using the frequency response curve and doing a tone match, kinda like the frequency convolution used by IR based cab sims. I could be wrong though. We'll see with more profiles, from mine I'm preferring the standard profiles to the ones made this new way which feel like a subtle step backwards in authenticity, but if I were playing live then the new approach would be the way to go with a real guitar cab as a monitor as that side is much better than before now, the audience likely wouldn't notice much difference FOH.
This. I believe test tones have not been changed a lot.
It's a shame that he profiler can't accept two input sources simultaneously, e.g. monitor in and spdif in, so you could send both signals at the same time to the KPA. But I'm guessing two channel AD will be part of the second generation KPA (as and when that comes along) to facilitate this.
What would the use be?
It would mean you wouldn't need to profile twice for this feature. Instead it could grab and process both input signals (DI and mic) simultaneously, reducing complexity.
Other potential uses with upgraded CPU would be of course stereo path in, or dual amp paths.
Oh, you meant while profiling!
But the return of the two signals would mix anyway, making it impossible to discriminate what generated what, if I got you right.
The two sequences would need to be sent one after the other anyway, saved time would be almost irrelevant (30"?).
A question:
What DI boxes can be used at the Speaker Output that do not include a Speaker Simulator on the XLR output?
Oh, you meant while profiling!
But the return of the two signals would mix anyway, making it impossible to discriminate what generated what, if I got you right.
The two sequences would need to be sent one after the other anyway, saved time would be almost irrelevant (30"?).
Actually the KPA does have two channel AD right now I realized, the return/alt input and the main input/spdif are separate, so they oculd feasibly do this right now in a single pass.
The signals would not be mixed, they'd be separate signals, just as they are now.
A question:
What DI boxes can be used at the Speaker Output that do not include a Speaker Simulator on the XLR output?
I use a THD Hotplate for this.
Since the Hotplate allows a line to be sent to a mixing board - it must has a speaker simulator circuit in it?
No. Line level doesn't mean speaker sim, what you do with the signal is up to you, it's not there specifically to go to some mixing board, it's just a line out jack (not XLR).
The line out is simply a second attenuated circuit after the first, it's has it's own control so you can dial it down to not overload anything attached which it will easily do as it's simply attenuating the speaker output, which is why you have to be careful about setting that level if you connect anything (including the Kemper) to the output of a THD hotplate. I do not recommend messing around with this stuff if you are not experienced with working with these setups.
Understood - I'm used to all manner of DI boxes and amp attenuators (I've been using tube amps since they were cheap). That's why I asked the question - I'm wondering what specific DI boxes Kemper had in mind when they designed this process.
The THD manual indicates the XLR output is intended for sending a signal direct to a mixing board, making amp miking unnecessary. If it sounds like there is no EQ applied to that output, the sound technicians I know would rather use a microphone
Understood - I'm used to all manner of DI boxes and amp attenuators (I've been using tube amps since they were cheap). That's why I asked the question - I'm wondering what specific DI boxes Kemper had in mind when they designed this process.
The THD manual indicates the XLR output is intended for sending a signal direct to a mixing board, making amp miking unnecessary. If it sounds like there is no EQ applied to that output, the sound technicians I know would rather use a microphone
Might be a new model. Mine is a few years old now and quite simple, no XLR output on it, just a jack.
The sound coming out is pretty brutal, not something I'd want to hear/use without some heavy processing and additional cab emulation. If that's their idea of speaker simulation then they haven't let on anywhere. It sounds like a DI, not a speaker.
Then again stuff going through an IR frequency convolution still sounds like a DI to me too, so what would I know.
I'd also assume that they would want to market/advertise the line out as having speaker emulation on it if they did, and they don't.
So far i use a cheap Behringer Ultra G-100, its possible to disable the cab sim
Because you now make two profiles, at different points in the signal chain and combine them.a. A profile of a DI signal taken post power amp (which is used to fill the amp block).
b. A standard profile via mic (who's cab is used to fill the cabinet block).The DI profile is taken via an attenuator that's basically run on the speaker cable from your amp, so this means that the amp is still reacting to the speaker. So all that's missing from this picture is the speaker (and mic) itself, it's basically already split by you at this point, not algorithm involved, all that's missing is the speaker & mic component.
When you merge the cab from the second profile with the first (which has no cab) then it just tries to get the result to match the sound of (b), what's required in order for that to happen is the separated sum total of the cab & mic. Therefore perfect separation.
.
I guess that we get total separation for the Cabinet + Speaker + Mic part here.
I still wonder if the Amp can be separated from the DI + Speaker impedance influence ?
Matti
So far i use a cheap Behringer Ultra G-100, its possible to disable the cab sim
Thanks for another example
Another good reason why the new profiling separation is good for us FRFR users: more cabinets. As a natural corollary of the better separation of amp, there will be better separation of cabinets. This is good, since till now, blending of cabinets with amplifiers has always been hit and miss due to the amplifier characteristics being buried in there.
It means more choice for us, so definitely a step forward. I wonder if the 1000 profile limit includes amps + cabinets or just amps or just cabinets.
Well, I must be old school then For em the main advantage of Kemper amp is that I can profile whole chain and be done with it - I don't think I ever swapped a cab or done anything except turning treble/bass/etc. around for 5 minutes (I mostly use my own profiles though). I don't think I'll be using that feature - reasons: I prefer playing over fooling around with gear and I have no desire to use Kemper live, this is what real amps are for. I'll check it out anyway.
Having said that, I am a bit confused where this development is leading to - IMO IRs are dead end and almost always sound bad and unresponsive, I am not sure how different this new feature is from slapping an IR over DI profile.
