It sounds great...but how does it work?

  • It seems to me the patent describes that the KPA sees the amplifier and cab as one system which introducés non-linear distortion on the incoming guitar signal. And that makes sense with the real world I guess (?:-) :( the speaker movement causes distortion in the output tubes dependent on frequency and amplitude of the output signal, where as the input signal introducés intermodulation distortion. So the KPA doesn't use IR at all but by having a relation of the in coming signal to the non-linear amplification and the relation of the outgoing signal on the non linear amplification you could describes the behavior of the whole amplification/cab unit. To differentiatie between an 'amplifier' part and a 'cab' part is actually synthetic but could be done by drawing a line somewhere between these two influences of input and output signal.
    Another reason why i think it must work this way is that the KPA makes no difference between profiling an amplifier like a mic pre-amp and an amplifier /cab system: both can be described the same way with this influences on input and output signal ( but in case of the mic- pre- amp the output signal has hardly any effect at all).
    The axe fx and l6 gear can do non-linear amplification like the KPA does but they can't easily capture all the subtle differences s of different amps. They have to program the very complex interactions between these signals and the non-linear amplification. Should that be the explanation for the sound differences? It's too complicated....
    the Boss was here :) I wished he enlightened us a bit :)

  • from the Profiling Guide:
    "Start by choosing a Rig in Browse Mode. It’s a good idea to choose a Profile that is somewhat related in character to the reference amp. This will give you the chance to compare the existing Rig with that of the setup you are about to profile, thereby ensuring that you get a result that is even better suited to your needs."


    ;)


    I totally hear you Don; I often see you trying to get folks to RTFM.


    Trouble is, I have the same problem as Booyah:


    We have a manual???? hahaha just couldn't resist...


    Mind you, if I had a printer I'd read the manual. I just cannot justify reading such a long PDF on my studio computer. Can't use mobile devices, otherwise I'd have done that (big-time EM/BT/WiFi/Cell-phone sensitivity).


    If a hard-copy manual (the full one/s) were available I'd be happy to buy it and eager to read it. It'd beat the Hell out of our local TV programming, IMHO.


    Sorry Don for my insolence, and thank you for pointing out that as usual, it was in the manual. That very damned good manual.

  • I googled a litlle to further on the issue, but i think it is too complicated for me to understand But i did find some interesting articles f.e. http://www.academia.edu/232522…ime-invariant_convolverAs far as I understand now, every non-linear system is mathematically calculable (there is a very complex formulation for it, see above article). But that is way too CPU-hungry to perform in real time without latency. To overcome this problem all kind of 'tricks' have been used to try do diminish calculation needed, but all with trade-offs in accureteness of simulating a real world amp. I don't know which method Kemper is using. I guess every different approach results in a kind of different tonal flavor, exactly as is heard in de modellers of different brands.In the above mentioned article there is reference to the VST plug-in of acustica audio ( http://www.acustica-audio.com/ ) which claim to use a very high quality way of modelling (with four models merged into its algoritm : http://www.acustica-audio.com/…cle&id=14&Itemid=247).Has anyone experience with this plug-in? It seems of very high quality and I wonder if anybody has tried it out with sampling a guitar amp?

  • I googled a litlle to further on the issue, but i think it is too complicated for me to understand But i did find some interesting articles f.e. http://www.academia.edu/232522…ime-invariant_convolverAs far as I understand now, every non-linear system is mathematically calculable (there is a very complex formulation for it, see above article). But that is way too CPU-hungry to perform in real time without latency. To overcome this problem all kind of 'tricks' have been used to try do diminish calculation needed, but all with trade-offs in accureteness of simulating a real world amp. I don't know which method Kemper is using. I guess every different approach results in a kind of different tonal flavor, exactly as is heard in de modellers of different brands.In the above mentioned article there is reference to the VST plug-in of acustica audio ( http://www.acustica-audio.com/ ) which claim to use a very high quality way of modelling (with four models merged into its algoritm : http://www.acustica-audio.com/…cle&id=14&Itemid=247).Has anyone experience with this plug-in? It seems of very high quality and I wonder if anybody has tried it out with sampling a guitar amp?


    That plugin is a nightmare!
    BTW, it doesn't do distortion.

  • I thought all Acustica's stuff employed convolution (IR processing), where multiple snapshots and interpolation between them were employed in order to approximate all settings.


    At least, this is how I've always viewed its "models" of outboard-analogue hardware such as compressors and preamps.

  • I don't understand how Bias Head is different from the CK's patent linked here before...


    If I'm not mistaken then BIAS actually just works with a tonematch. So it mimics the EQ but nothing else. You have to set up the amp yourself to sound close to your reference amp, then you match it to get closer. While this works quite well, profiling is superior on many levels.

    MJT Strats / PRS Guitars / Many DIY Guitars -- Kemper Profiler Rack / Kemper Remote / InEar

  • There is a tiny sorcerer inside doing quantum magic. CK catches them in local forest then builds miniature houses for them and puts them inside Kemper so they can live in there comfortably.

  • i found another interesting read: https://www.google.nl/url?sa=t…M4Yd1WRAlLnFL4Vsz4X1Xa4Kg
    I think this must be very similar to how the Kemper works? It is a bit counter intuitive you can capture a chain of different devices (pre-amp, power amp, cab, mike) in one formula. When this is indeed the way the Kemper works , some things are more clear: with this method you only need about 10 to 20 numbers to fill in the formula which leads to a very accurate moddeling of input. So no need for big IR data strings. It would also mean that what has always been stated by Kemper that there is no signature sound is true ( at least definately totally not in the way modelers based on linear modeling have) , and that there are no 'basic amps' stamped in to the memory of the Kemper.
    In this article an error of 0.7% was achieved in modelling a tubescreamer. Although i think i can 't hear such an error, the article seems to imply the profiling method (which the authors use, i don"t mean the Kemper method) can be upgraded. It also implies that with more computerpower results could be even better.
    But what i can 't comprehend is how the Kemper team has found a way in transforming parameters which are linker to (tube) amps (like bias, defenition, compression) in adequate changes in this formula.
    Well, in the end only sound matters and the Kemper is great at it

  • i found another interesting read: https://www.google.nl/url?sa=t…M4Yd1WRAlLnFL4Vsz4X1Xa4Kg
    I think this must be very similar to how the Kemper works? It is a bit counter intuitive you can capture a chain of different devices (pre-amp, power amp, cab, mike) in one formula. When this is indeed the way the Kemper works , some things are more clear: with this method you only need about 10 to 20 numbers to fill in the formula which leads to a very accurate moddeling of input. So no need for big IR data strings. It would also mean that what has always been stated by Kemper that there is no signature sound is true ( at least definately totally not in the way modelers based on linear modeling have) , and that there are no 'basic amps' stamped in to the memory of the Kemper.
    In this article an error of 0.7% was achieved in modelling a tubescreamer. Although i think i can 't hear such an error, the article seems to imply the profiling method (which the authors use, i don"t mean the Kemper method) can be upgraded. It also implies that with more computerpower results could be even better.
    But what i can 't comprehend is how the Kemper team has found a way in transforming parameters which are linker to (tube) amps (like bias, defenition, compression) in adequate changes in this formula.
    Well, in the end only sound matters and the Kemper is great at it

    The article is great, and I understand its implication. I admit though I'm a not skilled mathematician enough to comprehend all the steps.

  • Doesn't matter who said it.

    If this is the case you'll never be able to gather useful information from the interwebz.


    Such a statement would be more credible if somebody with inside knowledge (which Sinmix clearly doesn't have) said it.
    So many false rumors have been spread already.


    But heck, if you think the technology is so easy and you find the price of the Profiler questionable, why don't you build your own?

  • I think many users of this forum who come here for news, interesting tips, ideas etc are getting a bit fed up with this string of criticism that appears all the time, it appears in sinmix's name and doesn't help anything or anyone.


    It's all getting very monotonous and is bringing the forum down, a forum that many state is very helpful and friendly.


    Please keep the invective down.

  • I think many users of this forum who come here for news, interesting tips, ideas etc are getting a bit fed up with this string of criticism that appears all the time,

    Thanx so much..Yes me for example..


    But who am I;Do I belong to the few who are "fed up" as you put it nicely;Or do I belong to the "silent majority";


    I dont know..but for sure I know that I am not alone.