The Aliasing Problem

  • Cristoph Kemper said some time ago that fixing the aliasing would be no problem, but would require to speed up the algorithms, and he was hoping to save the processing power for future features. So the computing power is indeed to be treasured ;)


    Actually it was fixed back then on the next firmware update after he posted that (1.6.1 I believe), so at that time they did make the algorithm more intensive to solve the aliasing issue and now the CPU cycles are being used for this rather than other effects right now, so the scope for further improvement is probably non-existent at this point :/


    Are you 100% sure the aliasing effects you're hearing aren't pickup warble and frequency interference? A real amp will give the same results in this case.

    Edited once, last by Per ().


  • Actually it was fixed back then on the next firmware update after he posted that (1.6.1 I believe), so at that time they did make the algorithm more intensive to solve the aliasing issue and now the CPU cycles are being used for this rather than other effects right now, so the scope for further improvement is probably non-existent at this point :/


    Are you 100% sure the aliasing effects you're hearing aren't pickup warble and frequency interference? A real amp will give the same results in this case.

    Hi Per, yes, indeed I read these lines in a thread about aliasing back then, before 1.6.1 was released.
    I am 100% sure, I profiled my amp and compared the profile with my amp: in the first case there were signs of aliasing, in the second case none.
    Also, this has been acknowledged by kemper support: I opened a ticket, they could reproduce the problem, solved it and they're gonna include the fix in the next release.

  • Laimon,


    I never take it for granted that people understand concepts like digital aliasing. I didn't intend to imply that anyone in particular didn't understand it, I just wanted to make sure that people that haven't ever created a digital algorithm or been forced to wade through the calculus needed to solve equations associated with time to frequency domain still had a point of reference in the discussion.


    I am sure that you have produced a very valid test case that proves there are artifacts created by your profile. I have heard them and agree. I am just pointing out that there are several causes of digital artifacts. I am equally sure that the smart guys at Kemper amps are going to be able to fix it with the hardware they have chosen to design the Kemper profiler with.,

  • Well, release 2.5 beta is out, and the problem seems to be solved. Can't crank up the volume too much now to figure out, but couldn't hear any problem with the profile KM Andromeda, which previously showed a loooooot of aliasing. So, for now, great! :)

  • I read through this thread and I might have a similar/the same problem. I'm on 3.0.2 and all profiles except clean ones have a fizzy sound on top which increases with increasing gain.
    I'm wondering if this is just the way the Kemper sounds or if there's a problem with my unit. Depending on which monitoring I'm using, these high frequencies can get really annoying.
    The noise disappears with notes slowly decaying.
    By using a low pass filter, the noise can be reduced.


    I checked input and output levels and they seem okay, no red or even yellow lights.
    I recorded some examples with factory profiles. The first one is Gundy - Brian'sStrayDogs profile, the second one is the same profile with a low pass filter.

    External Content soundcloud.com
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.

    External Content soundcloud.com
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.

    The other examples are recorded with the mentioned KM Andromeda profile, one with and one without a low pass.
    External Content soundcloud.com
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.

    External Content soundcloud.com
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.


    Can anyone help me? Can you check if these profiles sound the same on your KPA?


    Edit: Here's another one: KM - Bog People. It nearly sounds like not using a cab simulation.

    External Content soundcloud.com
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.

  • I used a Stratocaster but the fizzy noise occurs with my other guitars as well. Until now, I didn't have the chance to listen the file which mdeerocks recorded but I will listen to it and compare as soon as I get home.
    Perhaps this isn't a problem at all and I'm just not used to the sound over headphones or frfr.
    However, I think these are frequencies higher than 8kHz which a usual guitar speaker cannot play and this a profile of a real amp shouldn't have such frequencies. Correct me if I'm wrong..

  • Don't know, here is bog people for you.


    https://soundcloud.com/mdee-rocking/bog-ppl-track-1/s-y0vYx


    I don't have any other profiles you've recorded.


    Thank you for recording this profile. It sounds similar to my recordings so I guess my profiler doesn't have a problem at all. In my examples the fizzy sound is louder and more present but I guess the reason for that is I used single coils.
    Anyway, this sound annoys me on some profiles and like I already mentioned, these are frequencies above the spectrum of a guitar speaker.
    Especcially on one PA system I'm using, this is pretty annoying.

  • However, I think these are frequencies higher than 8kHz which a usual guitar speaker cannot play and this a profile of a real amp shouldn't have such frequencies. Correct me if I'm wrong..


    The point is that, depending on the rig, you're probably listening to a closely mic'ed cab. This is more or less what you hear if you put your ear where the mic is (and what audience hears at any concert, unless someone EQs the strip differently).
    On a side note, pretty sure a guitar cab is like EQ'ing out those frequencies among other things.


    The solution I see:
    choose rigs according to the mics used for profiling the amp;
    choose them according to the mic's positioning;
    use a guitar cab;
    EQ things you don't want to hear out.


    HTH :)