Blind Test: Axe-Fx III vs. Kemper Profile

  • I'm never quite sure how to take some of the comments. The same people marvel at the KPA's accuracy while simultaneously talking about how a profile is "thicker and smoother", or earlier the suggestion that the profile sounded "more real". Should the profile not faithfully replicate the source tone, no matter how "unreal" or "thin and jagged" it sounds? If you're suggesting the KPA makes things sound "more real" than they actually are, isn't that an unwitting admission that profiling imposes its own qualities? We can't have it both ways.


    The circumstances are a bit different given this is an Axe-FX, not a Marshall. The profile also had multiple gain stages that the KPA struggles with and was tweaked. Regardless, we should be able to approach it with a little intellectual honesty.

  • I'm never quite sure how to take some of the comments. The same people marvel at the KPA's accuracy while simultaneously talking about how a profile is "thicker and smoother", or earlier the suggestion that the profile sounded "more real". Should the profile not faithfully replicate the source tone, no matter how "unreal" or "thin and jagged" it sounds? If you're suggesting the KPA makes things sound "more real" than they actually are, isn't that an unwitting admission that profiling imposes its own qualities? We can't have it both ways.


    The circumstances are a bit different given this is an Axe-FX, not a Marshall. The profile also had multiple gain stages that the KPA struggles with and was tweaked. Regardless, we should be able to approach it with a little intellectual honesty.

    I hear profiles in which the Kemper is spot on to the original amp. I'm not concerned with the Kemper's ability to profile the Axe or not. I have a Kemper because I've never heard anything come out of a Fractal product I liked.


    I don't think the Axe's patch sounded good at all regardless; not my style of music.

    I definitely wouldn't want to profile it.


    It's also easy to make the Kemper sound bad with just a few parameter changes.

    The key to everything is patience.
    You get the chicken by hatching the egg, not by smashing it.
    -- Arnold H. Glasow


    If it doesn't produce results, don't do it.

    -- Me

  • I'm not going to debate your preferences, those are yours. I may think there's a limit to what we expose ourselves to to suggest we've never heard anything come from a product that we liked. Given all the options and ways to manipulate gear, I've heard all sorts of things I thought I didn't like made to sound great, sometimes surprisingly so by just exposing myself more to them.


    But my point was really about the mental gymnastics I sometimes see when it comes to comparisons, which isn't exclusive to any one product people are enthusiastic about. In this case because it's compared with an Axe-FX, a competing device, people are careful to focus on Kemper-friendly critiques. Perhaps the biggest stretch is the insinuation that the Kemper profile of an Axe-FX makes the Axe-FX "more real". On one hand we suggest the KPA is "spot on" to a real amp, and on the other hand suggest its profiles are 120% better. There's a clear disparity in standards there.


    I'm probably being overly pedantic for some people, but these are really interesting discussions and tests that I wish were more objectively focused. Given, it's a Kemper forum. Any manufacturers forum behaves much the same way.

  • word of advise: if you want to make a serious comparison, maybe don't choose soundcloud to host it. don't you realize how much soundcloud degrades the quality of the recording? why talking details when you look at things through a filter which possibly even masks some of the broader strokes?

    Get in touch with Profiler online support team here

  • word of advise: if you want to make a serious comparison, maybe don't choose soundcloud to host it. don't you realize how much soundcloud degrades the quality of the recording? why talking details when you look at things through a filter which possibly even masks some of the broader strokes?

    Below are two samples that were saved at 48 KHz / 24-bit. One of them is the original source audio, which was recorded and saved directly from my DAW. The other is the same audio, but uploaded / streamed from SoundCloud and recorded direct to DAW. Can you tell which is which? What differences do you hear that you feel would serve as an impediment to assessing the sound quality of the originally posted clips in this thread?


    Sample 1

    Sample 2

  • ColdFrixion

    Soundcloud streams with 128kbps unless you're a Go+ subscriber which can offer streams with a 256kbps data transfer rate. Where I come from, 128kbps is no option for critical listening.

    Okay, well, give a critical listen to the samples I posted and you tell me if you hear a notable difference. If you do, can you describe it, and are you able to identify which sample is from the original source and which was streamed from SoundCloud?

  • Okay, well, give a critical listen to the samples I posted and you tell me if you hear a notable difference. If you do, can you describe it, and are you able to identify which sample is from the original source and which was streamed from SoundCloud?

    no, I won't. but it would be a good idea if you would put your comparison up on your dropbox and replace the soundcloud links.

    Get in touch with Profiler online support team here

  • no, I won't. but it would be a good idea if you would put your comparison up on your dropbox and replace the soundcloud links.

    If the differences between the samples I posted are notably audible to you, then it stands to reason it shouldn't be difficult to identify which one is the more inferior of the two. If they're not notably audible to you or anyone else, I see no reasonable basis for not using SoundCloud.

  • They're talking about null tests. Null tests are excellent at determining whether audio is identical. However, they're not necessarily great at determining whether one sample is inferior to another. For example, try this:


    Record a DI guitar sample.


    Now, reamp that sample and record it onto one track in your DAW. Then, create another track and make another recording of the reamp'ed sample on to that track. You'll now have two reamped samples of exactly the same DI recording. The two tracks will sound identical. But are they? If you sync both tracks perfectly and reverse the phase of one of them, you'll find that they do not null. In fact, you'll hear more residual noise from a reamped distorted guitar track than you will when comparing the residual noise from an 128kbps MP3 and WAV file of a distorted guitar track.

  • http://www.noiseaddicts.com/20…und-quality-test-128-320/


    I got this correct. You can hear the slight difference in the shaker. But honestly I really had to focus.

    I can hear a difference, but it's the same difference I typically hear with low bitrate vs. high bitrate MP3's. The cymbals and high-end are often the biggest giveaway; swishy sounding, especially the cymbals. The lows and mids are also affected, but it's easiest to hear it in the high-end.


    Fortunately, SoundCloud's encoding process offers a higher quality conversion than standard 128kbps MP3's. I converted the source sample I posted above to a 128kbps MP3 and then compared the residual noise using a null test between the SoundCloud sample and the 128kbps MP3. The MP3 had more residual noise than the SoundCloud sample, and I also didn't hear the artifacts associated with the low bitrate MP3 in the SoundCloud sample, either.

  • Below are two samples that were saved at 48 KHz / 24-bit. One of them is the original source audio, which was recorded and saved directly from my DAW. The other is the same audio, but uploaded / streamed from SoundCloud and recorded direct to DAW. Can you tell which is which? What differences do you hear that you feel would serve as an impediment to assessing the sound quality of the originally posted clips in this thread?


    Sample 1

    Sample 2

    Sample 1 feels a tad more lively. Original source?


    The difference isn't so massive that they're worlds apart and can't be used for simple comparisons sake IMO.

  • The first file sounded livelier to me too, actually. I'd have to import in daw and see if I could consistency tell them apart though. I'm skeptical of how detrimental SoundCloud streaming quality can be In this context, but perhaps it's good policy for us to provide the highest quality audio files to go along with tests.

    The bonanza

  • I preferred 2 and thought it must be the axe because it's newer technology and due to the infamous kemper high end boost . I found the harmonics of clip 1 (the axe) too pure and not authentic almost too sine like....... But as the kemper is trying to emulate the axe it's failing too reproduce this so I am confused.............. as I am as to the purpose of this test.

  • Here's the frequency spectrum from both samples. In my opinion, negligible to the point of irrelevance.

    A spectrogram doesn't necessarily show anything in terms of distortion or possible phase issues caused by a bad codec. Also, my initial comment was a generic comment for those who like to compare things and use Soundcloud for their audio files. If I would have wanted to direct it to you specifically, I would have tagged you in my post. As I said, it is a bad starting point to use a low quality player for all that. This doesn't mean that in your particular example, the difference is obvious to everybody. Especially not on every speaker.

    Get in touch with Profiler online support team here

  • A spectrogram doesn't necessarily show anything in terms of distortion or possible phase issues caused by a bad codec.

    I was using an RTA (real-time analyzer) rather than a spectrogram, however the latter is useful in its own right. For phase issues, I'd use a correlation meter. An RTA can help in analyzing and/or diagnosing frequency response issues.


    my initial comment was a generic comment for those who like to compare things and use Soundcloud for their audio files.

    Regardless whether the comment was directed at me specifically or not, I felt it was applicable to me since I used SoundCloud to post the original test in this thread, thus my response.


    As I said, it is a bad starting point to use a low quality player for all that. This doesn't mean that in your particular example, the difference is obvious to everybody. Especially not on every speaker.

    Well, the difference between 48 KHz and 96 KHz isn't obvious to everyone, and especially not on every playback system. However, there are those who claim to be able to hear an obvious difference between both and thus recommend recording at 96 KHz whenever possible. However, just because someone deems 48 KHz to be inferior to 96 KHz hardly means 48 KHz is a bad starting point.


    That said, in the case of basic (free) SoundCloud users, all samples uploaded to the service are processed through the same encoders and will thus share the same qualitative attributes, which means any sonic differences between the samples will generally be related to something other than the encoding. In this case, that something would be the profiling process.


    For the record, I'm not opposed to someone using a lossless format for blind testing if that's what they prefer, however I find SoundCloud's encoding quality completely sufficient for exposing discrepancies imposed by the profiling process.


    Further, at least some mastering studios use SoundCloud. This one, for instance.