Test Your Ears - Kemper vs. The Real Thing - TEST RESULTS

    a bit late to the party , but...


    the ultimate KPA vs the real amp would be:


    after doing the profile, using a split box so the guitar signal could go direct to the real amp and the KPA ,
    recording both signals simultaneous



    imho

    Yup - if the manual is correct, this can be done by sending "git analogue" signal out of the kemper into the amp. Supposedly, this should be the exact same signal as hits the kemper input (analogue split at unity gain).

    Quote from Cederick

    Of course that would be nearly impossible to tell.
    But you're missing the point we're trying to make over and over :whistling:

    I think I got everyone else's point. Can you sum up yours in a few words?


    Also, I noticed you referenced how I "played the riffs". I didn't create those clips. I simply spliced them up.


    Anyway, I'm not trying to make a test that's easy. If I were, I would've simply posted two samples back to back in sequential order, per usual.


    For a follow-up test, I might create the same type of clip without the markers or visual aids. However, the only requirement would be that participants have to specify at which intervals the change in samples takes place.

    Quote from Michael_dk

    Yup - if the manual is correct, this can be done by sending "git analogue" signal out of the kemper into the amp. Supposedly, this should be the exact same signal as hits the kemper input (analogue split at unity gain).

    Yes, this would be smart - but also needs to be used in the context of the 3 clip format I've described so that users can hear the original differences. The point is not to see if two clips can be created that fool people, the point is to take a profile that differs from the reference amp and demonstrate in a 3rd clip that something can be done to make it sound like the amp. And to share how it was accomplished.


    If you start out with a profile that already sounds nearly identical then an A/B test isn't accomplishing anything meaningful. It has already been acknowledged that sometimes the KPA can reasonably match the reference amp. It is the case where it does not do so where there is something to be learned.


    Sonic

    Quote from SonicExporer

    If you start out with a profile that already sounds nearly identical then an A/B test isn't accomplishing anything meaningful.

    That should come as news to the Boston Audio Society, who've conducted listening tests using ABX testing to investigate things like the audibility of various filters applied in high-resolution wideband digital playback systems.

    Quote from SonicExporer

    If you start out with a profile that already sounds nearly identical then an A/B test isn't accomplishing anything meaningful.

    So, if the complaint is that the KPA's profiling doesn't sound nearly identical to the referenced amp, then I'm puzzled at how doing an A/B test where you can't pick out the difference isn't meaningful. If you are complaining that the samples are too good, then that's basically an admission that the KPA can profile something nearly indistinguishable. At this point it feels like you're trying to move the goalposts to embolden a predetermined conclusion.


    All the people hollering that this isn't meaningful seem to miss the point (and yes, I've read this and the other marathon thread to know everyone's perspective). The fact that you can barely, or not even pick out which is which in a mix is entirely the point.


    Yes, back to back you can hear the differences. Yes, we would all like the KPA to be more accurate with those differences. Yes, we can discuss what those differences are and how to best mitigate them when profiling. Yes, we can also acknowledge the differences while keeping perspective. Yes, tests that make it hard to pick out those differences are reasonable and valid. Yes, there are several ways to accomplish said test beyond what has been done.


    In the end, you're not happy to engage @'ColdFrixion''s point because you're insistent on having the conversation precisely in the place you want it. And let's not forget that your point has been acknowledged repeatedly, but once someone tries to say, "but consider how well the KPA does in spite of that", you again start barking that they don't understand your point. We do, but that doesn't mean we have to agree and echo everything you say.


    This thread is about testing your ears. If you want it done a different way, provide that test yourself. It's not the threads responsibility to teach you anything. For that you should either consult support directly or create a thread about profiling experiences and tips.

    Quote from MementoMori

    So, if the complaint is that the KPA's profiling doesn't sound nearly identical to the referenced amp, then I'm puzzled at how doing an A/B test where you can't pick out the difference isn't meaningful. If you are complaining that the samples are too good, then that's basically an admission that the KPA can profile something nearly indistinguishable. At this point it feels like you're trying to move the goalposts to embolden a predetermined conclusion.
    All the people hollering that this isn't meaningful seem to miss the point (and yes, I've read this and the other marathon thread to know everyone's perspective). The fact that you can barely, or not even pick out which is which in a mix is entirely the point.

    No, you misunderstand the complaint, it seems like you are having the same confusion as @ColdFrixion is having, although the reason eludes me.


    I've stated over, and over again, that nobody is arguing the KPA can sometimes capture the reference amp to a reasonably indistinguishably degree. However too often there are differences, and they range from subtle to not subtle. In a recording environment this can matter. There is a very specific issue we've identified (go back and read the first few pages of the other long Cameron comparison thread if you need to refresh). So performing blind tests or A/B comparisons is nearly useless unless the 3-clip format is used. The idea is to have a profile and reference clip that differs, so people can hear the difference that was there initially, and the a 3rd clip that now more closely matches the original profile clip. And share what was done to improve things, be it some technique during re-profiling, or refining, or tweaking in the KPA, etc.


    These utterly pointless blind tests or A/B comparisons do near zero to serve the purpose as almost nobody is arguing the KPA cannot sometimes achieve reasonably indistinguishable results. And almost nobody is arguing that most of the time the KPA still creates a real amp type of sound. But what it doesn't do is re-create the reference amp every time, and the results are sometimes definitely not trivial, especially in a recording environment.


    Sonic

    Quote from MementoMori

    This thread is about testing your ears. If you want it done a different way, provide that test yourself. It's not the threads responsibility to teach you anything. For that you should either consult support directly or create a thread about profiling experiences and tips.

    Yes!! It's going to happen just not on here. Anyone interested in testing and coming up with ideas PM me. No arguing, just for fun and maybe results.

    Quote from SonicExporer

    No, you misunderstand the complaint, it seems like you are having the same confusion as @ColdFrixion is having, although the reason eludes me.
    I've stated over, and over again, that nobody is arguing the KPA can sometimes capture the reference amp to a reasonably indistinguishably degree. However too often there are differences, and they range from subtle to not subtle. In a recording environment this can matter. There is a very specific issue we've identified (go back and read the first few pages of this long thread if you need to refresh). So performing blind tests or A/B comparisons is nearly useless unless the 3-clip format is used. The idea is to have a profile and reference clip that differs, so people can hear the difference that was there initially, and the a 3rd clip that now more closely matches the original profile clip. And share what was done to improve things, be it some technique during re-profiling, or refining, or tweaking in the KPA, etc.


    These utterly pointless blind tests or A/B comparisons do near zero to serve the purpose as almost nobody is arguing the KPA cannot sometimes achieve reasonably indistinguishable results.


    Sonic

    No one here has failed to understand your point. It was understood from the get-go. Your point is that without a reference sample to compare a profiled sample with, the test is pointless because there's no baseline for comparison, but that's precisely the point. You shouldn't need a baseline if the differences are significant, as some people in the other thread claimed they are. We've already established that differences in a test involving samples played back to back can often be heard. This thread is an attempt to show that all of the differences that people claim are non-trivial are actually quite trivial when they're forced to identify them in a serious ABX listening test. However, if you don't like the format of this test, feel free to create a thread with your own.

    Quote from ColdFrixion

    This thread is an attempt to show that all of the differences that people claim are non-trivial are actually quite trivial when they're forced to identify them in a serious ABX listening test.

    How can you state such a thing when nobody claimed in the first place the clips being used in this thread were non-trivially different? Besides which the origin of these 3rd party clips is hardly reliable for testing purposes.. Are you starting to appreciate the largely pointless nature of these kinds of tests yet?

    Quote from SonicExporer

    ... nobody is arguing the KPA can sometimes capture the reference amp to a reasonably indistinguishably degree...

    Usually.


    Quote from SonicExporer

    ... almost nobody is arguing the KPA cannot sometimes achieve reasonably indistinguishable results.

    3rd time, including the other thread, that you've said this, mate. I'm only chiming in 'cause I feel this is an unfair-and-inaccurate representation of the hit : miss ratio of the Kemper when Profiling.


    Quote from SonicExporer

    And almost nobody is arguing that most of the time the KPA still creates a real amp type of sound.

    Actually, not "almost"; I'd say nobody.


    Perhaps I should've pointed this sort of thing out before, Sonic (there have been many examples, and no, I'm not going to trawl through both threads to quote them to prove anything), but I, uugghh... don't want to be involved in this; I just enjoy reading it.


    I'm only pointing out the appearance of bias, whether legitimate, unintended or unknowing, on your part, mate. ;)

    Quote from SonicExporer

    How can you state such a thing when nobody claimed in the first place the clips being used in this thread were non-trivially different?


    I answered this in an earlier post. I could hear the same characteristics in the original clips I used for this test that I was able to hear in other samples that people consider non-trivial. When the two samples were level matched and spliced together, the differences became fairly transparent.


    Quote from SonicExporer

    Besides which the origin of these 3rd party clips is hardly reliable for testing purposes.


    They're no less reliable than any of the other third party samples that were posted and evaluated by members in the other thread and lamented as evidence of the profilers inaccuracy.


    Quote from SonicExporer

    Are you starting to appreciate the largely pointless nature of these kinds of tests yet?


    No more than you apparently appreciate the pointless nature of such inane questions.

    Quote from Monkey_Man

    Usually.

    Well, that depends. It's very possible to make a profile that sounds like shit, but that has more to do with the user and their tools. Point being, there are plenty more distinguishable profiles than there are uncanny ones. Hence, commercial profilers exist with access to greater production and equipment than I would have in my living room. This also doesn't take into account how some people EQ their profiles to fit in their specific setup with their specific speakers, etc. I recently imported a ton of profiles into my Kemper and maxed out at 999. I'm currently going through and deleting massive amounts of them, most of which don't sound all that good to me compared to the handful that I really prefer.


    Sonic is right in a way that the test isn't helpful to a person who is finding it near impossible to profile his or her amp. However, he's arguing a point that wasn't presented in this thread, and everyone admits that this test is based on 1 specific result, profile, and performance. But likewise, any other test, including ones that the KPA struggles with would be met with the same complaint and comparison.

    Quote from Monkey_Man

    Usually.

    3rd time, including the other thread, that you've said this, mate. I'm only chiming in 'cause I feel this is an unfair-and-inaccurate representation of the hit : miss ratio of the Kemper when Profiling.

    Actually, not "almost"; I'd say nobody.
    Perhaps I should've pointed this sort of thing out before, Sonic (there have been many examples, and no, I'm not going to trawl through both threads to quote them to prove anything), but I, uugghh... don't want to be involved in this; I just enjoy reading it.


    I'm only pointing out the appearance of bias, whether legitimate, unintended or unknowing, on your part, mate. ;)

    You are simply highlighting one of my very points. That said, I think you are getting the language and intent twisted up, the phrases from the two quotes above are intended to mean essentially the same thing, as I think most readers clearly understood in context. And as for the last quote I used "almost" because there is at least one person on this forum who IIRC mentioned they can almost always tell the difference in a mix. If I hadn't used the terms "sometimes" and "almost" then I'd have someone else chiming in claiming things were stated too absolutely. :rolleyes:


    Don't kid yourself, an objective blind and deaf person would see FAR more bias coming from the Kemper disciples than anything perceived from me. The KPA has shortcomings matching a reference amp, they are not trivial at times, and the disciples simply refuse to accept this fact.