Hm, don't like the sound from FW 1.6 (Edit Fixed now)

  • Here is my test. Used the amp factory JCM 800 drive rig disabling noise gate and effects.


    1.5.4 -> https://dl.dropbox.com/u/74444515/test%20kpa%201.5.4.wav
    1.6.0 -> https://dl.dropbox.com/u/74444515/test%20kpa%201.6.0.wav


    I recorded the guitar dry and then reamped using spdif.


    Just to bring these up again MadH. I've brought these into Audition to look at a little more closely, the prominent feature to me is that the 1.60 version is more compressed, the overall volume is the same at peak and the same frequencies are there but you can see substantial difference in volume when the input is quieter and during transients.


    This is born out by the graphs, now that's not to say that 1.60 itself is any different, as we all know compression can result in subtle frequency shifts which would account for the ozone graphs being off. So is at all possible that your input sens settings were different? Assuming that you reamped without adjusting any parameters anywhere on your DAW between updates, i.e. you didn't raise the volume for the 1.60 re-amp then the only thing I can think of is those input sense settings changing somehow, unless the amp curves were adjusted (which wouldn't make sense).


    More compression would of course the very slight changes in frequency and of course to feel especially around transients. The most likely fcause that I can think of is increased volume of the sound going in to the unit. But I'm not sure how to test that as I believe sense is only used by the amp block right? So you need to have a clean amp in there perhaps and re-amp that instead of a distorted one. Sorry I can't do any more thorough tests right now cos I'm in the middle of work and away from the KPA, but if anyone has the time to try that out we could see if maybe just adjusting input sense settings fixes the problem (in which case maybe the 1.6 update just needs to mess with that slightly).


    Anyhow here's some images. First an animated gif showing the waveform + frequency spectrograph flipping between 1.54 and 1.60, the fatter waveform is 1.60 (edit: oops the animation didnt' export from fireworks.. brb with a fixed one).


    [Blocked Image: http://www.per-anders.net/general/kpa_madh_reampfw160.gif]


    And then a difference image so you can see where changes have occurred, you can see that the 1.60 re-amp is not uniformly louder than 1.54, but it is louder.


    [Blocked Image: http://www.per-anders.net/general/kpa_madh_reampfw160dif.jpg]


    The section is just the beginning of the reamp tracks.

  • Per, could you try reamping my guitar dry track with one of your own rigs and compare results? If so please use a rig with a considerable amount of gain (similar to mine).


    https://dl.dropbox.com/u/74444515/test%20kpa%20dry.wav


    MadH, why don't you reamp 2 tracks in 1.5x and compare the differences? If your position is correct, they should be the same. This would be called the "control" sample. If there are differences between 2 samples from the same firmware (as i predict there will be due to on the fly latency adjustments of 1.5x) you will see that you are chasing your tail...

  • Okay CK - ONE LAST SHOT at this after extensive loading and playing today.


    I used an original 1968 Strat direct into the KPA then from the KPA direct into a Mackie SRM450 V2 at all times.
    The Mackie was positioned as a foldback monitor and approx 2 metres from my head/ears


    Let me start off by saying yes, 1.6 "Beta" is a cleaner playing experience than 1.54. It's certainly more refined, polite & polished.
    I can fully accept that for many this is a welcomed & pleasant thing that no doubt makes their
    playing/recording experience "nicer".


    Yes, you've added some amazing and much appreciated additions and I have no doubt bettered its overall stability so once
    again a big thank you is in order. (I sincerely mean that)


    Down to business.


    Without ANY doubt OS 1.54 is more raw, immediate, percussive and more like a real amp would react when playing with high gain and output volume.


    You asked someone earlier were they a professional musician, may I ask do you play guitar (with many years of high end guitar amp experience) and have you personally played a guitar into both OS versions extensively? (code writing & testing is very different to the end user's experience as I have no doubt you're aware)


    Where to from here?


    I now accept that 1.6 is here to stay and that many people are again very happy, so challenging that is a waste of everyone's time


    Can you tell me what I and others need to change in the menu to bring back 1.54 re the attack, percussiveness and dirtiness? Clean sense etc .....?


    Can we discuss offline providing an OS (1.54) with the 1.6 updates such as midi, delay ducking etc... without any of the 1.6 audio / latency /aliasing /frequency /Db changes?


    I look forward to hearing from you.

  • MadH, why don't you reamp 2 tracks in 1.5x and compare the differences? If your position is correct, they should be the same. This would be called the "control" sample. If there are differences between 2 samples from the same firmware (as i predict there will be due to on the fly latency adjustments of 1.5x) you will see that you are chasing your tail...


    I will do that depending on Per results but I don't think the latency has anything to do with it because the way I did the comparison in Ozone is not affected by that. Anyway what I see on what Per posted does not seem to be caused by possible latency differences either.

  • I know some of the users in here own two KPAs, has anyone done an A/B comparison between two units, one running 1.54 and the other 1.60? I mean switching the input while playing with an A/B box and recording both Kempers, you know? So we can hear the switching between 1.54 and 1.60, back and forth, in the middle of a phrase. Preferably IMHO this should be done with a few profiles at different gain settings, without stompboxes (since there were some changes in them).


    Just an idea.


  • This profile is very dark - nothing above 7KHz.

    (All trademarks are trademarks of their respective owners, which are in no way associated or affiliated with soundside.de)


    Great Profiles --> soundside.de


  • Really, what are you hoping to accomplish at this point? Do you really think questioning CK's abilities are going to further your cause?


    I will do that depending on Per results but I don't think the latency has anything to do with it because the way I did the comparison in Ozone is not affected by that. Anyway what I see on what Per posted does not seem to be caused by possible latency differences either.


    Any "scientific" sample comparing 2 subjects requires a 3rd control subject to show the basis of the comparison. If you can not prove that 2 clips from 1.5X are exactly alike, then a variance with 1.6 means nothing.

  • Wow what a thread!


    Sounds the same to me I must admit but I am a bit deaf in one ear.


    We did some blind tests (compression algorithms for video) at work a while ago and used exactly the same source material in one of the tests, it was great listening to the users arguing about which was better and the different artifacts they believed they could see :)


  • Any "scientific" sample comparing 2 subjects requires a 3rd control subject to show the basis of the comparison. If you can not prove that 2 clips from 1.5X are exactly alike, then a variance with 1.6 means nothing.


    In the digital domain a digital source uses to have the same digital result after being processed unless there is some sort of analog/digital conversion in between or randomness coded into the algorithms. I guess CK would have mentioned that already if that was the case but maybe you are right.

  • My Kemper sounds different since I upgraded to 1.60.


    If the reqson is not the firmware, I'll have to find out what it is, maybe levels, maybe effects, maybe my poweramp.


    I'll downgrade to 1.5.4 again and see what is really happening, because the difference is big, not a subtle change....

    Proud Kemper+Axe FX II user.....yes, you can hook'em together, they WON'T explode.

  • In the digital domain a digital source uses to have the same digital result after being processed unless there is some sort of analog/digital conversion in between or randomness coded into the algorithms. I guess CK would have mentioned that already if that was the case but maybe you are right.


    I bet Will is correct.

  • Really, what are you hoping to accomplish at this point? Do you really think questioning CK's abilities are going to further your cause?

    You may want to re-read my post, particularly "Where To From Here?" you may comprehend the intention this time around so you know what I'm trying to accomplish.


    Questioning CK capabilities? I have no idea what you're talking about. I asked if he played guitar and had much experience with it, the question is still in play.


    Regarding my playing experience findings, we can't all be wrong who feel as I do.


    With your response of (usually) no value what are you trying to accomplish? If it's to get me offside you've achieved it.


    Let me close by saying as many others do, don't take the above personally. :D

  • My Kemper sounds different since I upgraded to 1.60.


    If the reqson is not the firmware, I'll have to find out what it is, maybe levels, maybe effects, maybe my poweramp.


    I'll downgrade to 1.5.4 again and see what is really happening, because the difference is big, not a subtle change....


    check your input sense, and also try the latency switch at the end of the master menu
    jumping around firmwares for some reason changes my clean sense, i rarely touch it.also twice going to 1.6 my volume into the DAW would be skyhigh without touching anything