less latency (i compared the Kemper, the Axe FX and the Line6 Helix)

    today, a friend of mine and myself compared the kemper and the helix and the axe fx...


    ( i use the kemper and real amps, and my friend the kemper and the helix and the Axe Fx belongs to another friend so yeah )


    ...and what i found out was that the helix and the kemper had slightly more latency, it was pretty good hearable ... does anybody know the exact time ? (i guess 6-8ms?, i really don't know)
    and when we switched to the axe fx the latency was way less ... we two both agreed on that but maybe we're just two idiots, who knows :D


    so, long story short:


    so is this a physical limitation of the processing power or is there a way to make things faster ... ?
    if there is a way to make it faster i'd be very happy, maybe by having a turbo mode, that deactivates routing options and the stomps ?



    oh and if anybody is interested it is soooo hard to make out a total winner of these 3 products, they are totally great ... well ... but my buddy profiled the helix so for me the kemper is still number one :D :D
    and you could profile the axe fx too...but the amount of effects and routing options on the axe fx are insane, but it is too easy to get lost in tweaking sounds ... nevermind :D

    Recoreded a guitar with a splitter

    • direct signal (analog)
    • kemper main out (analog)
    • spdif out (digital)

    Signal two comes 3,5 ms later, spdif out 3,0 ms later than the direct signal.
    Running here with Firmware 4.0.2 and constant latency off and using a studio profile with reverb.
    Wonder what's your signal chain and latency?

    A latency of 7 or 8 milliseconds isn't hard to notice if you're playing anything fast. 4 or 5 is right around where most people can feel that something isn't right, but it's not until you get past 10 that's easy to hear.


    Personally, half the time I forget that I've left Constant Latency on, and I'm an all-metal-all-the-time guy. Of course, I also grew up having to compensate the aim on my sniper rifle in Team Fortress and Counterstrike by about 350ms because my parents had dialup, so to me 5ms is nothing.

    Quote from Lokasenna

    A latency of 7 or 8 milliseconds isn't hard to notice if you're playing anything fast. 4 or 5 is right around where most people can feel that something isn't right, but it's not until you get past 10 that's easy to hear.


    Well, that would mean that most people would feel that something is wrong if they play eg. a Kemper's analog out (3.5ms from above's measurement?) through another latencywise comparable digital device (that would add another few milliseconds). I must say I doubt that this is true.

    I'm just putting this out there: I am having a very difficult time believing anyone is actually perceiving latency when using the Kemper. I don't care how fast they are playing. So Kemper should focus R&D resources on reducing non perceivable latency by 2 ms?


    Are we really having this discussion?

    Husband, Father, Pajama Enthusiast

    Like 8

    I find it amusing that the whole latency issue wasn't an issue before we all had DAWs and reducing latency became a sport similar to squeezing an extra 1/4HP out of a car engine. You never heard guys complaining that they couldn't stand more than 7 feet away from their amp on stage (7ms latency) because the lag became noticeable.

    I hate emojis, but I hate being misunderstood more. :)

    Like 5

    Of all the processors I've owned in the past 25 years (since POD 1.0) - L6, Digitech, Ensoniq etc., the Kemper has the least-percetible latency. That is to say, I can't perceive it, and cannot imagine how anyone could whilst playing. From what I've heard the Axe is in this league too, to be fair, although I've never owned one... and likely never will.


    Caveat: I've never turned Constant Latency™ on, and likely never will unless I need it during reamping, in which case it would only be employed temporarily. The 4.9ms it inflicts is still only like standing 5ft from your tube amp, and in the real world shouldn't bother anyone IMHO.


    I'm tired of getting into arguments with metal-rhythm heads who claim it messes up their timing. IMHO, the only solutions for those guys are to wear their cabinets on their heads, use headphones or IEMs exclusively or to take Chill Pills™... a bunch of 'em.

    Quote from mrs-zambesi


    Well, that would mean that most people would feel that something is wrong if they play eg. a Kemper's analog out (3.5ms from above's measurement?) through another latencywise comparable digital device (that would add another few milliseconds). I must say I doubt that this is true.


    I personally witnessed this. I was showing my Profiler to Umberto Fiorentino, a pro guitar player (member of the fusion band Lingomania among other things). We first tried it through the return of his Femder amp. Then we used a CLR, which is known to have a DSP. After the first three notes Umberto said he felt some latency.

    I never notice the latency myself (I usually play with headphones, through spdif to my soundcard). But I'm not very sensitive to latency either - some people are, though. Hopefully, it isn't a problem for them.


    It's true that there is latency when playing a real amp as well due to speed of sound. That also means that when you're playing with a cab/monitor connected to the kemper, latency will be longer still. So it's absolutely possible that while the kemper has very low latency, some people will have problems.

    Quote from Monkey_Man

    Of all the processors I've owned in the past 25 years (since POD 1.0) - L6, Digitech, Ensoniq etc., the Kemper has the least-percetible latency. That is to say, I can't perceive it, and cannot imagine how anyone could whilst playing. From what I've heard the Axe is in this league too, to be fair, although I've never owned one... and likely never will.


    Caveat: I've never turned Constant Latency™ on, and likely never will unless I need it during reamping, in which case it would only be employed temporarily. The 4.9ms it inflicts is still only like standing 5ft from your tube amp, and in the real world shouldn't bother anyone IMHO.


    I'm tired of getting into arguments with metal-rhythm heads who claim it messes up their timing. IMHO, the only solutions for those guys are to wear their cabinets on their heads, use headphones or IEMs exclusively or to take Chill Pills™... a bunch of 'em.


    Just keep in mind that if you are 5ft away of your cab with your kemper, with 4.9ms of latency' it is like your are 10ft away with a tube amp. The internal latency of the kemper comes in addition of the sound latency that you have with the tube amp.


    But for my side, I didn't notice any latencies

    Thanks guys for the answers, it is nice to find out about the actual latency in ms ...


    Like i said, two guys in the home studio, direct out of the axe fx, helix and the kemper into the same interface with direct monitoring.
    And it wasn't somekind of esotheric guessing where the latency was more, it was easy to hear. Just by listening to the actual acoustic sound of the guitar and the
    the sound out of the speakers ...


    And the thing that makes it quite obvious for me is the the axe fx had no feel- and hearable latency, otherwise i maybe haven't noticed it at all because, like i said teh helix and the kemper were very similar.
    (maybe it comes down to the fact the the axe fx sounds way more static and undynamic, so that the tone stands much faster)


    and if you deactivate the direct monitoring and use the monitoring in cubase, so that the latencies of the interface adds to it, it gets even more obvious.


    Maybe there is a way around that latency, maybe not ... imo the kemper sounded best and the latency is no deal braker, just something i found out...

    Quote from viabcroce


    I personally witnessed this. I was showing my Profiler to Umberto Fiorentino, a pro guitar player (member of the fusion band Lingomania among other things). We first tried it through the return of his Femder amp. Then we used a CLR, which is known to have a DSP. After the first three notes Umberto said he felt some latency.


    Well, my statement was referring to most people. It would be interesting to investigate that "Fiorentino phenomenon" :P further because at this point it is not clear that what he perceived was actually latency and from what I remember (you wrote about that in another thread) it was an in no way scientific test. @Lokasenna however claimed (if I understood him correctly) that most people would notice said latency and that I strongly doubt.

    In regards to early guitar amp modellers, manufacturers used to quite heavily compress sounds to make them feel more immediate and to combat the inevitable latency. Could it be that the patches you played through on the Axe had more compression than the Helix or Kemper? It hints at that, as your description of it sounding less dynamic and more lifeless.

    Perceived fastness at these latency levels is really more about tone and compression. Add a green scream in the signal chain and see how you feel about the latency then. The Axe and Pod both have quite compressed and exciting (I.e.e treble heavy) tones, this makes them feel very fluid and "instant", but also less like a real amp.

    i tried to make the sounds as similar as possible and i have also some profiles of the axe fx and the helix on the kemper ... but the difference is still hearable


    ... the interesting thing is that the profiles of the axe fx sounded more dynamic then when i turned down the volume on the guitar then on the axe fx itself, maybe you're right and the more dynamic and realisitc sounding tones can add to that feeling of having latency ... but i'm pretty sure at the moment that it has to do with latency too because i tried gates and compresson and so on too (i play with a 8 string guitar) to get similar results but yeah

    Latency issue will be discussed as long as there is DSP present. A/D, D/A, DSP operations, all take time = latency. Question is, what is the real value of latency? Comparing latency by ear is IMO not the best option. As it was mentioned before, compression and other factors can shift our perception of latency. @Bommel If you could measure the latency for Axe, Helix and Kemper and post results here, it will shed more objective light on the issue. great tool for measuring latency (if you are on Windows) can be downloaded here: https://centrance.com/downloads/ltu/

    Quote from Bommel

    but i'm pretty sure at the moment that it has to do with latency


    Quote from PETERFR

    Recorded a guitar with a splitterdirect signal (analog)


    kemper main out (analog)


    spdif out (digital)


    the procedure to measure the latency is quite simple and was done repeatedly by users when the Profiler was relatively new, but this hasn't really been reported here in the last 4 years, except for the already mentioned measurements and the constant latency feature for reamping.
    Please make the necessary recordings and compare.