Posts by Mark1964

I just spent the whole weekend (my wife was away) trying to get Kemper to work for me. But I just can't.

The real issue is still latency: it just DRAGS.

In contrast, my "analogue" silent recording chain comprises a Marshall 2203 into a THD Hotplate, and record the DI out of the Hotplate. Put up with the fizzy sound whilst tracking and then insert a 'speaker emulation on the recorded track.

Whilst tracking, the above has an immediacy and a precise response that the Kemper seems to lack. And the sound is, well, I think it's much better in every respect.

So at this point I give up, and my Kemper is going on eBay.

So, in the interests of being helpful to other users, here's what I've come up with. The aim was to USE the Kemper. I spent a lot of money on this, and I really want it to work... ...and loads of other people say it's really good! So the problem must be me.

On the issue of a rather bass-y, flat sound: a big improvement seems to be to use a lighter plectrum. I use a Jim Dunlop 0.60mm nylon. I know it's unfashionable to use such a whimpy pick with quite overdriven sounds (which is what I'm doing) but it works for me. In fact, this is what I always used to do, and then for some reason I changed to much heavier picks. I guess the lesson (re-)learnt is: don't under-estimate the effect of the pick on your sound.

Second, on latency. What seems to work - for solos - is: plug the cans direct into Kemper and listen to your guitar with one can on one ear, but the other ear "open". The "open" ear listens to the backing track only on the studio monitors. And then do the take. After a dozen failed attempts listening to everything through headphones via my audio interface, this particular solo got done 1st take with the new approach. Hmmm, maybe this is psychological, but maybe not. But it works for me.

So, the good news is - a success with Kemper. An entire track finished. And btw the profile I ended up using is the GT75 Scream stock one. This seems to work just fine with my Strat and Les Paul and sounds similar to my Marshall 2203, except - I have to say in a home recording environment - it sounds better. The distortion rolls off with the guitar volume similar to the real amp. Maybe this is because it's a merged profile. I seem to find myself general liking these more than studio profiles.

I know I said a few days ago that I'd posted my last post...

But here's an update after a few more days experimenting. I bought my Kemper just before Christmas btw.

The music I'm doing at the moment is various classic rock covers with band line-up of vocals / bass / drums / guitar / keyboards. All predictable stuff, and the guitar is pretty varied depending on the song. In most cases there's a lot of organ (Hammond XK1) to blend with. Also, it's a "patch-up" of a live gig recorded as a multi-track a few years ago. We were fortunate in having a great singer guesting for us (this guy's band had a 5 album deal in the 80s, and he is still going strong), so the mission is basically bringing everything else up to the standard of the lead vocals. This wouldn't have been even remotely possible a few years ago, but with today's technology it really is - sample-replacing the drums, re-recording pretty much everything else and "fitting" it to the vocals - a bit like some songs on Queen / Made In Heaven. Then add lots of applause... A fun and interesting project from a playing and technology perspective, and the results are fooling even the band members!

The constraint though is - because it's "live" there's no double (or triple or quadruple) tracking of guitar allowed: the guitar sound needs to stand up on its own as single performance, and "sit" in the mix with everything else.

So, I've tried:

1) real guitar / amp mic'd up in another room (and variation thereof: amp sound recorded via THD Hotplate, and then put through a cab emulation or IR in the DAW)

2) POD (1998 original)

3) Kemper

4) Using various guitar amp sims. (as stand-alone apps., not via Logic) on my laptop: Waves GTR3, AmpliTube4, Guitar Rig 5.

All of this btw is being recorded to a completely separate outboard mixer / recorder (not my laptop and not Logic).

Each of these has pros and cons.

1) is good, negligible latency, but there's a lot of wires, hassle, reproducibility is a problem... ...and it usually ends up sounding a bit like guitar recorded in a bedroom. (And the variation doesn't seem to work very well, possibly due to the Hotplate attenuation mangling the sound).

2) also has pretty low latency, and it's very quick and simple to set up + certain patches still sound OK even 20 years on.

3) is, like POD, great from a simplicity standpoint. It has IMHO a latency problem but also, of all the above, it doesn't seem to sit in the mix: no attack, too bass-y, lack of sustaining harmonics relative to the real thing (and the other sims.) - so it's not just latency (and I know most of the posters on this forum aren't cursed like I am, and thus can't detect "dragging" latency), there are some other issues as mentioned right at the start in my OP.

4) was a bit of a revelation. The overall latency is about the same as Kemper, and the sounds go from being dreadful to amazingly good, depending on the patch. But the real difference is that you can hear these sounds in the mix straight away - if you find a decent patch - without any real fiddling about.

I think 4) is the way to go.

Larry Mullen of U2 quotes a figure of 6mS for detectable latency leading to playing problems (see his wikipedia entry). He seems like a 1/2 decent musician...

The start of this thread was a cry for help, not an invitation for criticism. And the OP was not only on latency.

But on latency specifically: it is a problem if you're trying to play - and record - accurately, and what you're hearing in the cans is 5mS late vs. what you're actually playing and feeling. It just is. Kemper feels like it's dragging. It just does.

I only quote the Yamaha article as an independent source which disagrees with "the crowd" - " the crowd" as exampled on this forum.

This "crowd" says things like I play "standing 10 feet away from my cab..." and "1 foot=1 mS..." and uses such arguments to make the case that latency doesn't matter and you can't hear it. Well, I - and Yamaha - say it does matter. And I can hear / feel it - and I think any half decent musician can too.

Trying to track accurately in a recording context with Kemper and its inherent latency is not easy. For me, it's too much trouble - but I grew up in an analogue world where nobody had even heard of latency.

These latency problems don't occur DI-ing a clean guitar, or indeed using my 1998 POD, or a real amp for that matter (which mic'd up and monitored through headphones doesn't have this sort of latency).

Also, I don't think it's just latency. The attack of the Kemper isn't like a real amp and the sound lacks sustaining upper harmonics - please see the start of this thread.

Anyway, I'm a bit disappointed that there doesn't seem to be a solution from Kemper to the latency issue. It feels like I've inflamed a sensitive spot (that spot being Kemper latency), because nobody has made any suggestions on the other issues I have with Kemper in my OP. I thought forums such as these were supposed to be helpful - rather than scoffing.

Never mind.

This will be my last post.

I thank the minority of forum relpy posts that were helpful / understanding on these important topics.

And I wish everybody the very best of luck and happiness with their musical endeavours.

Best regards,

Mark 1964

And just to add a few things:

1. I think Kemper is really great idea i.e. modelling (sorry, "profiling") the entire signal chain, rather than components (pre-amp, power amp, cab, speakers, mic etc.) like other software guitar rigs do. And the convenience and simplicity vs. a real rig are great too. Especially for home recording. In principle at least...

2. The relative lack of tweakability of Kemper vs. others is, IMO, a good thing because it's possible to waste a lot of one's life tweaking. With Kemper, if you like it great, if not try another profile. And it's not like there's a shortage of profiles to choose from.

3. But WHY is the latency so high? Even with EVERYTHING (stomps, stack, FX) DISABLED? My experience with Kemper - and I've had mine for a couple of months now - is it's unusable as a direct recording tool for anything remotely precise. This is due to the cumulative latency of Kemper + "direct monitoring" with a relatively high-end interface. 5mS of latency (of which 3.5mS from Kemper) is an issue. At least it is to me. Everyone will tell me "What's wrong with you? 5mS latency?! You've gotta be kidding!" Everyone that is, except my ears and the "feel" when I play the guitar. Oh and Yamaha as well...

Thx christianbad for your very helpful response.

I'm not using Logic "software monitoring". In this test, the CC went directly from the VS2480 into an Apogee input and I recorded what came out of the Apogee's headphone output back onto the VS2480. This produced a latency for the Apogee (adjusting for the VS2480's inherent latency) of ~1.3mS. By comparison, at a 44.1kHz sample rate and 64 samples I/O buffer, Logic tells me its round trip and output latency are 5.8mS and 2.5mS respectively.

If I go down to 32 samples I/O buffer Logic RTL becomes 4.3mS (output latency=1.8ms). And if I go up to a 192kHz sample rate AND 32 samples buffer, Logic says RTL is still 4.3mS but output latency decreases a tiny bit to 1.5mS.

And in the real world, tracking at 32 samples I/O buffer and 192kHZ doesn't work due to glitching, even if - as I do - I record into a very simple tracking project with just a handful of basic audio tracks and no plug-ins whatsoever.

I think Logic latency is therefore not relevant here, and all that matters is the combination of a) Kemper's latency and b) the latency of "direct monitoring" using my Apogee interface. On b), the issue is - I suspect - that "interface latency" in this scenario is almost (?all?) due to the Apogee's ADC / DAC convertors. Therefore, increasing the sample rate and reducing the buffer size in Logic has a no impact. And perhaps to prove this the Apogee "direct monitoring" latency I'm measuring of ~1.3mS is less than any of the numbers quoted by Logic (even at 32 samples / 192kHz).

The problem with latency seems to be that cumulative effect is what matters. In total (Kemper + direct monitoring through Apogee) monitoring latency is about 5mS. According to Yamaha, 5mS is on the cusp of where "playing becomes difficult" and this is my experience. I'm not a virtuoso musician by any means, but I did all the piano grades as a kid and I've been playing the guitar for a long time so I'm not completely unmusical and - to me - Kemper drags: it just does.

The real issue is that Kemper latency is most of the problem: 3.5mS out of 5ms = 70%.

And if there's no solution for this, well... ...oh dear...

Anyway, thx again, christianbad, for your very helpful response.

Question: is it possible to improve Kemper's latency performance?

And here's some data as background.

I did some tests, using a Logic Klopfgeist click recorded onto my old Roland VS2480. This, if you like, was the "Control Click". I then bounced the CC using various routings to test the latency thereof. Results obtained by really zooming in on the waveform and measuring are as follows:

1. Out of VS2480 headphone output, back into VS2480 input. Result: 1.67mS delay (this is the "inherent" delay of one loop through the VS2480, obviously including it's DAC /ADC converter cycle, which is what you get when you record and monitor with this device).

2. Out of VS2480 master output, back into VS2480 input. Result: 1.67mS delay i.e. the same as 1, above (this was just to test that there's no difference between using the headphones out and the master out of the VS2480).

The following results are all stated both "gross" i.e. the latency of the whole chain, and "adjusted" i.e. the latency of just the component under test, after deducting the VS2480's "inherent" latency.

3. Out of VS2480 master output, into Apogee Duet connected to MacBook pro and with Logic booted up @44.1kHz sample rate and 64 samples I/O buffer. Then straight back out of the Apogee's headphone out i.e. not using software monitoring. The idea here is to test how good Apogee's "direct monitoring" is. Result: 3mS delay. Adjusting for the VS2480's 1.67mS inherent delay, this means Apogee - even in direct monitoring mode - is adding another ~1.3mS of latency.

4. Out of VS2480 master output, into 1998 POD on "POD Clean" patch. Result: 3.3mS delay (~1.6mS adjusted).

5. As for 4. but using a different POD patch: this time the "Brit Hi Gain" patch. Result: identical to 5. above, so it seems POD patch has little effect on latency.

6. Out of VS2480 master output, into Kemper with Stomps, Stack and FX all disabled. Result: 4.3mS latency (~2.6mS adjusted).

7. Out of VS2480 master output, into Kemper on "AC30 Clean SM57" patch (I think the 3rd patch that comes up on the factory default list). Result: 5mS latency (~3.3mS adjusted).

8. As for 7. above, but using AS Mars MP Gain 8 patch. Results: identical to 7. above.

So what?

Well, first, this means that in a "real world" recording scanario (recording using Kemper into an Apogee Duet using Apogee's "direct monitoring") total latency is ~3.3mS for Kemper (the manual I think says 3.5mS) plus a further ~1.3mS of latency for the Apogee Duet's "direct monitoring" i.e. a total of 4.9mS.

Latency seems to be a bit like having a drink. You don't notice the first one, or even the second. But after a few drinks you definitely notice the cumulative effect.

4.9mS is right on the cusp of when no less an authority than Yamaha say "playing starts to become difficult" due to latency (i.e. hearing what you're playing fractionally late verses your actual playing). Here's a link to that Yamaha article:

http://www.yamahaproaudio.com/…ter5/05_absolute_latency/

and, in summary, this is what it says.

signal path latency for in-ear monitor systems

1.15 - 2 ms Playable without any big problem.

2 - 5 ms Playable, however tone colour is changed.

5 - 10 ms Playing starts to become difficult. Latency is noticeable.

>10 ms Impossible to play, the delay is too obvious.

So, in summary, with a real world recording chain, the latency of playing using Kemper is such that "playing becomes difficult". This is my experience.

Back to my original question: is there any way of improving Kemper latency? I ask because my 20 year old POD has about HALF the latency of Kemper (~1.6mS vs ~3.3mS). And POD plus Apogee Duet comes in at ~2.9mS which is less than listening to Kemper with headphones plugged directly into Kemper.

To those that say "but 5mS of latency is like standing 5 feet away from a real amp / cab" I say: take a simple midi drum loop and shift e.g. the Kick backwards (or forwards) 5mS and then then try playing along to it. Or leave the Kick and instead move the Snare by 5mS. Or leave the entire loop as is, but add a click track mis-aligned by 5mS and try playing along listening to both. Then repeat the whole thing using 10mS instead of 5mS to see what Yamaha mean.

To me, Kemper drags. It just does. And I think latency is part of, but not all of, the problem. But it would be nice at least to try to improve the latency bit.

Thx!

Apologies. I meant to write "And I think latency is part of, BUT NOT ALL OF, the problem. But it would be nice at least to try to improve the latency bit."

Thx again!

Hello fellow Kemperers... been away for a few days helping my mum with my Dad, the latter having just had a knee replacement. Both in good shape, which is good! Had a bit of time to do some thinking and analysis on the above thread.

Here's what will probably win the award for the longest, most boring post by some random middle-aged bloke (if such an award exists...).

First, my current project is a "live" version of ZZTop's Gimme All Your Lovin' - recorded very badly at a Christmas gig on a multitrack back in 2010. Among other things, I'm redoing the guitar part - so the "guitar" context is a part that's mainly (but not always) 2 strings, 4ths and 5ths and sometimes 3rds (always a good test of whether on overdrive amp sounds "good"), quite rhythmic etc. with an overdriven sound.

When I play little chords like this on the Kemper and look at what a frequency analyser shows, typically it's a sort of skewed hump (skewed to the right i.e. the higher frequencies) BUT as the chord decays, you're left with about 3 quite sharp peaks which are the lower harmonics of the chord i.e. there's a burst of high frequency distortion at the start of the sound, but this quickly abates leaving, well, essentially a "clean" sound (and a fairly dull one at that). I know, as someone once said, you "engin-ear" not "engin-eye" but this is interesting, because a real overdriven valve amp seems not to do this. It somehow "smudges" those first 3 or so lower harmonics so they are much less prominent as the note decays, and the upper harmonics persist for much longer.

This, I think, is what's at the root of the "flat" Kemper sound. To my ears at least.

So, I get back home and decide to have a go at re-creating Billy Gibbons' finest. Thus, Gibson Explorer into my old Marshall 4210 combo used as head, into my Marshall 1936 2x12 mic'd with an SM57. And on the mic-ing I did something slightly different to normal i.e. mic more towards the edge of the 'speaker and mic angled so it's at about 90 degrees to the plane of the 'speaker cone i.e. off axis. This, predictably, sounds nothing like Billy Gibbons, and neither does my playing! But it is a nice warm sound, with enjoyable sustaining harmonics in the tone. And btw I did this two ways: one using the boost channel on the Marshall, and the other using the clean channel on a, well, clean-ish setting (everything on about 1/2) but boosted with a Maxon TS808 (drive=off, balance=max). Oh, and all this went into my old Roland VS2480 to make sure there were no latency issues.

And to me the above sounds great!

Given that I bought the Marshall combo in 1987, a fair criticism would be why I didn't figure this procedure out many years ago! The answer is: I did, but it's a bit of a hassle (see below). Plus - all the recent technology - including Kemper - maybe makes one not just more discerning about guitar sound, but actually more motivated (and able) to do something about it.

And on latency. This set up - recording onto the VS2480 - is almost latency free. And it feels it. The VS2480 has been criticised for it's pre-amps and converters - and this may or may not be an issue - but it is pretty darn good when it comes to latency: no adjusting the I/O buffer or setting plug-in latency compensation and so on. Kemper in contrast, to me, has latency: when I hit a string the "thump" of that string hit from the body of the guitar in my chest is in time with my playing: the thump from the Kemper profile is late. I know I'm sensitive to this, and it's worse if I record through Apogee into Logic on my Mac. Most people don't notice latency, but I do.

Now, here's the thing.

Hassle.

Doing all this reminded me of what a hassle setting up a real recording rig is. Some really enjoy setting gear up: in fact, so do I. But even so, all those wires, messing around with the amp controls, mic postioning, making sure the wife is out etc. And usually, by the time it's sounding "good" it's time to put all the toys away and go and feed the dogs. This to me, is one area where Kemper absolutely excels. It's just so easy: plug-in and play, and if you want a clean AC30 rather than an ODd Marshall it's there at the press of a few buttons. No wires, no noise, no hassle. It's a really great tool therefore for figuring out what guitar sounds / parts you want on a project, and "roughing them out" as it were. However, I haven't yet been able to make the jump to using Kemper as a substitute for real amps. Maybe I will get there in the end, and maybe I won't.

Wheresthedug, if you can isolate the gene that lets you cope with this amount of latency, and develop a treatment for me - I am a buyer!

And on the actual sounds, if by placebo effect you mean the "guitarist buys new gear, worries why it doesn't sound like their old gear, then tries to get it to sound like their old gear" effect, you may well be right. And it may well just be that POD v2.0 is familiar.

What I'd like - I think - is just some simple, basic profiles of classic amps with an SM57 and / or MD421 a few inches back, gain settings that suit me and no room mics or ribbons. These seem to be hard to find. Perhaps I'm not looking in the right places. In fact, I may try and do the profiling myself with my remaining collection of amps - go to a local studio and spend a few hours doing this systematically. I have a feeling my wife would appreciate a profiled version of my Brian May AC30 rather than the real thing!!

Thx for all the input.

Michael_dk Thx for this.

In my experience the opposite is true. It's often better NOT to hear the electronic version of the sound through headphone but instead ONLY hear the acoustic strum of the guitar strings. That way, what is - to me at least - what you're playing (and I guess after 40 years the sound of an acoustically strummed electric guitar is something I'm used to. Hence I often find myself taking one can off one ear and using this ear to listen to my acoustic strumming, with the other can listening to the backing track (with no electric version of what I'm playing).

If I do listen to Kemper, the combined latency of Kemper + the ADC / DAC route of my audio interface (in direct monitoring) mode is about 5mS of latency - in other words what I hear is 5mS after I've played it. This - to me - drags and makes playing hard.

This is just me - and it works for me.

The strange thing about latency is - I am genuinely surprised that more musicians aren't worried by this sort of thing. For example, take a quantised midi drum track and randomly re-position the Kicks / Snare by +/- 5 ticks (=+/- 2.5mS @125bpm). Then try playing along accurately to it. To me, it's impossible.

Thanks.

It's more like 5mS latency when you take account of using an audio interface, even in direct monitoring mode.

And it's not just me. This is what Yamaha say:

http://www.yamahaproaudio.com/…ter5/05_absolute_latency/

In summary:

signal path latency for in-ear monitor systems

1.15 - 2 ms Playable without any big problem.

2 - 5 ms Playable, however tone colour is changed.

5 - 10 ms Playing starts to become difficult. Latency is noticeable.

>10 ms Impossible to play, the delay is too obvious.

I'm surprised more musicians don't notice this.

Finally - and this is the real killer - the latency performance of Kemper is materially worse than a 20 year old POD 2.0.

My bad though - all of this is in the public domain, caveat emptor and all that!

Atlantic. Even if I plug the guitar and headphones both directly into Kemper, that 3.5mS latency - to me - drags. The sound is behind my playing and I notice it, and everything deteriorates. Let's call this scenario 1.

If (and let's call this scenario 2) I plug Kemper and headphones into my Apogee interface and listen through the Apogee in direct monitoring mode, it's a bit worse due to an extra round of ADC / DAC conversion in the Apogee (direct monitoring is not zero latency because of the ADC / DAC). I reckon the cumulative latency in scenario 2 is about 5mS which, according to sources like Yamaha (please see earlier in this thread), is where latency starts to be a problem for musicians having to deal with it. This is my experience (it's a problem...).

A test is to do all this with all amp / cab / FX sections in the top 1/2 of Kemper disabled, thus in theory you're just getting a clean DI sound. Kemper still - to me - drags in both scenarios (and obviously scenario 2 is a little bit worse).

When amp / cab etc. get switched on, to me - and this may well be subjective - it's worse again because you can't hear the attack of the note properly. A real amp and cab - even on a high gain setting - has some "thwack" when you hit the strings. This, to me, is missing in Kemper and - again for me - it makes it hard to play properly in time.

Thanks for the suggestions. Yes - I did experiment with all these parameters. Noe of them of course solve the Kemper latency problems (3.5mS or I think it's 4.5mS using "constant latency") - which is materially worse than my 1998 vintage POD 2.0.

And all these tweaks (and Tube Shape, Pick, Clarity and so on to my ears just "digi-fy" the sound even more. The quesion I have is - why do so many (all the mdeium / high gain ones I've tried) profiles have this problem in the first place? My POD doesn't - it may have other issues but you can clearly hear the attack of the note. Not so with Kemper.

OP said: "I couldn’t for the world get a true three- dimensional tone, with depth, balls and precision".

The "and precision" is exactly what I'm finding in a recording context. Kemper seems to drag - partly latency and partly (IMHO) the lack of transient attack on many sounds, particularly high gain ones.

It's not anything to boast about, but I definitely can (notice a few mS of latency). I think it's the combination of the Kemper 3.5mS / 4.5mS and a bit more once you add an audio interface, even using direct monitoring, where the extra ADC / DAC stages push the total over 5mS. We're talking about recording with headphones here, and endeavouring to "get it right" in terms of playing. Not live with back line, drummers etc. (that's different, obviously). FWIW, I don't think it's just the Kemper latency, it's the issues with the Kemper sound (lack of a proper transient attack on driven sounds etc., all as mentioned above).

Also as mentioned, I'm probably an average musician. I started playing the piano aged 6 and did all the grades, but I realised I was never going to be a concert pianist. I took up the guitar aged 14 and it was a lot more fun! The digital recording innovations of recent years mean I can actually get some of the sounds out of my head and onto disc (which, back in the days of Tascam 144s etc. didn't really happen!).

BUT, average or rubbish musician, I notice latency. And I'm not alone as per that Yamaha latency article earlier in the thread.

Anyway, it is what it is - and Kemper is not as good latency-wise as a 1998 POD. It just isn't.

Btw, one thing I really like about the Kemper is that you can't - at least I can't - really tweak the profiles effectively. If it doesn't sound good at the outset, find another profile and move on. This is not the case with even plug-in amp. emulators, where the temptation and opportunities to tweak are irresistible. For someone like me who should never go to a restaurant with a long menu (otherwise I will never decide what to order...), too much tweakability is a bad thing.

So, to cut to the chase, what I think I need is some SM57 and / or MD421 profiles, no room mics, no ribbons, sensible levels of gain for an old timer like me, a few classic amps, SM57 not too close and bass not on 11. Then I suspect all the re-amping palava will be worthwhile.

Wheresthedug - I am impressed! 20mS of latency is 40 ticks at 125bpm. Or to put this in real world terms, that's a perceptible snare flam or like someone clapping slightly out of time.

Here's another experiment for you to try:

Set up a midi drum loop at 125bpm. Then move ONLY the Snare backwards (or forwards) 40 ticks (=20ms) and try playing along with this. Alternatively, move ONLY the Kick (backwards or forwards 20mS). Or if you prefer leave the Kick and Snare and only move the Hi-hat 20mS. Then try playing along with any of these permutations.

Another variation is to have a click going and have the entire drum midi either +/-20mS and try tracking with both click and offset midi drums.

If you can cope, as I said earlier you're blessed. I can't!

People say to me: "what the heck, 20mS is 1/50th of second". I reply, "yes". But equally, at 125bpm a semiquaver (1/16th note for US friends) lasts 120mS. So 20ms is 1/6th of a semiquaver or, put another way, 20mS of latency is like having semi-quavers played approx. 17% out of time. When I think of latency this way, I feel less bad about my inability to deal with it!

Here's what some other people - Yamaha - say about latency:

http://www.yamahaproaudio.com/…ter5/05_absolute_latency/

In summary, to save the hassle of booting up the link:

Signal path latency for in-ear monitor system:

1.15 - 2 ms Playable without any big problem.

2 - 5 ms Playable, however tone colour is changed.

5 - 10 ms Playing starts to become difficult. Latency is noticeable.

>10 ms Impossible to play, the delay is too obvious.

Kemper's latency, according to the manual, is 3.5mS or (I believe) if you hit the constant latency button 4.5mS. And this is plugging a guitar directly into Kemper and listening on headphones also plugged directly into Kemper. If one then goes via an audio interface, even with direct monitoring another cycle of ADC / DAC conversion is added and you're up to the circa. 5mS where, according to the Yamaha-latency-Beaufort-scale, latency starts to become a problem. This is Kemper.

For comparison, the declared latency of AmpliTube 4 is a mere 0.2mS. However, the problem with AT4 (and its ilk) is that to use it you need to software monitor. And for my set-up (probably not untypical) round-trip latency @64 samples I/O buffer is 5.8mS (even at 32 samples it's 4.3mS). Add the 0.2mS AT4 latency and (at 64 samples) you're up to 6mS cumulative latency and into storm-brewing-Yamaha-Beaufort-scale latency.

The only solution I've found is to split my guitar signal (analogue-ly) and send one split to my 1998 POD and the other to the Kemper. Each split then goes into the Apogee on Direct Monitoring and I monitor only the POD sound and record the Kemper. POD is not latency-free, but it's better than Kemper. And this produces monitoring sound which is (to my ears) in time with my playing. I can then re-position the Kemper track to compensate for latency afterwards. OR I can take it one stage further, and monitor with POD and record a dry DI and re-amp the latter (with AT4, GR5, Waves GTR3, Kemper or real amp). But, at the end of the day this is 1) a hassle, and 2) sobering that latest technology Kemper has an effective cumulative latency performance that is worse than a 20 year old POD.

I know I seem to have a gene in me which objects to latency, and this is a cross I must bear! But, equally, the above monkey maths are the facts.

More subjective is that, in dusting down my old POD for the above operation, I was struck by the other problems mentioned above in relation to the Kemper sound: lack of attack transient, flat sound, too much bass, too little harmonics and so on. POD may be 20 years old, but it stands up remarkably well. It it not perfect by any means - for example, there is a weird notch EQ dip about 1.3kHz on some Marshall presets that you don't get on the real thing or AT4 / GR5 or indeed Kemper Marshalls. This is a bit like that 750Hz notch on a bass SansAmp Bass DI which - to some tastes - mangles the sound of bass (and the new SA Bass DI now has a mid control added). Strange.

Anyway, thanks for all the posts. It seems like a) I am in fact not doing anything fundamentally wrong, b) there is a latency problem with Kemper if, like me and Yamaha, you notice these things, and c) the "sound issues" are subjective.

On c), if anyone knows of profiles that are recorded with just an SM57, or SM57/MD421 pair or similar i.e. the old fashioned way (no ribbon mics, or room condensors etc.) please let me know. Also, it would be nice if the SM57 wasn't completely jammed up against the cab grill (check out the SM57 manual - it recommends 6-12 inches distant for mic-ing up a guitar amp) and without the amp bass control on 11. Just a few decent Marshalls at sensible levels off gain, an AC30 from clean to flat out and maybe some clean Fenders for my occasional forays into funk playing - that's all I really need. Any and all such profile suggestions gratefully received.

And then I can use Kemper as intended when I bought it, which is to record guitar silently with headphones, and without wires, mess, hum and buzz... ...and without annoying my wife or disturbing my kids' studies!

I won't be around for the next few days, as I'm off to visit my Dad who has just had a knee replacement operation. I am going to enquire about getting my latency aversion gene removed, but I'm not sure they do this on the NHS!

Thanks for all the comments. I read somewhere that guitarists often buy new gear, and then worry about why it doesn't sound like their old gear. And then they try and MAKE their new gear sound like their old gear. Maybe there's some of this going on with me... HOWEVER, I find it impossible to play a semi-quaver (or 1/16th note as our transatlantic friends would call it) Nile Rogers-style funky guitar part accurately with the Kemper. I find it similarly impossible to chug in time - properly in time - at 120bpm with a standard rock overdrive sound. The culprit in both cases is Kemper latency and, IMHO, that problem with the lack of attack on the note. Either of the above musical endeavours is totally doable with my old 1998 POD. The Kemper to me, doesn't sound anything like my Brian May AC30, My Marshall TSL, my Cornford Harlequin, my Marshall 4210 combo and so on and so on. One difference though is, it doesn't hum, it doesn't have room pings, resonances and all the, errrrm, quirks of recording a real amp. Neither does it have the wires, mess and noise of a real amp. In the past, sometimes I've got a great recorded guitar sound with a real amp, but often I don't. And even if I do, reproducibility is a problem. All this is why I bought the Kemper. It's therefore sobering to find out that, after a couple of months of pretty systematic testing of it, that Kemper basically has all the same issues as a 100 quid software plug-in: i.e. latency, lack of defined attack, flat sound, lack of harmonics etc. etc. And - from the comments so far - it sounds like I'm actually not really doing anything "wrong", it's just that "digital isn't for me". So, no practical tips on switching off Pure Cabinet, or setting "Space" to Headphones only, or messing around with Tube Shape, switching cabinets assuming I'm using a merged profile etc. etc. etc. Oh well, looks like I wasted my money...!

And separately on your other points - it may well be the case that many recorded guitar sounds these days are so processed by the time they're on a record that they've stopped sounding like real amps. This raises another issue which is: I wonder whether the reason some of the sounds on my 1998 vintage POD sound - to me - better than Kemper is because the developers at the time would have probably been about the same age as me at the time i.e. around 30 +/- and therefore grew up listening to the same sort of music I did and tried to emulate those guitar sounds. Whereas now, who cares what a "good guitar sound" was like in the late 70s and 80s? Who even knows how to do it these days? Some of the mics used on the Kemper profiles weren't invented in the 70s/80s. And it's rare to find a profile with e.g. an SM57 and MD421 (a classic combination). And so on.