Kemper Profiler MK 2

[DLC86]

The issue with using loudness-adjusted measurements for accuracy testing is that how humans perceive things has little to do with how similar two audio signals are.

It has a lot to do with how similar those audio signals are if the goal is to make them sound similar - or hopefully identical - to our ears.
The alternative is to use raw dBFS peak values if you just want to analyze the differences at the single sample level and with no weigthing or averaging... but that produces results that might seem huge differences for stuff that's actually inaudible (like short transients or 5 Hz content), so I surely prefer LUFS there.

As you mentioned, LUFS highly favors the mid- and high-end, so low-end inaccuracies will have little impact on the resulting measurement, even though those differences will be completely audible.

Again, the K-weigthing low end response closely follows the sensitivity of our ears (at 80 phons iirc), so if something has little impact on the measurement it also has little impact on our perception.
Something that has little impact on the LUFS measurement and makes you go "OMFG THAT'S A HUGE DIFFERENCE" simply can't exist, that's the whole point of LUFS and LKFS: make the measurement adhere to our perception (unlike dBFS peak values).

But wait, we listen to signals with our ears, right? Problem is, the fact that humans perceive bass as less loud than treble at a given level doesn't mean bass goes away entirely.

Never said that.

The bass region for two different profilers could measure the exact same in LUFS, yet sound wildly different.

Sure, and in that case the LUFS measured on the null test between those two profilers will reflect that difference. If it's huge to your ears it will be huge in the measurement as well.

Just do a quick search on his YouTube channel I like how Leo bothers to include these clips in his shootouts.

Will do.

 

[Lysander]

Something that has little impact on the LUFS measurement and makes you go "OMFG THAT'S A HUGE DIFFERENCE" simply can't exist, that's the whole point of LUFS and LKFS: make the measurement adhere to our perception (unlike dBFS peak values).

Sorry, that's just... plain wrong. You seem to confuse applying a loudness EQ curve to a recorded result, with trying to measure frequency responses with those curves baked in. These are very different things - otherwise, two signals measuring the same LUFS should sound the same to our ears.

No point in arguing this over forum posts though. There's plenty of good bibliography on audio weighting curves online.

 

[paisleywookiee]

Because if they can’t win on accuracy or sound quality, maybe ease of use across different applications would work in its favour.

NAM and ToneX were the nail in the coffin for Kemper for me. Not only do they sound much better, but the convenience of being able to run them as a plugin meant that there is no situation where a Kemper made any sense for me.

I can't believe Neural DSP hasn't come out with a $100 "capture player" that's free for QC/NC owners, and $100 for everyone else, that plays their captures and has some effects, as a plugin. That'd be really cool.

 

[laxu]

I can't believe Neural DSP hasn't come out with a $100 "capture player" that's free for QC/NC owners, and $100 for everyone else, that plays their captures and has some effects, as a plugin. That'd be really cool.

They haven't done a lot of things I thought they would, like a marketplace system for captures and presets. Seemed so likely when the QC was released.

Maybe the Nano Cortex is their idea of that sort of product, where they want to keep you buying NDSP hardware, maybe start yearning for a QC after getting a Nano.

 

[OneEng]

First, WTF @Orvillain, click bait for me? Awww. You must have missed me. And I didn't get you anything so I am feeling bad.

while that is true in theory it's rarely true in practice.
E.g. if I take an audio file and rotate the phase by 90° on all frequencies, that will produce a very bad null test result but the difference to our ears will be negligible.

But IME, specific cases like that never happen when comparing amp captures, in every test I've done and seen, the one producing the worst null was also the one sounding the most different compared to the source.

Fair. I think the biggest issue with Kemper's capture is that it requires a lot of manual intervention to get it to sound the same as the source. It's more of an art than a science.... which it shouldn't be. Others do it without all the hand waving and sacrificing a virgin to a pagan God.

That is precisely what it means.

Poor wording. Should have said "Audible". My thanks to the grammar police.

The thing with Kemper’s accuracy is that it doesn’t need null tests or any kind of visual representation to hear how far off it is, because you can hear it when listening without much trouble. Those kinds of things are more helpful when it’s impossible to distinguish and you still need to split hairs. Kemper just tends to have audible giveaways in comparisons, ToneX and NAM also have differences but they can be harder to notice.

Now that is fair. Particularly at the bottom Kemper has capture issues. Typically, I see people eq this out post profile, but it still makes your point.

A nulltest covers the entire frequency spectrum. So, what if one of the candidates is horribly off in the lower end but pretty damn near to perfect in the entire rest of the frequency spectrum whereas the other contestant is mildly off all over the frequency range? The one with the "wrong" low end will likely lose in a plain dB based nulltest - but it might still be the better one for any real life application because it's doing better in the relevant frequency range.

This brings up the other point being discussed. Does it sound good?

Lots of metal heads thrive on a metric crap ton of bottom end (even playing a 7 string to get more) where all other genres the guitar tends to be a mid range instrument for the most part. While most of us perceive a stand along guitar with more low end as being pleasing, it generally doesn't work well in the mix in a live setup (except for metal). Note: For most music played at bars and festivals.

Additionally, I agree with Sascha. Analysing a null test based on your metering in the DAW is really roughshod, and isn't the appropriate way to do it. The appropriate way is to perform the null within the spectral domain, just as I did here:

Ok, this is surprisingly articulate for a Richard Cranium. Kudo's.

Analyzing a null test in the frequency domain makes a metric shit ton of sense. This isolates where the differences are and allows us to easily throw out the parts that aren't heard anyway (arguably, metal heads would say that the lower frequencies are "felt", but no one can reasonably argue that >20KHz is of any use to humans).

Do you think that the frequency domain analysis eliminates phase shift issues as well? Just changing the phase in a time domain null test makes it fail completely even though the source is audibly identical.

So, apparently "MKII" profiles can only be created if your Kemper is hooked up to Rig Manager?! You can see @GuitarJon 's heart break in real time as he explains this.

Wow. This makes a great deal of sense if my guess is correct.

Given that:

1. The DSP and SRAM for the DSP are identical in MK1 and MK2
2. The Application CPU has been upgraded
3. All other hardware limitations of different variants are the same (ie, Toaster and Rack don't have WiFi or True Impedance added)

I have speculated that Kemper is looking to provide a paid upgrade to MK1 users so they can get the "full resolution" playback of the profiles created with MK2.

What if it is ALSO possible to create an MK2 profile on an MK1 despite the application processor being slower? Now it becomes quite difficult to prevent some EE hack (like me perhaps) from locating the instructions in the binary that have artificially locked out the function and removing it (and posting it on the web).

Requiring a link to the web can enable a very robust authentication method that would be virtually impossible to crack.

This would ensure that ONLY MK2 users could CREATE a new profile using the new algorithm.

In all honesty though, I wouldn't bother trying to hack it if they offered a paid upgrade at a reasonable price. Not worth my time.

Not to my ears, and I have them side by side.
And in 10 years of profiling my own amps…they always popped out great.
And 15 years of YT blind tests where people can’t tell A from B…those “eartests” made the mark also.

So “simply fails audible” can’t possibly be true in all cases.
Maybe certain amps, idnk. And sure there’s vids where it is audible, but that’s also the case for QC, and for both 10 others where no one can tell amp from capture by ear.

I only profiled my VHT and Fender BF before I sold them .... but I found that I actually like other peoples profiles better than anything I created. It isn't that my profiles didn't sound accurate, it is just that I now have more to select from and find I like other sounds better in the mix.

I owned a Nano for 28 days and returned it. Clumsy as a performance tool. MIDI is unevolved on it, as is the hard FX chains. Fixable yes, but released before it was finished. Sounded on par with tonex, but with better FX, even if they're inflexible.

... and this is a VERY good point. For those of us that gig, all these minor quibbles over capture accuracy are FAR down on the list of wants and needs. The ergonomics of the unit and EFX quality all rank far above "capture accuracy". I would guess that >90% of Kemper owners never create a profile.

Many people see the convenience of the Kemper in being a mature gigging rig. Nothing either ToneX or NAM qualifies for.

Absolutely!

I'll always pick Kemper over the other capture offerings in this space. Accuracy be damned. I like their effects and cherry picking profiles you know you like can work.

Not only this, but getting to a tone you want for a particular song is VERY easily achieved.

Wow that a new level of dumb
So that explains how mk1 can do the MK2 profiles it not the hardware but all through software
So MK 2 is really just a nice new shade of on a

I'll bet they are gearing up for a paid update to MK1. It will likely be a license only update; however, I wouldn't discount a paid hardware update as well since only a single DIMM slot card would need to be replaced. I doubt they would let the consumer do it, but a shop for sure.

The need to connect to the web would prevent people from just buying the DIMM card and updating themselves (which is exactly what I might do if it was overly expensive). The authentication to the server would prevent this.

Captures/profiles have a harder time making improvements without moving into things like parametrized models. Which might be something that an end user cannot easily do, and is more suited for things like Neural DSP's TINA process for data collection.

The ability to tweak after capture is one of the big strengths of Kemper IMO. This is particularly true since so few Kemper owners ever make their own profiles.

But the stupid thing is that why do you need a MK2 series in the first place, if the upgrade is entirely software?

Well, it isn't ENTIRELY software. They are updating the Application CPU. This is the processor that is more like a little PC on a card that you see in the video breakdowns. It is responsible for the boot up, GUI and interface management ..... but NOT the sound. The sound is all done in the DSP chip ..... which hasn't changed.

What this DOES do is offer a much faster boot up. There might be some other capabilities baked in that I am unaware of (I have heard that the USB now has an improved interface? This makes sense as it is connected to the Application CPU as well).

 

[Orvillain]

Analyzing a null test in the frequency domain makes a metric shit ton of sense. This isolates where the differences are and allows us to easily throw out the parts that aren't heard anyway (arguably, metal heads would say that the lower frequencies are "felt", but no one can reasonably argue that >20KHz is of any use to humans).

Do you think that the frequency domain analysis eliminates phase shift issues as well? Just changing the phase in a time domain null test makes it fail completely even though the source is audibly identical.

When you an STFT over an audio signal, you're sliding a window across the data and converting from the time domain to the frequency domain. If you also extract the phase information, you end up with magnitude and phase data for each frequency bin.

So it will account for any existing phase relationships.

However, the STFT can introduce a phenomenon known as FFT smearing, which is essentially a loss of time or frequency resolution due to the fixed size of the window. This is a result of the time-frequency trade-off, whereby a narrower window gives you better time resolution but worse frequency resolution, and vice versa.

You can tackle this by choosing an appropriate window size for your signal, by having an overlap of each window to reduce discontinuities, changing your window type (Hann versus Blackman-Harris for example), or more advanced multi-resolution techniques; like Wavelet analysis.

I'm not a Richard Cranium. But I am quite a fan of irreverant humour. I also like sexy burds.

 

[ejecta]

First, WTF @Orvillain, click bait for me? Awww. You must have missed me. And I didn't get you anything so I am feeling bad.


Fair. I think the biggest issue with Kemper's capture is that it requires a lot of manual intervention to get it to sound the same as the source. It's more of an art than a science.... which it shouldn't be. Others do it without all the hand waving and sacrificing a virgin to a pagan God.

Poor wording. Should have said "Audible". My thanks to the grammar police.

Now that is fair. Particularly at the bottom Kemper has capture issues. Typically, I see people eq this out post profile, but it still makes your point.

This brings up the other point being discussed. Does it sound good?

Lots of metal heads thrive on a metric crap ton of bottom end (even playing a 7 string to get more) where all other genres the guitar tends to be a mid range instrument for the most part. While most of us perceive a stand along guitar with more low end as being pleasing, it generally doesn't work well in the mix in a live setup (except for metal). Note: For most music played at bars and festivals.

Ok, this is surprisingly articulate for a Richard Cranium. Kudo's.

Analyzing a null test in the frequency domain makes a metric shit ton of sense. This isolates where the differences are and allows us to easily throw out the parts that aren't heard anyway (arguably, metal heads would say that the lower frequencies are "felt", but no one can reasonably argue that >20KHz is of any use to humans).

Do you think that the frequency domain analysis eliminates phase shift issues as well? Just changing the phase in a time domain null test makes it fail completely even though the source is audibly identical.

Wow. This makes a great deal of sense if my guess is correct.

Given that:

1. The DSP and SRAM for the DSP are identical in MK1 and MK2
2. The Application CPU has been upgraded
3. All other hardware limitations of different variants are the same (ie, Toaster and Rack don't have WiFi or True Impedance added)

I have speculated that Kemper is looking to provide a paid upgrade to MK1 users so they can get the "full resolution" playback of the profiles created with MK2.

What if it is ALSO possible to create an MK2 profile on an MK1 despite the application processor being slower? Now it becomes quite difficult to prevent some EE hack (like me perhaps) from locating the instructions in the binary that have artificially locked out the function and removing it (and posting it on the web).

Requiring a link to the web can enable a very robust authentication method that would be virtually impossible to crack.

This would ensure that ONLY MK2 users could CREATE a new profile using the new algorithm.

In all honesty though, I wouldn't bother trying to hack it if they offered a paid upgrade at a reasonable price. Not worth my time.

I only profiled my VHT and Fender BF before I sold them .... but I found that I actually like other peoples profiles better than anything I created. It isn't that my profiles didn't sound accurate, it is just that I now have more to select from and find I like other sounds better in the mix.

... and this is a VERY good point. For those of us that gig, all these minor quibbles over capture accuracy are FAR down on the list of wants and needs. The ergonomics of the unit and EFX quality all rank far above "capture accuracy". I would guess that >90% of Kemper owners never create a profile.

Absolutely!

Not only this, but getting to a tone you want for a particular song is VERY easily achieved.

I'll bet they are gearing up for a paid update to MK1. It will likely be a license only update; however, I wouldn't discount a paid hardware update as well since only a single DIMM slot card would need to be replaced. I doubt they would let the consumer do it, but a shop for sure.

The need to connect to the web would prevent people from just buying the DIMM card and updating themselves (which is exactly what I might do if it was overly expensive). The authentication to the server would prevent this.

The ability to tweak after capture is one of the big strengths of Kemper IMO. This is particularly true since so few Kemper owners ever make their own profiles.

Well, it isn't ENTIRELY software. They are updating the Application CPU. This is the processor that is more like a little PC on a card that you see in the video breakdowns. It is responsible for the boot up, GUI and interface management ..... but NOT the sound. The sound is all done in the DSP chip ..... which hasn't changed.

What this DOES do is offer a much faster boot up. There might be some other capabilities baked in that I am unaware of (I have heard that the USB now has an improved interface? This makes sense as it is connected to the Application CPU as well).

 

[laxu]

Well, it isn't ENTIRELY software. They are updating the Application CPU. This is the processor that is more like a little PC on a card that you see in the video breakdowns. It is responsible for the boot up, GUI and interface management ..... but NOT the sound. The sound is all done in the DSP chip ..... which hasn't changed.

What this DOES do is offer a much faster boot up. There might be some other capabilities baked in that I am unaware of (I have heard that the USB now has an improved interface? This makes sense as it is connected to the Application CPU as well).

The processor on the Player is like 200 MHz while the old one was something <100 MHz. While this might seem like a huge upgrade...it shouldn't really make a difference for anything but faster boot time and more responsive GUI. We are still talking about a chip that is less capable than what you find in a Tonex One.

If the DSP is the same, why can't it run those new profiles on the MK1 hardware too at the same quality? That's what's confusing to me.

 

[LP59]

If the DSP is the same, why can't it run those new profiles on the MK1 hardware too at the same quality?

Profit on sales of new units? There's a pretty good margin if the biggest change to manufacturing is retooling for a new color of paint.

 

[LP59]

First, WTF @Orvillain, click bait for me? Awww. You must have missed me. And I didn't get you anything so I am feeling bad.

Another post from you about aliasing would be a wonderful gift. Endless amusement. (I stopped reading your post after the first line, so maybe you did say something about aliasing).

 

[Sascha Franck]

Sure but that’s only one use/market for it. Why not open it up at this point to plugins too, given no one is going to steal the tech when there are open source options that beat it. STL and Overloud already essentially started eating their slice of the pie, Kemper can still cash in what’s left if they want to.

Not sure how well that'd turn out for them. All of a sudden they'd have to take care about copy protection and what not.
If at all, it'd possibly be cool if they did it the way Line 6 did, as in having a 100% HW compatible plugin. Which is just fantastic with HX Native (even if I only use it for special occasions myself).

Yep, ever since day 1.

Oh right, now I seem to remember. I even considered trying it out myself one day - but something else must've gotten in the way, so I even forgot how it was like.
Anyhow, as long as there's no decent HW NAM player apart from the Dimehead one (which isn't my COT UI-wise), it's not that interesting for me anyway.

 

[Mongillo19]

What's most annoying about this whole release is the lack of transparency. Having to hear potential features/functionality through random youtubers and rumor?

It just piles on to the idea this was a last second idea to generate sales. Why wouldn't you have atleast an example of the new profiling, or how the profiling process has changed? And that to me is the minimum. The fact it didn't launch with the new profiling is ridiculous

 

[laxu]

Profit on sales of new units? There's a pretty good margin if the biggest change to manufacturing is retooling for a new color of paint.

The MK2 chassis is different, made from aluminium so it just looks largely the same.

It'll be interesting to see what MK1 vs MK2 amounts to when they release the new profiling.

 

[Sascha Franck]

What's most annoying about this whole release is the lack of transparency. Having to hear potential features/functionality through random youtubers and rumor?

It just piles on to the idea this was a last second idea to generate sales. Why wouldn't you have atleast an example of the new profiling, or how the profiling process has changed? And that to me is the minimum. The fact it didn't launch with the new profiling is ridiculous

Open job offer at Kemper: Senior Marketing Advisor.

 

[Dom Mcsweeney]

I think this is nothing more than an exercise in Dopamine marketing.
Get people hyped up for something that will drop later on, in the meantime people will furiously argue for months until the release, thus fuelling 'the algorithm', for absolutely no cost.

In doing this they are winning regardless because they are satisfying both the algorithm with a constant bickering fest, and keeping everyone guessing - a bit like how Christmas works for kids. It's straight out of the religious playbook, and works in the same way on the same kind of people, and doesn't do anything to improve the world (or in this case, technological advancement).

 

[nicolasrivera]

I can't believe Neural DSP hasn't come out with a $100 "capture player" that's free for QC/NC owners, and $100 for everyone else, that plays their captures and has some effects, as a plugin. That'd be really cool.

I can't believe Line 6 has no touchscreens yet! we live in wierd times!

 

[Orvillain]

This launch is total gobshite, whatever anyone says. Basic information is just not obtainable.

You don't "soft launch" hardware. You simply don't do it.

 

[LP59]

I think this is nothing more than an exercise in Dopamine marketing.
Get people hyped up for something that will drop later on, in the meantime people will furiously argue for months until the release, thus fuelling 'the algorithm', for absolutely no cost.

In doing this they are winning regardless because they are satisfying both the algorithm with a constant bickering fest, and keeping everyone guessing - a bit like how Christmas works for kids. It's straight out of the religious playbook, and works in the same way on the same kind of people, and doesn't do anything to improve the world (or in this case, technological advancement).

Maybe. But not having a marketing push at the time of the Mk2 launch means they are losing all their momentum at the most crucial time, and getting a lot of the wrong kind of attention.

 

[MirrorProfiles]

Not sure how well that'd turn out for them. All of a sudden they'd have to take care about copy protection and what not.
If at all, it'd possibly be cool if they did it the way Line 6 did, as in having a 100% HW compatible plugin. Which is just fantastic with HX Native (even if I only use it for special occasions myself).

where would they lose out? Every company releasing software has to deal with piracy and protecting their software. The benefit for Kemper is as soon as anyone wants to use their tone elsewhere, they have to buy the HW. If it brings more people to their platform, that’s a win. It’s worked well for NAM and ToneX….

 

[ian_dissonance]

6 months from now when you can unlock MkII profiles for your MkI for the low price of $250. So many times the internet reminds really smart people can say stuff that means absolutely nothing, lol.