Kemper Profiler MK 2

[2dor]

The idea about keeping the input stage properly set and hence using the hardware Kemper is pretty clever imo.

Well it's what having the same hardware does. That's why stuff like ToneX and NAM desperately rely on proper signal calibration for a good user-experience across different chains. Kemper (and Quad Cortex as well) have the benefit of the hardware underlay which is consistent.

 

[Jarick]

Really intriguing developments! Sounds like it’s going to be an improvement in all kinds of ways in terms of the profiles themselves.

 

[Deleted member 3656]

Really intriguing developments! Sounds like it’s going to be an improvement in all kinds of ways in terms of the profiles themselves.

No refining, and quicker profiling than before. Compare that to how long it takes for tonex to capture even with the update (especially with some processors).

 

[2dor]

Really intriguing developments! Sounds like it’s going to be an improvement in all kinds of ways in terms of the profiles themselves.

If those claims are true + liquid profiles, Kemper's gonna get a new lease on life.

 

[Mongillo19]

If those claims are true + liquid profiles, Kemper's gonna get a new lease on life.

What is a bummer though is all new profiles will need to be created right?

I imagine that will be a good amount of work for creators but then again, a good sales opportunity

 

[paisleywookiee]

No refining, and quicker profiling than before. Compare that to how long it takes for tonex to capture even with the update (especially with some processors).

Can the Kemper do batch processing, since it relies on a computer now anyway? Otherwise capture time is pretty irrelevant now, with Tonex's batch processing.

 

[2dor]

Can the Kemper do batch processing, since it relies on a computer now anyway? Otherwise capture time is pretty irrelevant now, with Tonex's batch processing.

No word on batch processing but with the 36s reamping / training time, churning a pack of profiles is gonna be pretty darn fast compared to anything else I know of.

 

[quadgear]

Surprising quote from Christoph Kemper that he hadn't tried the 2.0 profiling yet. That would suggest it's very new and will take more time to release....

"Just curious because I hadn't had the chance to play around with it yet."

 

[paisleywookiee]

No word on batch processing but with the 36s reamping / training time, churning a pack of profiles is gonna be pretty darn fast compared to anything else I know of.

Sure, I'm just saying time comparisons aren't as crucial as before, IMO, due to the batch processing of Tonex. I guess for "content creators" it would be. But for the average Joe, they'll do a bunch, let it cook, and be done.

 

[Deleted member 3656]

Sure, I'm just saying time comparisons aren't as crucial as before, IMO, due to the batch processing of Tonex. I guess for "content creators" it would be. But for the average Joe, they'll do a bunch, let it cook, and be done.

There's also situations where people in are the studio and want results quickly. But sure, I agree that the tonex batch processing also has its strengths.

 

[Chocol8]

I guess for "content creators" it would be. But for the average Joe, they'll do a bunch, let it cook, and be done.

Content creators will do the same thing. They may grumble a bit about Tonex, but the customer base is so much bigger than Kemper these days, the small differences are not going to change which platforms they favor. They are going to make the stuff that sells. Right now that is Tonex QC and NAM.

In the future Proxy might make a big dent, and Kemper 2 certainly has a shot, but if it doesn't run on version 1 hardware, I don't think capture vendors are going to trip over themselves to recapture their amps for a small potential customer base limited to only the MKii hardware.

 

[2dor]

Content creators will do the same thing. They may grumble a bit about Tonex, but the customer base is so much bigger than Kemper these days, the small differences are not going to change which platforms they favor. They are going to make the stuff that sells. Right now that is Tonex QC and NAM.

In the future Proxy might make a big dent, and Kemper 2 certainly has a shot, but if it doesn't run on version 1 hardware, I don't think capture vendors are going to trip over themselves to recapture their amps for a small potential customer base limited to only the MKii hardware.

Kemper may still have a chance though if their new profiling delivers. 36s is crazy for a good quality profile when you compare it to anything else out there. Put "liquid profiling" on top of that & I could see the ecosystem still being attractive especially if the Kemper Player can do profiling now (albeit connected to a PC).

 

[Deleted member 3656]

Kemper may still have a chance though if their new profiling delivers. 36s is crazy for a good quality profile when you compare it to anything else out there. Put "liquid profiling" on top of that & I could see the ecosystem still being attractive especially if the Kemper Player can do profiling now (albeit connected to a PC).

It's speculation, but I believe the regular Kemper profiling fills out a bunch of parameters in a very flexible amp sim and then EQ matches. Something of that sort.

That can be very fast. But the kind of machine learning tonex does requires a lot more horse power.

Then again, I wonder how the new profiling works, if it's really that different from before, assuming the above is correct. Considering how quick it is, I would imagine it's based on some similar underlying tech, but maybe a more developed version of it?

But again. Speculation to the max.

 

[2dor]

It's speculation, but I believe the regular Kemper profiling fills out a bunch of parameters in a very flexible amp sim and then EQ matches. Something of that sort.

That can be very fast. But the kind of machine learning tonex does requires a lot more horse power.

Then again, I wonder how the new profiling works, if it's really that different from before, assuming the above is correct. Considering how quick it is, I would imagine it's based on some similar underlying tech, but maybe a more developed version of it?

But again. Speculation to the max.

I've read that hypothesis on a few forums along the years but we don't really know for sure what's going on under the hood.

In truth, I think users shouldn't really care & I'd be firmly planted in that camp if the results are consistently faithful to the source.

 

[dws2384]

No refining, and quicker profiling than before. Compare that to how long it takes for tonex to capture even with the update (especially with some processors).

Sure but let’s hold off to see how good it is. There’s likely a reason why NAM and Tonex take the time they do. They are very very accurate. Nothing is a free lunch.

 

[Deleted member 3656]

I've read that hypothesis on a few forums along the years but we don't really know for sure what's going on under the hood.

Sure. I don't think anyone outside Kemper knows for sure.

It's just that when I got into EQ matching, I saw how easy it was to match a lot of amp sims to the source that way.

It made sense to me if Kemper had automated other things, like the right levels of gain. Aside from that, EQ matching can go very far.

In truth, I think users shouldn't really care & I'd be firmly planted in that camp if the results are consistently faithful to the source.

I care, to some extend, mostly because it's interesting to see how different units accomplish things.

But I don't think users should be put off by the Kemper approach, assuming it's even what some think it is, if results are great.

 

[BenIfin]

.... They are very very accurate. Nothing is a free lunch.

Not trying to flog a dead horse, but only in terms of a Null Test. There is a lot of other factors that are way more important.

 

[Chocol8]

The time it takes to train a capture isn't going to be a deciding factor for a lot of buyers and liquid profiling isn't really a killer feature either. It is attractive to people with a modeler mindset, but not really a game changer as far as the ability to get a good tone.

Kemper is going to have to deliver something a little more significant to stay relevant, but I still see little evidence that they even want to. I think the Virus is the best evidence of Kemper's business plan.

 

[BenIfin]

- 36 seconds profiling time.
- no refining
- PC required for new 2.0 profiling
- 2.0 profiles can be liquid profiles
- on-the-unit profiling can still be done
- supposedly doing back-to-back 2.0 and legacy profiles shows a noticeable quality / accuracy difference between the 2

The idea about keeping the input stage properly set and hence using the hardware Kemper is pretty clever imo.

Well it's what having the same hardware does. That's why stuff like ToneX and NAM desperately rely on proper signal calibration for a good user-experience across different chains. Kemper (and Quad Cortex as well) have the benefit of the hardware underlay which is consistent.

Really intriguing developments! Sounds like it’s going to be an improvement in all kinds of ways in terms of the profiles themselves.

No refining, and quicker profiling than before. Compare that to how long it takes for tonex to capture even with the update (especially with some processors).

If those claims are true + liquid profiles, Kemper's gonna get a new lease on life.

The plot is starting to thicken up very nicely ...... bring on the "classic" vs "2.0" profiles of the same Amp comparisons !

 

[Jarick]

The time it takes to train a capture isn't going to be a deciding factor for a lot of buyers and liquid profiling isn't really a killer feature either. It is attractive to people with a modeler mindset, but not really a game changer as far as the ability to get a good tone.

Kemper is going to have to deliver something a little more significant to stay relevant, but I still see little evidence that they even want to. I think the Virus is the best evidence of Kemper's business plan.

Disagree on liquid profiling I think that’s still a really great feature. Having one or a few profiles to cover the range of an amp over having a couple dozen is huge. Especially if the profiles are well made so the gain scales properly.

I do think UI is going to be much more of a driving factor for success than profile accuracy. I’ve said before but Tonex is smaller and cheaper for standalone modeling and QC is better to use for a modern UI.