Kemper Player - 1st [ Paid for ? ] Update coming very soon ?

I mean in theory a dimehead pedal could do the capturing/processing but typically need a GPU.

That's why I'd suggest to do some mass capturing on a simple device and then do all the heavy lifting once the device is connected to a computer/cloud.
 
Just to continue our friendly point-counter-point debate on this :) ... it is my view that ^^ these 2 words ^^ are the very reason why NAM will never ever be any sort of Industry Capture Standard.

There has been / is / and will continue for some time a "default de-facto" standard ... which is Kemper Profiles ... but again, for obvious reasons, they too will never ever be any sort of Industry Capture Standard.

In fact, I would argue - theoretically - that if there were to be another widely-adopted-format for the emerging near future, Tonex would have the lead by a mile ...... I don't think a product has exploded and spread so widely in the M.I in my memory in recent time .... but even Tonex will never be that.

Just my 11c :)
The only relevant difference between NAM vs Tonex is that Tonex is proprietary to IK. They are both just different takes on PyTorch based machine learning.

NAM is already on the way to becoming a standard as there are multiple devices and software that can run it. Open source is not an issue for this, it's a plus because you are not beholden to some company keeping it updated. You simply fork your own version and take the functionality as is, or augment/alter it to your needs.

I think you are confusing standards with market leaders here. Kemper is a market leader, but their profiling is not a standard, or suited to become one unless they make it available to others.
 
They are both just different takes on PyTorch based machine learning.

To be clear, that's not much different than saying two songs are the same because they are in the same key. Pytorch covers a lot of ground and is a basic tool with many different ML algorithms and supports arbitrary topologies. And even then, it's often the data and feature selection that make the biggest difference to outcomes.

That said, it's possible under he covers there's more similarities, but who knows.
 
The only relevant difference between NAM vs Tonex is that Tonex is proprietary to IK. They are both just different takes on PyTorch based machine learning.

NAM is already on the way to becoming a standard as there are multiple devices and software that can run it. Open source is not an issue for this, it's a plus because you are not beholden to some company keeping it updated. You simply fork your own version and take the functionality as is, or augment/alter it to your needs.

I think you are confusing standards with market leaders here. Kemper is a market leader, but their profiling is not a standard, or suited to become one unless they make it available to others.
ToneX uses Tensorflow, but I get what you mean.

There aren't any standards with this stuff. You've got weights, you've got biases. But each ML framework uses a slightly different approach. Because of commerical realities, it isn't really feasible to expect all companies to pull together and agree on a standard.

At the moment in the HISE community, there is a push to support NAM models. But NAM relies on a different ML framework than HISE does. So they're trying to bridge the gap. Once it is done, you're going to see a butt ton of NAM based plugins released by companies like Korneff and the like.

The conversation is here if you're interested:
 
WAV/AIF turned into the standard for IRs as well. Really not all *that* much different.
They didn’t turn into the standard, they’re literally just what impulse responses are and always have been. A capture is not as simple to play back in the way that a typical .wav is to convolve. It really isn’t similar to adopt.
 
Last edited:
ToneX uses Tensorflow, but I get what you mean.

There aren't any standards with this stuff. You've got weights, you've got biases. But each ML framework uses a slightly different approach. Because of commerical realities, it isn't really feasible to expect all companies to pull together and agree on a standard.

At the moment in the HISE community, there is a push to support NAM models. But NAM relies on a different ML framework than HISE does. So they're trying to bridge the gap. Once it is done, you're going to see a butt ton of NAM based plugins released by companies like Korneff and the like.

The conversation is here if you're interested:


I've messaged the link to that thread to Steve, maybe he'll have time to check in. We'll see.
Definitely would be cool if Hise did manage to get NAM support implemented.
 
Hmm, I could have sworn I found PyTorch libraries bundled with the Tonex stuff. Maybe I just remember incorrectly.
Not sure. This is why I see:
1727699382773.png
 
I've messaged the link to that thread to Steve, maybe he'll have time to check in. We'll see.
Definitely would be cool if Hise did manage to get NAM support implemented.
Yeah it would be a pretty good thing for the community, because Hise does not require JUCE or C++ knowledge in order to build a plugin. So pretty basic JavaScript (Hise uses HiseScript, which is inspired by JS) knowledge and their built in ScriptNode system would be enough to start building an amp sim.
 
Yeah it would be a pretty good thing for the community, because Hise does not require JUCE or C++ knowledge in order to build a plugin. So pretty basic JavaScript (Hise uses HiseScript, which is inspired by JS) knowledge and their built in ScriptNode system would be enough to start building an amp sim.

yeah, that does sound pretty ideal and hopefully it works out
 
I feel like people are confusing upgrade with update.
Kemper updates are still free from what I can tell. The upgrades cost money, as is normal for any upgrade for anything ever
This is an update, though. They’re just turning stuff on that the hardware could already do. And it’s not even kind of the same thing as model packs from L6 back in the day. The X3 could still do all of its hardware functions and run a full “load” on the DSP. The is more like you bought a 8GB graphics card for full price and then NVIDIA emailed you to say to an extra $700 they’ll let you use the other 8GB of processing that’s already there.
 
You all can say what you want. For 13 years the Kemper has been the best deal on the market because it has always sounded good (best to many), the free fw updates add cool new features and don’t mess with legacy tones, and because the founder, an innovator technologically but also for his resistance to trending corporate practices, made a conscious decision to forgo self-dealing and establish a cut-free platform for third-party vendors, who in turn create affordable supplemental products, AND all original hardware is still supported. Etc etc, zero physical gotchas, no planned obsolescence, and the sh*t werks. Now moving forward the $1200 Player is part of that ecosystem.

If you be getting your tones elsewhere, or prefer the QC which is a little bigger but still small to the Player, or anything else, power to you. No one cares! Go for it!

For my money, and I’ve owned everything but Fractal, Kemper has always been a solid ick-free (premium it’s true) investment.

ymmv
 
You all can say what you want. For 13 years the Kemper has been the [...]
I think we've spotted the problem.

I actually agree with you to some extent. Kemper did good for a long time. But, for the past few years/versions, a lot of duct tape and chewing gum has been applied to keep things trundling along, and it's really starting to show. That's where most of the outrage is coming from. Not $299 for fresh new functionality, but $299 for a rickety structure that's straining and bulging the seams of the base architecture.
 
I'm not necessarily opposed to DLC, but I don't like the idea of tiered product functionality like the Player is doing.

The concept of DLC originally was that you bought a video game, and after the fact they added more content, and you paid for that incremental product. To me that's fine, it's above and beyond what you got originally. So let's say IK Multimedia is doing a Friedman Amps bundle for Tonex, and you can get a bunch of new amps and cabs for $50. That to me is legit. It's new content, separate licensing, applies to whatever hardware you have.

My problem with the Player is that this functionality already existed and is free in other products. They just chose to disable it and then later enable at a price. That to me is really shady. Same thing if NDSP comes out in a few months and says "hey you can now have all the effects in QC but it costs $300". Shady activity.

There's a grey area I suppose where if you price the product low enough, and then tier the additional functionality, there's a benefit to the customer if they don't need everything.
 
Just wait 'til someone mentions how we used to do the subscription model back in the day too.
1727713153981.png

That design style of the Guitar Port makes me feel nostalgic. Part discarded Transformer codpiece, part googie motorcycle fuel tank? Apple has made things so much more about function, which is overall for the best but everything looks so boring these days.
 
Last edited:
Back
Top