Kemper User Guide

[Armin]

Just found it:
Profiler | Keep up-to-date

Here is something about 'how it works':

Under the hood

In case you enjoy reading about technical details, below is a description of what the KPA is doing during the profiling process:

During the first phase, you will hear white noise with a rising amplitude. The KPA is now collecting data about the frequency response of the reference amp. The frequency response will change dramatically as the gain increases. This is how the KPA learns about the circuitry of the reference amp and the frequency response of the cabinet. Also, the characteristic impedance curve of the speaker, including its feedback to the power amp, is detected in fine detail.

In the next phase, slowly pulsating white noise is sent to the reference amp. The volume of the white noise is set to a level at which the reference amp starts to distort. This is how the KPA learns about the dynamic distortion curve of the tubes in the reference amplifier. Using this information, the KPA is able to recreate that curve with the highest possible accuracy. This is also true for transistor-based and digitally-modeled distortions.

In the third step, the KPA sends a complex tonal texture that follows a mathematically-based set of rules to the reference amp. This texture creates unique interference patterns that allow the KPA to take a “fin- gerprint” of the DNA of the reference amp’s particular sound. The distortion of the speaker, along with the partial pattern of the loudspeaker diaphragm (also known as “cone breakup”) are excited by this tonal mixture. They complete the characteristic interference pattern that the KPA will reproduce faithfully, once the measurements have been taken.

If the reference amp is clean, the KPA skips the third phase (because there is no distortion to be mea- sured).

 

[quonsar]

Ah yes, the old 'fingerprinting the DNA' approach.

 

[FractalAudio]

Shame that they are resorting to those marketing tactics. That, and much of what they claim is mathematically impossible. Looks like a good product, they don't need to resort to that nonsense.

 

[Armin]

Yes but they change it from:

'to MAGICALLY take a “fingerprint” of the DNA of the reference amp’s'

to the less marketing bla bla version:

'to take a “fingerprint” of the DNA of the reference amp’s'

Lets see how it sounds ....

 

[Satch]

Shame that they are resorting to those marketing tactics. That, and much of what they claim is mathematically impossible. Looks like a good product, they don't need to resort to that nonsense.

If what you say is true, they are doing "false advertising" and there is quite strong regulations against that in Europe and North America.


I read the following long time ago on their website that seemed believable to me:

How does the profiling technique compare to classic modeling?

It took us several years of research and development to have the Profiling algorithm up and running. This chapter might be of interest for the more technically-minded amongst you.

Initially, we focussed our attention on existing and well-documented techniques, most of which fall under the heading “Dynamic Convolution”. Even though these approaches seemed very promising at first glance, and certainly useful for recreating simple, soft distortion circuits, they were by no means adequate for complex guitar amplifiers with heavy distortion. Reality has shown that no other company has managed to create a profiling amp by these methods so far.

Other well-known modeling methods, such as a distortion devices framed by a pre- and post-equalizer, have been used in modeling amps of the first generation. They deliver reasonable results, but the sound still lacks complexity, depth and feel; furthermore, these methods don’t qualify for an automated profiling procedure.

So, we had to go much deeper and, in the process, developed our own technology. A tube amp has dozens of dimensions in sound; consequently, the equation of the profiling algorithm needed dozens of variables to exactly recreate the sound and feel of the original amp.

We spent considerable time researching the dynamic behavior of the tube. It was also necessary to study the interactions between the tube and surrounding circuits, which become very complex when the tube starts to distort. We were lucky to find a way to extract this exact tube behavior by analyzing the intermodulation products of crossing sine sweeps. The relation between the intermodulations and the unprocessed sine sweeps tell us the full story about the distortion shape and dynamic of the tube and the behavior of the surrounding circuit. It can even deliver the information about several distortion stages in a row, so long as there is only one stage significantly distorting. This measurement is independent of the frequency response of the amplifier. Once the distorting part has been analyzed, it can be separated from the transfer function. The frequency response of the cabinet is then easy to determine and separate.

Aside from the ease of profiling the whole amp, there is another good reason not to analyze the amp part by part: we capture the sound of every component at the place where it belongs, including all interactions between the components.

As a potential customer I would like to know if I am being lied to.

 

[MaxTwang]

After reading the Kemper Basics & profiling manuals I was put off. I've got a KPA on order, but I'm thinking it's best to wait and see how well it's received. For the price it doesn't have the stereo connections I would expect (as found on the 11r and Fractal products). And no computer connection, just file transfer with a USB thumb drive.

The profiling section of the manual seemed concerned that the customer's expectations not be set too high, like it documented more situations that would make you "unsatisfied" than ones that would make you "satisfied". For example:

You might also run into trouble when profiling a sound in which both the pre- and power amps of the reference amplifier are driven into distortion. If the resulting profile sounds unsatisfying, try to reduce the volume of the power amp.

WTF, good thing guitar players don't like preamp and power amp distortion at the same time :roll

The KPA Beta Testers can stop hyping the "you won't need an Axe, the KPA can profile the Axe"

You can also profile solid-state amps and modeling amps, although the results from profiling modeled amps can sometimes be underwhelming.

You can profile with a distortion pedal, but

"However, there is one exception: some distortion pedals use a special design that cannot be profiled accurately, for instance the Tube Screamer."

No Tube Screamer type pedals, what's left?

I do love those green LED's around the knobs and the clips of the Fuchs kick ass.