Liquid Profiling .... Specific Question (?)

I’d say it’s something else:

Quote from Eric Klein - Chief Product Design Architect- Line6

Yep. This is awesome. We've been talking about something like Liquid Profiling for years and of course Christoph pulls it off.

Glückwunsch, mein Freund!

Quote from benifin

When he wrote this, I took that to mean L6 had been considering a Modeling + Profiling idea (?)

Ben

Why would it be that? Quad Cortex has had both for several years. No reason to congratulate anyone.

I'd guess this is some kind of proprietary technique. I don't thin Kemper would do component-level modeling, that's just not how they have done things so far (supposedly). Although I'm sure they have looked at the schematics in the process.

Quote from benifin

Hi all !

Been a while. Very impressed with what I've read so far about Liquid Profiling.

I know its not out yet, but I think enough is known, but I cant find a clear definitive answer.

I have - what I think is a simple question - but cant find a clear answer:-

In Liquid Profiling - at the moment / on release - there will be some 40+ "amps" or "channels" as Kemper call them - my question is:-

=> have these 40+ "amp channels" all been component level modeled [like Fractal, Helix etc..] so their full Gain and Pre-Amp Stages and Full EQ Stacks are fully component level modeled (?) -or- is it some other proprietary "technique" (?)

Thanks,

Ben

Display More

What do you want the answer for ? What can it change for us ?

It's just a question of terms (and design)....

The only thing i know is that tomorrow i will be able to turn the mid button from 8 to 3 on my fender Twin profile and it will react exactly the same as the actual one....

It would be more interesting to know what is your goal.... Cause the technique will not be necessarily the way to achieve it.

I wanna mean, Kemper could have modeled every electrical circuit resistor and so what ?

I assume you have a demand....

If a patent has been granted, it’s something different and proprietary. He also wouldn’t have announced it in 2011, shelved it and come back to it.

Simply combining existing technologies doesn’t qualify for a patent.

….and why would the lead architect of a competing product openly congratulate him for some ho-hum idea?

I think the simple answer is - we don't know how they have done it and probably never will.

...and why care, the real question/test will be does it work/how effective is it

Well, you can figure it out - just read the patent 😁

Does liquid profiling model specific amplifier gain structures?

For example, could I take a clean Fender Deluxe profile and crank it up in a realistic manner?

Or same thing with a Marshall JCM 800, can I take a slightly crunchy amp and turn it all the way up to sound like the same amp on 10?

I’d be shocked if the gain was now modeled. It never has been before.

Quote from Jarick

Does liquid profiling model specific amplifier gain structures?

For example, could I take a clean Fender Deluxe profile and crank it up in a realistic manner?

Or same thing with a Marshall JCM 800, can I take a slightly crunchy amp and turn it all the way up to sound like the same amp on 10?

No. The fact that the sound does not change when you apply Liquid Profiling to existing old profiles should make it clear that there is no new modeling of gain structures.

The Liquid Profile gain knob now replicates the effect of the bright cap on real amps. Yes, this will help a Fender Deluxe or Marshall JCM Liquid Profile to crank up in a more realistic manner.

This is all explained quite thoroughly in the interview CKemper did with ToneJunkie.

Quote from Michael_dk

I'd guess this is some kind of proprietary technique. I don't thin Kemper would do component-level modeling, that's just not how they have done things so far (supposedly). Although I'm sure they have looked at the schematics in the process.

I respectfully disagree to a point. All their FX are modeled, so there is a basis to assume the tone stacks are coded. We'll see soon enough.

Quote from madmarcus1960

I respectfully disagree to a point. All their FX are modeled, so there is a basis to assume the tone stacks are coded. We'll see soon enough.

nobody is saying they aren’t coded merely that they aren’t component level modelled. It is easy to see the output from any tonestack based on component values using software that does component level modelling (check out the Duncan Amps tone stack model software for an example). All Kemper need to do is code the output results they don’t need to get those results by actually modelling the circuit which proced them. Basically all profiles are currently models it’s just that they are only models of the output without the need for processing power capable of doing the heavy lifting of all the component level calculations that produced that outcome.

Quote from Michael_dk

I'd guess this is some kind of proprietary technique. I don't thin Kemper would do component-level modeling, that's just not how they have done things so far (supposedly). Although I'm sure they have looked at the schematics in the process.

Quote from Wheresthedug

nobody is saying they aren’t coded merely that they aren’t component level modelled. It is easy to see the output from any tonestack based on component values using software that does component level modelling (check out the Duncan Amps tone stack model software for an example). All Kemper need to do is code the output results they don’t need to get those results by actually modelling the circuit which proced them. Basically all profiles are currently models it’s just that they are only models of the output without the need for processing power capable of doing the heavy lifting of all the component level calculations that produced that outcome.

Here’s some basics.

What is a bright cap? A resistor and capacitor which functions as a filter.

What is a tone stack? Resistors and capacitors that make up a set of filters.
How are these analog filters going to get modeled in digital signal processing? With digital filters.
Is digitally modeling the behavior of resistor-capacitor filters component modeling? Yes, by definition.

Liquid Profiling is very clever and will work great. But there’s no mystery at this point, no need to make up some “secret sauce” or some “modeling of components that’s not component modeling” oxymoron.

If none of this makes sense to you, just go watch CKemper’s interview. He literally says Liquid Profiling is a marriage of profiling and modeling.

My take is LP will be modeled filters. The whole profiling process figures out how the tubes are reacting to the input. So there is no need for the Kemper to try and model those components.

The simplest thing to do is model the EQ controls from the real amp. This is important because those EQ values are not the same the Kemper uses now with its fixed value EQ.

Add to this the fact most amp filters interact with each other. Changing the mid or bass may actually change how the treble sounds. So boosting the 120 Hz (bass) on a graphic EQ will give you more bass around 120 Hz only. Changing the bass on an amp may also add some 900 Hz because you are altering the circuit and the bass components are part of the mid components. They feed into each other.

The current Kemper EQ can only give one type of EQ curve. The new EQs being added will give you a lot of alternatives. And they will be some what predictable if you know the amp. You could already get very close using the Studio EQ. But it would take some knowledge, testing, and the values will NOT interact like the real amp. With LP it is a one click operation.

Modelers try to nail down all of the gain stages and their interactions. I predict the Kemper will stick to the same single stage gain structure it has now. Just use more accurate EQs.

Quote from madmarcus1960

I respectfully disagree to a point. All their FX are modeled, so there is a basis to assume the tone stacks are coded. We'll see soon enough.

The tone stacks are models. We know this because he’s said it.

There is no way they’ve removed the Profile’s gain structure code to replace it with a model. You could never guarantee that the sound wouldn’t change, and per CK - it doesn’t unless you fiddle with the knobs.