Posts by RosboneMako

Quote from OneEng1

...

Yes, that is what I was thinking.

2048/44100 = .04644s Inverse = 1/.04644s = 21.53 Hz
2048/96000 = .02133s Inverse = 1/.02133 = 46.875 Hz

So increasing sample rate to 96k could actually sound worse on the low end.

As far as the 100k speaker samples/points, There are algorithms for large IR based reverbs that could be applied where you bin samples after a certain point. So the speaker could actually have the full reverb of the cab/room etc. Not sure if that is good or bad. But Line6 always had some "room" verb, so maybe it would be a cool thing? Exciting to see what they come up with. At 44.1 kHz that could be 2.27 seconds of verb. And maybe this requires a faster processor or more RAM to store the extra samples when first acquired? Thus requiring new hardware and robbing us Mk1 folks of an update

I am not smart enough to fully comprehend the aliasing. Since clean guitar cant really create freqs past the 20 kHz range or audibly need to, the aliasing is lost on me. Because the aliasing would come from the sampling, which is well below 20k. So there should be no aliasing in the sampling action.

When creating amp clones for my VST, the Kemper would always fall apart above 7 kHz. My guess is the input could be filtered to 7 kHz anyway. Well below the 20 kHz (22.05 Nyquist). I have not tested it. I just noticed the Kemper did it, but real amps did not. They worked all the way out to 20k and beyond. Again this reinforces that the Kemper should not have aliasing issues via input sampling. You could probably filter down to 2-5k and still get a natural sound?

A quirk of software is that it can be creating something that did not exist before. Necessity is the mother of invention. You can't solve a problem you did not know existed.

This means that Mk2 was ready a while ago. But there may be new things being looked at, now that it is in peoples hands. The engrs have maybe thought of cool new things to add or adjust. The more we talk about what it could be and what our hopes are, the better ideas they may get.

Another thought I had was people wishing the Kemper was 48 or 96 kHz instead of 44.1 kHz. The lower the sample rate, the more accurate the IRs are. They will have more low freq info in them. A side effect of a faster processor could be post effect rate adjustment. Could still process at 44.1 but upsample to 48/96.

Another thought on DI type boxes: The Kemper unit sounds like it directly samples the voltage on the speaker. That voltage would not necessarily translate to sound. The mass of the voice coil, speaker material, etc is part of what cuts off the high end. So a direct sample may be much brighter than the actual speaker. The voltage may be applied to the speaker but the mass/surround tension/cab pressure may stop the speaker from moving.

Speakers are also not linear in their movement. The louder the sample is, the less accurate it may be.

All this said, it is like the discussion in the other thread about null testing. With speakers, all of the errors and problems can be what give it the character you are looking for. So a DI should always be in your arsenal, in case that is the sound you are trying to get.

I have made many IRs from simple digital filters. This should mean they have perfect phase etc. A real IR with all of its problems usually sounds better.

For me I like my IRs in between perfect and hairy messes. It probably comes down to how you are hearing the sound. If your monitors are coloring the sound, you may want the IRs to be less hairy and closer to perfect. If you monitors are not coloring the sound, you may want IRs with all the problems embedded in them. Luckily, Kemper gives you a control to go between perfect and hairy.

Like everything Kemper, the due diligence has been done and some form of solution is provided.

Quote from Davkemp1994

I do use Rig Manager and get profiles from there. I just wish they had an option on there where you could just click on the profile and hear it and decide if you like it and want to download it.

Sorry, I am confused. If you are in Rig Manager, double clicking the profile loads it into the Kemper and you can play it. It takes like 1 second.

It would take longer to download a wave/mp3 file sample and it would make Rig Manager do something it is not intended too. Play music files.

It is a cool concept, but seems unnecessary. it is already super easy.

Quote from taylorphototexas

I will say this though, I'm going back to profiling with cabs. I like the results better.

I am a dumb guy but I play a smart one on the internet.

A speaker is probably more than just some electrical components. There is the surround material that acts like a spring. A spring that oscillates the voice coil thru the magnetic field.

A speaker in a cab is also under pressure which would push/pull once voltage is released.

The changes are probably a very small amount and is therefore negligible?

Just curious if that would explain why you may like a live cab better?

Quote from jaeger 28

In some other forums (TGP, TGF, ..) it's a commonly accepted truth that Kemper is old, sucks and sounds like a cocked wah, Tonex and NAM are way more accurate and QC is kinda acceptable.

Interesting. While writing my VST, I found when I swept amps the input EQ bass was boosted. If you look at the high end on a Kemper it seems to die off around 7 kHz. My VST has Low Cut and High Cut filters and when you over do them you get a wah type sound. I have never noticed it on the Kemper though.

To a laymen, a null test is the only scientific way to say "this sounds correct". And I applaud a scientific way. However, it really only applies to NAM or some other AI type model. You cant apply this to a Kemper any more than you can a Helix, Fractal, etc. These units are not trying to match it perfectly. They are getting the DNA of the amp only.

And as OneEng1 keeps saying, the Kemper goes far beyond simply copying. It gets the basics and then lets you adjust those basics. I doubt an AI platform can do that. Yet.

I wanted to add that as the gain goes up, the Cab IR becomes the most important part of the sound. You could probably switch thru several amps while using the same IR and not notice much of a difference.

This is because the high gain causes any EQ before the tubes to get flattened out. Think of it like you have a 12 dB boost at 500 Hz. Then raise the gain by 12 dB. That boost is now gone and you are back to flat.

So your best bet for high gain is good IRs from sounds you like. Maybe cruise thru Rig Exchange, etc. Tone matching your favorite band may be a good place also.

And dont forget if you start with a wave file IR, you can EQ it in audio editing apps like Goldwave to tweak it further.

But a good IR is the best place to start because there will be weird things happening with phase in the IR that give it that signature sound. Then post gain/cab EQ from there. This could be why a good paid for IR may be the most reliable solution. They have the studio and equipment to post EQ it and get it right.

Mini D Version 4.10 posted to fix Compressor bug.

I doubt any of those differences are from hardware. That looks purely software. And personally it looks good for me because it looks less compressed. Which is my only complaint about my Stage.

As Ruefus said, you do not want the power amp to ever create any distortion. If it did you would be hearing the power amp and not the Kemper. So the idea is to play it at 20-50 watts, like a normal amp, so it never distorts.

You want the power amp sound to come from the Kemper profile.

For any modeler you want a speaker capable of more freqs than a normal 12". And an amp that is flat and never distorts. If either of these are not true you are not fully utilizing the Kemper.

This should be explained in the manual so RTFM

You need a decent sound card. Preferably one with multiple outputs. The audio latency may become an issue if it lags too much.

For a Stage you will:
Guitar --> Kemper
Kemper SEND --> Audio Card In (1 left)
Audio Card Out (1 left) --> Kemper Return

If you do not have extra audio outs you need to monitor the sound some other way because your output will be in use.
You could also Use headphones one the Kemper if your audio output is tied up with the Kemper.

For my setup I use:
Guitar --> Kemper
Kemper SEND --> Audio Card In (1 left)
Audio Card Out (3 left) --> Kemper Return
Audio out 1+2 go to studio monitors

Then you just follow the normal profiling steps.

My hints would also be:
1) Make any distorted profiles at edge of break up level instead of full on distortion. The results will be much better.
2) A tiny bit of noise gate on the Bias may help with noise and improve dynamic range in the Kemper.

The only difference between High Gain and Low Gain is how much bass is present.

That is why DEFINITION rolls off the bass. If you want more control use an EQ.

Clean amps are generally Low and high freq heavy (50Hz, 10 kHz).
Mid Gain to Blues solo is Mid freq heavy (700 - 1000 Hz).
High Gain has no bass and a lot of highs. Usually a slope up to 2000 Hz or so.

After making Mako Distortion 2, the goal was to post the code. However, that code is a nightmare. I am not a C++ or an object oriented programmer and was learning as I went. I tried to clean it up a little by making a limited version of the app. Most things are forced, hacked, and feel like rooting around in someones attic.

GitHub - RosboneMako/MakoMiniD: Mako Mini Distortion

Mako Mini Distortion. Contribute to RosboneMako/MakoMiniD development by creating an account on GitHub.

github.com

The code is posted and included in the ZIP file. If you are on Windows and have Visual Studio it should be easy to get it up and running. Juce will default the build to the Standalone app which makes coding and debugging 10x easier. Switching the build mode to RELEASE generates the Exe and Vst3 files in the release folders. Read the Manual for a quick rundown of how to do it.

Using the code to build will also remove doubts about unsafe things happening. Since you can read thru the code and see nothing nefarious is in there.

If I get crazy I will try to post some videos about how to use the app and maybe some coding stuff.

Quote from DavidTCIC

But man, i'm extremely frustrated about the kemper algo not being able to DI capture some amps

KNOWING WHAT IT IS TRYING TO DO
A typical clean amp might have:
-> preAmp EQ -> Gain stage -> Power Amp

A high gain amp may be:
-> preAmp EQ -> Gain stage -> Post Gain EQ -> Power Amp

For a modeller this gets boiled down to:
-> preAmp EQ -> Gain stage -> Cabinet IR

As far as frequency goes, the Kemper needs to do magic to get the PreAmp EQ. The Power Amp, post gain EQ, speaker, mic etc end up being in the Cab IR.

The Kemper will not be able to get the PreAmp EQ accurately if the Gain Stage is set too high. It is not possible. The Gain needs to be clean with a low level signal and distorted with a high level signal. So edge of breakup area.

This transistion lets you see the pre EQ because the EQ gets flattened with more gain. The difference of low to high gain = the preAmp EQ. If you have too much power amp/speaker distortion, the preAmp EQ gets obscured and flattened out even more.

If your amp has a much more complicated layout the Kemper may struggle. If you have EQ, then gain, then more EQ, then more gain, etc it is almost impossible to figure out what stage is doing what.

DEFINITION ZERO
As far as the issue with the Zero def, I have never seen this. Mine is always the other way around Def of 10.

A def of zero means there is NO BASS in your amp. If it is a fuzz turd that means you need to turn the def up. Because a Def of zero adds bass. Bass into gain = fuzz tone.

Make a profile and skip the last step where you play guitar into it. Set the def to 5 (noon) and that should sound exactly like your amp when profiled.

HELPED?
Hope that helps get you thinking the right way and shines a light on what could be the issue. It is best to think of the Kemper as a PreAmp capture device. Then let the Sag, Compression, Tube Shape, etc controls create the magic from the power amp. It is good to get a little power amp in there so it can understand it, but not so much it flattens out the EQ and obscures the preAmp EQ.

The Kemper could really do with a few more power stage controls to tune up the captured preAmp section.

My thoughts on the mid gain capture method in detail:

Thread

Best profiling gain setting

1) WHAT IS THIS POST ABOUT
Since trying to create my own version of amp cloning, I have come to an understanding of how the Kemper may go about making a profile™.

My process is to do a frequency sweep of the amplifier applying a high volume signal. Then sweep again with a very low volume signal. The difference of these two sweeps will reveal the input EQ of the amplifier. This EQ is where most of the amps sound comes from. IMO, It is the critical piece of cloning an amp.

Over the years of making…

RosboneMako

January 31, 2025 at 9:35 PM

Quote from robtguitar

I think they HAVE to get rid of this format. This is so limited now compared to the competition.

The whole point of the Kemper is this signal chain setup:
- It makes it a no brainer for people who are not computer/phone savvy.
- You need to change a setting, click the button and edit. Simple and easy to do on stage.
- The controls are actual switches. Not some sensitive annoying touch screen.
- It mimics a real world setup, not some imaginary studio wiring nightmare.
- Users do not need to understand CPU usage and number of blocks working.

It is designed to hit 95% of use cases and be as simple to use as possible. That is main draw to this product.

What you said has some merit. I agree, I personally would have fun if it did more. But it sort of defeats its core purpose.


TOUCHSCREEN TIRADE
Personally, I hate touchscreens. They are hard to work with in many cases. Work differently if you are sweating, its raining, wearing gloves, etc, etc, etc. I can press a switch with any part of my body even if its wet or clothed.

A miss-click on a touchscreen can take you several menus away from where you want to be or make radical changes you do not want. If I do not hit the switch perfectly, nothing happens.

If I am in a weird menu with a touchscreen, I need to find some magic BACK or RETURN button and mash that a bunch of times praying to the spaghetti monster I do not inadvertently hit the wrong area with a knuckle/palm. If you need to change the Mod on the Kemper, press MOD. Done. Simple.

Then you have weirdos like the ding dongs at Apple who have decided that the alignment of your finger has to match the screen orientation. So you cant touch a button from any angle other than straight on and perfectly aligned. Again, a mechanical switch does not care what angle your finger tip is when you hit it.

In my mind a touchscreen is a loaded gun with no safety. Great when you need it but is prone to causing many accidental deaths.

Quote from pippopluto

Guys, I am trying to find the place where CK discusses this perspective problem in the way the Profilers reproduce amps.

Thanks for your comprehension!

Hope you find it. But the priofiling system figures out what the pre gain frequency response is. Any Wah before the amp gets merged into that response.

Lets say the pre gain freq response is perfectly flat. Add a wah with a 700Hz 9db boost and a Q of 2. The freq response the Kemper stores will be the Wah response (700 and 2).

This post I made explains the process of detecting the pre gain response.

Thread

Best profiling gain setting

1) WHAT IS THIS POST ABOUT
Since trying to create my own version of amp cloning, I have come to an understanding of how the Kemper may go about making a profile™.

My process is to do a frequency sweep of the amplifier applying a high volume signal. Then sweep again with a very low volume signal. The difference of these two sweeps will reveal the input EQ of the amplifier. This EQ is where most of the amps sound comes from. IMO, It is the critical piece of cloning an amp.

Over the years of making…

RosboneMako

January 31, 2025 at 9:35 PM

Quote from pippopluto

Does anyone remember (or can point me at) where the comment was?

The wah sound is just a simple bandpass filter that is before the gain stages (Amp). There are two BP filters in the STUDIO EQ. Set the frequency between 400-900Hz, set the Q to 1.4 - 3.0, and set the gain to 6dB or more.

A Wah pedal in general is just a foot adjustable bandpass filter. You move the frequency up/dn with your pedal.

Or you can use the WAH effect. You have less control but have cool things like adding gain dynamically.

Quote from Ruefus

You can jump brands for reasons other than sounds.

One big plus with the Helix is using it in a studio situation. With multiple inputs and routings it makes a flexible audio interface. A Kemper Player with a new Helix could be a perfect storm for these users.

Quote from JedMckenna

How much time have you even spent with any Line 6 effects? So much nonsense already in this thread

I am a Line 6 fan boy. I never had issue with my Pod XT Live, UX8, or Pod HD effects. The Helix effects are ok, not great.

It could be that I expected more and was let down. And that some effects seemed to be missing. Or the fact that I like the Kemper effects more.

I am assuming you have problems from others trashing Line 6. Because OneEng1 always has a logical engineering non-personal take. He stated "is considered" meaning the general feel from internet forums. Pointing out that Line 6 needs something new or to retouch those effects to really make a big splash. Which it appears they have not.

I still have all of my Line 6 gear and love it all. But the Helix has sat idle for years while I play the Kemper. I personally find the Kemper effects to be much better sounding. Just one opinion out of millions.

Quote from BayouTexan

You can easily record a cover song by simply downloading the cover and dropping the MP3 or Wav into a DAW track. If you are serious about recording then a DAW is required. Anything else will give you amatuer results.

Open Broadcaster Software (OBS) is the free program 80% of the world uses to stream/record video. Twitch, YouTube, etc. The problem is it doesn't use ASIO, or at least my old version didn't. So the easy answer is to use another interface that has good Windows drivers. but the OP is asking if there is another way.

I have not played with voicemeter, but it probably has the same issues as OBS. That is a cool audio mixing/control program streamers use.

If you can afford a Kemper, you can afford a $60 audio interface was the way I looked at it.