Hoping to get amp/cab captures in Fractal gear in the future, does that appeal to anyone?

I'm not sure this is correct as I understood that captures (in Tonex at least) are based on the IK pre prepared series of sounds that one can hear the process digest for a few minutes during capture. I don't think the guitar used is even taken into account during capture other than for level setting - afaik, its derived from the target device' settings + IK's standard amalgam of sounds thrown at it.
Perhaps. But if true, the profile would sound identical whether using a Tele, Filtertrons, or overwound active pickups. And they don't, in my experience. You can't "get there from here" if you define "there" without knowing where your "here" (the starting point) is, unless you're damn lucky. And since the profiles sound different with different inputs, the odds of your nailing a tone with precision are small.

If what is desired is a guitar "mellotron", input-agnostic, that produces Jimmy Page's 2/26/79 tone on a given tune, whether played on a Tele or an active EMG shred machine, then all we need is a controller and enough samples (based on pitch, vibrato, attack speed and intensity) of that specific guitar sound that can realistically be "copied", regardless of the player, his technique, and gear.

Not sure how you decouple input parameters from output sonics, when the sonics are fixed, and the input wildly varies.
 
Last edited:
Perhaps. But if true, the profile would sound identical whether using a Tele, Filtertrons, or overwound active pickups.

Yeah, that's not true. The input signal is different from different pickups. The profile does not create the same output from any input signal. The profile is created using frequency sweeps that cover the whole guitar frequency range. Even if the profile is applying a fixed EQ curve (like tone matches do in Fractal gear), the underlying differences in the input signals between guitars or pickups would still come through and drive the profile differently.

You are correct in that profiles will only match the source tone of the profile using the same gear setup as was used to originally dial in that tone. For example, if someone dials in a killer SRV tone using their own Vibroverb amp, TS-808, and vintage Strat and then creates a profile of that rig. It's only going to sound the same if you also use a similar voiced Strat. If you play through the profile with a Les Paul with hot humbuckers, it's going to sound radically different.

That's why so many owners of profilers find themselves endlessly searching through premade profiles when chasing a very specific tone. The profile does not compensate for differences in input from different guitars and player technique.
 
Last edited:
f true, the profile would sound identical whether using a Tele, Filtertrons, or overwound active pickups
It doesn't sound the same for all those inputs because IK's amalgam of pre-prepared sounds (that gets played to the target device to derive the profile) covers all those various input bases.

Not sure how you decouple input parameters from output sonics, when the sonics are fixed, and the input wildly varies.
As above, it's decoupled because IK's amalgam of pre-prepared sounds covers those bases. Afaik (and it makes sense based on my (+ what I see of others') usage of the product so far), the profiles are derived just from IK's sound samples fed into the device at whatever settings.

Would be nice if IK gave a more in depth explanation of how the capture process works but I don't see any thing there (I can' get AI to corroberate but I don't think much of AI so won't go there lol!).
 
Last edited:
Just like an amp will sound different with different guitars plugged into it, a capture will sound different with different guitars plugged into it.
 
That's why so many owners of profilers find themselves endlessly searching through premade profiles when chasing a very specific tone. The profile does not compensate for differences in input from different guitars and player technique.
That's why I bought some captures from a pro who describes what his interface related level compensation was with every capture pak, and who describes the settings for each capture, and, with audio samples - what guitar was used etc. Worth it imo.

But ya, if the capture was of an amp/cab set up for a strat, and I use an LP into it, it's gonna sound different - just like if I set up my
amp/cab irl for my vintage strat, and while I'm on p break somebody plugs into my amp/cab with an LP without changing any settings, it's gonna sound different (+ prob not so good)
 
As above, it's decoupled because IK's amalgam of pre-prepared sounds covers those bases. Afaik (and it makes sense based on my usage of the product so far), the profiles are derived just from IK's sound samples fed into the device at whatever settings.

I'd wager what they are actually doing is EQ matching the input signal and the amp separately. That would help somewhat normalize the frequency content and level of the guitar input signal before it gets to the amp part of the profile, giving you more consistent output response between guitars. What that also does is completely destroy the tone of your own guitars and pickups. When I switch guitars or pickups, I want to hear it. Otherwise what's the point. An amp model or profile that sounds exactly the same with any guitar or pickup couldn't be further from what I actually want. I'd be pissed if my $3000 guitar gave the exact same results as a $100 junker. That's an epic fail in my book.
 
Yeah, that's not true. The input signal is different from different pickups. The profile does not create the same output from any input signal. The profile is created using frequency sweeps that cover the whole guitar frequency range. Even if the profile is applying a fixed EQ curve (like tone matches do in Fractal gear), the underlying differences in the input signals between guitars or pickups would still come through and drive the profile differently.

You are correct in that profiles will only match the source tone of the profile using the same gear setup as was used to originally dial in that tone. For example, if someone dials in a killer SRV tone using their own Vibroverb amp, TS-808, and vintage Strat and then creates a profile of that rig. It's only going to sound the same if you also use a similar voiced Strat. If you play through the profile with a Les Paul with hot humbuckers, it's going to sound radically different.

That's why so many owners of profilers find themselves endlessly searching through premade profiles when chasing a very specific tone. The profile does not compensate for differences in input from different guitars and player technique.
Yup, exactly. The profile is input-sensitive, and what that does to the output is variable, so the profile will sound different with other than identical creator and consumer rigs.
 
That's why I bought some captures from a pro who describes what his interface related level compensation was with every capture pak, and who describes the settings for each capture, and, with audio samples - what guitar was used etc. Worth it imo.

But ya, if the capture was of an amp/cab set up for a strat, and I use an LP into it, it's gonna sound different - just like if I set up my
amp/cab irl for my vintage strat, and while I'm on p break somebody plugs into my amp/cab with an LP without changing any settings, it's gonna sound different (+ prob not so good)

Yeah, when I was farting around with the NAM plugin before, almost none of the profiles I found online gave any indication of what was used to capture them. Many didn't even specify the amp's settings that were used. Total guessing game and the results followed suit.
 
Yeah, when I was farting around with the NAM plugin before, almost none of the profiles I found online gave any indication of what was used to capture them. Many didn't even specify the amp's settings that were used. Total guessing game and the results followed suit.
Yeah, I'm not sure how people manage to find anything there.
 
There were a couple of entries on Tone3000 that included multiple captures at different settings and those were tolerable, but still felt very limiting. Adjusting the plugins knobs felt absolutely nothing like the real amp. It was more like post EQ at the mixer. Totally not the same.
 
It doesn't sound the same for all those inputs because IK's amalgam of pre-prepared sounds (that gets played to the target device to derive the profile) covers all those various input bases.


As above, it's decoupled because IK's amalgam of pre-prepared sounds covers those bases. Afaik (and it makes sense based on my (+ what I see of others') usage of the product so far), the profiles are derived just from IK's sound samples fed into the device at whatever settings.

Would be nice if IK gave a more in depth explanation of how the capture process works but I don't see any thing there (I can' get AI to corroberate but I don't think much of AI so won't go there lol!).
So the player's entire rig is sensed by the IK software, and it adjusts the profile's output tones to taste? On a "dumb" 56 Tele (aka with zero onboard self-instrumentation), you believe IK can "learn" pickup impedances, scale lengths, radii, etc., --and then adjust the profile to sound like the created profile did on a vastly different rig? Doubtful. I'd love to see the code that accomplishes this with dumb appliances, because it would be revolutionary.
 
So the player's entire rig is sensed by the IK software, and it adjusts the profile's output tones to taste? On a "dumb" 56 Tele (aka with zero onboard self-instrumentation), you believe IK can "learn" pickup impedances, scale lengths, radii, etc., --and then adjust the profile to sound like the created profile did on a vastly different rig? Doubtful. I'd love to see the code that accomplishes this with dumb appliances, because it would be revolutionary.
I never said any of that! Those are your conjectures - I'm simply saying: afa ik/can see, Tonex profiles are captured using IK's amalgam of sound samples fed into the target device being captured at whatever settings - the source guitar does not appear to influence the capture during the actual capture process. This imo, makes some sense and aligns with what I've seen the product do - I get a resulting capture of the target device at given settings, which yes, seems to be good enough to be able to sound different with different guitars playing into it (not sure how realistically or why I'd wan't to do that if the amp was set up for a specific guitar type), but which, as discussed above, does not really react as realistically to varying inputs (ie boosts, pickup changes, volume changes, preeq ...) as what we are used to with Fractal white box modelling (not a terrible thing - it's just another tool, not a replacement).

That's it - and it makes logical sense to me - maybe there's more to it that's beyond my view, but I can't find any concrete info.
 
Last edited:
An amp model or profile that sounds exactly the same with any guitar or pickup couldn't be further from what I actually want. I'd be pissed if my $3000 guitar gave the exact same results as a $100 junker. That's an epic fail in my book.
Agree totally. But at least adjusting an amp model in a modeler is (more or less) going to reflect that amp with that adjustment made. But tweaking a profile to any great extent results in a total crapshoot; it does NOT react like a high-quality amp model, more like a funhouse mirror So if yr profile works as-is, great! However, all bets are off when you significantly increase your profile's gain, or presence, etc.
 
So the player's entire rig is sensed by the IK software, and it adjusts the profile's output tones to taste? On a "dumb" 56 Tele (aka with zero onboard self-instrumentation), you believe IK can "learn" pickup impedances, scale lengths, radii, etc., --and then adjust the profile to sound like the created profile did on a vastly different rig? Doubtful. I'd love to see the code that accomplishes this with dumb appliances, because it would be revolutionary.
Why would it need to sense anything? Maybe I misunderstand something but essentially the capturing process need to send known signals and measure what comes out of the device you’re capturing, so you need different levels, different frequencies, and different attack times to see response in frequency and time domains.

The issue isn’t capturing itself, but the settings of the device while capturing. Whoever is doing that will set it so that it sounds good for them with a particular guitar, and whether it’ll work for you is an open question.
 
It just dawned on me...for the people making a deal about this...How is NAM all that different than Tone Matching which we've had since the Axe-FX II. I don't think it really is. Instead of having Tone Matching on a next-gen unit...we'd just have NAM instead. Same difference.
 
Back
Top Bottom