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

Wallets right now… just in time for halloween!

Homer Simpson Reaction GIF by hoppip
 
Did you turn down the variac?

Note I greatly prefer the Brown sounds in the Fractal over Tonex.

The Tonex system of captures is a hot mess to me. Doom scrolling capture snapshots to find a tone … not fun. The Brown sound collection has a giant pile of one man’s ‘authentic’ captures. Which one is real one? They’re radically different even within a collection? Some are flat out horrible.

None are any better than what you can quickly dial in on the Fractal. Expect the Fractal lets you find your Brown tone for your
Rig.

That’s what Fractal is all about and why it has no equal. No gimmicks, just tone.
 
Funny plot twist in the thread, kind of an unfortunate turn of events in general.

But I guess it’s hard to ignore the clear market demand.
 
I've farted around with the NAM plugin a bit. Meh. Some tones are pretty good but you're stuck with whatever settings the capture maker used. No accurate way to adjust for different guitars, styles, preferences, etc.

Sifting through tons of pre-made profiles gets old really quick. For a free plugin and free profiles, it's pretty cool. It's certainly no replacement for Fractal models for my uses.
 
mr_fender said:
Yeah to me it's like the difference between sifting through tons of legacy cabs vs using DynaCabs and just moving the mic where I want it.
Click to expand...
TBH, I don’t care for Dynacabs and don’t use them. I have a few of IRs I’ve been using for years, to me all of this is a waste of space on the device, unneeded actions during firmware upgrades, and additional load time for no purpose (in my case). Neither do I need those thousands of legacy IRs, but at least I don’t need to do anything about them most of the time.

But as a matter of principle yes, I’d probably prefer Dynacabs if I needed to mess with cabs at all.

Guess it just illustrates that there’s no such thing as free lunch, there are opportunity costs all around.
 
FractalAudio said:
There are no plans to offer this in the current generation products.

Next-generation products will support NAM.
Click to expand...
If NAM support is fully implemented , there are two realistic approaches—each with pricing implications. Because standard NAM files can’t run in real time on traditional audio DSPs and instead require an ARM-class processor:
  1. Go all-ARM: Build the unit around a powerful multi-core ARM processor. This would mean rewriting essentially 100% of the current FAS amp/FX code for ARM. (Eventide did something similar for its flagship with good results, so it’s feasible.) Upside: potentially a lower price, since FAS could drop the costly legacy DSP line.
  2. Hybrid ARM + DSP: Pair a high-end (expensive) audio DSP with an ARM processor, since no single chip today offers both at the required performance level. This would push the price higher—likely in the €/$4k range for the next Axe-Fx—acceptable for pro audio, but less attractive to bedroom or semi-pro players.
 
swiss studio said:
This would push the price higher—likely in the €/$4k range for the next Axe-Fx—acceptable for pro audio, but less attractive to bedroom or semi-pro players.
Click to expand...
Devices like QC do both modeling and capturing, don’t they? The new Line6 devices are also announced to do that. And there’s at least one Chinese modeler that runs NAM models, albeit in some concerted form. None of these push 4K.
 
swiss studio said:
If NAM support is fully implemented , there are two realistic approaches—each with pricing implications. Because standard NAM files can’t run in real time on traditional audio DSPs and instead require an ARM-class processor:
  1. Go all-ARM: Build the unit around a powerful multi-core ARM processor. This would mean rewriting essentially 100% of the current FAS amp/FX code for ARM. (Eventide did something similar for its flagship with good results, so it’s feasible.) Upside: potentially a lower price, since FAS could drop the costly legacy DSP line.
  2. Hybrid ARM + DSP: Pair a high-end (expensive) audio DSP with an ARM processor, since no single chip today offers both at the required performance level. This would push the price higher—likely in the €/$4k range for the next Axe-Fx—acceptable for pro audio, but less attractive to bedroom or semi-pro players.
Click to expand...
On the other hand, Dimehead NAM Player is very good for NAM. Low latency, very small in size and easy to use even though there is no editing program for the computer. In fact, it has quite nice convolution echoes and you can load your own echoes if you want. I think the delay is also good. Tremolo is good. It is possible to open two NAM files. For example, a recording from a distortion pedal. Chorus and compressor blocks, among others, are in the works and when they come, it actually has all the effects I need.
 
Actually FAS has a very good option for next gen AXEFXIII with both enough DSP for white box modeling and FX AND a powerfull enough ARM core :
Texas Instruments
TDA4VM Processors AND THEY COST A LOT LESS!

1 Features

Processor cores:

• C7x floating point, vector DSP, up to 1.0GHz, 80

GFLOPS, 256 GOPS

• Deep-learning matrix multiply accelerator (MMA),

up to 8 TOPS (8b) at 1.0GHz

• Vision Processing Accelerators (VPAC) with Image

Signal Processor (ISP) and multiple vision assist

accelerators

• Depth and Motion Processing Accelerators

(DMPAC)

• Dual 64-bit Arm® Cortex®-A72 microprocessor

subsystem at up to 2.0GHz

– 1MB shared L2 cache per dual-core Cortex®-

A72 cluster

– 32KB L1 DCache and 48KB L1 ICache per

Cortex®-A72 core

• Six Arm® Cortex®-R5F MCUs at up to 1.0GHz

– 16K I-Cache, 16K D-Cache, 64K L2 TCM

– Two Arm® Cortex®-R5F MCUs in isolated MCU

subsystem

– Four Arm® Cortex®-R5F MCUs in general

compute partition

• Two C66x floating point DSP, up to 1.35GHz,

40GFLOPS, 160GOPS

• 3D GPU PowerVR® Rogue 8XE GE8430, up to

750MHz, 96GFLOPS, 6Gpix/sec

• Custom-designed interconnect fabric supporting

near max processing entitlement
 
Dave Merrill said:
Personally I'd suggest letting Cliff work out the trade-offs and best options. He's vastly better informed about all this, and intimately familiar with the requirements of Fractal-style dsp.
Click to expand...
But as he won’t share those part before the official announcement it was kind of interesting to speculate about the feasibility of this SOTA white and black box modeling. Knowing that this new TI chip could easily fit all requirements for less money than AF3 current TI both DSP is a great hope for the next product.
Also the proof that the owner of FAS wants the best optionS for his customers. Even Line6 didn’t fully stated NAM support and instead will offer Proxy, a closed source stuff, like ToneX.
I like NAM because it’s the best capture today (even if I can’t make any difference with ToneX v2 tech today…) AND it’s open source meaning the captures I make today will still be supported in many decades regardless of what a private company decides to do.
Also leaning that FAS is opened on machine learning that could be a smart way to refine White box modeling, especially for vacuum tubes and electronic behavior that can be far from easy to shrink into simple algorithms.
 
mr_fender said:
I've farted around with the NAM plugin a bit. Meh. Some tones are pretty good but you're stuck with whatever settings the capture maker used. No accurate way to adjust for different guitars, styles, preferences, etc.

Sifting through tons of pre-made profiles gets old really quick. For a free plugin and free profiles, it's pretty cool. It's certainly no replacement for Fractal models for my uses.
Click to expand...
I've been messing around with NAM in the last year or so too and I totally agree, for amp tones I vastly prefer classic modeling tweakability to sifting thru thousands of profiles.
But NAM is invaluable for certain things like capturing your esoteric amp and taking it in the box.
And not only amps, e.g. you can capture your favourite fuzz, od or distortion pedal, even studio gear like mic preamps, channel strips and compressors, or make dynamic speaker models to use in place of IRs which replicate speaker compression/drive as well as any saturation introduced by the mic or the preamp (some people have already done all of that).
Basically everything that can produce a non-linearity can be captured quite faithfully.

So this represents a great expansion of possibilities and it also means that Fractal will need to pay less attention to requests like "I want that amp" or "I want that pedal" and can concentrate their development on more interesting stuff.
 
DLC86 said:
I've been messing around with NAM in the last year or so too and I totally agree, for amp tones I vastly prefer classic modeling tweakability to sifting thru thousands of profiles.
But NAM is invaluable for certain things like capturing your esoteric amp and taking it in the box.
And not only amps, e.g. you can capture your favourite fuzz, od or distortion pedal, even studio gear like mic preamps, channel strips and compressors, or make dynamic speaker models to use in place of IRs which replicate speaker compression/drive as well as any saturation introduced by the mic or the preamp (some people have already done all of that).
Basically everything that can produce a non-linearity can be captured quite faithfully.

So this represents a great expansion of possibilities and it also means that Fractal will need to pay less attention to requests like "I want that amp" or "I want that pedal" and can concentrate their development on more interesting stuff.
Click to expand...
Totally agree, I have a lot of my AF3 amp/cabs in NAM format for plugins ease of use and also to share the tone for some projects in logic as the modeling may change within the (usually great) modeling updates. But sometimes a sound is just perfect for me and I like to save it in .nam format. Also I use ToneX captures too because I don’t want to get my precious AF3 out my home,
 
DLC86 said:
it also means that Fractal will need to pay less attention to requests like "I want that amp" or "I want that pedal"
Click to expand...
Exactly what I'm afraid of: profiling diverting significant attention away from what they've been successfully doing and what I enjoy in the product. If next gen is only about profiling, sadly, I'll be out since my interest in Fractal is the improvement of the white-box models and new additions of. I'm happy with Tonex for profiling if/when I want to do that - not sure NAM adds anything new.
 
Last edited:
sprint said:
Exactly what I'm afraid of: profiling diverting significant attention away from what they've been successfully doing and what I enjoy in the product. If next gen is only about profiling, sadly, I'll be out since my interest in Fractal is the improvement of the white-box models and nee additions of.
Click to expand...
No need to worry about that, if you know Cliff just a little, you know he won't abandon white box approach
 
You must log in or register to reply here.
Back
Top Bottom