• We would like to remind our members that this is a privately owned, run and supported forum. You are here at the invitation and discretion of the owners. As such, rules and standards of conduct will be applied that help keep this forum functioning as the owners desire. These include, but are not limited to, removing content and even access to the forum.

    Please give yourself a refresher on the forum rules you agreed to follow when you signed up.

New BluGuitar Amp X - Holy $#!%

666was999

Power User
I could play everything with 3 amp channels only, the problem is to find an amp with the right ones. In the tube amp market the right combination of my 3 channels never existed. So lets hear this one, maybe this time they are in there.
 

merlin17

Power User
Dynamic IRs might mean interpolation between two other IRs, but if so, that's not really how microphone recording works unfortunately.
For example: a mic at 2" from the center will not sound the same as a 50/50 blend of a mic that's on center and one at 4" from center.
In the info it says speakers don't sound the same at different volumes and that's what dynamic IR's would incorporate...
 

jefferski

Fractal Fanatic
"...offers for the first time the groundbreaking possibility to edit and save circuit parameters at component level"

Um, isn't that what all the AFX advanced controls do?
 

666was999

Power User
"...offers for the first time the groundbreaking possibility to edit and save circuit parameters at component level"

Um, isn't that what all the AFX advanced controls do?
The signal path isn"t digital. It's digital controlling device does these edits to the all analog signal path.
Then it will be an analog modelling unit, right?
 

deathbyguitar

Experienced
I could play everything with 3 amp channels only, the problem is to find an amp with the right ones. In the tube amp market the right combination of my 3 channels never existed. So lets hear this one, maybe this time they are in there.
Exactly. I love digital modeling but I only ever use like 2 amp models.
 

MikeyB59

Power User
I have an Amp 1 and I think it sounds damn good. It's very simple and straight forward to operate, too. If this thing has an equivalent level sound, it'll be a great piece of gear.
 

Chris Hurley

Experienced
Did Amp X change from the regular "nanotubes" it used before to the Korg nutube? The previous versions (Amp1) didn't use the Korg nutubes.
 

peteri

Experienced
Interesting some people like them - I had the Amp 1 maybe 3 or 4 years ago when it hadn't been out that long.

Form factor was great, sat very nicely on a medium size board meaning the board was all you ever needed and it wasn't too heavy.

But for me the sound was disappointing - the speaker emulation not much better (if at all) than the H&K Red Box and the whole thing just felt like a solid state device, meaning how it responded to pedals, the drive etc.

Actually one of the most disappointing pieces of gear I'd had in recent years, promised a lot and sounds good in the demos (don't they all?) but for me - just like the Vox ToneLab.

YMMV
 

666was999

Power User
It's Gen1 was meant for vintage tones.
I didn't like that too. Also I don't own a noodled down worn out strat with lush shimmering highs, another part of the vintage equation.

Gen2 (Mercury and Iridium) were not available 3 or 4 years ago, it's rather modern tuned (mercury) or even metal (iridium). I played the mercury and it was instant fun.

I wouldn't go away from my III for it, it's way limited in it's abilities in comparison, but it's at least that tiny piece of gear you could take to the living room for some practising while watching football or like that.
 

peteri

Experienced
It's Gen1 was meant for vintage tones.
I didn't like that too. Also I don't own a noodled down worn out strat with lush shimmering highs, another part of the vintage equation.

Gen2 (Mercury and Iridium) were not available 3 or 4 years ago, it's rather modern tuned (mercury) or even metal (iridium). I played the mercury and it was instant fun.

I wouldn't go away from my III for it, it's way limited in it's abilities in comparison, but it's at least that tiny piece of gear you could take to the living room for some practising while watching football or like that.
Thanks - that's an interesting difference I didn't know about, time to google!
 

MikeyB59

Power User
It's Gen1 was meant for vintage tones.
I didn't like that too. Also I don't own a noodled down worn out strat with lush shimmering highs, another part of the vintage equation.

Gen2 (Mercury and Iridium) were not available 3 or 4 years ago, it's rather modern tuned (mercury) or even metal (iridium). I played the mercury and it was instant fun.

I wouldn't go away from my III for it, it's way limited in it's abilities in comparison, but it's at least that tiny piece of gear you could take to the living room for some practising while watching football or like that.
I didn’t know that. I’m definitely someone going for more vintage or at least vintage gain level type sounds, so it makes sense that I’d like my Gen 1
 

plexi59

Power User
I didn't. All three of those things are "neural" in 2020. In the 90's and 00s neural networks were the "second best thing to do anything in machine learning". Now they are the best thing and by quite a wide margin, in some cases exceeding human performance on the task. When this stuff works, it's like magic. I know because I make my living by making it work for practical problems.

There's zero chance of anything "neural" going on in this device. :)
 

plexi59

Power User
On a more serious note, there's a TON of neural network-specific hardware coming to the market right now. All those things do is basically convolution, and they can do it at a staggering speed, tens of teraflops per chip is not uncommon (theoretical of course, practical throughput is substantially less). This happens to be the heaviest thing something like AxeFX has to do as well. I wonder if there's opportunity for technology reuse here.
 

metalaman

New Member
IMHO, If it ain't learning and improving, it ain't machine learning. Not that I'm against this tech(I've had my hands dirty with it), I think it's very useful and cool, but the best part about it is that it learns and becomes better over time.
Frankly, I just can't comprehend how modelers using ML can learn over time.
 

plexi59

Power User
But it is improving. The promise is there - neural networks provably can approximate any function, no matter how complex. One such function is how to transform the signal from the guitar to the tone you want to hear. The devil is in the details, but theoretically it's doable. It should theoretically be possible to transform my playing into something Paul Gilbert would not be ashamed of, as well. :) But that's way out there in the future. All those "style transfer" demos with images where you scribble something and it gets redrawn in the style of Van Gogh - the principle could be the same.

Technically, in 2020 an iPhone 11 Pro has 6 teraops of int8 throughput on its TPU, and ~1.3 TFLOPs of floating point on its GPU. And that's before you even get to the CPU, which too has more GFLOPs than AxeFX3, although they are harder to harness in a general purpose chip.

That's not what I'm advocating for here though. Neural DSP is in its infancy it will be a while (a decade perhaps) before we get anything practical out of it. In the meanwhile I wonder if neural network chips (which are essentially very powerful convolution chips and little else) can be used for off-label guitar processing. Basically just make them do convolutions and nonlinearities - bread and butter of a guitar modeler.
 
Last edited:
Top Bottom