When will these people realize that there's no way to accurately simulate an all-tube guitar amp? Power tubes are fundamentally different from transistors in the way they shape the sound and interact with the guitar, and are an actual part of the musical instrument. You can simulate the sound they make and play back that sound, but on stage, in a live performance, solid state amps don't feedback the way tube amps do, they don't distort the way tube amps do, and they don't interact magnetically with the guitar. Most rock and metal musicians need that as a part of their setup.
Only if you don't need those exact characteristics of an all-tube amp, then you can recreate the sound digitally and have that flexibility.

Joris,
You seem to have missed the whole point of this article, and the objective of the guy's design. This is an all-tube amp. It is not a "solid-state amp". It uses tubes for amplification and rectification, not transistors.
The solid-state and computer bits allow you to re-configure the amp to different tube-amp circuits. The solid state stuff is not involved in the actual sound-processing at all.

it's a shame the amp didn't make it through the customs in time for the show. They probably thought it was some terrorist equipment designed to blow up America.
it's also disappointing not to have a video with a demonstration.
So it's likely to undergo some radical changes before it's released. Presumably that means it doesn't work very well at the moment. I always feel it's better to try and perfect something before you announce it.
As far as using it for gigs, you've got to lug around the pad, as well as the amp. It still seems strange that you cannot replicate the sound of valves.

I think this is a intriguing amp to say the least. I do hope the inventor's vision is realized and the amp lives up to all who play guitar. After all this is the 21st century and technology is changing by leaps and bounds.

@Dmoy: yea I seem to have missed that part... That's what happens when you scan through an article on "yet another digital tube amp". But this one is really different.

The "tube sound" has been debunked by an article in Electronic Design. Yes, they sound different, but it's due to damping factor and filtering through a transformer. If you feel like lugging them around, fine.
http://electronicdesign.com/analog/tubes-versus-solid-state-audio-amps-last-word-or-house-fire-part-2

@Captain Obvious: the sound has indeed been debunked, but the use of a tube amp as part of the musical instrument setup is not. Tube amps really are different. See my first comment.

@Joris van den Heuvel: It's really no more different than the difference between two tube amps or the difference between two transistor amps. It is entirely, both in theory and realistically, possible for a digital circuit or a computer program to simulate everything tube amps did. In fact, one can do more with transistor logic than with vacuum tubes, with significantly less power and space usage.
The problem is how complicated the circuit or software has to be. As for being part of the instrument and interacting magnetically with the guitar, that can be done digitally too. Would you like to stick a gyroscope in your guitar and pair it to a computer too? Could dedicate that to some weird effects.
TL;DR: This amp is a nice proof of concept of a digitally controlled tube amp, and would be a nice point at which the amp war just... Ends. It's annoying, guys. The answer to your search for the best amp is "preference". Not "tubes", not "transistors", nor "large", nor "small". Just, "Preference".

Yes it is. The way a tube amp couples with a guitar pickup differs from how a transistor amp does it. There's no denying in that: it's a simple fact. You can't simulate the stray magnetic fields and saturation effects of the output transformers, nor can you simulate a high output impedance and how all that affects the speaker resonance and feedback with the guitar, and even the magnetic and acoustic coupling with the speaker cabinet.

"You can't" is not the words you are looking for, my friend. "It is not feasible to" would be more appropriate. The amount of processing power and memory that would take would make the amp too power hungry. But yes, it is, in fact, all entirely possible. One must think outside the box. Noise can be heard, as well as interpreted. In addition, sensors can be added, and anything that can't be accommodated by those two, can be simulated entirely in software. If one were to go that far, it may be easier to get a tube amp.
But that is only useful if you specifically want or need that tube amp sound. Otherwise it is cheaper to get a transistor amp, which would also be lighter.
As for output transformers, one could go through two identical transformers, one to bring it up and one to bring it down, with the same combined impedance that they would want from one transformer in their tube amp, and may get close. Possibly. I wouldn't hold my breath, but it would be cheaper than the hardware and the man-hours for the software to simulate that.
Really, if we did more research, we might be able to reproduce the sound. But to what end? If we did come up with the solution, who would even buy the resulting product? Firstly, everyone would doubt it. If they don't dismiss it outright, they'd be uncertain, and any tube amp users who are attracted to it at all, would fear "converting". This is an example of FUD, which keeps Microsoft and Apple afloat in the software market. Welcome to humanity, bro.