The IR-X does have the typical Friedman sound: good high-gain modded Marshall sound and mediocre low-headroom clean sound that doesn't even crunch particulary well and generally struggles to keep up. In that sense, channel one is good as a pedal platform for drive pedals, but don't expect much more of it - what you're buyng here is the high-gain channel two, which is actually pretty good at medium-gain sounds too. The sound is and does feel like true tube, as much as any device that's not an actual tube amp can. However, definition in the higher gain range could be better, and, personally, no matter what guitar i plugged into it (single coils, low, medium or high output humbuckers), I never achieved the balance of sustain and clarity that I'm looking for in a high-gain sound, especially when it comes to recording, which is the main purpose I got it for. The boosts, while they work, are subpar as compared to most standalone overdrive pedals (a TS MINI in the front did a much better job in getting a good lead sound out of channel two than the internal boost circuit).
The power amp sim and IR's are decent enough, although most positive reviews of the pedal (like Pete Thorn's, etc.) actually use externally loaded IR's for a reason. The DI out is good. The headphone out was inevitably noisy, I never quite understood why.
The foot switches have a latency, which in my opinion makes the device barely usable as a live tool, if you would want to go between channels one and two "on the spot". In fairness, this is not a problem unique to this device (I have also encountered that with one of the bigger Boss pedals that have the same type), and it was explained to me that, because of the noiseless design, they trigger the changes not upon being engaged, but upon being released. Personally, I doubt it's just that and couldn't help feeling like the device was just too slow in "computing" the change. Anyway, I found it counterintuitive to my playing and couldn't really get used to it at all.
The controls work fine and have a decent amount of range to them, but the knobs are too light for my taste and kept on moving upon the slightest unintentional contact (by a cable, my foot, etc.).
The latter two observations lead me to qualify this more as a desktop rather than as a live tool, which is stupid, considering it has the infrastructure to work like one.
Overall, this is certainly one of the better attempts at amp simulation we have seen in recent years, but has its inherent limitations and some design and quality flaws, which made me feel like getting far less than what I expected from a Friedman, bearing a Friedman price premium. And here is my last but not least gripe with this: I bought it in the first months after its release for 595 €. Less than a year later, it was already selling way under 500 €. It is now 444 €. Yes, I know you sometimes pay a novelty premium immediately after a product is released and prices may see some adjustment down the road ... but this is huge! I mean, it's huge like it's almost insulting. 25% of first-hand price erosion in less than 24 months for a big-name product like that is to me the last in a series of bitter lessons to never again go for the hype of big names and YouTube "marketing" like that.