They matter only when you expect people to buy another GPU every 2 years because yes. AI can't damage older GPU market just because those GPU's lack the capability to do the task, and as you know mining so well, you would already know that happened before. There were GPU's for mining that raised their price artificially (yes, artificially), and others that were not used for that crap that stayed or slightly got a price increase.
Define slightly.
Supply gets constrained, and market demand fluctuates. Everybody feels the higher prices directly on GPUs and other components in the PC market, and used components are more likely to have been cooked, abused, or have been stressed running flat out in gaming applications, compared to the console market or big OEMs that get their components in bulk or at scale.
The point is that arguments around 600 being too much, because people can build their own PC or have the used prices dialed in low is ridiculous, because prices are skyrocketing on the supply side now and older used junk will follow suit. Meanwhile, the value proposition of this OEM box will likely only increase if the used market goes up.
Shortages of cutting edge new GPU product will lead to the older GPUs going up, whether or not they meet high end tasks, or low end ones directly competitive with weaker Steam Machine.
It's almost ridiculous, trying to follow what you are even arguing.
Regarding the Steam Machine performance, I also believe you know secondary school math, so I will also skip the slip. a 50% more powerful means the Steam Machine would be 150% as powerful of the total power of a Series S. You're saying 200% and 50%.
I said: "We haven't benchmarked the Machine yet, but it's expected to be double (200%, not 50%) the Series S."
I'm saying double. A 100% increase over the Series S, but we don't know because there are no comparative benchmarks.
If the 50% would mean the same as I said, then the 200% would mean the Steam Machine is 100+200% = 300% the power of a Series S, so you would be saying that the Steam Machine would be even slightly faster than a Series X, and it's been confirmed it won't, specially when it's not a device dedicated for gaming like a console.
And no, it won't smoke the CPU of any of the consoles because simply put, you're saying that a CPU of 6c/12t would be way faster than a CPU two generations older having 8c/16t. Sure, it will be faster in single core operations, but not that much faster in multicore stuff. We are talking abour maybe 15-20% faster, nothing more.
I said: "It will likely smoke the xbox in cpu bound applications."
Implying that games heavily cpu bound will outperform the xbox, which in itself is arguably pointless considering any heavily cpu bound applications on the xbox console are knee-capped behind the dev mode hypervisor.
Anyways, saying that it will emulate everything is nonsense, comparing a console to a PC for emulation is a win-win for a PC, there is not even competition on that. Like saying PS5 can play 99% of PS4 catalogue without issues and Steam Machine can not. What statement is that? simply out of place.
For emulation we have infinite devices, even those second hand PC's for less than 300€ that would run Bloodborne without a hiccup. You want to include everything in the Steam Machine because it is a PC? sure, you're totally right, but let's not see it as a Steam Machine, let's see it as a PC with a custom case.
I have used an Xbox Series S almost exclusively for emulation in both retail and developer mode.
People literally buy that SKU just to fuck around and play Gamecube or PS2 games. It can run 360 emulation, but not well.
Me saying: "Emulation on the gabecube will likely be running 360, PS3, Switch, and PS4 games like Bloodborne." is not out of place. It's just stating what it will be capable of. We are comparing the value proposition here. This thing is a PC, a more powerful Steam Deck, and it will smoke the consoles in CPU bound applications, like some emulators.
It's like you're trying to be ultra defensive about this thing potentially being called a console by somebody.
And finally, my point is: what is your strategy on selling a product you know it will sell differently depending on your approach. You prefer to get more profit for unit, price it higher and think only in a smaller part of the population who think like sheeps and understand videogames as something that needs to have ultra detailed graphics, or you're going to get less per unit, do a machine for reaching a wider audience because you have a big demand on 1080p market where the user prioritises stability over detail? at the end, it is something Sony will create and has to be a different approach to what PS5 is, because we are not in 2020 anymore, and many people is moving to PC because they are tired of what consoles have become, plus they can play almost everything on it.
You're saying consoles are too expensive and need to be cheaper. You're also saying this PC won't sell at 600 because it's too much, despite that literally translating to 400 dollars in 2006 where PS3 and 360 blew the doors off everything, and created the contemporary online networked gaming platform paradigm shift.
Yes. People have jumped ship to PC, but next-gen Playstation, Nintendo and Xbox will still sell tens of millions of next gen consoles even if priced at 600+, and this PC console hybrid probably won't reach that far.
Valve probably won't market this thing in stores. We don't know the price, and your point is completely obfuscated by your communication.
You or anyone can say whatever they want, but Xbox Series S was a brilliant idea coming from the wrong company.
OK?
