No. I said that Tom's review is for ordinary SSD with name Optane. Which use nand flash chips and have same problems like any other SSD when moving really big files and fullfil it's "SLC" buffer.
Generally speaking, what my colleagues and I have in mind is a board with a controller and non-volatile memory chips, which, unlike the one in SSD, is not nand flash, despite the exterior similarity.
I don't know why all the fuss, since all PCIe 5.0 comes from the CPU i/o chiplet with ZEN 4, and from the fall with ZEN 5 architecture. So, at least in theory, it wouldn't be a problem for all motherboards with an AM5 slot to get such specifications on at least the near-CPU PCIe and m.2 slots...
It makes a lot of sense for an entirely new architecture. Which out of the box will support the next version of Microsoft DirectX that will arrive with the next version of Windows in 202*(?).
Ok, so we've written some stuff that may or may not be helpful. What do you think about the next AMD architecture that will be after rDNA 4. It is also mentioned in the article and there is an obvious fear in the comments to write anything other than trolling AMD. Can it really be a new...
No one has mentioned war in this discussion. Something is worth what it can be sold for. If it produces only slightly more rays than the rtx 5060(which will probably cost $400), consumes 50% more electricity, and comes in at $700, no one will even consider buying the rx 8800.
You should think of the rDNA 4 "flagship" as the rx 7800 xt on RT steroids. Given that despite the expected improvements in RT, it will likely struggle for the kind of performance that is in the space between the rtx 5060 and rtx 5070... The price would have to be really low to have any chance...