I meant with that b3d is a more serious tech discussion forum then some others are, where insiders originate from.
Copy that.
I think it’s best to just slow down and to stop reiterating the points already made unless you have a stronger more nuanced perspective that may be missed.
there are other strategies to reasoning and logic that can be employed to showcase items; and its critical to do it in this way. You want redundancy without repetition. This is how you formulate arguments that are stronger without annoying people
for example can you make a strong argument without github leaks?
This process is straight forward; the first insider leaka indicates that PS5 and XSX difference was the difference between XBO and X1S. This difference is approximately 7% on the GPU side iiRc.
Since then all the insider leaks have been the same vein; no real leaks on actual spec numbers, just an emphasis of PS5 being more powerful by 7%+. By the logic we are using. Say up to 15 max. If we cast a wide net on leaks.
We still have no numbers at this time from the insiders mind you. Everyone is afraid to give out the official spec number they are looking at. But w/e.
somehow people believe this relative number to leak is critical but it’s not.
Because we are able to strongly determine the power footprint of XSX without the need of github leaks.
We are told it is twice as powerful. We have determined it’s size of 405mm^2. These 2 days points corroborate a power profile of close to approximately 12.
so by large PS5 should be between 12.84 to 13.8 TF by leakers accounts.
still have yet a credible leaker who said earlier that PS5 was more powerful by 7% to say the obvious Yes XSX is 12 TF.
They already told us the relative difference. Anyone here can do math. X*107% = Y
Pretty simple. Yet we don’t. Because if the answer to that formula is not the same as the spec sheet they are looking at; that XSX cannot be 12.
that’s where the Github leaks becomes a problem. It corroborated the size of the chip with the large number of CUs before we saw the size of the chip.
We know with relative confidence intervals that XSX will be ~12