How do you come to this conclusion based on one, poor image of the die? Do you even know what die was shown?
Idle speculation is one thing but idle speculation followed by statements such as "And i said that it' does seem great but it obviously helps ATi with this Rxxx generation" are pointless.
-FUDie
omg Did you read thread subject
You dont read my posts, righ? Well I'll try for smarties like you to explain myself for the 3rd time.
Looking how many problems just 800SPs have in power distribution and patch they issue as cap-ring on RV790 i'd say all these are just cheap design, and for more building parts they need to leave their
taifun flood concept and rearrange the die a little bit. It's hellawa cheaper than a cap-ring on die
And these
cheap checkerboard solution will yet againg be just enough for their <1500 as it seems SP units.
I was stating that the grainy visual evidence indicated that the situation was not as simple as you claimed, and some evidence might potentially go against it.
I also stated that devoting such a large fraction of die area to cache, as your claim asserts, would be very unusual for a GPU.
How unusual? When RV770 also had
a huge cache just differently spread , and G200 has a huge cache in that
cross surrounded by 10 cores.
Possibly, but is this heat spread out evenly?
There will be two hot and two cold corners of the chip.
RV770 might be somewhat cooler on the edges, and in the one section with other logic, but why would this be less even than what you've claimed?
That's what i was trying to point out. There's
better and cheaper quick solution than cap-ring that they needed to patch R700 series. But somehow i don't believe that these
hot-spots as you call them would make any problem for them, cause they have one large
hot-spot in the center these days
And yes that's why i point out that nVidia's G200 seems like really greener solution from the design standpoint, but not perfect ofc. And that's why i mention Elpida's solution as a great one in it industry in recent years. Better performing , maybe just for memory blocks, than weird Prescott fragmented design. Yep Prescott has cache on one side but it was fragmented on die-shot unlike other
classic CPU cache designs.
And i said that it' does seem great but it obviously helps ATi with this Rxxx generation ("It's not a lot in fact but seems it works well for AMD")
I'm not clear what this part means.
Doesn't mean anything when you cut it this way
this Rxxx = RV870 based generation whatever they call it.
I don't think Nvidia had the time to change their design once RV770's specs were known to them.
The chips were released pretty close to one another, and GT200 had some amount of delay. The details for the design would have been set in stone for a long time, probably in the year prior to release. It might be more, the process takes 4-5 years these days.
Not such a long time ... maybe few month, but it's better for them to just add two more cores (uneven placement of building blocks on pretty symmetrical G200 die). Anyway who add some parts after whom .... it's just speculative but G200 shouldn't be 584mm2, but around 500mm2 (472mm2 iirc, initial web specs-ulations in jan-feb 2008). And yep it's harder to nVidia than ATi to add two more
processing cores, in my opinion too as i stated somewhat before, as they have lighter SP units.
