The G92 Architecture Rumours & Speculation Thread
- Thread starter Arun
- Start date
- Status
Well, I haven't seen one personally, and it's not inside my 60 day rule yet (to my knowledge), but the tea leaves and the whispers in the wind all point in that direction. And without another high-end chip in sight, seems fairly logical to me. I can tell you that around the time of G80 release we asked about a GX2, and we're told then that there was unlikely to be one for that generation, but it'd probably make a comeback the next. G92 is next.
PConline measured in their Review of 8800 GT a die-size of 330mm².
Put a dual-slot cooler on it, rise VGPU to 1.25-1.3V and clock it up to ~800/2400MHz and buy some GDDR4 at Samsung with 1.2GHz+.
Summarized you will get a solution, which can beat 2xRV670 in most cases and will have a higher margin than G80-SKUs and have the advantage of 384Bit in bandwidth and 768MB in memory-size, which is optimal for enthusiast-gaming at the moment and upcoming games, till the next high-end SKU comes.
If NV will really goes GX2 again they must be insane or there is a strong reason, that does not allows higher clocks, like above...
Why is G80@65nm = G92 not a high-end-chip?
Put a dual-slot cooler on it, rise VGPU to 1.25-1.3V and clock it up to ~800/2400MHz and buy some GDDR4 at Samsung with 1.2GHz+.
Summarized you will get a solution, which can beat 2xRV670 in most cases and will have a higher margin than G80-SKUs and have the advantage of 384Bit in bandwidth and 768MB in memory-size, which is optimal for enthusiast-gaming at the moment and upcoming games, till the next high-end SKU comes.
If NV will really goes GX2 again they must be insane or there is a strong reason, that does not allows higher clocks, like above...
Last edited by a moderator:
Maybe he expected "A Perfect 10"?
Cherry-picked benchmarks are one thing; flat-out lies about competing products are another. The latter is what Charlie thinks Nvidia is guilty of: that's what's got him so riled.
Personally, i think both G92 and RV670 have a perfect feature set/performance balance for the market segment they're aimed at.
Then why does he stay quiet about it whenever it's AMD/ATI doing it ?
If you read his posts (i wouldn't call them "stories") there's always a sense of almost personal grudge against Nvidia, for reasons i can't quite ascertain.
I just love how he gets so riled up over NV using AA when benchmarking a high end GPU. Oh the humanity!
Pre-G80, did Nvidia stress on turning on AF all the time on their high end GPUs?
Jawed
Legend
Blimey - I can't make out if the calipers are measuring the die or the die+sealant. But if it's really 330mm2 then 289mm2 was an entertaining diversionPConline measured in their Review of 8800 GT a die-size of 330mm².
and makes the comparison with RV670's 194mm2 even more fun.That memory would be too slow to achieve 8800GTX performance (though it'll be interesting to see how close). And there's still the question of whether the chip can support GDDR4.
I think you're relying too much upon it being 384-bit.
Jawed
No, Charlie has been an NV hater for over 1 year now and to date has never reported 1 bad thing about ATI.
First you say they needed to upgrade the specs to compete, now you are implying they downgraded the specs for yields... make up your mind. Furthermore, unless you can cite the MSRP of the GT's competition, you're in no position to determine if anything is or is not getting "blown out of the the water."CJ said:
Last edited by a moderator:
He caught his wife having a threesome with Nvidia and Vista.
On the zoom-pictures it looks like they only measured the die.
Why should NV give this advantage of their architecture up?
- Status
Similar threads
