Nvidia G92 is GeForce 8700 GTS（VR-Zone）
根據這篇的說法，單晶片的時候是65nm、256bit、512MB GDDR3、8層PCB。

Salty specs of G92 and G98 hit the web（TechConnect Magazine）
G98與G86一樣規模為1TPC、封裝為64bit；另外，G84 65nm版本維持編號。

現在的問題就是G92單晶片本身的規模了，這關係到G92 MCM(or single die dual-core)的成本問題….聽說如果G80還撐得住，高階有可能繼續讓G80挑大梁。

[EDIT]
由於有7TPC的規格出現，看來G92似乎與G80一樣都是8TPC的規模？

另外，die size方面大約知道與G71類似，所以NVIDIA目前看來打算重複7950GX2的產品策略，也就是沒有MCM而是雙晶片+雙PCB。

這樣最大的缺點自然是難以避免地會有SLI失效時單卡性能下降的問題….優點則是沿用設計設計所帶來的成本改善；為了彌補這點，會不會有類似現有G80高階定位的D8p單晶片產品(256bit以上)則是另一個問題。

http://gametomorrow.com/blog/index.php/2007/09/05/cell-vs-g80/
well，IBM又拿他們的iRT來說嘴了XD
能不能拿些別的應用出來….XD

這篇是引用薩爾大學的 Philipp Slusallek教授拿G80作ray-tracing的論文，Stackless KD-Tree Traversal for High Performance GPU Ray Tracing的數據，與CELL跑iRT的比較。非常有趣的是，iRT的作者正好是 Philipp Slusallek的學生。

這完全就是重複當初RacingPHT兄自己做的Ray-Tracing demo on G80的狀況，只是當初是預估，這次是實際跑完。

從左至右：

2.6 GHz AMD Opteron – Saarland Ray-tracer
Nvidia GeForce 8800 GTX – Saarland Ray-tracer
Sony Playstation3 (partial 3.2 GHz Cell processor running Linux) – IBM iRT
3.2 GHz Cell Processor – IBM iRT
IBM QS20 Blade (Two 3.2 GHz Cell Processors) – IBM iRT

論文本身有提到因為Register footprint的關係，G80運作的效率並不高，有待CUDA本身的改善。

Although we believe a performance of over 16M rays/s to be already quite impressive for the CONFERENCE scene, we expected much higher performance: The G80 with its 128 scalar arithmetic units running at 1.3GHz should deliver over 160GFlops, meaning that tracing one ray costs about 10,000 cycles. We suspect the main bottleneck to be the large number of registers in the compiled code, which limits the occupancy of the GPU to less than 33%. Unfortunately, although the program requires much less registers, the CUDA compiler is not yet mature enough and cannot aid in reducing their count. An option would be to rewrite the whole CUDA code in PTX intermediate assembly.

16M rays/sec，考慮他們用的主要是scalar unit，至少應該有160GFLOPS，等於每個ray要跑10000cycles！不過即使是考慮33%的效率，仍與6SPE的CELL有相當的距離，更別提完整8SPE的CELL與內建雙CELL的QS20 blade，這應該與每個SPE都具備相對多數的register有關….然後IBM的人對Larrabee放話了。XD

iRT ray tracer已經於9/5開放下載。
http://gametomorrow.com/blog/index.php/2007/09/05/interactive-ray-tracer-irt-available-for-download/

下載位址為
http://www.alphaworks.ibm.com/tech/irt

內含有兩個demo scene，各為33萬3千個三角形的法拉利、以及六萬九千個三角形的史丹福兔寶寶。

在這兩個scene底下，PS3可以達到720p解析度下超過40fps。
PS3有6SPE、QS20有16個SPE之故，效能與SPE數量成線性關係，後者能達前者的2.5倍性能。

—-
說真的，來點別的應用吧。TGS07希望多點東西….
謠傳TGS07會有HOME for JP、firmware 2.0、新的線上AV服務、新定價策略與同捆包，甚至是FF7….

http://gametomorrow.com/blog/index.php/2007/09/05/cell-vs-g80/
well，IBM又拿他們的iRT來說嘴了XD
能不能拿些別的應用出來….XD

這篇是引用薩爾大學的 Philipp Slusallek教授拿G80作ray-tracing的論文，Stackless KD-Tree Traversal for High Performance GPU Ray Tracing的數據，與CELL跑iRT的比較。非常有趣的是，iRT的作者正好是 Philipp Slusallek的學生。

這完全就是重複當初RacingPHT兄自己做的Ray-Tracing demo on G80的狀況，只是當初是預估，這次是實際跑完。

從左至右：

2.6 GHz AMD Opteron – Saarland Ray-tracer
Nvidia GeForce 8800 GTX – Saarland Ray-tracer
Sony Playstation3 (partial 3.2 GHz Cell processor running Linux) – IBM iRT
3.2 GHz Cell Processor – IBM iRT
IBM QS20 Blade (Two 3.2 GHz Cell Processors) – IBM iRT

論文本身有提到因為Register footprint的關係，G80運作的效率並不高，有待CUDA本身的改善。

Although we believe a performance of over 16M rays/s to be already quite impressive for the CONFERENCE scene, we expected much higher performance: The G80 with its 128 scalar arithmetic units running at 1.3GHz should deliver over 160GFlops, meaning that tracing one ray costs about 10,000 cycles. We suspect the main bottleneck to be the large number of registers in the compiled code, which limits the occupancy of the GPU to less than 33%. Unfortunately, although the program requires much less registers, the CUDA compiler is not yet mature enough and cannot aid in reducing their count. An option would be to rewrite the whole CUDA code in PTX intermediate assembly.

不過即使是考慮33%的效率，仍與6SPE的CELL有相當的距離，更別提完整8SPE的CELL與內建雙CELL的QS20 blade，這應該與每個SPE都具備相對多數的register有關….然後IBM的人對Larrabee放話了。XD

—-
說真的，來點別的應用吧。TGS07希望多點東西….
謠傳有HOME for JP、firmware 2.0、新的線上AV服務、新定價策略與同捆包，甚至是FF7….

http://gametomorrow.com/blog/index.php/2007/09/05/cell-vs-g80/
well，IBM又拿他們的iRT來說嘴了XD
能不能拿些別的應用出來….XD

這篇是引用薩爾大學的 Philipp Slusallek教授拿G80作ray-tracing的論文，Stackless KD-Tree Traversal for High Performance GPU Ray Tracing的數據，與CELL跑iRT的比較。非常有趣的是，iRT的作者正好是 Philipp Slusallek的學生。

這完全就是重複當初RacingPHT兄自己做的Ray-Tracing demo on G80的狀況，只是當初是預估，這次是實際跑完。

從左至右：

2.6 GHz AMD Opteron – Saarland Ray-tracer
Nvidia GeForce 8800 GTX – Saarland Ray-tracer
Sony Playstation3 (partial 3.2 GHz Cell processor running Linux) – IBM iRT
3.2 GHz Cell Processor – IBM iRT
IBM QS20 Blade (Two 3.2 GHz Cell Processors) – IBM iRT

論文本身有提到因為Register footprint的關係，G80運作的效率並不高，有待CUDA本身的改善。

Although we believe a performance of over 16M rays/s to be already quite impressive for the CONFERENCE scene, we expected much higher performance: The G80 with its 128 scalar arithmetic units running at 1.3GHz should deliver over 160GFlops, meaning that tracing one ray costs about 10,000 cycles. We suspect the main bottleneck to be the large number of registers in the compiled code, which limits the occupancy of the GPU to less than 33%. Unfortunately, although the program requires much less registers, the CUDA compiler is not yet mature enough and cannot aid in reducing their count. An option would be to rewrite the whole CUDA code in PTX intermediate assembly.

不過即使是考慮33%的效率，仍與6SPE的CELL有相當的距離，這應該與CELL SPE都具備相對多數的register有關….然後IBM的人對Larrabee放話了。XD