qcmadness

http://www.anandtech.com/show/10 ... force-2016-liveblog

引用:

09:40PM EDT - 1080 is faster than 980 SLI
09:44PM EDT - Back of the envelope calculation says that GTX 1080 should be around 25% faster than GTX 980 Ti (SLi), using NVIDIA's numbers

09:44PM EDT - Though for NVIDIA, they're going to want to focus on getting GTX 700 series users to upgrade

qcmadness

引用:

10:17PM EDT - The new king: GTX 1080
10:18PM EDT - $599
10:19PM EDT - $699 founder's edition
10:19PM EDT - Available May 27th

10:20PM EDT - GTX 1070, $379, June 10th

dom

Wow 早過 A 記舖貨

qcmadness

引用:

原帖由 dom 於 2016-5-7 11:50 發表
Wow 早過 A 記舖貨

You should use NVIDIA for the next 10 years

Puff

1.6-1.7 GHz

qcmadness

引用:

原帖由 Puff 於 2016-5-7 12:29 發表
1.6-1.7 GHz

即係粒chip既規模比想像中細

dom

引用:

原帖由 qcmadness 於 2016/5/7 11:53 發表

You should use NVIDIA for the next 10 years

GTX1070

Puff

引用:

原帖由 qcmadness 於 2016-5-7 12:32 發表

即係粒chip既規模比想像中細

可能大過 23x Pol 10


希望 Pol 10 perf/watt 好睇
無論嚟緊六月呢水有冇 GDDR5x 都好

qcmadness

引用:

原帖由 Puff 於 2016-5-7 15:30 發表

可能大過 23x Pol 10


希望 Pol 10 perf/watt 好睇
無論嚟緊六月呢水有冇 GDDR5x 都好

2xx?
你會唔會太樂觀

我估計GP100最小400sqmm.

kevinyuri

引用:

原帖由 Puff 於 2016-5-7 12:29 PM 發表
1.6-1.7 GHz

turbo max 2.1ghz

Puff

引用:

原帖由 qcmadness 於 2016-5-7 15:39 發表

2xx?
你會唔會太樂觀

我估計GP100最小400sqmm.

呢粒邊喺 GP100 呀大佬

https://devblogs.nvidia.com/parallelforall/inside-pascal/
Tesla P100 3584  (not fully enabled) 610mm2
GTX 1080 得 2560 仲要 99.9% 冇 1/2 DPFP
GM204 398 mm2 28nm 2048.
當 60% area shrink 衰極都 300 左右咋化


[ 本帖最後由 Puff 於 2016-5-7 18:37 編輯 ]

qcmadness

引用:

原帖由 Puff 於 2016-5-7 18:29 發表

呢粒邊喺 GP100 呀大佬

https://devblogs.nvidia.com/parallelforall/inside-pascal/
Tesla P100 3584  (not fully enabled) 610mm2
GTX 1080 得 2560 仲要 99.9% 冇 1/2 DPFP
GM204 398 mm2 28nm 2048.
...

Memory pad無得縮

dom

GTX 1070 好期待

Puff

GP104 314 mm2
同 Pol 10 有段距離