http://www.itmedia.co.jp/enterprise/articles/0905/29/news041.html

GPUコンピューティング戦争に勝機を見いだすNVIDIA

http://pc.watch.impress.co.jp/docs/column/kaigai/20091023_323529.html

NVIDIAが目指す究極のプロセッサへと進む「Fermi」アーキテクチャ

　「以前のG80アーキテクチャは、コンピューティングにも優れたグラフィックスプロセッサだった。しかし、Fermiはグラフィックスにも優れた並列コンピューティングプロセッサだ。設計思想はFermiで変わった」とNVIDIAのTony Tamasi(トニー・タマシ)氏(Senior Vice President, Content and Technology, NVIDIA)は語る。

G80是「兼具通用運算的GPU」、但是Fermi是「兼具繪圖的平行處理器」，還真的講了….XD

「G80の構造は、伝統的なGPUからの大きな革新だったが、まだグラフィックスアーキテクチャを引きずっている面もある。

例えば、G80のプロセッサクラスタをよく見ると、NV40の再構成的な部分も見えてくる。

演算ユニットの構成だけを見ると、NV40の2つのプロセッサを解体して、演算ユニット部分を1個のSMへと編成したのがG80だ。」

GT200出來之前的確大家都會這樣想….NV40/G70的單一個pixel shader重組之後變成G80，單元數量規模沒有顯著變大，但是指令編排之類的東西就大幅改變了，吞吐性能大幅改善則是因為ROP。只是單單這樣看很難聯想到規模一口氣大了2.5倍，雖然CPU也有一個northwood大幅膨脹成Prescott的例子。

—-

http://nvidia.fullviewmedia.com/GPU2009/1001-belvedere-1001.html

From Brook to CUDA

http://nvidia.fullviewmedia.com/GPU2009/1002-crystal-1070.html

Face Recognition for Photographs and Video

http://developer.nvidia.com/object/nexus_features.html

NVIDIA Nexus – Features

http://www.techcrunch.com/2009/10/21/how-the-iphone-is-blowing-everyone-else-away-in-charts/

How The iPhone Is Blowing Everyone Else Away (In Charts)

http://jp.techcrunch.com/archives/20091021how-the-iphone-is-blowing-everyone-else-away-in-charts/

iPhoneの一人勝ち、グラフで如実に―モルガン・スタンレーの報告、続編

不過 「Apple Unshacked Mobile Developers from Carrier Walled Gardens」這句話就爆笑了，因為後面接 「Leveraged 100MM+ iTunes Users Trained to Purchase Digital Goods Apple’s way」…. XDa

解放之後自行佔領，這大概是大部分的掠奪者的常識….於是我可不可以說Android快點站起來XD

http://www.itmedia.co.jp/enterprise/articles/0910/21/news012.html

VerizonのAndroid採用でAppleが神経をとがらせる10の理由

說不定這兩家的不同就差在遇到正面挑戰的反擊style….

http://plusd.itmedia.co.jp/pcuser/articles/0910/21/news025.html

アップル、世界初をうたうマルチタッチ対応マウス「Magic Mouse」

http://pc.watch.impress.co.jp/docs/news/20091021_323201.html

アップル、ボタンレスのマルチタッチマウス「Magic Mouse」

記得前幾禮拜M$才在講說研究中，其實Apple早就搞定了嗎…

當然他們本來就比較早出Multi-touch device說起來跑比較快是理所當然。

這讓我想到Project Natal，不管東西到底怎樣，發表會只是搞出來出來打空氣的。

其實可能要能動根本就是下一代主機的事情….結果SONY傻傻地拿辛苦研究的東西出來奉陪，Apple乾脆就直接產品化還外加把專利也拿了。

http://pc.watch.impress.co.jp/docs/news/yajiuma/20091007_320067.html

【やじうまPC Watch】Microsoft、マルチタッチに対応した「Mouse 2.0」を開発中

—-

http://play60gamer.wordpress.com/2009/10/21/ffxiii-on-the-ps3-to-have-no-install/

FFXIII on the PS3 to have no Install

http://uk.ps3.ign.com/articles/103/1037157p1.html

Final Fantasy XIII Update

「

Toriyama also shared some technical details on the game. Save data will take up 500KB. Trophy data takes up 3,080KB (yes, it looks like the game supports Trophies!). In case you were stocking up munitions for a big system wars battle, there won’t be a hard disk installation.

Worried about load times? Don’t be. The only time you’ll see load times is when you transition from town to field. Once you’re out on the field, the game is seamless. Toriyama explained that the game’s tech staff tackled loading early on in order to keep the experience as smooth as possible.

」

哇哩….莫名其妙….居然跑出這種莫名的自信….

然後反正最後一定會遇上360把一堆片子全塞進硬碟之後和PS3無install來對比…..

http://www.brightsideofnews.com/news/2009/10/16/exclusive-sony-psp2-arriving-in-20102c-features-powervr-sgx543-graphics.aspx

Exclusive: Sony PSP2 arriving in 2010, features PowerVR SGX543 graphics

雖說PC上的PowerVR SGX535(GMA500)拜Intel對自家driver莫名的信心所致性能有封印的疑問，不可小覷；不過PowerVR的fillrate基本上是等效數據不是實際的fillrate….所以SGX543MP16要說可以和GTX260-216拼有點早了。不過400MHz的SGX543MP4應該真的表現會不錯….

PowerVR series5XT(SGX)的MPcore家族在65nm底下即使是16core都不會很大，搭一搭外部的輔助單元，看起來其實desktop市場也頗有競爭力（thin client限定就是了）….而且這個小core還有一個好處是即使和GS放在一起似乎也不會有很大的負擔，畢竟即使SGX有DX10.1等級的繪圖功能可用，PSP2要是沒辦法上市第一天就完整相容PS2和PSP的軟體的話那一定會被罵到臭頭….

當初才讓人嘆息DC的PowerVR2和PS2的EE搭起來可能當時會多棒多棒，現在SONY似乎真的要用PowerVR….世界總是這麼奇妙。

http://www.evga.com/forums/tm.asp?m=100973314&mpage=1&key=&#100973314

Fermi’s folding ability…

「

Some additional explanations:

I was forced to use usb monitor as GPUs haven’t any video output (this engineering samples of Fermi are Tesla like, but they have 1.5GB of memory each like GT300 will).

Because of the new MIMD architecture (they have 32 clusters of 16 shaders) i was not able to load them at 100% in any other way but to launch 1 F@H client per cluster and per card. Every client is GPU3 core Beta (Open MM library). I supose it is much more efficient then previous GPU2. In addition they need very little memory to run. Having 16GB of DDR3 and using Windows 7 Enterprise I’ve managed to run 200 instances of F@H GPU and 4 CPU (i7 processor HT off). The 7th card is not fully loaded. This could also be an issue with EVGA X58 mobo.

I use together two Silverstone Strider PSU’s 1500W each that is probably too much but

now I experiment with overclocking (cards are factory unlocked). Max power consumption

I’ve noticed was 2400W.

The whole system is cooled by my own construction of liquid CO2 which is heavy an inconvenient and I have to supply a new cylinder every 5 days.

That’s It. I wish everybody to improve your folding speed using GT300 soon!

Just keep folding…

」

也就是說，Fermi在folding上大概有GT200的四倍快，大過sp與記憶體頻寬增加的幅度。因為GPU3 core beta目前沒有針對Fermi做過修改，所以應該是得益於新的cache subsystem？

遺憾的是這和繪圖部份性能沒什麼關係….XDa

補充：

想想上面忘記一件事情，因為FAH實際上是用倍精度在跑，所以GT200和Fermi的合理差距是8倍。(GT200：1TFLOPS/8 = 125GFLOPS、Fermi：2TFLOPS/2=1TFLOPS)

如果實際上只跑4倍的話，那代表瓶頸可能卡在別的地方。