C++博客-Der See der Vergessens-最新评论http://www.cppblog.com/Jedimaster/CommentsRSS.aspx 忘却之湖zh-cnWed, 28 Feb 2007 15:05:03 GMTWed, 28 Feb 2007 15:05:03 GMTcnblogsre: veryCD让中国骡丢脸 流氓软件扬名国外http://www.cppblog.com/Jedimaster/archive/2008/02/03/11322.html#42417拖拉拖拉Sat, 02 Feb 2008 19:04:00 GMThttp://www.cppblog.com/Jedimaster/archive/2008/02/03/11322.html#42417我没发现VC版的eMule有什么吸血或滥发的行为。这个消息里说的是Vagga，它假冒VC的版本名称的。

拖拉 2008-02-03 03:04 发表评论
]]>re: 找工作真难http://www.cppblog.com/Jedimaster/archive/2008/01/30/10969.html#42225天使天使Wed, 30 Jan 2008 07:28:00 GMThttp://www.cppblog.com/Jedimaster/archive/2008/01/30/10969.html#42225http://hi.baidu.com/miao8888

天使 2008-01-30 15:28 发表评论
]]>re: World Of Warcraft Server Source Topichttp://www.cppblog.com/Jedimaster/archive/2007/09/21/13373.html#32633游荡者游荡者Fri, 21 Sep 2007 09:19:00 GMThttp://www.cppblog.com/Jedimaster/archive/2007/09/21/13373.html#32633

游荡者 2007-09-21 17:19 发表评论
]]>re: World Of Warcraft Server Source Topichttp://www.cppblog.com/Jedimaster/archive/2007/09/21/13373.html#32634游荡者游荡者Fri, 21 Sep 2007 09:19:00 GMThttp://www.cppblog.com/Jedimaster/archive/2007/09/21/13373.html#32634

游荡者 2007-09-21 17:19 发表评论
]]>re: 利用SAH实现kD树快速分割模型实践http://www.cppblog.com/Jedimaster/archive/2007/09/03/18812.html#31440freeskyfreeskyMon, 03 Sep 2007 01:14:00 GMThttp://www.cppblog.com/Jedimaster/archive/2007/09/03/18812.html#31440

freesky 2007-09-03 09:14 发表评论
]]>re: Pure GPU Computing Platform : NVIDIA CUDA Tutorialhttp://www.cppblog.com/Jedimaster/archive/2007/08/23/18939.html#30674DimitrisDimitrisThu, 23 Aug 2007 07:18:00 GMThttp://www.cppblog.com/Jedimaster/archive/2007/08/23/18939.html#30674As long as accuracy is concerned, cuda complies with the error margin specifications (except for doubles i believe). There is are alot of SDK examples on their site with reference execution comparison, only once it failed to pass but the accuracy threshold was pretty tight! I'm using it now on computational physics thesis.

For starters i just ported a simple random number generator to the gpu, the results were identical to the digits i cared about (6th crucial). Mind though that i had to convert it to float!
Double should be working on CUDA 1.0 i think, but they didnt work for me, it wouldnt assign the value at all when double.
I didnt even bother parallelizing it, just a 1x1 block and It was WAY OVER a magnitude faster. I'm impressed. All that with a 8600GTS.

Dimitris 2007-08-23 15:18 发表评论
]]>re: Pure GPU Computing Platform : NVIDIA CUDA Tutorialhttp://www.cppblog.com/Jedimaster/archive/2007/08/15/18939.html#30101sdfsdfsdfsdfsdfsdfWed, 15 Aug 2007 12:02:00 GMThttp://www.cppblog.com/Jedimaster/archive/2007/08/15/18939.html#30101

sdfsdfsdf 2007-08-15 20:02 发表评论
]]>re: GPU还可以做什么 —— Brook for GPUs,Stream Computing On GPUshttp://www.cppblog.com/Jedimaster/archive/2007/08/09/13689.html#29654m17m17Thu, 09 Aug 2007 10:06:00 GMThttp://www.cppblog.com/Jedimaster/archive/2007/08/09/13689.html#29654
嘿嘿，对笔者的个人发展方向很感兴趣哦，公司最近在召各方面的三维人才哦，不知道你有没有兴趣哦

m17 2007-08-09 18:06 发表评论
]]>re: veryCD让中国骡丢脸 流氓软件扬名国外http://www.cppblog.com/Jedimaster/archive/2007/06/05/11322.html#25562liliTue, 05 Jun 2007 05:08:00 GMThttp://www.cppblog.com/Jedimaster/archive/2007/06/05/11322.html#25562

li 2007-06-05 13:08 发表评论
]]>re: Pure GPU Computing Platform : NVIDIA CUDA Tutorialhttp://www.cppblog.com/Jedimaster/archive/2007/02/26/18939.html#19005JedimasterJedimasterMon, 26 Feb 2007 11:46:00 GMThttp://www.cppblog.com/Jedimaster/archive/2007/02/26/18939.html#19005前几天去了下德国BOINC论坛，把这个想法和德国人交流了一下，引用了一篇回复。

Zitat:
Zitat von Jedimaster Beitrag anzeigen
NVIDIA has released the CUDA, a parrallel library use NVIDIA GPU.
I also got interested in it but unfortunately it seems to work solely with GeForce 8 series.

Zitat:
if we can supply client program use GPU & CPU, maybe we can highly improve our speed.
Sure we could but when some projects released their applications as open source some people started to recompile them optimized (making use of MMX, SSE, all kinds of technologies the original binaries still lack). This had mainly 2 effects:

1. People started using "optimized" core-clients (manipulated to demand a multiple of credits since they are calculated by CPU time which has been decreased by optimizations). From my point of view these people just did not understand the credit system although demanding more credits for completing 2 WUs in the time of one unoptimized may seem reasonable.

2. and more dramatic: Some projects noticed a large discrepancy in the returned results. I think it was Einstein@Home that first asked their users not to use optimized clients. That caused problems at validation when erroneous results should have been sorted out. I don't know what the accuracy of GPU-based calculations is.

Zitat:
Sorry for my poor German, entschuldigung.
Shouldn't matter too much for most users here.


Kurz nochmal auf Deutsch: Jedimaster schlägt vor, daß man die kürzlich von NVIDIA freigegebene Bibliothek CUDA für GPU-basierte Berechnungen von z.B. Fouriertransformationen etc. benutzen könnte um die Anwendungen drastisch zu optimieren. Ich habe daraufhin geantwortet, daß CUDA nur mit der GeForce 8-Reihe zusammenarbeitet und derartige Berechnungen wie einst die MMX/SSE-Optimierungen Einbrüche bei der Genauigkeit zur Folge haben könnten, die zu denselben Problemen wie schon bei Einstein im Validator-Prozess führen könnten. Darüberhinaus würden wieder mehr "optimierte" Core-Clients zum Einsatz kommen.

德国人认为把CUDA等等技术用于分布式计算，在数值的有效性上还欠妥，甚至连专门为CPU优化的程序都不推荐实用。结合目前G8系列还没有上市，而且相关的中间件还不够成熟，可能开发相应的程序还不是很现实。

为了应付游戏还是有些浪费，这边物理卡刚出来，Havok就又来凑热闹了，SM2人们用的正顺手，SM3早就到来了。还是把优化工作做好先。

Jedimaster 2007-02-26 19:46 发表评论
]]>