{"id":364,"date":"2009-09-30T19:26:17","date_gmt":"2009-10-01T01:26:17","guid":{"rendered":"http:\/\/www.realtimerendering.com\/blog\/?p=364"},"modified":"2009-09-30T19:26:17","modified_gmt":"2009-10-01T01:26:17","slug":"nvidia-announces-fermi-architecture","status":"publish","type":"post","link":"https:\/\/www.realtimerendering.com\/blog\/nvidia-announces-fermi-architecture\/","title":{"rendered":"NVIDIA Announces Fermi Architecture"},"content":{"rendered":"<p>Today at the <a href=\"http:\/\/www.nvidia.com\/object\/gpu_technology_conference.html\">GPU Technology Conference<\/a> (the successor to last year&#8217;s <a href=\"http:\/\/www.nvidia.com\/content\/nvision2008\/index.html\">NVISION<\/a>), <a href=\"http:\/\/www.nvidia.com\/page\/home.html\">NVIDIA<\/a> announced Fermi, their new GPU architecture (exactly one week after <a href=\"http:\/\/www.amd.com\/us\/Pages\/AMDHomePage.aspx\">AMD<\/a> <a href=\"http:\/\/www.realtimerendering.com\/blog\/first-directx11-gpu-ships\/\">shipped the first GPU<\/a> from <em>their<\/em> new Radeon HD 5800 architecture).\u00a0 NVIDIA have published a Fermi <a href=\"http:\/\/www.nvidia.com\/content\/PDF\/fermi_white_papers\/NVIDIAFermiArchitectureWhitepaper.pdf\">white paper<\/a>, and writeups are popping up on the web.\u00a0 Of these, the ones from <a href=\"http:\/\/www.realworldtech.com\/page.cfm?ArticleID=RWT093009110932\">Real World Technologies<\/a> and <a href=\"http:\/\/www.anandtech.com\/video\/showdoc.aspx?i=3651\">AnandTech<\/a> seem most informative.<\/p>\n<p>With this announcement, NVIDIA is focusing firmly on the GPGPU market, rather than on graphics.\u00a0 No details of the graphics-specific parts of the chip (such as triangle rasterizers and texture units) were even mentioned.\u00a0 The chip looks like it will be significantly more expensive to manufacture than AMD&#8217;s chip, and at least some of that extra die area has been devoted to things which will not benefit most graphics applications (such as improved double-precision floating-point support and more general programming models).\u00a0 With full support for indirect branches, a unified address space, and fine-grained exception handling, Fermi is as general purpose as it gets.\u00a0 NVIDIA is even adding C++ support to CUDA (the first iterations of OpenCL and DirectCompute will likely not enable the most general programming models).<\/p>\n<p>Compared to their previous architecture, NVIDIA has shuffled around the allocation of ALUs, thread scheduling units, and other resources.\u00a0 To make sense of the soup of marketing terms such as &#8220;warps&#8221;, &#8220;cores&#8221;, and &#8220;SMs&#8221;,\u00a0 I again recommend <a href=\"http:\/\/s09.idav.ucdavis.edu\/talks\/02_kayvonf_gpuArchTalk09.pdf\">Kayvon Fatahalian&#8217;s SIGGRAPH 2009 presentation on GPU architecture<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Today at the GPU Technology Conference (the successor to last year&#8217;s NVISION), NVIDIA announced Fermi, their new GPU architecture (exactly one week after AMD shipped the first GPU from their new Radeon HD 5800 architecture).\u00a0 NVIDIA have published a Fermi white paper, and writeups are popping up on the web.\u00a0 Of these, the ones from [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[87,57,211,41],"class_list":["post-364","post","type-post","status-publish","format-standard","hentry","category-resources","tag-cuda","tag-gpgpu","tag-gpu","tag-nvidia"],"_links":{"self":[{"href":"https:\/\/www.realtimerendering.com\/blog\/wp-json\/wp\/v2\/posts\/364","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.realtimerendering.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.realtimerendering.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.realtimerendering.com\/blog\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.realtimerendering.com\/blog\/wp-json\/wp\/v2\/comments?post=364"}],"version-history":[{"count":7,"href":"https:\/\/www.realtimerendering.com\/blog\/wp-json\/wp\/v2\/posts\/364\/revisions"}],"predecessor-version":[{"id":371,"href":"https:\/\/www.realtimerendering.com\/blog\/wp-json\/wp\/v2\/posts\/364\/revisions\/371"}],"wp:attachment":[{"href":"https:\/\/www.realtimerendering.com\/blog\/wp-json\/wp\/v2\/media?parent=364"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.realtimerendering.com\/blog\/wp-json\/wp\/v2\/categories?post=364"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.realtimerendering.com\/blog\/wp-json\/wp\/v2\/tags?post=364"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}