{"id":79063,"date":"2015-07-27T01:30:41","date_gmt":"2015-07-26T23:30:41","guid":{"rendered":"http:\/\/mygaming.co.za\/news\/?p=79063"},"modified":"2015-07-27T01:37:39","modified_gmt":"2015-07-26T23:37:39","slug":"nvidia-pascal-cards-sport-17-billion-transistors-and-32gb-of-hbm2-vram","status":"publish","type":"post","link":"https:\/\/mygaming.co.za\/news\/hardware\/79063-nvidia-pascal-cards-sport-17-billion-transistors-and-32gb-of-hbm2-vram","title":{"rendered":"NVIDIA Pascal cards sport 17 billion transistors and 32GB of HBM2 VRAM"},"content":{"rendered":"

If you combine the transistor count of NVIDIA’s GeForce GTX Titan X<\/a> (GM200<\/a>) and AMD’s Radeon R9 Fury X<\/a> (Fiji XT<\/a>), that’s 8 billion and 8.9 billion transistors respectively, you’d get a total 16.9 billion transistors.<\/p>\n

Even then, you still don’t quite have the absurd 17 billion transistors that NVIDIA’s Maxell 2 successor, Pascal, packs on its silicon. That’s according to a report by Fuzilla<\/a>.<\/p>\n

Going from a 28nm process, as used in the development of Maxwell 2, to a 16nm FinFET process has done wonders for the GPU and the sort of density NVIDIA has been able to accomplish.<\/p>\n

As a result, Pascal and its GP100 beastly compute orientated microarchitecture will be used in the creation of new cards for NVIDIA’s GeForce<\/a>, Quadro<\/a> and even Tesla<\/a> ranges. The Quadro and Tesla being their industrial-grade workhorse GPUs.<\/p>\n

\"NVIDIA<\/a>

NVIDIA G100 Pascal GPU Module<\/p><\/div>\n

It’s all well and good increasing the compute performance of your GPU multiple times over, but if the GPU cannot deliver that information via the necessary bandwidth, then it’s really a rather pointless endeavour.<\/p>\n

To accommodate the massive transistor count, Pascal is going to pack a lot of VRAM, and we mean a lot.<\/p>\n

Thanks to high-bandwidth memory 2 (HBM2<\/a>), the 4GB limitation of HBM1 (as seen on the AMD’s Fury range), is no longer an issue.<\/p>\n

In fact, the increased density of HBM2 allows NVIDIA to stack enough VRAM on the GP100 that we’re expecting between 16GB and 32GB on their top-end cards.<\/p>\n

Keep in mind that as large a pool of memory as that is, it’s also connected to a 4096-bit bus and a heck of a lot of bandwidth. According to WCCF Tech<\/a>:<\/p>\n

With 8Gb per DRAM die and 2 Gbps speed per pin, we get approximately 256 GB\/s bandwidth per HBM2 stack. With four stacks in total, we will get 1 TB\/s bandwidth<\/strong> on NVIDIA\u2019s GP100 flagship Pascal which is twice compared to the 512 GB\/s on AMD\u2019s Fiji cards and three times that of the 980 Ti\u2019s 334GB\/s.”<\/p><\/blockquote>\n

A 1 TB\/s bandwidth is nothing to scoff at. It should eat 4K and even 8K resolutions for breakfast.<\/p>\n

\"The<\/a>

The performance benefits of Pascal over Maxwell 2.<\/p><\/div>\n

The GP100 Pascal cards will also come equipped with NVLink, a new interconnect technology we describe in our Pascal piece last week<\/a>.<\/p>\n

It essentially offers 5 to 12x faster transfer speeds of vanilla PCIE 3.0. It only makes sense to improve the throughput of the GPU with the rest of the PC.<\/p>\n

It makes little to no sense to create this graphical beast and then have it unable to send its visual and computational concoctions to the CPU and RAM at a sufficient speed to prevent severe bottlenecking.<\/p>\n

And if this is what Pascal is capable, imagine what NVIDIA is cooking up for Pascal’s successor, Volta.<\/p>\n

Good luck AMD, that’s all we have to say. You may have dibs when it comes to HBM2 chips, but you’d better make one hell of a GPU to boot.<\/p>\n

\n

Source: Fudzilla<\/a><\/strong><\/h4>\n

In other hardware news<\/strong><\/h3>\n

GeForce GTX 950\u2019s specifications confirmed: launching August 17th<\/a><\/p>\n

Radeon R9 Nano crushes the R9 290X in Unigine Heaven benchmark<\/a><\/p>\n

i7-6700K reaches 6.5 GHz with liquid nitrogen: crushes i7-4790K in synthetic benchmarks<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"

And don’t forget an incredible 1 TB\/s bandwidth.<\/p>\n","protected":false},"author":221,"featured_media":79067,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_sma_x_autopost_status":"idle","_sma_x_autopost_error":"","_sma_x_post_id":"","_sma_facebook_post_id":"","_sma_instagram_post_id":"","_sma_x_attempts":0,"footnotes":""},"categories":[7],"tags":[14943,2420,14877,47,1771,1085,14879,1969,45,14875,14560],"class_list":["post-79063","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-hardware","tag-17-billion-transistors","tag-geforce","tag-gp100","tag-gpu","tag-graphics-cards","tag-hardware-2","tag-hbm2","tag-news-2","tag-nvidia","tag-pascal","tag-specifications"],"_links":{"self":[{"href":"https:\/\/mygaming.co.za\/news\/wp-json\/wp\/v2\/posts\/79063","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mygaming.co.za\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mygaming.co.za\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mygaming.co.za\/news\/wp-json\/wp\/v2\/users\/221"}],"replies":[{"embeddable":true,"href":"https:\/\/mygaming.co.za\/news\/wp-json\/wp\/v2\/comments?post=79063"}],"version-history":[{"count":1,"href":"https:\/\/mygaming.co.za\/news\/wp-json\/wp\/v2\/posts\/79063\/revisions"}],"predecessor-version":[{"id":79075,"href":"https:\/\/mygaming.co.za\/news\/wp-json\/wp\/v2\/posts\/79063\/revisions\/79075"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/mygaming.co.za\/news\/wp-json\/wp\/v2\/media\/79067"}],"wp:attachment":[{"href":"https:\/\/mygaming.co.za\/news\/wp-json\/wp\/v2\/media?parent=79063"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mygaming.co.za\/news\/wp-json\/wp\/v2\/categories?post=79063"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mygaming.co.za\/news\/wp-json\/wp\/v2\/tags?post=79063"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}