Printable View
JHH's GTC12 keynote begins
http://smooth-las-akam.istreamplanet...d1/player.html
http://blogs.nvidia.com/2012/05/live...kicks-off-gtc/
Note that this is the live stream: http://smooth-las-akam.istreamplanet...d1/player.html
IMO, VGX (a cloud-based gaming initiative) is BOUND to be announced.
GK110 confirmed:
"The NVIDIA Tesla K20 GPU is the new flagship of the Tesla GPU product family, designed for the most computationally intensive HPC environments. Expected to be the world's highest-performance, most energy-efficient GPU, the Tesla K20 is planned to be available in the fourth quarter of 2012.
The Tesla K20 is based on the GK110 Kepler GPU. This GPU delivers three times more double precision compared to Fermi architecture-based Tesla products and it supports the Hyper-Q and dynamic parallelism capabilities. The GK110 GPU is expected to be incorporated into the new Titan supercomputer at the Oak Ridge National Laboratory in Tennessee and the Blue Waters system at the National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign."
http://www.marketwire.com/press-rele...da-1657561.htm
If the Tesla version arrives in Q4,guessing the desktop version arrives with with Win 8 .
awesome link jjj thanks
http://www.abload.de/img/desktop_2012_05_15_214jdfy.png
largest and most powerful gpu we've ever built...
wow even bigger than GT200?
http://www.abload.de/img/desktop_2012_05_15_213ouyq.png
kepler is world's first gpu designed for the cloud, to be deployed into cloud data centers worlwide. it does this with:
--virtualized gpu
--no longer does it need to connect to a display, it can render and stream instantaneously right out of chip to a remote location
--super energy efficiency, so it can be deployed in a massive scale
A BIT more info: http://www.hardwarecanucks.com/news/...ng-in-q4-2012/
Plus, GeForce GRID info: http://www.hardwarecanucks.com/news/...-for-everyone/
Decisions, decisions.Quote:
Originally Posted by jjj
Wait for the GTX 780 which is looking more and more likely to be a spring 2013 release, or upgrade to three GTX 680 4GB + waterblocks when they finally become available? Or, maybe two GTX 690s with waterblocks?
I also can hardly resist my self from selling my two GTX580 and buy two GX670 :(
The only problem I see with this is finding a place that has a pair of 670'sQuote:
Originally Posted by koc
Two GTX 690 with waterblocks :DQuote:
Originally Posted by UrbanSmooth
A 30 second search found 4x Galaxy GTX 670's and 2x Zotac GTX 670's... http://www.microcenter.com/search/se...x=0&submit.y=0Quote:
Originally Posted by THX1138
You're still assuming NVIDIA is going to release a compute oriented consumer product when they have some very compelling reasons not to.Quote:
Originally Posted by UrbanSmooth
They make more money on smaller chips, and the consumer market just told them that it doesn't place high priority on compute. (we've seen STEAM stats, Skymtl, and neoseeker saying that GTX680s are outselling 7970s by a great margin, and if the market wanted compute, the 7970 would be selling better as it leads the 680 in this area)
I wouldn't be shocked if NVIDIA split their product lines going forward. I wouldn't be shocked if they went back to the old ways either, but I can see some reasons they would not.
newegg has not been out of stock on all brands since they launched to my knowledge. They're down to one brand in stock right now, but others have been coming in and out of stock since launch.Quote:
Originally Posted by THX1138
I've seen them at Amazon and on Ebay for MSRP/close during that time as well, people can buy them if they want one.
If AMD actually plays the price war game, which is sounds like they do at this point. In that they place their product line up at similar price points as their previous 6000 series, Nvidia would be forced to drop prices too. If there was enough of a pricing area between the 680 and a potential GK110 I don't see why Nvidia wouldn't release it. Even if it was $600 plus. They would still be able to sell it easily.Quote:
Originally Posted by Rollo
For your sake, since I can't be bothered to go through your browsing history, I hope you weren't one of those guys shouting 'PhysX and CUDA' from the rooftops for Thermi. Also, interesting how PhysX is barely mentioned nowadays.Quote:
Originally Posted by Rollo
Damien Triolet from Hardware.fr talks about start 2013 for GeForce GK110 (end of the year for Tesla version).
What's wrong with CUDA? People in the professional / HPC world are still "shouting it from the rooftops" as you put it. Currently, there isn't a better, more adaptable GPU compute language on the market. OpenCL surely has the chance to make it big but being an open format, there is still very little directed focus on prioritizing many inefficiencies.Quote:
Originally Posted by cegras
So the real question is -I'd guess- whether we're going to see the GK110 line on the desktop or whether we're going to have two separate lines of products from now on and there-after.
If we do, it will be a first to have lesser chips on the gaming arena, I guess side-effect from the console dominance...
the is the same as the last two big chips. they were tweaked to be good for non-gaming work too. the timing of their enterprise work just means it got announced this way first. these supercomputer contracts are a big deal for them, they are official a government contractor now! good work if you can get it! this may also be why they've held it back, to make it just right for those contracts.Quote:
Originally Posted by Stevethegreat
all the enterprise stuff in the big chip makes it less power efficient for gaming than the smaller chips. no big deal. 580 is still faster than 560.
I don't think anyone knows for sure... For all we know, it may never come out as a Geforce card... It's not like gamers need compute, and that's what this chip is about.Quote:
Originally Posted by Olivon
Sort of like SB-E to SB, larger and more expensive and mostly makes sense only for professional apps. But let's wait and see. :yepp:
Yeah but we did get the gtx 580, this time aroind we may not get *the* 780, or rather if we do it will be the gk114 (midrange) chip instead of the gk110. A bit of how we got the gk104 as this generation's high end instead of the gk100...Quote:
Originally Posted by bamtan2
I recorded the Press Conference Q&A after the Keynote :) http://bit.ly/GTC2012
Actually, the Keynote is up on Youtube:
http://www.youtube.com/watch?v=8FPQT...7ED9CB13A00DFD
ALSO, FOR EVERYONE...... KEPLER GK110 WHITE PAPER: http://www.nvidia.com/content/PDF/ke...Whitepaper.pdf
Very interesting stuff in there. Pretty much better than any possible article. ;)
I didn't say Keynote :) I know the Keynote was streamed and uploaded to YouTube which is why I didn't upload it. I only uploaded the Q&A session after the Keynote which was a press conference. Press conferences usually have a lot of more info than the Keynotes do ;)Quote:
Originally Posted by SKYMTL
I wasn't referring to your post.
However, this time I found the keynote had more info than the Q&A, but I guess that's just me....