NVIDIA Congratulates Turing - But Might Delay new upcoming Turing graphics cards

by Hilbert Hagedoorn on: 06/25/2018 08:10 AM | source: | 22 comment(s)

So, over the weekend two things happened. First up, it was Alan Turing's birthday, the man behind cracking the Enigma code, the one mathematical name in the industry looking to be the codename to be used by NVIDIA for their GPUs. And guess what?

So with the man having its birthday, it's always fun for a company that if you would use the name on one of your products, at least to do something with that on his birth date. And yeah, that's exactly what NVIDIA did as they posted this tweet:

It's more of a hint really, Turing seems to be the real codename for NVIDIA's upcoming GPU series or at least one of them. It's tradition for NVIDIA to make use of a product codename based on a mathematician, Pascal, Fermi, Kepler etc.

Right, then there is the second bit of news, word out on the street is that NVIDIA will be further delaying the new graphics cards. The mining arena seems to have collapsed, there would be much stock left based on current inventory, and since NVIDIA really doesn't 'need to' release a new series of cards. They might sit it out a little while longer to clear up inventory mentions SemiAccurate. Also, earlier this month it was announced that Nvidia would hold a presentation on Hot Chips mentioning "the next generation of mainstream gpu's" in the conference agenda. As it now seems, this presentation has been canceled and replaced by one from Intel. That is another indication that introduction of Turing seems to have been postponed, although we are not sure.

Random photo: GTX 1060 MSI

Related Stories

NVIDIA CEO Hands out Special CEO Edition Titan V GPUs to AI Researchers - 06/21/2018 03:21 PM
At the Computer Vision and Pattern Recognition conference in Salt Lake City Wednesday, NVIDIA&rsquo;s Jensen Huang told top AI researchers he wanted to help them achieve a very different kind of p...

NVIDIA CEO Makes Interesting Remark, New Gamers GeForce is a long time away from now - 06/04/2018 02:04 PM
Earlier today the first press conferences started in Taiwan, today Nvidia held an invite-only press-conference with Jensen Huang. Nothing ground-breaking or new for the consumer market has been intro...

NVIDIA Confirms G-Sync Ultra HD HDR monitors launch this month (updated) - 05/21/2018 07:04 AM
It's been a topic of long discussion and even longer wait, market availability of Ultra HD 144Hz HDR G-Sync that is. Well, it's now confirmed. In a series of media briefings this week Nvidia shar...

NVIDIA CEO comments on GeForce Partner Program - 05/15/2018 07:12 AM
Nvidia recently held a conference call on their latest financial results, one person in the call was Jen-Hsung Huang, and he is the CEO of Nvidia. He was asked about the cancellation of the Geforce...

Nvidia cautious that possible cryptocurrency demand might drop - 03/13/2018 10:04 AM
GPU demand from the cryptocurrency mining industry is showing signs of slowdown recently and Nvidia has started taking measures to minimize possible damage, according to some market sources, reports d...

2 pages 1 2

Jericho2013
Member

Posts: 52
Joined: 2013-04-26

#5560223 Posted on: 06/25/2018 04:26 PM

Pascals are still wiping the floor with all the games so I'm in no rush with my Titan Xp's.

JamesSneed
Senior Member

Posts: 642
Joined: 2017-02-14

#5560282 Posted on: 06/25/2018 08:56 PM

I don't think Nvidia was ever intending on shipping 12nm gaming GPUs. The only 12nm variant that brings anything is 6.5T and its roughly 10% increase in switching speed/power. Their FFN variant was for reticule sizes above 600mm2 and that was only available in 7.5T which brings no other benefits over 16nm. I never understood why people seemed so convinced that next gen would be 12nm when 7nm was slated for mass production 1H this year for over 2 years now.

It states right after the HUT 8 part that he improved on what the Polish had already done:

The Polish originally cracked specific implementations of the enigma machines - the germans improved them overtime and the method the polish were using became ineffective as complexity increased. Turing stepped in and came up with a more "generalized" method that sped up the process and made it effective again.

Semantics a bit here. The polish were "first" and did hand over what they had . The polish made a method to essentially brute force attack the message to determine the keys where Pascal improved that by essentially making a "crib" attack. Also the polish never cracked the latest enigma machines they were cracking the commercial ones that the germans used as a starting point(Not that they couldn't have if they were not overrun by Germans) but the Polish never did crack the improved enigma machines. So to say Pascal is first isn't wrong and to say the Polish were first isn't wrong either because they both had firsts at cracking certain versions of the enigma machines.

From the CIA link included:

"When Poland was overrun by Germany in September 1939, the Polish as well as French cryptanalysts shared everything they knew about ENIGMA with the UK, which allowed the cryptanalysts at Bletchley Park, including the famous Alan Turing, to finally crack the ENIGMA ciphers."

https://www.cia.gov/news-information/blog/2016/who-first-cracked-the-enigma-cipher.html

Noisiv
Senior Member

Posts: 6664
Joined: 2010-11-16

#5560287 Posted on: 06/25/2018 09:46 PM

So to say Pascal is first isn't wrong

No.
Tesla was first

alanm
Senior Member

Posts: 9022
Joined: 2004-05-10

#5560309 Posted on: 06/26/2018 12:09 AM

blazngun
Member

Posts: 23
Joined: 2017-11-24

#5560313 Posted on: 06/26/2018 12:28 AM

If Nvidia delays these GPUs in order to refine their performance, I'd be fine with that.

They would be stupid to not experiment while waiting for inventory to move. Surprised no one has speculated they could be waiting for better gddr6 for launch to the masses.

fantaskarsef
Senior Member

Posts: 11151
Joined: 2014-07-21

#5560329 Posted on: 06/26/2018 06:31 AM

I'm seeing some disturbing news re Nvidias new NDAs.
https://videocardz.com/76645/nvidias-new-non-disclosure-agreement-leaked

German site Heise.de has refused to sign it as its a long term (5 years) contract which could limit even speculation about technical aspects of Nvidia products unless its "beneficial" to Nvidia.
https://www.heise.de/newsticker/meldung/In-eigener-Sache-Nvidia-NDA-als-Maulkorb-fuer-Journalisten-4091751.html

I wonder why now? Could the timing also be related to how Turing may turn out? That it may not live up to expectations?

I wouldn't say that has to give us any specific clues to the chip itself, but more to how Nvidia tries to control the environment around them. Same as with GPP, now they're just doing it with tech journalists and not board partners.

But this time around there's actually a paper to read so everybody knows what's up, not just hearsay. I wish back with GPP such a document would have been leaked too.

Good find though, thx for sharing.

Fox2232
Senior Member

Posts: 9840
Joined: 2012-07-20

#5560330 Posted on: 06/26/2018 06:45 AM

I understand the enthusiast point of view but I am also sure that at the Nvidia headquarters they try to focus on the 250 watt tdp limit for their gaming cards so how much power the new theoretical 12nm architecture may need to be capable of 120 hz at 4k?
At this moment it seems that only the Titan V is capable to do what you would like to see in the mainstream segment.
I think you missed the point of "New Architecture", and ignored it for 12nm.
12nm may not do much, likely same achievable clock at 100mV less, so around ~10-15% smaller power consumption, or bit better OC. But point of Turing is in changes inside building blocks of GPU itself. Some things may be done within fewer GPU cycles. It may have those new Tensor cores.
Maybe better intra frame voltage gating for parts which are done their work and are waiting for next batch of data. (which may bring even bigger energy efficiency over 12nm)
There are really many things next architectural step can bring. What if nVidia improved HW level compression for textures and next generation had effectively 20% more space for textures on card with same amount of physical memory?

2 pages 1 2

Post New Comment

Click here to post a comment for this news story on the message forum.