Nvidia GTC16 Webcast Live Stream@5pm/9am PDT Jen looks great in leather.

drunkenmaster · 5 Apr 2016 at 20:26

Jono8 said:
Well by exaggerating I meant the x2 980Ti performance for £200 line

Oh yeah, that stupid x70 £250 thing was equally hilarious but yeah less people, however a lot really did think there would be proper volume launch in April.

Purgatory · 5 Apr 2016 at 20:27

Simon Belmont said:
Ordered 2.

June ain't far off.

June will be the paper launch.. of the x70 and x80 at this rate.. so good luck

stulid · 5 Apr 2016 at 20:29

D1craig said:
any chance of updating the original post with whatever info we have? its hard work reading all this junk in here...

Jen looks good in leather, that's all that there is to know.

bru · 5 Apr 2016 at 20:29

Those clock speeds are pretty high for a Tesla part, I mean the P100 base clock is higher than the kingpins boost clock :eek:

for a non consumer part that is ridiculous.

easyrider · 5 Apr 2016 at 20:30

stulid said:
Jen looks good in leather, that's all that there is to know.

Too portly for my tastes.

Rroff · 5 Apr 2016 at 20:31

stulid said:
Jen looks good in leather, that's all that there is to know.

That and that there is zero chance of Pascal working and its the world's most expensive brick.

CAT-THE-FIFTH · 5 Apr 2016 at 20:32

bru said:
Those clock speeds are pretty high for a Tesla part, I mean the P100 base clock is higher than the kingpins boost clock for a non consumer part that is ridiculous.

Simon Belmont · 5 Apr 2016 at 20:33

CAT-THE-FIFTH said:
I think it indicates that Nvidia are quite worried about the Intel compute cards,whch are slowly creeping into the supercomputer market.

Intel compute card could change everything. If Nvidia want to stick around they need to keep happy the very people that have keep them in business for so long "The PC Gamer"

RichardH754 · 5 Apr 2016 at 20:34

CAT-THE-FIFTH said:

lol its back!

CAT-THE-FIFTH · 5 Apr 2016 at 20:36

Simon Belmont said:
Intel compute card could change everything. Nvidia need to keep happy the very people that have keep them in business for so long "The PC Gamer"

The profit margins are quite high in that area though - they are a significant contribution to the companies bottom line and what makes it economically to build some large GPUs in the first place.

We would never have had the GK110 based cards in 2012 or 2013 IMHO if Nvidia could not sell them in the commercial markets.

CAT-THE-FIFTH · 5 Apr 2016 at 20:38

RichardH754 said:
lol its back!

The legend never dies!!

Rroff · 5 Apr 2016 at 20:38

pmc25 said:
I don't think the Drive stuff is in production either. The card in the glass case had unmarked heatsinks with Tegra decals on them, to prevent identification of what it was (prior showing was quickly identified as Maxwell, despite them claiming it to be Pascal). Almost certainly means they still don't have Drive PX2 test chips, even if it is going to be the first Pascal product to reach production.

With nVidia that doesn't really mean much - they've done major shows before with just slides and/or mock ups and then released a new line up shortly after.

tommybhoy · 5 Apr 2016 at 20:38

D1craig said:
any chance of updating the original post with whatever info we have? its hard work reading all this junk in here...

Will that do?

Purgatory · 5 Apr 2016 at 20:39

drunkenmaster said:
There is a decent amount of information on the Pascal architecture from the link someone put up earlier. They've given shaders per SM, number of SMs and some fairly detailed numbers, number of registers, number of threads, clock speeds/performance for GP100. Assuming that Pascal is the same up and down the range in base design then that is a lot of Pascal architecture info. It's heavily FP64/FP16 performance increase focused and FP32 performance increase is pretty painfully low.

https://devblogs.nvidia.com/parallelforall/inside-pascal/

CAT-THE-FIFTH · 5 Apr 2016 at 20:43

Someone on AT forums made this observation:

In that respect Pascal is more closer to GCN the ever in terms of on chip memory resources and occupancy and it even has 64K of shared mem per SM (64 wide) just like GCN CU (64 wide) having 64K of shared mem! (The same ALU/shared mem ratio!)

GCN and Pascal might be more a like than we think!!

Rroff · 5 Apr 2016 at 20:44

drunkenmaster said:
It's heavily FP64/FP16 performance increase focused and FP32 performance increase is pretty painfully low.

FP32 isn't that bad - sure the hardware units hasn't gone up much but the clock speeds are a fair jump - the actual GFLOP increase isn't to be sniffed at.

The CUDA core per SM is an interesting change for those who've been keeping up

Purgatory · 5 Apr 2016 at 20:44

stulid said:
Jen looks good in leather, that's all that there is to know.

He actually looked very nervous, made a good few mistakes and even at one stage looked very annoyed. Clearly something not right. Was not his greatest presentation but then again all they had to show was slides and a render and a $130k box of promises.

mikeym · 5 Apr 2016 at 20:44

stulid said:
Jen looks good in leather, that's all that there is to know.

What about the other guy and his hair? :confused:

Rroff · 5 Apr 2016 at 20:47

Purgatory said:
He actually looked very nervous, made a good few mistakes and even at one stage looked very annoyed. Clearly something not right. Was not his greatest presentation but then again all they had to show was slides and a render and a $130k box of promises.

The number of times he repeated/started to repeat himself caught my attention - something clearly not right but might be personal not business.

Ian Evey · 5 Apr 2016 at 20:48

so no fruit pascals anytime soon for me

Competitor rules

Nvidia GTC16 Webcast Live Stream@5pm/9am PDT *Jen looks great in leather.*

Nvidia GTC16 Webcast Live Stream@5pm/9am PDT Jen looks great in leather.