Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Nvidia Pascal GPU to Feature 17B Transistors and 32GB HBM2 VRAM

wccftech.com

Post reply on HN
Print view View original

21–30 of 97 posts

Re: Nvidia Pascal GPU to Feature 17B Transistors and 32GB HBM2 VRAM

#21

Post by dman »

raverbashing wrote: post #20
bryanlarsen wrote: post #17

Earlier quoted context omitted.

It could be argued that AMD does have it together. If you look at the benchmarks for pretty much any particular price point, AMD & nVidia cards are roughly comparable. AMD is better at some games and resolutions, and nVidia is better at others. Rarely is the difference more than 10%. nVidia used to have a sizable power/noise advantage, but Fury appears to close that gap. AMD even seems to be closing the driver stabil…

Not sure about their graphics division, but AMD has stopped innovating in the CPU department...

They are working on Zen - which is a ground up rearchitecture. They have limited resources so the existing designs (Opterons and APUS) are only receiving minor incremental updates until Zen lands. Zen is expected in 2016, AMD's continued existence as a serious CPU maker depends on it.

Re: Nvidia Pascal GPU to Feature 17B Transistors and 32GB HBM2 VRAM

#23

Post by noir_lord »

bryanlarsen wrote: post #17
blinkingled wrote: post #5

AMD get it together, pretty please. Nobody wants to live in a world where only NVidia produces discrete GPUs! That 32GB seems too high for cost constrained consumer market though - may be they will have a leaner variant for desktops/gaming.

It could be argued that AMD does have it together. If you look at the benchmarks for pretty much any particular price point, AMD & nVidia cards are roughly comparable. AMD is better at some games and resolutions, and nVidia is better at others. Rarely is the difference more than 10%. nVidia used to have a sizable power/noise advantage, but Fury appears to close that gap. AMD even seems to be closing the driver stabil…

I've used http://www.videocardbenchmark.net/high_end_gpus.html and https://www.cpubenchmark.net/high_end_cpus.html for years.

It's some what asinine to reduce the complexity down to a single number but I've found that they are usually pretty reflective of what you find in practice.

Re: Nvidia Pascal GPU to Feature 17B Transistors and 32GB HBM2 VRAM

#24

Post by noir_lord »

norea-armozel wrote: post #3
Zekio wrote: post #2

Looks like the next few generations of GPUs are gonna be fast!

I think this may be specs for their Tesla cards. I doubt they would throw 32 HBM2 even on a Titan unless they can keep the price point the same (even for folks who can spend 1k USD have limits on their budget or at least their perception of a budget). But I wouldn't doubt by Q2/2016 Nvidia will be bringing more competition to the market with chips based on HBM usage. IMO, AMD pulled the trigger a tad too soon on thei…

Surprises me that RAM hasn't increased more, the machine I'm on is 5 years old and it has 2GB of RAM on the GPU (HD6950) which I paid ~200 quid for 5 years ago.

Re: Nvidia Pascal GPU to Feature 17B Transistors and 32GB HBM2 VRAM

#25

Post by bryanlarsen »

raverbashing wrote: post #20
bryanlarsen wrote: post #17

Earlier quoted context omitted.

It could be argued that AMD does have it together. If you look at the benchmarks for pretty much any particular price point, AMD & nVidia cards are roughly comparable. AMD is better at some games and resolutions, and nVidia is better at others. Rarely is the difference more than 10%. nVidia used to have a sizable power/noise advantage, but Fury appears to close that gap. AMD even seems to be closing the driver stabil…

Not sure about their graphics division, but AMD has stopped innovating in the CPU department...

So has Intel. Skylake (2016) and Sandy Bridge (2010) have roughly comparable single core desktop benchmarks. Power/performance ratios have increased drastically, but that's mostly because of process improvements. AMD has been stuck on 28nm for a long time, but that's not AMD's fault.

Re: Nvidia Pascal GPU to Feature 17B Transistors and 32GB HBM2 VRAM

#26

Post by gwern »

> 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 on NVIDIA’s GP100 flagship Pascal which is twice compared to the 512 GB/s on AMD’s Fiji cards and three times that of the 980 Ti’s 334GB/s.

> The Pascal GPU would also introduce NVLINK which is the next generation Unified Virtual Memory link with Gen 2.0 Cache coherency features and 5 – 12 times the bandwidth of a regular PCIe connection. This will solve many of the bandwidth issues that high performance GPUs currently face.

To point out the obvious, this sounds like it could be fantastic for deep learning. Not just is the RAM big enough to hold a lot of current datasets, it'll help alleviate the latency bottlenecks in updating stuff.

Re: Nvidia Pascal GPU to Feature 17B Transistors and 32GB HBM2 VRAM

#27

Post by varelse »

icegreentea wrote: post #16
whoisthemachine wrote: post #9

> The 17 Billion transistors on the Pascal GPU are twice the transistors found on the GM200 Maxwell and the Fiji XT GPU core which is literally insane Seems like Moore's law is alive and well in the graphics/attached processor space.

That's because GPUs were frozen at the 28nm node since like 2011. It'll be ~4 years of no die shrinks at the top end of GPUs when they finally transition to 16nm. If anything, Moore's law is behind schedule in the GPU space. Note, that both NVidia and AMD rely on TSMC to manufacture their chips, so they're completely constrained by TSMC's ability to implement new process nodes.

2011: GTX 580, 1.5 TFLOPS

2012: GTX 680, 3.0 TFLOPS (~2.0 attainable)

2013: GTX Titan, 4.4 TFLOPS (~3.2 attainable)

2014: GTX 980, 4.6 TFLOPS

2015: GTX Titan X, 6.7 TFLOPS

Looks to me like they're doubling perf roughly every 2 years.

Meanwhle, my Core i7-5930k's SOL is <1/2 of 2011's GTX 580 at 672 GFLOPS and it still doesn't have fast approximate transcendentals. Skylake begins to fix this, but c'mon, GPUs have had these for almost a decade now...

Re: Nvidia Pascal GPU to Feature 17B Transistors and 32GB HBM2 VRAM

#28

Post by x3sphere »

bryanlarsen wrote: post #17
blinkingled wrote: post #5

AMD get it together, pretty please. Nobody wants to live in a world where only NVidia produces discrete GPUs! That 32GB seems too high for cost constrained consumer market though - may be they will have a leaner variant for desktops/gaming.

It could be argued that AMD does have it together. If you look at the benchmarks for pretty much any particular price point, AMD & nVidia cards are roughly comparable. AMD is better at some games and resolutions, and nVidia is better at others. Rarely is the difference more than 10%. nVidia used to have a sizable power/noise advantage, but Fury appears to close that gap. AMD even seems to be closing the driver stabil…

As someone who has bought AMD cards for the past 5 years for gaming, I switched over to Nvidia this time.

You've overlooked that Maxwell / 980 Ti has a ton of overclocking headroom. OC to OC, you're looking at 20-30% performance difference at more commonly used resolutions like 1440p and 1080p. The gap narrows only at 4K.

Voltage is locked on Fury right now and AMD isn't saying why. The best OC I've seen on one is 10% higher clocks, with Maxwell 30% is common. Huge difference considering both cards are in the same price bracket.

Lastly pricing is also an issue with the Fury cards. Usually AMD undercut Nvidia, but this time the flagship matches the 980 Ti in cost. As the underdog, AMD is going to have a tougher time winning people over at the same price point. This happens in any market.

I like to support the underdog, but I couldn't this time given the large disparity in OC performance.

Re: Nvidia Pascal GPU to Feature 17B Transistors and 32GB HBM2 VRAM

#29

Post by bd »

gwern wrote: post #26

> 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 on NVIDIA’s GP100 flagship Pascal which is twice compared to the 512 GB/s on AMD’s Fiji cards and three times that of the 980 Ti’s 334GB/s. > The Pascal GPU would also introduce NVLINK which is the next generation Unified Virtual Memory link with Gen 2.0 Cac…

Yup, this is exactly the marketing angle Nvidia CEO was using in his GPU Technology Conference presentation (10x speed-up on deep-learning vs their current Maxwell architecture):

http://blogs.nvidia.com/blog/2015/03/17/pascal/

Re: Nvidia Pascal GPU to Feature 17B Transistors and 32GB HBM2 VRAM

#30

Post by curiousjorge »

I've been look at gtx 750 ti which has maxwell processor in it. I don't care too much about playing games on ultra but the sheer quiet, cool temperature on load, and low power consumption is what makes me attracted to it. I wonder if the new Pascal GPU will make room for more such low power consumption and quiet cards that pack plenty of punch?
Post reply on HN