News for hackers. Stuff that matters.
Re:Caffe2 adds 16 bit floating point training support on the... by throwaway87213 on Friday May 12, 2017 @02:47AM (#14321936)
I can see why they launched 1080 Ti early. Had I not seen this I'd definitely not be waiting for Volta.

How are things in the red camp ? There was some HIP thing where Fiji was as good as Pascal.

Re:Caffe2 adds 16 bit floating point training support on the... by dharma1 on Friday May 12, 2017 @07:02AM (#14322889)
Still no good CuDNN equivalent from AMD for machine learning that would make their cards competitive for that use case
Re:Caffe2 adds 16 bit floating point training support on the... by Symmetry on Friday May 12, 2017 @08:38AM (#14323430)
In consumer terms AMD's got a compelling product for mid-range GPUs where most of the money is but NVidia's got the top of the range locked up. Hopefully we'll see some good chips out of AMD soon which combine a genuinely good CPU core with a good GPU in one package which'll be really nice in the low end of the market.

As for scientific computing I suppose there are workloads where having your GPU and CPU on one die might give you latency savings that are worth it but mostly NVidia is going to keep owning that market for the next few years.

Re:Caffe2 adds 16 bit floating point training support on the... by slizard on Friday May 12, 2017 @12:57PM (#14325638)
The new "Radeon Instinct" cards are about to be released: http://www.anandtech.com/show/10905/amd-announces-radeon-ins...

The mid-range has a lot of potential as it's based on Fiji which NVIDIA can't match in terms of memory BW with any of their existing ML products (nor near future ones). This is important IMO because ML inference is memory-bound and GDDR5 on NVIDIA's cards is no match for even 1st gen HBM.

The big cards is announced to be 12.5/25 Tflops SP/HP, so as long as it's not "tensoring" it should be in between P100 and V100. If they get the price right (and given that even P100 is currently still $5.5-6 w/o tax), it will hopefully gain traction.

A lot hinges on their fully OSS (!) software stack, but they are getting things done, just had a new release: https://rocm.github.io

Re:Caffe2 adds 16 bit floating point training support on the... by philjohn on Friday May 12, 2017 @07:16AM (#14322958)
Waiting for Vega, which is AMD's answer to the 1080
Re:Caffe2 adds 16 bit floating point training support on the... by slizard on Friday May 12, 2017 @01:02PM (#14325679)
That's a blanket-statement and ends up being incorrect/inaccurate. The top Vega card is HBM2, not even in the same ballpark as the 1080.

Also, recent Linux patches hint that there will be dual-chip water-cooled Vega boards too which is likely not even 1080 Ti regime: https://www.techpowerup.com/233208/linux-drivers-point-to-up...

Re:Caffe2 adds 16 bit floating point training support on the... by NTripleOne on Friday May 12, 2017 @07:58AM (#14323176)
it seems increasingly common that that's the team red mantra nowadays, "Wait for x".

Wait for ryzen.

Wait for vega.

Wait for ryzen again.

Re:Caffe2 adds 16 bit floating point training support on the... by quickben on Friday May 12, 2017 @09:01AM (#14323591)
Amd cpu line is rolling out. You can buy ryzen.

Vega is coming out this year. Volta isn't consumer based. Go check the enterprise prices.

Consumer Volta is mid 2018, so this, from a consumer perspective, is a paper launch.

Re:Caffe2 adds 16 bit floating point training support on the... by eleitl on Friday May 12, 2017 @04:14AM (#14322234)
Any idea on the timeline for consumer Volta?
Re:Caffe2 adds 16 bit floating point training support on the... by dogma1138 on Friday May 12, 2017 @04:48AM (#14322367)
Early 2018 depending on GDDR6 supply. Seems like Hynix will start mass production on in late 2017.
Re:Caffe2 adds 16 bit floating point training support on the... by distances on Friday May 12, 2017 @04:48AM (#14322369)
Phoronix said "Volta desktop graphics cards are expected to succeed Pascal in late 2017 or early 2018" in https://phoronix.com/scan.php?page=news_item&px=NVIDIA-Volta....
Re:Caffe2 adds 16 bit floating point training support on the... by tanderson92 on Friday May 12, 2017 @07:13AM (#14322948)
I was told Q4 2017 for non-graphics consumer (what kind of consumer are you referring to), but that is earlier than most others are saying. https://news.ycombinator.com/item?id=14310633
Re:Caffe2 adds 16 bit floating point training support on the... by DrNuke on Friday May 12, 2017 @06:38AM (#14322807)
The GTX 1070 for laptops is destined to be the most interesting price/performance opportunity for a while imho, you can even undervolt the CPU -0.100 to -0.150 to reduce overheating.
Re:Caffe2 adds 16 bit floating point training support on the... by v4n4d1s on Friday May 12, 2017 @05:59AM (#14322637)
German news portal heise.de said Q2 2018.