Radeon RX Vega Confirmed launching at SIGGRAPH

Published 2017-07-03 08:15 by Hilbert Hagedoorn

Click here to post a comment for Radeon RX Vega Confirmed launching at SIGGRAPH on our message forum

Evildead666

2017-07-04 12:43

Agreed, this should have been out before Christmas. I'm not sure who is willing to buy a product that sits between a year old 1070/1080 if we go by the FE results. the price difference between the two is not even that high anymore. unless they go and undercut the gtx 1070 price... HBM2 sounded lovely back when we still had plan GDDR5 now we got GDRR5x hitting 12ghz on some of the models, so the bandswidth gap is tiny

I've been waiting for an AMD GPU thats quite a bit faster than my 290, for quite some time. the 580 is about 'on-par', or a little bit better, so i wasn't willing to make the jump. I'll be quite happy with a reference RX Vega (either the top one, or the second one) and slap a waterblock on it.

#5449524

Redemption80

2017-07-05 00:02

It's cool that you think better technology is a bad thing.

I think he meant it's a bad thing for AMD as a for profit company.

#5449528

PrMinisterGR

2017-07-05 00:07

Out of all of this, we don't actually have any indication that HBM is the issue here. Everybody who preordered Vega FE got it, and Koduri himself in the RTG AMA said that unlike HBM1, they have two suppliers for HBM2. It's also easier to manufacture than HBM1. HBM is a bit of a meme at this point I think.

#5449532

Redemption80

2017-07-05 00:26

I was meaning that it is more expensive and has helped contribute to the delays if we believe what has been written. Not that i think it could have been released much sooner as it's clearly still nowhere near ready, unless it is ready and the hardware is a failed experiment. I kinda doubt that, but it's possible.

#5449536

PrMinisterGR

2017-07-05 00:45

The only current failure we see is 100% on the software department. Until that clears out, a lot of things are open for the hardware.

#5449542

Denial

2017-07-05 01:57

Nvidia has their own kind of Infinity Fabric coming soon, and Intel will surely follow suit if they want to stay relevant. http://research.nvidia.com/publication/2017-06_MCM-GPU%3A-Multi-Chip-Module-GPUs The PDF is an interesting read for anyone who wants it. VEGA will need lots of optimization and at launch will not be 100% ready. Will likely take 6 months for the drivers to mature to the point that it can be close to 1080 Ti in some small cases, and by then Volta will arrive.

Wow, yeah that paper is awesome for anyone interested in GPU tech.

#5449577

Exascale

2017-07-05 07:43

That is a really good read. Guess we just found out what Nvidias post Summit and Sierra true exascale architecture will look like from that paper. Good find. That L1.5 cache makes the GPM a very interesting NUMA node in itself and helps solve the biggest problem for making exascale architectures, which is pJ/bit cost of moving data around.