News Our first (almost proper) look at Vega

And the new 'Frontier' card. So basically just a reference card. But still:

Now, lets hope for an RX Vega Nano.

Top is liquid cooled.

Read more here:


aka James
Gold Supporter
Feb 22, 2015
The current PCIe spec only allows up to 1x 8-pin + 1x 6-pin. Dual 8-pin is technically out of spec, but was notably used on the 295x2.


SFF Lingo Aficionado
Feb 11, 2016
The Palit 1080 Ti HOF has three 8-pins and it's new to the market!

Back on topic though, the Frontier Edition is definitely a single GPU solution.
The suggestion is that the Gold Edition's CLC will either allow it to be clocked higher or maintain those clocks without being thermally throttled.

With RTG's new enterprise first, client later approach, I'm not holding my breathe for a consumer Vega release in Q2.



Cable-Tie Ninja
Nov 19, 2016
Gnng. They look terribly cheap. :-(
I wonder what forced them to add that plastic cube with the R's to one corner. It can't been voluntary.

Maybe they are so confident in their performance/price that they will be bought despite their looks?


SFF Lingo Aficionado
Feb 11, 2016
I don't think its a great sign for power efficiency though that one of the first Vega cards uses an AIO for cooling.
Even with the new manufacturing process now being utilised for their high end offering I wouldn't be surprised if this has a TDP in the range of of 250-300W which doesn't bode well for a cut down consumer card.

With all being said, I'm still excited to see what RTG bring to the table, particularly if there is a Vega Nano iteration.


Virtual Realist
Gold Supporter
May 11, 2015
Dual x8 power connectors, is this a dual-GPU card?
Single GPU, just one Fury-sized (or slightly bigger) die. Same Stream Processor count, and the raw TFLOP rate increase is almost identical to the clock speed increase. Outside of the Single Precision performacne doubling to account for packed operations, 'big Vega' is looking to be a Fury clock-bump when it comes to consumer performance. The big question is memory: it's stated to have 16GB of HBM2, but the bus width compared to Fury is half (2048bit vs. 4086bit) which would indicate only two HBM2 dies vs 4 HBM dies. But neither SK-Hynix nor Samsung have even advertised 8-stack (8GB) HBM2 yet. Memory bandwidth takes a slight hit compared to Fury, but for consumer workloads memory bandwdith has never really been a limiting factor.
I can't think of any reason this card would have a TDP below the PCIe max of 300W, being aimed at the performance-above-all 'deep learning' market.
With RTG's new enterprise first, client later approach, I'm not holding my breathe for a consumer Vega release in Q2.
With this card just scraping in at the end of the '1H2017' shipping windows for Vega, I wouldn't bet on consumer cards coming until Q4. Possible a Titan-esque high-margin consumer 'halo' card using binned dies in Q3.


Lord of the Boards
Editorial Staff
Gold Supporter
May 9, 2015
It's their workstation-class professional GPU, it's blue to differentiate from the red for the consumer Radeon. The Radeon Pro Duo was also blue by the way :)
Though Intel Xeon Phi cards are also blue, though those usually end up in rack server enclosures and not workstations or desktops.
  • Like
Reactions: SeñorDonut