After what appears to be a very unusual false start, AMD has now formally launched their new Radeon Vega Frontier Edition card. First announced back in mid-May, the unusual card, which AMD is all but going out of their way to dissuade their usual consumer base from buying, will be available today for $999. Meanwhile its liquid cooled counterpart, which was also announced at the time, will be available later on in Q3 for $1499.

Interestingly, both of these official prices are some $200-$300 below the prices first listed by SabrePC two weeks ago in the false start. To date AMD hasn’t commented on what happened there, however it’s worth noting that SabrePC is as of press time still listing the cards for their previous prices, with both cards reporting as being in-stock.

AMD Workstation Card Specification Comparison
  Radeon Vega Frontier Edition Radeon Pro Duo (Polaris) Radeon Pro WX 7100 Radeon Fury X
Stream Processors 4096 2 x 2304 2304 4096
Texture Units ? 2 x 144 144 256
ROPs 64 2 x 32 32 64
Base/Typical Clock 1382MHz N/A N/A N/A
Peak/Boost Clock 1600MHz 1243MHz 1243MHz 1050MHz
Single Precision 13.1 TFLOPS 11.5 TFLOPS 5.7 TFLOPS 8.6 TFLOPS
Half Precision 26.2 TFLOPS 11.5 TFLOPS 5.7 TFLOPS 8.6 TFLOPS
Memory Clock 1.89Gbps HBM2 7Gbps GDDR5 7Gbps GDDR5 1Gbps HBM
Memory Bus Width 2048-bit 2 x 256-bit 256-bit 4096-bit
Memory Bandwidth 483GB/sec 2x 224GB/sec 224GB/sec 512GB/sec
VRAM 16GB 2 x 16GB 8GB 4GB
Typical Board Power <300W 250W 130W 275W
GPU Vega (1) Polaris 10 Polaris 10 Fiji
Architecture Vega Polaris Polaris GCN 1.2
Manufacturing Process GloFo 14nm GloFo 14nm GloFo 14nm TSMC 28nm
Launch Date 06/2017 05/2017 10/2016 06/24/15
Launch Price Air: $999
Liquid: 1499
$999 $649 $649

Meanwhile AMD has also posted the final specifications for the card, confirming the 1600MHz peak clock. Sustained performance is a bit lower, with AMD publishing a “Typical clock” of 1382MHz. It’s worth noting that this is the first time AMD has used this term – they’ve previously used the term “base clock”, which is generally treated as the minimum clockspeed a card under a full gaming workload should run at. AMD is typically very careful in their word choice (as any good Legal department would require), so I’m curious as to whether there’s any significance to this distinction. At first glance, “typical clock” sounds a lot like NVIDIA’s “boost clock”, which is to say that it will be interesting to see how often Vega FE can actually hit & hold its boost clock, and whether it falls below its typical clock at all.

Feeding the GPU is AMD’s previously announced dual stack HBM2 configuration, which is now confirmed to be a pair of 8 layer, 8GB “8-Hi” stacks. AMD has the Vega FE’s memory clocked at just under 1.9Gbps, which gives the card a total memory bandwidth of 483GB/sec. And for anyone paying close attention to AMD’s naming scheme here, they are officially calling this “HBC” memory – a callback to Vega’s High Bandwidth Cache design.

As for power consumption, AMD lists the card’s typical board power as “< 300W”. This is consistent with the earlier figures posted by retailers, and perhaps most importantly, this is AMD’s official typical board power, not the maximum board power. So we are looking at a fairly high TDP card, and given that AMD has had a great deal of time to sit and work on their reference blower designs over the last few years, I’m anxious to see what that means for this initial air-cooled card.

For display outputs, the Vega FE devotes its entire second slot to airflow, so all of the display connectors are found on the first slot. Typical for AMD cards of the past couple of years, we’re looking at 3x DP 1.4 ports along with 1x HDMI port. AMD is also throwing in a passive DP to SL-DVI adapter in the box.

Moving on, let’s talk about the software setup for the Vega FE. As this is a card meant for (in part) game developers, AMD has opted to give the card access to both their pro and gaming drivers. Taking things one step further however, rather than making them separate downloads and installations, AMD has merged both drivers into a single install. Users can now download a single driver package and simply switch between driver modes in AMD’s control panel, allowing quick access to both driver types.

Unfortunately AMD hasn’t released much more in the way of detailed information on how driver switching works. In particular, it’s not clear whether switching requires a reboot or not; I would assume not, but it remains to be seen. Ultimately the primary purpose of this switch is for allowing game developers to switch modes for testing, going from pro driver mode for development to gaming mode for testing. The difference, I suspect, is less about driver code, and more about what driver optimizations are enabled. Games can get away with numerous hacks and optimizations in the name of performance, whereas professional applications need deterministic accuracy.

Otherwise, the driver situation touches on probably what remains one of the least-clear points of this product launch: who is the Radeon Vega Frontier Edition for? AMD is doing everything they can to encourage their typical Radeon consumer base to wait for the forthcoming Radeon Vega RX cards. In the meantime the company is stating that the card is “For Data Scientists, Immersion Engineers, and Product Designers” and certainly the pricing is closer to a professional card than a consumer card. Complicating matters is that AMD has been posting performance figures for SPECviewperf, Creo, other distinctly professional workloads, the kinds that typically go hand-in-hand with certified drivers. And at least for the moment, it doesn’t appear that AMD’s drivers have been certified (not that we’d expect them to be for a new architecture).

At a high level the Vega FE seems to compete with NVIDIA’s Titan Xp – and certainly that’s how AMD is choosing to pitch it – though this isn’t helped by the fact that NVIDIA isn’t doing a great job of establishing a clear market segmentation either, particularly since the launch of the GeForce GTX 1080 Ti. The Titan Xp is most certainly a partial gaming card (albeit a very expensive one), whereas AMD is focused more on professional visualization use cases that NVIDIA is not. Though where both overlap is on the compute front, where both the Vega FE and Titan Xp are essentially “entry-level” cards for production compute work. Otherwise, it may be better to treat the Vega FE as a beta testing card, especially given the “frontier” branding and the fact that AMD is clearly attempting to build out a more complete ecosystem for the future Vega RX and Instinct cards.

As for compute users in particular, AMD will be releasing the ROCm driver a bit later this week, on the 29th. Vega FE has a lot of potential for a compute card, thanks to its high number of SPs combined with equally high clocks. However serious compute users will need to code for its capabilities and idiosyncrasies to get the best possible performance on the card, which is all the more reason for AMD to get cards out now so that developers can get started. Compute will be the long-tail of the architecture: AMD can tweak the graphics performance of the card via drivers, but it's up to developers to unlock the full compute capabilities of the Vega architecture.

Wrapping things up, for anyone interested in picking up the Vega FE, AMD is currently only linking to Newegg’s storefront, where both the air cooled and water cooled cards are listed as “coming soon”. Otherwise SabrePC lists the cards in stock, albeit at prices over AMD’s MSRP.

Source: AMD

Comments Locked


View All Comments

  • jjj - Wednesday, June 28, 2017 - link

    Most likely.
    Would be nice if the lowest end Vega RXSKU would be clock limited instead of cores disabled but that doesn't work from a yield perspective.
    Maybe they should make a Green SKU that keeps clocks lower LOL.Not sure the world is ready for that but maybe paired with Radeon Chill, it could be marketable.
  • extide - Wednesday, June 28, 2017 - link

    You mean, like the Nano..? They already did that and it was reasonably successful, at least after they dropped the price.
  • jjj - Thursday, June 29, 2017 - link

    The marketing there was about form factor.
  • edzieba - Wednesday, June 28, 2017 - link

    Not only is it not huge progress, it's zero progress: for FP32, 8.6/13.1 TFLOP = 1600/1050 MHz. It's a direct scaling with frequency due to the same number of SPs. FP16 doubles from that number due to the use of packed math as seen in later Polaris chips. This very much appears to be a similar case to the Rx-4xx > Rx-5xx series clock bumps, though Vega also has a memory controller update (half the bus, at a bit less than double the pin data rate). It would not be unexpected to see real-world performance scale similarly from the Fury (around 1.52x), though the mention of 'Typical Clock' rather than 'Base Clock' may mean poorer scaling if power/temperature limiting is a factor.
  • jjj - Wednesday, June 28, 2017 - link

    We only have TDP numbers so we'll see.

    Typical Clock might indicate a variable base clock based on sensor input.
  • CiccioB - Wednesday, June 28, 2017 - link

    You can't evaluate architectural improvements using maximum TFLOPS values, as these are calculated synthetically by multiplying number of ALU * 2 * frequency to simulate the maximum throughput of FMA instructions (that executes 2 operations in a single cycle).
    Architectural improvements are those that allow general algorithms that are a mix and more complex fluxes of instructions, that are not all FMA and so cannot reach the peak TFLOPS, to be as close as possible to that.
    With number on paper you cannot know the improvements, that's why you need different benches with different scenery and algorithms (computing tasks are different to graphic tasks and both are different to gaming tasks),
  • psychobriggsy - Wednesday, June 28, 2017 - link

    That's how peak TFLOPS calculations work.

    Vega 10 will be smaller than Fiji due to the process shrink, despite the FP16 support that Polaris does not have (unless you count the PS4 Pro variant).

    What we do not know is if the IPC (efficiency) has improved with Vega. It is one of AMD's claims, so it should have.

    Whilst the HBM2 is half the memory bus, it is running faster. Whilst the on-paper bandwidth is a little lower than the on-paper bandwidth of Fiji, in reality Fiji could not make use of it all, and estimates are of between 350-400GB/s in the real world. In addition, Vega has advanced memory bandwidth saving techniques, which will aid a lot as well.

    The mention of typical clock is a concern, I would be hoping to see the SKU hit full clocks most of the time, at least in gaming. I could understand different behaviour in the professional profile.
  • bronan - Wednesday, June 28, 2017 - link

    Agree i am concerned about the change they made from base to typical which makes the water cooled card probably a much better choice to make.
  • msroadkill612 - Wednesday, June 28, 2017 - link

    A key number u would think is idle~ power. Who cares about power when u need the grunt, but its nice to have an affordable, always on pc.
  • CiccioB - Wednesday, June 28, 2017 - link

    If you do nothing all day with your PC it is important to have a low idle power, but if you use your GPU then how much it consumes is important as well, especially when high power consumption = high noise.
    BTW, Watts difference on idle power is measured on a single hand, when working you may need few dozens of them.

Log in

Don't have an account? Sign up now