Radeon Pro WX 8200 promises pros lots of bang-for-the-buck

The SIGGRAPH conference kicks off today in Vancouver, and AMD is marking the occasion by launching a fresh pro graphics card. The Radeon Pro WX 8200 houses a Vega 10 GPU with 56 compute units enabled, fed by 8 GB of second-generation SK Hynix HBM2 RAM, and it'll feed displays using four Mini DisplayPort outputs.

That refined memory gives the WX 8200 512 GB/s of memory bandwidth, up from 410 GB/s on the consumer RX Vega 56. Working back from AMD's specs, that HBM2 RAM is likely running at an effective rate of 2 Gbps per pin. AMD's specified theoretical FP32 rate of 11 TFLOPS suggests the WX 8200 has a boost clock range of 1530 MHz. Both of those figures are superior to the RX Vega 56.

AMD is aggressively pitting the WX 8200 against Nvidia's roughly-$1800-at-retail Quadro P5000 in a range of workloads. The company generally promises a decent lead in performance over the P5000 for renderers like Radeon ProRender and Blender Cycles, as well as pro apps like Nuke, Premiere Pro, and Maya.

AMD believes that the future of professional graphics work lies in two new frontiers of performance that traditional metrics like those above don't capture, however: VR and simultaneous rendering and multitasking.

The company's DirectX 12-powered VRMark Cyan Room synthetic VR test result is self-explanatory. Like the other internal tests AMD presented, the company believes the WX 8200 leads the Quadro P5000 for professional use. For pros who are previewing their designs or models in VR, that performance parity could be an important mark in AMD's favor.

Original image: Vizrt

AMD also made the point that pros don't need to worry about manually tuning their systems for the best performance with the WX 8200 under blended workloads when multitasking, as one might with the Quadro P5000 and its dedicated toggles in Nvidia's control panel for managing graphics and compute workloads or focusing entirely on graphics performance.

To drive this point home, AMD mixed graphics and computing workloads all at once by running the SPECviewperf 13 benchmark on top of a Blender Cycles render task. The company claims that the SPECviewperf suite of tests generally ran at rates professionals would find acceptable for interactive use, while the Quadro P5000 failed to deliver acceptable performance in even one of the SPECviewperf tests under the same mixed workload.

The biggest selling point for the WX 8200 may be its price tag: $999, or slightly more than the retail price of the Quadro P4000 for that potentially Quadro P5000-meeting or -beating performance. The card will go up for pre-order at Newegg tomorrow and should hit e-tail shelves in early September.

AMD will be showing off the WX 8200 at SIGGRAPH alongside an improved version of Radeon ProRender that can perform hybrid rendering with ray-tracing and rasterization, as well as a heterogeneous rendering approach that can put CPU power to work on difficult areas of a rendered scene when it's needed—perhaps an ideal task for the Threadripper 2990WX that's coming to retail tomorrow, as well. Interested pros in Vancouver should check out AMD's booth for more information.

Comments closed
    • Voldenuit
    • 1 year ago

    Did they set the Blender CPU priority to ‘High’ on the Quadro setup and ‘Low’ on the Radeon Pro setup?

      • NoOne ButMe
      • 1 year ago

      No.
      They did ISO because the are not (that) dumb

      [url<]https://www.anandtech.com/Gallery/Album/6608#27[/url<]

    • Mr_m_to_the_t
    • 1 year ago

    Any comparison to the AMD Radeon Pro Duo 32GB would be nice….

    • DancinJack
    • 1 year ago

    Hey only three years after the P5000!

      • ptsant
      • 1 year ago

      They had to have a token victory before the next generation from nVidia comes out (soon). Don’t expect much in the GPU space from AMD for at least 12-18months.

      • derFunkenstein
      • 1 year ago

      Two years, right? There weren’t any GP104 cards before 2016 were there?

      Still, it took quite a while.

    • willyolioleo
    • 1 year ago

    oh wow i didn’t know SIGGRAPH was in vancouver.

    maybe if i have some time off i’ll check it out

    [i<]one day price: $750 USD[/i<] [i<]full conference price: $1550 USD[/i<] never mind... anime convention at the same venue last week was, like, $75 for the weekend.

      • DancinJack
      • 1 year ago

      Yeah, most people that go to those things get passes through work. I’ve been to a few similar things through work and was glad at the time, I’m not paying 500+ for a single day of news when it isn’t mission critical for everything I do.

      • NoOne ButMe
      • 1 year ago

      Like me and hotchips. Although hotchips is far far cheaper.

      “just” $550 or so registering early for two days.

      For students Hotchips is 100% worth it I would think. $120 for the talks, an extra $60 for the tutorials (those I am unsure about what is offered).

        • jts888
        • 1 year ago

        I finally asked work to send me to Hot Chips this year, and to my happy semi-surprise they actually said yes. Intel and AMD are only presenting Cascade Lake and Raven Ridge, so the POWER9 might be the biggest detail unveiling in the big iron space. (I expect Zen 2 next year and Ice Lake in 2020/2021 to be great presentations, but not every year can have some singular spectacular topic.)

        But it’s not even the symposium ticket or the airfare price that seemed steep, but the hotel costs in Cupertino. I would definitely Air BnB this if I was paying for it myself…

          • NoOne ButMe
          • 1 year ago

          Congratulations, hope you enjoy!

          I live close enough to drive if I wanted… Although a cheapo hotel would probably be worth saving 4-6 hours of my time over the two days depending on the traffic.

      • RtFusion
      • 1 year ago

      Went to AniRevo again this year and it was a great!

      • the
      • 1 year ago

      Could pay that price, go to the nVidia keynote and walk out lucky with a Titan V CEO Edition.

      • USAFTW
      • 1 year ago

      For $1550, you could either:
      1. Go to 20.66 anime conventions; or:
      2. Go to the same anime convention 20.66 times.
      and watch the glow on Jen Hsun’s leather jacket in stunning 4K at youtube.

        • tipoo
        • 1 year ago

        I need my CEO Math™ in real time, damn it

      • Zizy
      • 1 year ago

      Eh, it is always like that.
      NT conference costs a lot (or used to, haven’t been in a while), but another MS organized event just slightly later in a nearby place had 0 fee for the weekend and gave prizes for the best apps… (sure, the prizes were Windows phones and tablets they couldn’t sell, but still).
      You pay to network mostly. The talks are just the cherry.

      • Action.de.Parsnip
      • 1 year ago

      Holy moly is it actually that much????

    • ronch
    • 1 year ago

    No mention of power consumption?

      • Anton Kochubey
      • 1 year ago

      Judging by power connectors, it’s limited to 300W (75W from PCIe slot, 75W from a 6-pin and 150W from a 8-pin).

      • Goty
      • 1 year ago

      Anandtech cites AMD and claims the “TDP/max power consumption” is 230W.

        • DoomGuy64
        • 1 year ago

        Now why can’t we get a desktop Vega this good? Dual slot, higher bandwidth, and lower power consumption makes me think AMD seriously botched the desktop versions, while this pro card is what Vega was supposed to be.

          • ptsant
          • 1 year ago

          The MSRP of VEGA 56 is less than half. Nevermind what the retailers have been charging, because of mining. There is a lot of component selection that goes into that.

          Plus they also mentioned that the RAM is specifically better and might not have been available at the time (in fact, HBM2 shortages were probably one of the most important obstacles in VEGA launch).

            • DoomGuy64
            • 1 year ago

            bah. That’s not an excuse for now. Vega 56 shouldn’t be worse than Fury, and yet it is. Refresh the lineup with better hardware. It doesn’t need to match the pro, just be viable desktop cards.

            If the price of this ever dropped to desktop levels, I’d easily pass over the desktop cards.

            • K-L-Waster
            • 1 year ago

            Must be nice to live in a world where speed + functionality can be delivered without increasing prices…

            • DoomGuy64
            • 1 year ago

            Stop putting words in my mouth. AMD is perfectly capable of refreshing Vega 56 AT THE CURRENT SPEED, but with lower power consumption and smaller coolers.

            There is NO EXCUSE for Vega 56 to not be currently made at the quality level of Fury cards, which means dual slot coolers and lower power consumption.

            Vega 56 is literally an updated Fury, and it is completely unacceptable to be less efficient. AMD needs to lower the clockspeed and voltage if that’s a problem, but this WX 8200 card shows it’s not an issue at this point in time. So just refresh 56 with whatever power consumption tweaks they’re doing with the 8200, and keep the lower clockspeeds if the HBM is lower quality.

            Quite frankly, none of you naysayers have any evidence that the WX 8200 is using any magical hardware that justifies $999 other than being a “WORKSTATION” branded card. It’s the same components as a regular Vega card, just like the “FRONTIER EDITION” card used the same components as a regular Vega, only marked up for being a workstation card. AMD can now manufacture Vega 56 with lower power consumption, and it is ridiculous to not do so if they want to be competitive with Nvidia. If they do not take this opportunity, there will end up being not a single point AMD has effectively fought Nvidia, especially after Nvidia releases new chips while AMD won’t have anything new for another year.

            Ridiculous. It’s like you people literally DON’T want AMD to compete.

            • NoOne ButMe
            • 1 year ago

            Because Vega 56 has a $150 lower MSRP and in the price of the HBM2 is in the $50-75$ more range than the Fury’s HBM costed.

            AMD bet on HBM2 with far different cost estimates than the market ended up at.

            • DoomGuy64
            • 1 year ago

            So? They’re still selling it, and selling it to both workstation and consumer. Just update the desktop part with lower power consumption. Hell, AMD could probably do that with existing hardware with a Driver or Bios update.

            • NoOne ButMe
            • 1 year ago

            To bad there is not a second BIOS which drops power 30-50W at the cost of a couple percentage points of performance…

            Or by site that showed how such a BIOS would preform.

            *cough* [url<]https://techreport.com/review/32391/amd-radeon-rx-vega-64-and-rx-vega-56-graphics-cards-reviewed/11[/url<] *cough*

            • BobbinThreadbare
            • 1 year ago

            Yes, professional level hardware has always carried a silly markup

            • DoomGuy64
            • 1 year ago

            It’s the drivers that command that markup, and understandably so. The hardware on the other hand is pretty much the same, especially on the AMD front, and doesn’t excuse AMD from not releasing a more power efficient Vega on the desktop.

            • NoOne ButMe
            • 1 year ago

            You seem to have missed it, so I’ll repost it:
            To bad there is not a second BIOS which drops power 30-50W at the cost of a couple percentage points of performance…

            Or by site that showed how such a BIOS would preform.

            *cough* [url<]https://techreport.com/review/32391/amd-radeon-rx-vega-64-and-rx-vega-56-graphics-cards-reviewed/11[/url<] *cough*

Pin It on Pinterest

Share This