Sapphire's Dual-X R9 280 OC Graphics Card Review

How We Tested Sapphire's Dual-X R9 280 OC

In this single-product review, we'd like to focus on how Sapphire's Dual-X R9 280 OC fares against its predecessor, the Radeon HD 7950 Boost reference card, and Nvidia's similarly-priced competition, the GeForce GTX 760.

Loading...

Two of the games we're testing have an option to use a Mantle code path, so we're running those benchmarks (Thief and Battlefield 4) with Mantle enabled and disabled to measure the API's impact.

Graphics cards like the Radeon R9 280 require a substantial amount of power, so XFX sent us its PRO850W 80 PLUS Bronze-certified power supply. This modular PSU employs a single +12 V rail rated for 70 A. XFX claims continuous (not peak) output of up to 850 W at 50 degrees Celsius.

We've almost exclusively eliminated mechanical disks in the lab, preferring solid-state storage for alleviating I/O-related bottlenecks. Samsung sent all of our labs 256 GB 840 Pros, so we standardize on these exceptional SSDs.

Loading...


Test System
CPU
Intel Core i7-3960X (Sandy Bridge-E), 3.3 GHz, Six Cores, LGA 2011, 15 MB Shared L3 Cache, Hyper-Threading enabled.
Motherboard
ASRock X79 Extreme9 (LGA 2011) Chipset: Intel X79 Express
Networking
On-Board Gigabit LAN controller
Memory
Corsair Vengeance LP PC3-16000, 4 x 4 GB, 1600 MT/s, CL 8-8-8-24-2T
Graphics
Sapphire Dual-X R9 280 OC
850/940 MHz GPU, 3 GB GDDR5 at 1250 MHz (5000 MT/s)

Reference AMD Radeon HD 7950 Boost

850/925 MHz GPU, 3 GB GDDR5 at 1255 MHz (5000 MT/s)

Nvidia GeForce GTX 760
980/1033 MHz GPU, 2 GB GDDR5 at 1502 MHz (5008 MT/s)
SSD
Samsung 840 Pro, 256 GB SSD, SATA 6Gb/s
Power
XFX PRO850W, ATX12V, EPS12V
Software and Drivers
Operating System
Microsoft Windows 8 Pro x64
DirectX
DirectX 11
Graphics Drivers
AMD Catalyst14.7 RC 1
Nvidia 340.52 WHQL
Benchmarks
Watch Dogs
Version 1.04.497, Custom THG Benchmark, 90-second Fraps run, Driving
Arma 3
V. 1.26.126.789, 30-sec. Fraps "Infantry Showcase"
Battlefield 4
Version 1.3.2.3825, Custom THG Benchmark, 90-Sec
Assassin's Creed IV: Black Flag
Custom THG Benchmark, 40-Sec
ThiefVersion 1.6.0.0, Built-in Benchmark
TitanfallVersion 1.0.5.7, Demeter Map, Custom THG Benchmark

Create a new thread in the US Reviews comments forum about this subject
This thread is closed for comments
37 comments
    Your comment
  • Why no noise measurements ?
    7
  • ill wait for tonga
    1
  • Quote:
    Why no noise measurements ?

    The one thing I was really interested to see
    6
  • This... as a newer, faster and more power efficient R9 285 comes out?
    What are the board partners thinking?
    0
  • They got the wrong memory bandwidth for the R9 280 and 7950 cards... it's 384-bit.

    Which is why I'm worried that the 285 won't be able to keep up. Particularly at higher resolutions.

    [Response by Cleeve:

    Doh! Fixed. :) [/Response]
    1
  • Wow mantle gives Intel 6 cores relevance in games. Wonder if AMD will design a 16 core now for the AM3? Cant wait to see mantle on Intels new 8 core CPU's.
    1
  • Am I the only one who finds it unfair to bench an Overclocked card and put it against a reference model?
    0
  • The mantle 4770 test for thief was -.1 min fps and only 1fps average increase. Clearly on the 6 core its getting more in both min fps and average.
    1
  • The battlefield 4 mantle test lost performance with the 4770 on 1080.
    1
  • 550730 said:
    Am I the only one who finds it unfair to bench an Overclocked card and put it against a reference model?


    This is a Sapphire Dual-X review, not a reference 280 launch.

    The Sapphire Dual-X R9 280 comes overclocked from the factory.
    0
  • 749236 said:
    This... as a newer, faster and more power efficient R9 285 comes out? What are the board partners thinking?


    This is simply a last look at the 280, relevant from the perspective that these will probably be discounted on the market for a little while.

    As far as performance, we don't believe the 285 will be much faster (possibly even slower), based on specifications.
    2
  • I agree with Shneiky. The small margins by which the 280 edges out it's win could easily be explained away by the fact that it's an aftermarket card vs reference cards.
    -1
  • What I mean is - this is OC 280 vs Stock 760. If you get a OC model of a 760 from Asus, EVGA or Gygabyte or whatever - the margin between 280 and 760 will shrink, turning it into a battle of power an acoustics, since you can get custom 280 and custom 760 for comparably the same price.

    P.S Just made a check. In most big on-line retailers in Netherlands, Germany and Belgium, the SAPPHIRE DUAL-X R9 280 3GB GDDR5 OC w. Boost cards are actually more expensive than OCed 760s.

    SAPPHIRE DUAL-X R9 280 3GB GDDR5 OC - 270 EUR

    ASUS R9280-DC2T-3GD5 - 245 EUR and it is clocked 40 Mhz higher

    GIGABYTE GV-N760OC-2GD - 230 EUR and it is clocked 100 Mhz and 120 Mhz more with Boost compared to reference model. That is more than 10% over-clock and it is enough to surpass the OCed 280.
    3
  • Quote:
    What I mean is - this is OC 280 vs Stock 760. If you get a OC model of a 760 from Asus, EVGA or Gygabyte or whatever - the margin between 280 and 760 will shrink, turning it into a battle of power an acoustics, since you can get custom 280 and custom 760 for comparably the same price. P.S Just made a check. In most big on-line retailers in Netherlands, Germany and Belgium, the SAPPHIRE DUAL-X R9 280 3GB GDDR5 OC w. Boost cards are actually more expensive than OCed 760s. SAPPHIRE DUAL-X R9 280 3GB GDDR5 OC - 270 EUR ASUS R9280-DC2T-3GD5 - 245 EUR and it is clocked 40 Mhz higher GIGABYTE GV-N760OC-2GD - 230 EUR and it is clocked 100 Mhz and 120 Mhz more with Boost compared to reference model. That is more than 10% over-clock and it is enough to surpass the OCed 280.


    I agree that the 760 reference was a little misleading along with not including the power and temperature measures from the 760 as well.
    0
  • 1063134 said:
    I agree with Shneiky. The small margins by which the 280 edges out it's win could easily be explained away by the fact that it's an aftermarket card vs reference cards.


    You can't buy a reference card on Newegg anymore. All the Radeon R9 280's come with custom coolers.

    There are still a lot of 760's with reference coolers available, however.

    Nevertheless, you guys are probably picking nits and missing the point. The margin of win/lose is so close that it's irrelevant, as I point out in the conclusion page. A random sample of different benchmarked games might show 760 strengths better than the newest titles that we chose for relevancy.

    The point is, they're so close it's almost irrelevant which you choose. Pick based on price and your preferred features, or your prevelance for Radeons or GeForces.

    As far as power/noise measurements, we'll be digging into that much deeper in an upcoming Radeon review. ;)
    0
  • Quote:
    ill wait for tonga


    I'm curious. Why? There seems to be no advantage to the R9-285 at all except lower draw power. That's it. Who cares about a higher clock on the memory when its gimped by a 256 bit memory bus. Do the numbers.

    You can see 176 GB/s memory bandwidth vs 240 from the Tahiti. Both have the same exact number of shader cores and Tahiti also has a slightly higher base clock. Seems to me that Tonga at the same price point has no benefit whatsoever. Don't understand this card.
    0
  • Quote:
    You can't buy a reference card on Newegg anymore. All the Radeon R9 280's come with custom coolers. There are still a lot of 760's with reference coolers available, however. Nevertheless, you guys are probably picking nits and missing the point. The margin of win/lose is so close that it's irrelevant, as I point out in the conclusion page. A random sample of different benchmarked games might show 760 strengths better than the newest titles that we chose for relevancy. The point is, they're so close it's almost irrelevant which you choose. Pick based on price and your preferred features, or your prevelance for Radeons or GeForces. As far as power/noise measurements, we'll be digging into that much deeper in an upcoming Radeon review. ;)


    I understand and acknowledge that. The point I was making is that you ( perhaps inadvertently ) put this 7950 Boost rebrand in the best possible light by using an aftermarket cooled and OC'd version against the stock clocked and cooled competition. But since there are no reference models of the card available , that may have been AMD's aim - to put this rebrand in the best case scenario. But you the reviewers should try and level the playing field.

    Honestly it's not a big deal. I understand where you're coming from , it's just my opinion is all :D
    0
  • 1063134 said:
    Honestly it's not a big deal. I understand where you're coming from , it's just my opinion is all :D


    I hear you, man.

    Like I said, we've got a thorough comparison coming in the near future with more details. This was more of a goodbye to the 280, really, which i believe will be a better option than the 285 unless you *REALLY* want to save power.
    0
  • 21257 said:
    As far as power/noise measurements, we'll be digging into that much deeper in an upcoming Radeon review. ;)


    That's what I am talking about. To me efficiency is nearly as important as performance assuming a near equal price and close performance. Even 10% less performance at 60 FPS is only 54 FPS something I can't really notice without fraps on.

    I would like to see how the new cards are stacking up to the GTX 7 series though. Since the GTX 760's have seemed to have not fallen in price I might think about selling mine off to get something new.
    1
  • you guys can wait and I will enjoy my fully under warranty msi twin frzrs i picked up off of ebay for $125 a piece.

    btw i already sent one into MSI and within 10 days it came back fully repaired. my daughter (autistic) opened up my rig and dumped juice on the GPU and MSI still replaced it without conflict or headache (I did clean with 99% alcohol). they have just earned my money for future purchases
    0