Latest News Posts

Social
Latest Forum Posts

The $109 Console-killer GPU: AMD’s Radeon R7 260 Graphics Card Reviewed
Bookmark and Share

AMD Radeon R7 260
Print
by Rob Williams on December 23, 2013 in AMD-Based GPU

No one should be surprised at the fact that testing out $500 graphics cards is fun, but with the right perspective, budget cards can be, too. Take the $109 AMD Radeon R7 260, for example, which has debuted following flagship console releases. With that in mind, let’s see what such an affordable GPU can pull off at the much-loved 1080p resolution.

Introduction

I can’t remember the last time I took a look at a graphics card that hovered around the $100 price-point, but for a couple of reasons, I couldn’t wait to dig into AMD’s $109 Radeon R7 260.

For starters, with the roll-out of the “next-generation” consoles having wrapped-up, my interest in modestly priced GPUs has recently piqued. As I’m sure is clear by now, neither the Xbox One or PlayStation 4 offer the sort of graphics we envisioned a couple of years ago for a “next-gen” model. When “price” and “size” are the major focuses during development, compromises have to be made, and those have been made quite obvious here.

Tying into that, I often see console fans around the Web claim that PC gaming is “expensive”. For a solid build, that’s true; the up-front cost will be more. But to compare the PC as a platform to consoles in an apples-to-apples manner is a bit unfair. Unlike consoles, PCs are part of an open platform – they can be upgraded, and otherwise poked and prodded in ways consoles can’t be (take game mods, for example). So, while a good PC might cost more than a console, it’ll offer a much higher level of flexibility.

AMD Radeon R7 260

For those reasons, AMD’s $109 R7 GPU comes at a great time, as it allows us to see what a truly affordable card can pull off. All of our benchmarks focus on the 1080p resolution, which is notable since many blockbuster games on the latest consoles are limited to 720p to preserve smooth performance.

There’s a Techgage reason for AMD’s R7 260 dropping at a good time, as well: This is the first GPU review where we’ve implemented a proper “Best Playable” feature. While we’ve dabbled with this idea in the past, we’ve worked over the mechanics and plan to include it in all of our future GPU reviews.

The reason this feature is important: Our goal is to find the best graphics configuration that will allow a game to run at 1080p with an average performance of 60 FPS – or at leastĀ close to it. And because a visual cue has never hurt anyone, we’ll be providing an ingame screenshot for each title which was snapped using the same graphics settings we provide. With these screenshots, you’ll be able to see exactly what you’d be getting, all with the realization that you’d be enjoying the game at a cool 60 FPS.

I’ll explain the feature a bit more on that page, but for now, let’s take a quick look at AMD’s current-gen lineup:

AMD Radeon Cores Core MHz Memory Mem MHz Mem Bus TDP Price
R9 290X 2816 1000 4096MB 5000 512-bit 250W $549
R9 290 2560 947 4096MB 5000 512-bit 250W $399
R9 280X 2048 <1000 3072MB 6000 384-bit 250W $299
R9 270X 1280 <1050 2048MB 5600 256-bit 180W $199
R9 270 1280 <925 2048MB 5600 256-bit 150W $179
R7 260X 896 <1100 2048MB 6500 128-bit 115W $139
R7 260 768 <1000 1024MB 6000 128-bit 95W $109
R7 250 384 <1050 1024MB 4600 128-bit 65W $89

The R7 260 utilizes the same architectural design as the R7 260X, but a couple of important changes have been made. The biggest one is a drop of GDDR5 density, from 2GB to 1GB. I don’t like seeing 1GB framebuffers in this day and age, so a secondary goal while testing will be seeing how much of a detriment it proves to be at our target resolution of 1080p.

The other changes include a slight drop in cores (from 896 to 768), and a drop in clock speed (from 1.1GHz to 1.0GHz). As would be expected, these changes also result in a lower power draw; -20W in this particular instance, resulting in a rating of 95W.

Thanks to the introduction of our Best Playable feature (and the swapping out of two games), I’ve had to start the process of rebenchmarking our current fleet of GPUs. So for this review, I’m just going to have results from both of these cards in question, which is all well and good given the next step up GPU-wise would bring us to the $200 price-point (I do not have an NVIDIA card on-hand which is priced like either of these two AMD cards, and current-gen models that would beĀ suitable are not available).

With all of that said, let’s proceed to our testing methodology page, and then get right into testing.


  • http://www.youtube.com/user/no6969el Noel Barcellos

    Not sure if the 260 has support for mantle like you suggested at the end of your article. Other than that great set of information, I think I might just get one for my lady.

    • http://techgage.com/ Rob Williams

      Here’s a slide out of the press deck showing support for Mantle on this GPU. It’s a smart move by AMD… it’s frustrating when companies lob off technologies like these on lower-end parts.

    • RainMotorsports

      GCN is 7730 on up and not sure what GCN version Mantle “requires” 7770 is 1.0 7790 is 1.1. But the R7 240 and above should support mantle if I am not mistaken.

  • christianh

    As I say on every review site… Why test this level card with such a high-priced CPU…?

    It will still be GPU limited with an i5 or Richland…

    • http://techgage.com/ Rob Williams

      The other option would be using a modest CPU and risking the introduction of some bottleneck. That’s not the point here. The PC itself is kept 100% identical; all that changes is the GPU. Which is important, because that’s the focus.

    • RainMotorsports

      In GPU testing nothing but the GPU and driver should ever change. If he swaps the CPU out the results should be considered invalid no matter what any other data says otherwise. Its not very scientific to change static variables.