|
|
| Introduction and Related Info | |||||||||||||
|
About six weeks ago, after an extended development cycle, AMD launched their latest flagship ATI-GPU based graphics card, the Radeon HD 2900 XT. The Radeon HD 2900 XT is built around the company's unified, DirectX 10-class, R600 GPU and offers a number of key features over an above the older X1000 series of products, like new anti-aliasing modes and support for shader model 4.0. At the time of the HD 2900 XT's debut, AMD also disclosed a multitude of details regarding mobile and mainstream GPUs derived from the R600 architecture. In our coverage of the Radeon HD 2000 series as it became known, we talked about not only the Radeon HD 2900 XT, but five other members of the family, including the Radeon HD 2600 XT (GDDR3 and GDDR4 versions), the 2600 Pro, and the Radeon HD 2400 XT and Pro. Unfortunately, cards weren’t ready in time to launch alongside the 2900 XT, but they are now and we’ve got a trio of them in house for a benchmarking throw-down. On the pages ahead, we’ll go into detail on the new Radeon HD 2600 XT GDDR4, the Radeon HD 2600 Pro, and the Radeon HD 2400 XT, and compare their performance to a number of current and previous generation cards. But first, here are the Radeon HD 2600 and HD 2400 series’ features and specifications, as per the folks at ATI...
There is some pertinent information related to today's launch available on our site that we recommend you read, to get familiar with ATI's R600 GPU, their previous GPU architectures, and their key features. The Radeon HD 2600 XT, 2600 Pro, and 2400 XT are derivatives of the R600, and such they have a number of key features in common that we've already covered in much greater detail. The article we suggest you peruse include:
If you haven't already done so, we recommend scanning through our 2900 XT / R600 coverage, our CrossFire Multi-GPU technology preview, the Radeon X1950 Pro with Native CrossFire article, and the X1K family review. In those four pieces, we cover a large number of the features offered by the new Radeon HD 2600 / 2400 series and explain many of benefits of DirectX 10. We recommended reading these articles because there is quite a bit of background information in them that'll lay the foundation for what we're going to showcase here today. |
| Our Test Systems and 3DMark06 | ||||||||||||||
|
HOW WE CONFIGURED THE TEST SYSTEMS: We tested all of the graphics cards used in this article on either an EVGA nForce 680i SLI motherboard (NVIDIA GPUs) or an Intel D975XBX2 board (ATI GPUs) powered by a Core 2 Extreme X6800 dual-core processor and 2GB of low-latency Corsair RAM. The first thing we did when configuring the test system was enter the BIOS and set all values to their default settings. Then we manually configured the memory timings and disabled any integrated peripherals that wouldn't be put to use. The hard drive was then formatted, and Windows XP Pro with SP2 and the June '07 DX redist was installed. When the installation was complete, we then installed the latest chipset drivers available, installed all of the other drivers necessary for the rest of our components, and removed Windows Messenger from the system. Auto-Updating and System Restore were also disabled, the hard drive was defragmented, and a 1024MB permanent page file was created on the same partition as the Windows installation. Lastly, we set Windows XP's Visual Effects to "best performance," installed all of the benchmarking software, and ran the tests.
We've broken up our graphs into singe- (top) and multi-GPU (bottom) configurations in an effort to make them easier to read. As you can see, according to 3DMark06, the new Radeon HD 2600 and 2400 cards trail most of other cards we tested. The 2600 XT managed to pull ahead of the 8600 GT by a small margin, but the 2600 Pro and 2400 XT couldn't quite keep up.
If we tunnel down deeping into the 3DMark06 results, we can see just where the Radeon HD 2600 and 2400 cards excelled, or faltered. Their performance in the shader model 2.0 tests were well behind the rest of the cards, but in the more taxing shader model 3.0 / HDR tests they did much better, relatively speaking. The 2600 XT was able to outpace everything but the 7950 GT and 8800 GTS in the SM 3.0 / HDR tests. |
| HL2: Episode 1 | ||||||
|
The new Radeon HD 2600 XT and Pro, and 2400 XT cards somewhat struggled with our custom Half Life 2: Episode 1 benchmark. The Radeon HD 2600 XT performed about on-par or just slightly ahead of the GeForce 8600 GT, with the 2600 Pro, and 2400 XT trailing behind by a few frames per second. Linking a pair of 2600 XT or 2400 XT cards in a CrossFire configuration resulted in some nice performance gains. In fact, the 2600 XT CrossFire configuration was able to pull ahead of a pair of GeForce 8600 GTS cards at the higher resolution. |
| F.E.A.R. Performance | ||||||
|
Running the F.E.A.R. benchmark with high levels of anti-alising and anisotropic filtering with the game configured in its highest quality mode proved to be too much for these new mainstream Radeon HD 2600 and 2400 series cards. Here, whether running in a single-card or CrossFire configuration, the new Radeons couldn't keep with any of the competition. |
| Quake 4 Performance | ||||||
|
The results from our custom Quake 4 benchmark tell essentially the same story as the F.E.A.R. results on the previous page. The Radeon HD 2600 XT, 2600 Pro, and 2400 XT trailed the competition here, whether running in single-card or CrossFire mode. |
| Prey Performance | ||||||
|
As you'd expect, considering both games are based on the same underlying engine, the new Radeon HD 2600 XT's and Pro's, and 2400 XT's performance in our custom Prey benchmark mirror those of Quake 4. The Radeon HD 2600 XT comes withing striking distance of the 8600 GT, still can't quite catch the GeForce. |
| S.T.A.L.K.E.R. Performance | ||||||
|
We saw more of the same with our custom S.T.A.L.K.E.R. benchmark. Here, the Radeons showed good performance scaling when switching from singe- to dual-cards, but once again their performance wasn't quite on the level of the entry level GeForce cards nor the previous generation X1950 Pro. |
| Overclocking the 2600 and 2400 | ||||
|
For our next set of performance metrics, we spent some time overclocking the Radeon HD 2600 XT, 2600 Pro, and 2400 XT using the clock frequency sliders available within ATI's Catalyst drivers, under the "Overdrive" menu.
To find each card's peak core and memory frequencies, we used the automated utility built-into Overdrive and then tried to raise the GPU and memory frequencies further until our test system was no longer stable.
When all was said and done, we were able to take the Radeon HD 2600 XT up from its stock GPU and memory speeds of 800MHz and 1100MHz, respectively, to 855MHz and 1175MHz. The 2600 Pro was able to hit a GPU speed of 660MHz and a memory speed of 550MHz, and the 2400 XT peaked at 700MHz for the GPU and 900MHz for the memory. While each card was overclocked, we re-ran a couple of high-resolution benchmarks and saw marginal gains of about 0% to 9%. |
| Power Consumption and Noise | ||||
|
Before we bring this article to a close, we'd like to cover a few final data points. Throughout all of our benchmarking and testing, we monitored how much power our test system was consuming using a power meter and also took some notes regarding its noise output as reported by our digital sound level meter. Our goal was to give you all an idea as to how much power each configuration used and to explain how loud the configurations were under load. Please keep in mind that we were testing total system power consumption at the outlet here, not just the power being drawn by the video cards alone.
Last month, when ATI first disclosed details regarding the Radeon HD 2600 and 2400 cards, they claimed power consumption would be very low with these GPUs, and it turns out they weren't kidding. While idling and while under a full 3D workload, all three of the new Radeon HD cards we tested proved to sip power in comparison their competition. Then again, their performance levels in our game tests were usually lower as well, so they should consume less power. With these relatively low power requirements, we suspect the mobile versions of these GPUs are going to be quite popular with ODMs. We should also note that all three of these new mainstread Radeon HD 2000 series cards remained nice and quiet throughout our testing. From about 1 foot away, our entire test rig generated between 47 and 49dB of noise. And the fans on all of the cards never spun up to full speed - even after hours and hours of testing. |
| Our Summary and Conclusion | ||||
|
Performance Summary: Throughout all of our in-game and synthetic testing the fastest of the three new mainstream Radeon HD 2000 series cards we tested in this article, the Radeon HD 2600 XT, performed about on par with or markedly behind a GeForce 8600 GT. The more affordable Radeon 2600 Pro came in a few percentage points behind the 2600 XT, and as expected the 2400 XT fell in behind the 2600 Pro. We also spent some time testing the AVIVO HD video engine in these new cards with a few SD and HD workloads, but weren’t able to compile all of the data in time for launch. We will be updating this article in the next day or so, with the results from our AVIVO HD testing as well.
ATI expects the new Radeon HD 2600 XT, 2600 Pro, and 2400 XT cards to hit store shelves in the next 2 -3 weeks; sometime in mid-July. The latest information we have on pricing puts various flavors of the 2600 XT in the $119 - $149 price range, the 2600 Pro in the $89 - $99 range, the 2400 XT in the $75 - $85 range, and the 2400 Pro (not tested here) in the $50 - $55 range. If history is an indicator, however, expect actual street prices for these cards to be about 5% to 15% higher than these suggested prices for a while. For example, Radeon HD 2900 XT cards have just recently begun selling at their MSRP of $399 after about 6 weeks on the market. Overall, the new Radeon HD 2600 XT, 2600 Pro, and 2400 XT cards should make for quiet, low-power upgrades from any integrated graphics solution and offer a relatively low-cost of entry into the world of DirectX 10. These cards are obviously not geared to hardcore gamers, but at lower resolutions without high levels of AA and anisotropic filtering enabled they’ll be adequate for casual gaming. These cards are also well suited to HTPC applications where video playback performance and low-noise output are of the utmost importance.
|