[nextpage title=”Introduction”]
Radeon HD 2600 XT is the new mid-range video card from AMD/ATI, supporting Shader 4.0 unified architecture (i.e., DirectX 10) and competing directly with GeForce 8600 GT from NVIDIA, both costing around USD 150. In this review we will compare Radeon HD 2600 XT to GeForce 8600 GT and also to several other mid-range boards from both ATI and NVIDIA. Check it out.
Figure 1: AMD reference model for ATI Radeon HD 2600 XT.
The main difference between Radeon X1000 and the new Radeon HD 2000 families is the adoption of DirectX 10 on Radeon HD 2000 family. What this means is that they will support the next generation of games to be released starting this year. It also means that instead of using separated shader units for each kind of shader processing (pixel, vertex, physics and geometry) video cards from this family use a unified shader architecture, where the shader engines can process any one of these tasks. On NVIDIA side, this unified architecture is used on GeForce 8 family – of course implemented differently inside the chip. On our ATI Radeon HD 2000 Series Architecture article you can find a more in-depth explanation about this.
There are two big problems with Radeon HD 2600 XT – and, for that matter, also with other members of the Radeon HD 2600 and HD 2400 families.
First, it isn’t available on the market yet. During AMD’s technical sessions to present Radeon HD 2000 to the media in April, they said that Radeon HD 2600 and Radeon HD 2400 would be available in late June. All we got was a paper launch on June 28th, 2007 announcing the two families, however we are already in July and no sign of these cards on the market. Unfortunately this is an old habit from ATI that AMD still couldn’t get rid off yet. According to AMD, Radeon HD 2600 XT will reach the market 2 weeks after its paper launch, so if everything goes fine we should see this video card on the stores by the end of the next week.
The second problem is that AMD there are two basic versions of Radeon HD 2600 XT, one using GDDR4 memories running at 2.2 GHz (1.1 GHz DDR) and another using GDDR3 memories running at 1.6 GHz (800 MHz DDR). And for Radeon HD 2400 they allow the manufacturer to set any clock rate they want. The problem is that the reference models we got for reviewing were running at the top clock speeds and other Radeon HD 2600 and HD 2400 boards you will find at the market may be running at lower clock rates and thus achieving a lower performance.
We think this is really nasty to the customer, as customers can be easily deceived. One can buy a Radeon HD 2600 XT card based on a review to discover later that the model he or she bought has a performance lower than the model featured on the review, for example.
We think that in name of transparency AMD should require all their partners to add some extra info on the product name to clarify what clocks and/or memory technology that particular product is using, for example Radeon HD 2600 XT xxx/yyy, where xxx is the chip clock and yyy is the memory clock. Or even use a different product name (e.g., 2600 XT for the model with GDDR3 and 2610 XT for the model with GDDR4). But of course we don’t live on a perfect world.
And another thing that is confusing is the future of ATI brand itself. All ATI employees are now officially AMD employees, the website is from AMD, AMD refers ATI products as being AMD products but they are still red and using ATI stickers.
On the Radeon HD 2600 XT reference board we got for reviewing the graphics chip was running at 800 MHz and had 256 MB GDDR4 memory chips running at 2.2 GHz (1.1 GHz x 2), using a 128-bit memory interface. This provides a maximum theoretical memory transfer rate of 35.20 GB/s. Here we must explain that both ATI and NVIDIA are now referring to their memory clocks with their correct clock rates, not the doubled clock rate – DDR and similar memory technologies work transferring two data per clock cycle and usually their clock rates are labeled with double their real clock rates. In order to make it easier to compare current products to older ones, we are keeping the old way to label memory clock rates.
As we mentioned, the suggested price for this video card is of USD 150, the same price range GeForce 8600 GT currently has. Another question is whether Radeon HD 2600 XT will really reach the market costing that. From our experience the prices of all products rise when they are launched, dropping after some weeks.
For a full comparison between Radeon HD 2600 XT and other chips from ATI, please read our tutorial ATI Chips Comparison Table. On NVIDIA Chips Comparison Table you can compare them to competitors from NVIDIA.
On Figures 2 and 3 you can see the reviewed reference model from AMD/ATI.
Figure 2: AMD reference model for ATI Radeon HD 2600 XT.
Figure 3: AMD reference model for ATI Radeon HD 2600 XT, back view.
[nextpage title=”ATI Radeon HD 2600 XT Reference Model”]
From the pictures posted in the previous page you could see that Radeon HD 2600 supports Native CrossFire technology.
On thing that is new on the Radeon HD 2000 series is the support for HDMI with digital audio on the same connector. Because of that Radeon HD 2600 XT video cards will come with a DVI-to-HDMI adapter. More information about this feature can be found on our ATI Radeon HD 2000 Series Architecture article.
You could also see from the pictures posted in the previous page, the reference model for Radeon HD 2600 XT uses a big cooler that dissipates the heat produced not only by the graphics chip and the memory chips, but also by the coils (chokes) found on the board. We removed this cooler to take a look and it has an aluminum body with a copper base, see Figure 4.
Figure 4: Cooler.
In Figure 5, you can see this video card without its cooler.
Figure 5: Radeon HD 2600 XT reference model without its cooler.
This video card uses four GDDR4 512-Mbit 0.91 ns chips from Samsung (K4U52324QE-BC09), making the 256 MB of memory this video card has. These chips can officially run up to 2.2 GHz (1.1 GHz x 2) and since on this video card the memory chips are already running at 2.2 GHz there is no headroom for overclocking the memory inside its specs. Of course you can always try to push the memory clock above the memory chip maximum official clock rate.
[nextpage title=”Main Specifications”]
- Graphics chip: Radeon HD 2600 XT, running at 800 MHz.
- Memory: 256 MB GDDR4 memory (0.91 ns, 128-bit interface) from Samsung (K4U52324QE-BC09), running at 2.2 GHz (1.1 GHz DDR).
- Bus type: PCI Express x16.
- Connectors: Two DVI and one S-Video output.
- Video Capture (VIVO): No.
- Number of CDs/DVDs that come with this board: N/A.
- Games that come with this board: N/A.
- Programs that come with this board: N/A.
- More information: https://ati.amd.com
- Suggested price for the US market: USD 150.00
[nextpage title=”How We Tested”]
During our benchmarking sessions, we used the configuration listed below. Between our benchmarking sessions the only different device was the video card being tested.
Hardware Configuration
- Motherboard: ASUS P5B (Intel P965, 0904 BIOS)
- CPU: Core 2 Extreme X6800 (dual-core, 2.93 GHz)
- CPU Cooler: Gigabyte Neon 775-BL
- Memory: 2 GB PC-1066/PC2-8500 (Corsair TWIN2X2048-8500C5 kit), configured at 1,066 MHz with 5-5-5-15 timings.
- Hard Drive: Samsung HD080HJ (80 GB, SATA-300, 8 MB buffer, 7,200 rpm)
- Power Supply: Zalman ZM-600HP
- Video Monitor: Samsung SyncMaster 1100MB
- Screen resolution: 1280x1024x32@85 Hz
Software Configuration
- Windows XP Professional installed using NTFS
- Service Pack 2
- Direct X 9.0c
- Intel inf driver version: 8.0.1.1002
- ATI video driver version: Catalyst 8.3 (Radeon HD 2600 and HD 2400)
- ATI video driver version: Catalyst 7.2 (all other ATI video cards)
- NVIDIA video driver version: 93.71 (GeForce 6 and 7 Family)
- NVIDIA video driver version: 158.22 (GeForce 8 Family)
Used Software
- 3DMark03 Professional Edition 3.6.0
- 3DMark06 Professional Edition 1.10
- Battlefield 2142 1.01
- Far Cry 1.4 with HardwareOC Far Cry Benchmark 1.7
- F.E.A.R. 1.08
- Quake 4 1.3
We adopted a 3% error margin; thus, differences below 3% cannot be considered relevant. In other words, products with a performance difference below 3% should be considered as having similar performance.
[nextpage title=”3DMark03″]
3DMark03 simulates DirectX 8 and 9 games. Even though this program may be considered “old”, we ran it to see how the tested video cards perform on older games. Since we are comparing mid-range cards, we decided to run this program in two resolutions, 1024×768 (which is considered low for today’s standards and thus providing a simulation for low resolution) and 1600×1200 (which provides a simulation for high resolution). At each resolution we simulated two scenarios, first with no image quality enhancements enabled (this scenario we called “low”) and then with anti-aliasing set at 4x and anisotropic filtering set at 4x (this scenario we called “high”). The results you check below. All video cards listed below were running with the default clock rates defined by the chip manufacturer.
| 3DMark03 Professional Edition 3.6.0 – 1024×768 | Score | Difference |
| GeForce 8600 GTS 256 MB (MSI) | 18236 | 26.54% |
| Radeon X1950 Pro 256 MB (PowerColor) | 18175 | 26.12% |
| GeForce 8600 GT 256 MB (MSI) | 15741 | 9.23% |
| GeForce 7600 GT 256 MB (XFX) | 14487 | 0.53% |
| Radeon HD 2600 XT 256 MB (AMD) | 14411 | |
| Radeon X1650 XT 256 MB (HIS) | 13663 | 5.47% |
| Radeon X1650 Pro 256 MB (HIS) | 10574 | 36.29% |
| Radeon X1600 XT 256 MB (HIS) | 10271 | 40.31% |
| GeForce 6600 GT 128 MB (NVIDIA) | 9557 | 50.79% |
| Radeon HD 2600 Pro 256 MB (AMD) | 9278 | 55.32% |
| GeForce 7600 GS 256 MB (XFX) | 8984 | 60.41% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 8935 | 61.29% |
| GeForce 8500 GT 256 MB (MSI) | 8124 | 77.39% |
| GeForce 7300 GT 256 MB (Zogis) | 7359 | 95.83% |
| Radeon X1300 Pro 256 MB (ATI) | 6385 | 125.70% |
| Radeon HD 2400 XT 128 MB (AMD) | 5953 | 142.08% |
| GeForce 6600 128 MB (Albatron) | 5593 | 157.66% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 4179 | 244.84% |
| 3DMark03 Professional Edition 3.6.0 – 1600×1200 | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 11222 | 52.64% |
| GeForce 8600 GTS 256 MB (MSI) | 10538 | 43.34% |
| GeForce 8600 GT 256 MB (MSI) | 8797 | 19.65% |
| GeForce 7600 GT 256 MB (XFX) | 8450 | 14.93% |
| Radeon X1650 XT 256 MB (HIS) | 7556 | 2.77% |
| Radeon HD 2600 XT 256 MB (AMD) | 7352 | |
| Radeon X1650 Pro 256 MB (HIS) | 5429 | 35.42% |
| Radeon X1600 XT 256 MB (HIS) | 5385 | 36.53% |
| GeForce 6600 GT 128 MB (NVIDIA) | 5255 | 39.90% |
| GeForce 7600 GS 256 MB (XFX) | 5086 | 44.55% |
| Radeon HD 2600 Pro 256 MB (AMD) | 4718 | 55.83% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 4497 | 63.49% |
| GeForce 8500 GT 256 MB (MSI) | 4313 | 70.46% |
| GeForce 7300 GT 256 MB (Zogis) | 4011 | 83.30% |
| Radeon HD 2400 XT 128 MB (AMD) | 3009 | 144.33% |
| GeForce 6600 128 MB (Albatron) | 2915 | 152.21% |
| Radeon X1300 Pro 256 MB (ATI) | 2712 | 171.09% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 2076 | 254.14% |
| 3DMark03 Professional Edition 3.6.0 – 1024×768 – AAx4, AFx4 | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 11262 | 85.63% |
| GeForce 8600 GTS 256 MB (MSI) | 10742 | 77.06% |
| GeForce 8600 GT 256 MB (MSI) | 9119 | 50.30% |
| GeForce 7600 GT 256 MB (XFX) | 7980 | 31.53% |
| Radeon X1650 XT 256 MB (HIS) | 7602 | 25.30% |
| Radeon HD 2600 XT 256 MB (AMD) | 6067 | |
| Radeon X1650 Pro 256 MB (HIS) | 5592 | 8.49% |
| Radeon X1600 XT 256 MB (HIS) | 5396 | 12.44% |
| GeForce 8500 GT 256 MB (MSI) | 4873 | 24.50% |
| GeForce 6600 GT 128 MB (NVIDIA) | 4844 | 25.25% |
| GeForce 7600 GS 256 MB (XFX) | 4635 | 30.90% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 4605 | 31.75% |
| GeForce 7300 GT 256 MB (Zogis) | 3894 | 55.80% |
| Radeon HD 2600 Pro 256 MB (AMD) | 3704 | 63.80% |
| Radeon X1300 Pro 256 MB (ATI) | 3376 | 79.71% |
| Radeon HD 2400 XT 128 MB (AMD) | 2699 | 124.79% |
| GeForce 6600 128 MB (Albatron) | 2617 | 131.83% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 1745 | 247.68% |
| 3DMark03 Professional Edition 3.6.0 – 1600×1200 – AAx4, AFx4 | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 6472 | 105.53% |
| GeForce 8600 GTS 256 MB (MSI) | 5612 | 78.22% |
| GeForce 8600 GT 256 MB (MSI) | 4639 | 47.32% |
| GeForce 7600 GT 256 MB (XFX) | 4276 | 35.79% |
| Radeon X1650 XT 256 MB (HIS) | 4110 | 30.52% |
| Radeon HD 2600 XT 256 MB (AMD) | 3149 | |
| Radeon X1650 Pro 256 MB (HIS) | 2714 | 16.03% |
| Radeon X1600 XT 256 MB (HIS) | 2691 | 17.02% |
| GeForce 7600 GS 256 MB (XFX) | 2320 | 35.73% |
| GeForce 6600 GT 128 MB (NVIDIA) | 2237 | 40.77% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 2217 | 42.04% |
| GeForce 8500 GT 256 MB (MSI) | 2162 | 45.65% |
| GeForce 7300 GT 256 MB (Zogis) | 1917 | 64.27% |
| Radeon HD 2600 Pro 256 MB (AMD) | 1903 | 65.48% |
| Radeon HD 2400 XT 128 MB (AMD) | 1362 | 131.20% |
| GeForce 6600 128 MB (Albatron) | 1212 | 159.82% |
| Radeon X1300 Pro 256 MB (ATI) | 1038 | 203.37% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 837 | 276.22% |
3DMark06 simulates DirectX 9.0c (Shader 3.0) games and it also puts HDR (High Dynamic Range) into the equation to calculate its final score. So it simulates the most high-end games available today. Since we were comparing mid-range cards, we ran this program in two resolutions, one low (1024×768) and one high (1600×1200). The results you check below. All video cards listed below were running with the default clock rates defined by the chip manufacturer.
| 3DMark06 Professional Edition 1.10 – 1024×768 | Score | Difference |
| GeForce 8600 GTS 256 MB (MSI) | 7002 | 15.74% |
| Radeon X1950 Pro 256 MB (PowerColor) | 6198 | 2.45% |
| Radeon HD 2600 XT 256 MB (AMD) | 6050 | |
| GeForce 8600 GT 256 MB (MSI) | 5743 | 5.35% |
| Radeon X1650 XT 256 MB (HIS) | 4292 | 40.96% |
| Radeon HD 2600 Pro 256 MB (AMD) | 4194 | 44.25% |
| GeForce 7600 GT 256 MB (XFX) | 4192 | 44.32% |
| Radeon X1650 Pro 256 MB (HIS) | 3369 | 79.58% |
| Radeon X1600 XT 256 MB (HIS) | 3295 | 83.61% |
| GeForce 8500 GT 256 MB (MSI) | 2927 | 106.70% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 2860 | 111.54% |
| Radeon HD 2400 XT 128 MB (AMD) | 2800 | 116.07% |
| GeForce 7600 GS 256 MB (XFX) | 2763 | 118.96% |
| GeForce 6600 GT 128 MB (NVIDIA) | 2213 | 173.38% |
| GeForce 7300 GT 256 MB (Zogis) | 2046 | 195.70% |
| Radeon X1300 Pro 256 MB (ATI) | 1920 | 215.10% |
| GeForce 6600 128 MB (Albatron) | 1357 | 345.84% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 741 | 716.46% |
| 3DMark06 Professional Edition 1.10 – 1600×1200 | Score | Difference |
| GeForce 8600 GTS 256 MB (MSI) | 4607 | 14.35% |
| Radeon X1950 Pro 256 MB (PowerColor) | 4496 | 11.59% |
| Radeon HD 2600 XT 256 MB (AMD) | 4029 | |
| GeForce 8600 GT 256 MB (MSI) | 3707 | 8.69% |
| Radeon X1650 XT 256 MB (HIS) | 2923 | 37.84% |
| GeForce 7600 GT 256 MB (XFX) | 2765 | 45.71% |
| Radeon HD 2600 Pro 256 MB (AMD) | 2738 | 47.15% |
| Radeon X1650 Pro 256 MB (HIS) | 2117 | 90.32% |
| Radeon X1600 XT 256 MB (HIS) | 2086 | 93.14% |
| GeForce 7600 GS 256 MB (XFX) | 1796 | 124.33% |
| GeForce 8500 GT 256 MB (MSI) | 1784 | 125.84% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 1781 | 126.22% |
| Radeon HD 2400 XT 128 MB (AMD) | 1681 | 139.68% |
| GeForce 7300 GT 256 MB (Zogis) | 1281 | 214.52% |
| Radeon X1300 Pro 256 MB (ATI) | 1100 | 266.27% |
[nextpage title=”Quake 4″]
Quake 4 uses the same game engine as Doom 3 and since we are comparing mid-range cards, we decided to run this program in two resolutions, 1024×768 (simulating a low resolution) and 1600×1200 (simulating a high resolution), first with image quality set at “low” and then with image quality set at “high”. We upgraded this game to version 1.3 and run the id_demo001 net demo that comes with this version. Click here for more details on how to use Quake 4 to benchmark a system. The results you check below and are given in frames per second. All video cards listed below were running with the default clock rates defined by the chip manufacturer.
| Quake 4 1.3 – 1024×768 – low | Score | Difference |
| GeForce 8600 GTS 256 MB (MSI) | 123.21 | 21.52% |
| GeForce 8600 GT 256 MB (MSI) | 114.53 | 12.96% |
| GeForce 7600 GT 256 MB (XFX) | 109.14 | 7.64% |
| Radeon X1950 Pro 256 MB (PowerColor) | 106.73 | 5.27% |
| Radeon HD 2600 XT 256 MB (AMD) | 101.39 | |
| Radeon X1650 XT 256 MB (HIS) | 85.44 | 18.67% |
| GeForce 7600 GS 256 MB (XFX) | 79.65 | 27.29% |
| GeForce 6600 GT 128 MB (NVIDIA) | 77.86 | 30.22% |
| Radeon HD 2600 Pro 256 MB (AMD) | 76.90 | 31.85% |
| GeForce 7300 GT 256 MB (Zogis) | 74.81 | 35.53% |
| Radeon X1650 Pro 256 MB (HIS) | 66.54 | 52.37% |
| GeForce 8500 GT 256 MB (MSI) | 61.90 | 63.80% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 55.69 | 82.06% |
| GeForce 6600 128 MB (Albatron) | 46.40 | 118.51% |
| Radeon HD 2400 XT 128 MB (AMD) | 42.63 | 137.84% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 35.03 | 189.44% |
| Quake 4 1.3 – 1600×1200 – low | Score | Difference |
| GeForce 8600 GTS 256 MB (MSI) | 77.30 | 43.49% |
| Radeon X1950 Pro 256 MB (PowerColor) | 67.88 | 26.01% |
| GeForce 7600 GT 256 MB (XFX) | 65.14 | 20.92% |
| GeForce 8600 GT 256 MB (MSI) | 61.34 | 13.87% |
| Radeon HD 2600 XT 256 MB (AMD) | 53.87 | |
| Radeon X1650 XT 256 MB (HIS) | 44.18 | 21.93% |
| GeForce 7600 GS 256 MB (XFX) | 41.01 | 31.36% |
| GeForce 6600 GT 128 MB (NVIDIA) | 39.59 | 36.07% |
| Radeon HD 2600 Pro 256 MB (AMD) | 37.66 | 43.04% |
| GeForce 7300 GT 256 MB (Zogis) | 37.63 | 43.16% |
| Radeon X1650 Pro 256 MB (HIS) | 33.65 | 60.09% |
| GeForce 8500 GT 256 MB (MSI) | 29.63 | 81.81% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 28.01 | 92.32% |
| GeForce 6600 128 MB (Albatron) | 23.02 | 134.01% |
| Radeon HD 2400 XT 128 MB (AMD) | 20.93 | 157.38% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 16.95 | 217.82% |
| Quake 4 1.3 – 1024×768 – high | Score | Difference |
| GeForce 8600 GTS 256 MB (MSI) | 120.14 | 29.18% |
| GeForce 8600 GT 256 MB (MSI) | 112.95 | 21.45% |
| GeForce 7600 GT 256 MB (XFX) | 110.23 | 18.53% |
| Radeon X1950 Pro 256 MB (PowerColor) | 100.20 | 7.74% |
| Radeon HD 2600 XT 256 MB (AMD) | 93.00 | |
| Radeon X1650 XT 256 MB (HIS) | 78.24 | 18.87% |
| GeForce 7600 GS 256 MB (XFX) | 75.27 | 23.56% |
| GeForce 6600 GT 128 MB (NVIDIA) | 71.89 | 29.36% |
| GeForce 7300 GT 256 MB (Zogis) | 70.76 | 31.43% |
| Radeon HD 2600 Pro 256 MB (AMD) | 69.01 | 34.76% |
| GeForce 8500 GT 256 MB (MSI) | 60.53 | 53.64% |
| Radeon X1650 Pro 256 MB (HIS) | 58.80 | 58.16% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 49.15 | 89.22% |
| GeForce 6600 128 MB (Albatron) | 42.87 | 116.93% |
| Radeon HD 2400 XT 128 MB (AMD) | 38.55 | 141.25% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 32.15 | 189.27% |
| Quake 4 1.3 – 1600×1200 – high | Score | Difference |
| GeForce 8600 GTS 256 MB (MSI) | 75.51 | 50.66% |
| Radeon X1950 Pro 256 MB (PowerColor) | 63.35 | 26.40% |
| GeForce 7600 GT 256 MB (XFX) | 62.23 | 24.16% |
| GeForce 8600 GT 256 MB (MSI) | 59.99 | 19.69% |
| Radeon HD 2600 XT 256 MB (AMD) | 50.12 | |
| Radeon X1650 XT 256 MB (HIS) | 41.05 | 22.10% |
| GeForce 7600 GS 256 MB (XFX) | 39.05 | 28.35% |
| GeForce 6600 GT 128 MB (NVIDIA) | 37.10 | 35.09% |
| GeForce 7300 GT 256 MB (Zogis) | 35.94 | 39.45% |
| Radeon HD 2600 Pro 256 MB (AMD) | 35.00 | 43.20% |
| Radeon X1650 Pro 256 MB (HIS) | 30.75 | 62.99% |
| GeForce 8500 GT 256 MB (MSI) | 28.93 | 73.25% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 25.62 | 95.63% |
| GeForce 6600 128 MB (Albatron) | 21.47 | 133.44% |
| Radeon HD 2400 XT 128 MB (AMD) | 19.56 | 156.24% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 15.76 | 218.02% |
[nextpage title=”F.E.A.R.”]
F.E.A.R. is a heavy game and we used its internal benchmarking module. We upgraded it to version 1.08 and since we are comparing mid-range cards, we decided to run this program in two resolutions, 1024×768 (simulating a low resolution) and 1600×1200 (simulating a high resolution). We set “computer settings” to “maximum” and then ran each resolution in two scenarios, first with “graphics card” set at “low” and then with this item set at “maximum”. Let’s take a look at the results, given in frames per second. All video cards listed below were running with the default clock rates defined by the chip manufacturer.
| F.E.A.R. 1.08 – 1024×768 – Low Quality | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 398 | 40.14% |
| Radeon HD 2600 XT 256 MB (AMD) | 284 | |
| Radeon X1650 XT 256 MB (HIS) | 268 | 5.97% |
| GeForce 8600 GTS 256 MB (MSI) | 247 | 14.98% |
| GeForce 8600 GT 256 MB (MSI) | 193 | 47.15% |
| GeForce 7600 GT 256 MB (XFX) | 186 | 52.69% |
| Radeon HD 2600 Pro 256 MB (AMD) | 182 | 56.04% |
| Radeon X1650 Pro 256 MB (HIS) | 167 | 70.06% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 139 | 104.32% |
| GeForce 7600 GS 256 MB (XFX) | 121 | 134.71% |
| GeForce 6600 GT 128 MB (NVIDIA) | 114 | 149.12% |
| Radeon HD 2400 XT 128 MB (AMD) | 109 | 160.55% |
| GeForce 8500 GT 256 MB (MSI) | 94 | 202.13% |
| GeForce 7300 GT 256 MB (Zogis) | 83 | 242.17% |
| GeForce 6600 128 MB (Albatron) | 66 | 330.30% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 38 | 647.37% |
| F.E.A.R. 1.08 – 1600×1200 – Low Quality | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 196 | 55.56% |
| Radeon HD 2600 XT 256 MB (AMD) | 126 | |
| Radeon X1650 XT 256 MB (HIS) | 123 | 2.44% |
| GeForce 8600 GTS 256 MB (MSI) | 115 | 9.57% |
| GeForce 8600 GT 256 MB (MSI) | 89 | 41.57% |
| GeForce 7600 GT 256 MB (XFX) | 89 | 41.57% |
| Radeon HD 2600 Pro 256 MB (AMD) | 83 | 51.81% |
| Radeon X1650 Pro 256 MB (HIS) | 75 | 68.00% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 62 | 103.23% |
| GeForce 7600 GS 256 MB (XFX) | 58 | 117.24% |
| GeForce 6600 GT 128 MB (NVIDIA) | 54 | 133.33% |
| Radeon HD 2400 XT 128 MB (AMD) | 49 | 157.14% |
| GeForce 8500 GT 256 MB (MSI) | 42 | 200.00% |
| GeForce 7300 GT 256 MB (Zogis) | 38 | 231.58% |
| GeForce 6600 128 MB (Albatron) | 31 | 306.45% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 17 | 641.18% |
| F.E.A.R. 1.08 – 1024×768 – Maximum Quality | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 77 | 126.47% |
| GeForce 8600 GTS 256 MB (MSI) | 62 | 82.35% |
| GeForce 8600 GT 256 MB (MSI) | 51 | 50.00% |
| Radeon X1650 XT 256 MB (HIS) | 50 | 47.06% |
| GeForce 7600 GT 256 MB (XFX) | 46 | 35.29% |
| Radeon X1650 Pro 256 MB (HIS) | 40 | 17.65% |
| Radeon HD 2600 XT 256 MB (AMD) | 34 | |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 33 | 3.03% |
| GeForce 8500 GT 256 MB (MSI) | 29 | 17.24% |
| GeForce 7600 GS 256 MB (XFX) | 27 | 25.93% |
| GeForce 6600 128 MB (Albatron) | 27 | 25.93% |
| GeForce 6600 GT 128 MB (NVIDIA) | 26 | 30.77% |
| GeForce 7300 GT 256 MB (Zogis) | 26 | 30.77% |
| Radeon HD 2600 Pro 256 MB (AMD) | 20 | 70.00% |
| Radeon HD 2400 XT 128 MB (AMD) | 13 | 161.54% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 10 | 240.00% |
| F.E.A.R. 1.08 – 1600×1200 – Maximum Quality | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 37 | 131.25% |
| GeForce 8600 GTS 256 MB (MSI) | 26 | 62.50% |
| Radeon X1650 XT 256 MB (HIS) | 24 | 50.00% |
| GeForce 7600 GT 256 MB (XFX) | 23 | 43.75% |
| GeForce 8600 GT 256 MB (MSI) | 22 | 37.50% |
| Radeon X1650 Pro 256 MB (HIS) | 17 | 6.25% |
| Radeon HD 2600 XT 256 MB (AMD) | 16 | |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 14 | 14.29% |
| GeForce 6600 128 MB (Albatron) | 13 | 23.08% |
| GeForce 7600 GS 256 MB (XFX) | 12 | 33.33% |
| GeForce 7300 GT 256 MB (Zogis) | 12 | 33.33% |
| GeForce 8500 GT 256 MB (MSI) | 11 | 45.45% |
| GeForce 6600 GT 128 MB (NVIDIA) | 11 | 45.45% |
| Radeon HD 2600 Pro 256 MB (AMD) | 9 | 77.78% |
| Radeon HD 2400 XT 128 MB (AMD) | 6 | 166.67% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 4 | 300.00% |
[nextpage title=”Far Cry”]
Far Cry is a heavy game based on the Shader 3.0 (DirectX 9.0c) programming model. We’ve updated the game to version 1.4. To measure the performance we run four times the demo created by German magazine PC Games Hardware (PCGH) and the results presented below are an arithmetic average of the collected data. We used the HardwareOC Far Cry Benchmark 1.7 utility to help us collecting data.
Since we are comparing mid-range cards, we decided to run this program in two resolutions, 1024×768 (simulating a low resolution) and 1600×1200 (simulating a high resolution). At each resolution we simulated two scenarios, first with no image quality enhancements enabled and graphics details set to “maximum” (this scenario we called “low”) and then with anti-aliasing set at 4x, anisotropic filtering set at 16x and graphics details set to “ultra” (this scenario we called “high”). On all scenarios we set the rendering engine to Shader 3.0. The results, given in frames per second, you check below. All video cards listed below were running with the default clock rates defined by the chip manufacturer.
| Far Cry 1.4 – 1024×768 – Maximum Details | Score | Difference |
| GeForce 8600 GTS 256 MB (MSI) | 141.84 | 40.74% |
| Radeon X1950 Pro 256 MB (PowerColor) | 139.29 | 38.21% |
| GeForce 7600 GT 256 MB (XFX) | 131.39 | 30.37% |
| Radeon X1650 XT 256 MB (HIS) | 128.86 | 27.86% |
| GeForce 8600 GT 256 MB (MSI) | 118.14 | 17.23% |
| Radeon HD 2600 XT 256 MB (AMD) | 100.78 | |
| Radeon X1650 Pro 256 MB (HIS) | 97.75 | 3.10% |
| GeForce 7600 GS 256 MB (XFX) | 89.96 | 12.03% |
| GeForce 6600 GT 128 MB (NVIDIA) | 85.77 | 17.50% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 81.60 | 23.50% |
| Radeon HD 2600 Pro 256 MB (AMD) | 70.20 | 43.56% |
| GeForce 7300 GT 256 MB (Zogis) | 67.03 | 50.35% |
| GeForce 8500 GT 256 MB (MSI) | 60.84 | 65.65% |
| Radeon HD 2400 XT 128 MB (AMD) | 54.53 | 84.82% |
| GeForce 6600 128 MB (Albatron) | 50.01 | 101.52% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 34.90 | 188.77% |
| Far Cry 1.4 – 1600×1200 – Maximum Details | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 112.29 | 117.36% |
| GeForce 8600 GTS 256 MB (MSI) | 78.86 | 52.65% |
| GeForce 7600 GT 256 MB (XFX) | 75.56 | 46.26% |
| Radeon X1650 XT 256 MB (HIS) | 71.84 | 39.06% |
| GeForce 8600 GT 256 MB (MSI) | 61.09 | 18.25% |
| Radeon HD 2600 XT 256 MB (AMD) | 51.66 | |
| Radeon X1650 Pro 256 MB (HIS) | 49.39 | 4.60% |
| GeForce 7600 GS 256 MB (XFX) | 46.17 | 11.89% |
| GeForce 6600 GT 128 MB (NVIDIA) | 45.07 | 14.62% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 40.88 | 26.37% |
| Radeon HD 2600 Pro 256 MB (AMD) | 35.33 | 46.22% |
| GeForce 7300 GT 256 MB (Zogis) | 34.52 | 49.65% |
| GeForce 8500 GT 256 MB (MSI) | 30.56 | 69.04% |
| Radeon HD 2400 XT 128 MB (AMD) | 27.11 | 90.56% |
| GeForce 6600 128 MB (Albatron) | 25.62 | 101.64% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 17.65 | 192.69% |
| Far Cry 1.4 – 1024×768 – AAx4, AFx16, Ultra Details | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 109.38 | 172.29% |
| GeForce 8600 GTS 256 MB (MSI) | 98.33 | 144.78% |
| GeForce 8600 GT 256 MB (MSI) | 79.72 | 98.46% |
| GeForce 7600 GT 256 MB (XFX) | 77.96 | 94.08% |
| Radeon X1650 XT 256 MB (HIS) | 74.08 | 84.42% |
| GeForce 7600 GS 256 MB (XFX) | 49.96 | 24.37% |
| Radeon X1650 Pro 256 MB (HIS) | 48.93 | 21.81% |
| GeForce 6600 GT 128 MB (NVIDIA) | 45.34 | 12.87% |
| GeForce 7300 GT 256 MB (Zogis) | 44.59 | 11.00% |
| GeForce 8500 GT 256 MB (MSI) | 41.31 | 2.84% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 40.86 | 1.72% |
| Radeon HD 2600 XT 256 MB (AMD) | 40.17 | |
| Radeon HD 2600 Pro 256 MB (AMD) | 27.62 | 45.44% |
| GeForce 6600 128 MB (Albatron) | 24.88 | 61.45% |
| Radeon HD 2400 XT 128 MB (AMD) | 21.35 | 88.15% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 17.97 | 123.54% |
| Far Cry 1.4 – 1600×1200 – AAx4, AFx16, Ultra Details | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 66.87 | 217.37% |
| GeForce 8600 GTS 256 MB (MSI) | 46.71 | 121.69% |
| Radeon X1650 XT 256 MB (HIS) | 42.10 | 99.81% |
| GeForce 7600 GT 256 MB (XFX) | 38.49 | 82.68% |
| GeForce 8600 GT 256 MB (MSI) | 38.33 | 81.92% |
| Radeon X1650 Pro 256 MB (HIS) | 24.81 | 17.75% |
| GeForce 7300 GT 256 MB (Zogis) | 21.23 | 0.76% |
| GeForce 7600 GS 256 MB (XFX) | 21.13 | 0.28% |
| Radeon HD 2600 XT 256 MB (AMD) | 21.07 | |
| GeForce 6600 GT 128 MB (NVIDIA) | 21.04 | 0.14% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 20.53 | 2.63% |
| GeForce 8500 GT 256 MB (MSI) | 18.78 | 12.19% |
| Radeon HD 2600 Pro 256 MB (AMD) | 14.31 | 47.24% |
| GeForce 6600 128 MB (Albatron) | 11.44 | 84.18% |
| Radeon HD 2400 XT 128 MB (AMD) | 10.90 | 93.30% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 8.47 | 148.76% |
[nextpage title=”Battlefield 2142″]
Battlefield 2142 is the latest member of the Battlefield franchise. We updated this game to version 1.01. We created our own demo based on Sidi Power Plant map (click here to download the demo we created for this test), which provided a very consistent number of frames per second. We ran it and measured performance with FRAPS. Click here to read in details how we benchmarked using Battlefield 2142.
Since we are comparing mid-range cards, we decided to run this program in two resolutions, 1024×768 (simulating a low resolution) and 1600×1200 (simulating a high resolution). First we ran our demo with image quality set at “low” (with texture manually set at its minimum level) and then with image quality set at “high” (with anti-aliasing manually set at 4x). Below you can see the results, given in frames per second. All video cards listed below were running with the default clock rates defined by the chip manufacturer.
| Battlefield 2142 1.01 – 1024×768 – low | Score | Difference |
| GeForce 8600 GTS 256 MB (MSI) | 335.30 | 20.14% |
| Radeon X1950 Pro 256 MB (PowerColor) | 330.39 | 18.39% |
| Radeon HD 2600 XT 256 MB (AMD) | 279.08 | |
| GeForce 8600 GT 256 MB (MSI) | 262.63 | 6.26% |
| GeForce 7600 GT 256 MB (XFX) | 254.29 | 9.75% |
| Radeon X1650 XT 256 MB (HIS) | 230.79 | 20.92% |
| Radeon HD 2600 Pro 256 MB (AMD) | 183.46 | 52.12% |
| Radeon X1650 Pro 256 MB (HIS) | 171.15 | 63.06% |
| GeForce 7600 GS 256 MB (XFX) | 164.71 | 69.44% |
| GeForce 6600 GT 128 MB (NVIDIA) | 152.17 | 83.40% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 142.72 | 95.54% |
| GeForce 8500 GT 256 MB (MSI) | 127.20 | 119.40% |
| GeForce 7300 GT 256 MB (Zogis) | 124.62 | 123.94% |
| Radeon HD 2400 XT 128 MB (AMD) | 119.98 | 132.61% |
| GeForce 6600 128 MB (Albatron) | 87.63 | 218.48% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 62.96 | 343.27% |
| Battlefield 2142 1.01 – 1600×1200 – low | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 184.41 | 45.48% |
| GeForce 8600 GTS 256 MB (MSI) | 171.89 | 35.60% |
| GeForce 7600 GT 256 MB (XFX) | 132.60 | 4.61% |
| GeForce 8600 GT 256 MB (MSI) | 131.81 | 3.98% |
| Radeon HD 2600 XT 256 MB (AMD) | 126.76 | |
| Radeon X1650 XT 256 MB (HIS) | 117.01 | 8.33% |
| Radeon HD 2600 Pro 256 MB (AMD) | 83.00 | 52.72% |
| Radeon X1650 Pro 256 MB (HIS) | 82.04 | 54.51% |
| GeForce 7600 GS 256 MB (XFX) | 78.45 | 61.58% |
| GeForce 6600 GT 128 MB (NVIDIA) | 75.60 | 67.67% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 68.07 | 86.22% |
| GeForce 8500 GT 256 MB (MSI) | 64.47 | 96.62% |
| GeForce 7300 GT 256 MB (Zogis) | 59.96 | 111.41% |
| Radeon HD 2400 XT 128 MB (AMD) | 55.77 | 127.29% |
| GeForce 6600 128 MB (Albatron) | 42.81 | 196.10% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 28.47 | 345.24% |
| Battlefield 2142 1.01 – 1024×768 – high | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 101.89 | 87.33% |
| GeForce 8600 GTS 256 MB (MSI) | 86.49 | 59.02% |
| Radeon X1650 XT 256 MB (HIS) | 70.57 | 29.75% |
| GeForce 8600 GT 256 MB (MSI) | 69.90 | 28.52% |
| GeForce 7600 GT 256 MB (XFX) | 63.95 | 17.58% |
| Radeon HD 2600 XT 256 MB (AMD) | 54.39 | |
| Radeon X1650 Pro 256 MB (HIS) | 46.82 | 16.17% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 38.99 | 39.50% |
| GeForce 7600 GS 256 MB (XFX) | 38.71 | 40.51% |
| GeForce 6600 GT 128 MB (NVIDIA) | 36.30 | 49.83% |
| Radeon HD 2600 Pro 256 MB (AMD) | 35.58 | 52.87% |
| GeForce 8500 GT 256 MB (MSI) | 34.52 | 57.56% |
| GeForce 7300 GT 256 MB (Zogis) | 30.55 | 78.04% |
| Radeon HD 2400 XT 128 MB (AMD) | 23.05 | 135.97% |
| GeForce 6600 128 MB (Albatron) | 20.32 | 167.67% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 13.96 | 289.61% |
| Battlefield 2142 1.01 – 1600×1200 – high | Score | Difference |
| Radeon X1950 Pro 256 MB (PowerColor) | 55.54 | 105.86% |
| GeForce 8600 GTS 256 MB (MSI) | 43.22 | 60.19% |
| Radeon X1650 XT 256 MB (HIS) | 35.05 | 29.91% |
| GeForce 8600 GT 256 MB (MSI) | 33.63 | 24.65% |
| GeForce 7600 GT 256 MB (XFX) | 31.16 | 15.49% |
| Radeon HD 2600 XT 256 MB (AMD) | 26.98 | |
| Radeon X1650 Pro 256 MB (HIS) | 19.85 | 35.92% |
| GeForce 7600 GS 256 MB (XFX) | 17.94 | 50.39% |
| Radeon HD 2600 Pro 256 MB (AMD) | 17.41 | 54.97% |
| GeForce 8500 GT 256 MB (MSI) | 16.95 | 59.17% |
| GeForce 6600 GT 128 MB (NVIDIA) | 16.67 | 61.85% |
| Radeon X1300 XT 256 MB GDDR3 (HIS) | 16.47 | 63.81% |
| GeForce 7300 GT 256 MB (Zogis) | 13.98 | 92.99% |
| Radeon HD 2400 XT 128 MB (AMD) | 10.88 | 147.98% |
| GeForce 6600 128 MB (Albatron) | 9.38 | 187.63% |
| GeForce 6200 128-bit 128 MB (Leadtek) | 6.49 | 315.72% |
[nextpage title=”Conclusions”]
In our review we were able to compare the new Radeon HD 2600 XT to GeForce 8600 GT and to a vast range of new and old mid-range PCI Express video cards, as you could see on our previous pages.
With both GeForce 8600 GT and Radeon HD 2600 XT with GDDR4 memory having a suggested price of USD 150, they are clearly targeted to users that are looking for the best cost/benefit ratio possible, i.e., to users that want to play games at a good performance but don’t want to buy an expensive high-end model.
So, which one is the best for Average Joe?
Well, we are pretty confident that GeForce 8600 GT still is the best video card for the average user that is willing to spend up to USD 150 on a decent video card. It is faster than Radeon HD 2600 XT in almost all situations, you can buy it right away and, best of all, even though its suggested price is of USD 150 you can easily find GeForce 8600 GT models costing less. Since Radeon HD 2600 XT is not available at this writing, it may reach the market initially costing more than GeForce 8600 GT.
It is also very important to keep in mind that another Radeon HD 2600 XT model will be available using GDDR3 memories running at a lower clock rate and at a lower price: it will have a suggested price of USD 130. So if you find a cheap Radeon HD 2600 XT being sold, beware, this particular model isn’t a direct competitor to GeForce 8600 GT. And cheap here will mean far slower.
For those interested, here is a summary of our bechmarking. GeForce 8600 GT was between 9% and 50% faster than Radeon HD 2600 XT on 3DMark03, between 8% and 20% faster on Quake 4 and between 17% and 98% faster on Far Cry. On F.E.A.R. Radeon HD 2600 XT was between 42% and 47% faster when no image quality settings were enable, but when we increase image quality GeForce 8600 GT was between 37% and 50% faster. And on Battlefield 2142 Radeon HD 2600 XT was faster than GeForce 8600 GT only at 1024×768 with no image quality settings enabled (6% faster), on other scenarios GeForce 8600 GT was between 4% and 28% faster.
The only program where Radeon HD 2600 XT was faster was on 3DMark06, with a performance advantage between 5% and 7%. Which isn’t really a significant advantage when we consider all other results.
Oh yes. Radeon HD 2600 XT supports HDMI with digital audio routed directly to the HDMI connector. That is definitely an advantage over GeForce 8600 GT if you are into high-def video.
Leave a Reply