The Fastest 3D Cards Go Head-To-Head

Evaluation Of The New Generation

Some readers have complained about the lack of clear recommendations and awards for the new graphics cards. Since Nvidia and AMD put out their products at different times, no direct match-up has been made until now. This test is a perfect basis for comparison, since all graphics chips have completed the same benchmarks with the same test system. We have evaluated 3D performance, power consumption, price, and noise levels. To keep the evaluation simple, we award from one to four points for each category—the better the card does, the more points it is given.

The power consumption in watts is evaluated separately for the 2D and 3D modes, so there are points for both categories. The performance per watt has not been evaluated, as we are only interested in the overall efficiency of the samples. In 3D mode, the Nvidia cards are insatiable, while in 2D mode, AMD does poorly. Although all of the cards drop their clock speeds, the power consumption of the HD 4850 and HD 4870 remain high. At the moment, the high clock speed of the GDDR5 RAM on AMD’s HD 4870 has a negative influence on 2D power levels.

We have heard rumors that current versions of the PowerPlay energy saving function do not work properly. The clocking rate is reduced, but the voltage remains the same, which might explain the high 2D consumption.

Swipe to scroll horizontally
1. Power consumptionwatts2D modeLow Power 3D3D modePoints2DPoints3D
GeForce GTX 280 (1024 MB)11713035231
GeForce GTX 260 (896 MB)11112533642
Radeon HD 4870 (512 MB)147028813
Radeon HD 4850 (512 MB)122023724

Power consumption, entire system, 2D = Desktop Vista, 3D = Full load

When it comes to noise levels, only the Radeon HD 4850 is reasonable. The HD 4870 is very loud, while the GTX 260 and GTX 280 generate extreme acoustic output under full load. The overclocked HD 4870 from MSI does not have a constant fan speed in 2D mode. Rather, the fan gets louder and then quieter again. A new VGA BIOS may solve this problem. The points are a mixed result of 2D and 3D.

Swipe to scroll horizontally
2. noise level dB(A) 2D dB(A) 3D dB(A) Points
GeForce GTX 280 (1024 MB)37.754.5 - 54.72
GeForce GTX 260 (896 MB)38.1 - 44.253.5 - 54.01
Radeon HD 4870 (512 MB)35.545.5 - 46.03
Radeon HD 4850 (512 MB)36.341.24

The overall performance is quite clear; the gaps between values are not evaluated.

Swipe to scroll horizontally
3. overall performancefpsPercentPoints
GeForce GTX 280 (1024 MB)3765.6127.04
GeForce GTX 260 (896 MB)3464.8116.83
Radeon HD 4870 (512 MB)3355.7113.22
Radeon HD 4850 (512 MB)2965.6100.01

This evaluation only shows the level of MSI overclocking in percent. The Radeon HD 4850 is supplied at default clocking, and is thus not included. To ensure that the evaluation is fair, the overclocked value is included in the end result once, and ignored once.

Swipe to scroll horizontally
4. OC power 1920x1200 AA (MSI)PercentPoints
GeForce GTX 280 (1024 MB)5.83
GeForce GTX 260 (896 MB)4.52
Radeon HD 4870 (512 MB)3.51

Price is also quite clear, with the gaps between the figures not evaluated.

Swipe to scroll horizontally
5. The pricedollarsPoints
GeForce GTX 280 (1024 MB)4201
GeForce GTX 260 (896 MB)2902
Radeon HD 4870 (512 MB)2703
Radeon HD 4850 (512 MB)1804

On points, the Radeon HD 4850 is the clear winner. If you want a little more performance, though, you would be well advised to go for the HD 4870 or GTX 260, both of which represent serious competition for AMd due to price.

Swipe to scroll horizontally
6. Evaluation of pointswithout OCwith OC
GeForce GTX 280 (1024 MB)1114
GeForce GTX 260 (896 MB)1214
Radeon HD 4870 (512 MB)1213
Radeon HD 4850 (512 MB)1515
  • San Pedro
    Looks like the results for SLI and Crossfire were switched with the single card results. . .
  • Duncan NZ
    Not a bad article, really comprehensive.
    My one complaint? Why use that CPU when you know that the test cards are going to max it out? Why not a quad core OC'ed to 4GHz? It'd give far more meaning to the SLI results. We don't want results that we can duplicate at home, we want results that show what these cards can do. Its a GPU card comparason, not a complain about not having a powerful enough CPU story.

    Oh? And please get a native english speaker to give it the once over for spelling and grammar errors, although this one had far less then many articles posted lately.
  • elbert
    No 4870x2 in CF so its the worlds top end Nvidia vs ATI mid to low end.
  • Lightnix
    It'd be a good article if you'd used a powerful enough CPU and up to date Radeon drivers (considering we're now up to 8.8 now), I mean are those even the 'hotfix' 8.6's or just the vanilla drivers?
  • elbert
    Version AMD Catalyst 8.6? Why not just say i'm using ATI drivers with little to no optimizations for the 4800's. This is why the CF benchmarks tanked.
  • at 1280, all of the highend cards were CPU limited. at that resolution, you need a 3.2-3.4 c2d to feed a 3870... this article had so much potential, and yet... so much work, so much testing, fast for nothing, because most of the results are very cpu limited (except 1920@AA).
  • wahdangun
    WTF, hd4850 SHOULD be a lot faster than 9600 GT and 8800 GT even tough they have 1Gig of ram
  • mjam
    No 4870X2 and 1920 X 1200 max resolution tested. How about finishing the good start of an article with the rest of it...
  • I agree, the 4870 X2 should have been in there and should have used the updated drivers. Good article but I think you fell short on finishing it.
  • @pulasky - Rage much? It's called driver issues you dumbass. Some games are more optimised for multicard setups than others, and even then some favour SLi to Crossfire. And if you actually READ the article rather than let your shrinken libido get the better of you, you'll find that Crossfire does indeed work in CoD4.

    Remember, the more you know.