Conclusions

Out of doubt, this product can be theoretically twice as fast as the 4870 card. But in practice, it's 1.5-1.8 times as fast, sometimes even slower. When you read graphs and performance gain values, you must keep in mind that dual-GPU cards (both from AMD or NVIDIA) have a big drawback: software distribution of load between processors. So maturity and performance of a given game depends solely on how well programmers finetuned the corresponding driver. There are a lot of games, where all these superpowerful dual-GPU cards demonstrate low performance and lags, to make things worse. It happens, when a processor has finished its previous task and gets an easy assignment, while the second processor is loaded "to the brims", so overall performance slumps. This way you get FPS jumping from 5 to 60 or higher. Visually, it will look as irritating lags. I repeat for those readers, who are in the habit of snatching phases out of context and then complaining about them in forums -- these lags are not bound to happen, it's just a possibility. It all depends on a given game, driver versions, and subjective attitude of a gamer.

AMD has tried to design the fastest graphics card in the world. Whether the company has succeeded or not is up to gamers to decide, because we must mention some spread of results. Besides, the new card is outperformed in a number of tests. So we cannot give a single-value estimate -- whether it's a king or not. As we can see, the drivers need some polishing.

Besides, we must stress that all winning results were registered only in universally recognized games equipped with built-in benchmarks, or in 3DMark, for which developers fine-tune drivers in the first place. But there are other games as well.

Nevertheless, the preliminary research shows that RADEON HD 4870 X2 has fewer problems with lags (we just played 2-3 games, where the 3870 X2 card apparently suffered from FPS jumping. Out of doubt, it's all possible owing to the new inter-GPU bridge and improved drivers. Anyway, we shall continue to look into this issue.

It's a pity that AMD stopped producing high-end single-GPU solutions comparable to similar products from NVIDIA. Now it only makes mid-end single-GPU cards, pairing them for the high-end segment.

Drawing a preliminary bottom line, we can state that R700 is actually very fast and, despite potential drawbacks, it seriously threatens top solutions from NVIDIA. We have already seen the Californian company cutting prices for GeForce GTX 280/260.

Afterword

A few words about our benchmarks.

In our updated article about FRAPS, we have illustrated how crude and inaccurate tests with this utility are. Testers do not have other tools, except for benchmarks built into games.

This article and its first part explain that it's sometimes possible to test games with integrated and identically looped demos. Although this implies lots of potential errors, since measurement accuracy depends on tester: whether he starts/stops FRAPS in time or does it too late/early.

But I have run across situations, when demo load changes abruptly in the very beginning or end. So, a half-second delay in starting/stopping the utility changes the average FPS by 15-20%. That's not a measurement error anymore - such test is a total waste of time. One time you delay the test, another time you start it too early (not intentionally, of course), and you end up with absolutely different performance results.

But even that's not the most important thing. The fact is, there are almost no games with built-in demos anymore. So, testers are forced to use a method that we deem totally unacceptable. They measure gaming performance by walking a straight line from the a starting point in a scene to a selected destination (the nearest fence, tree, etc.).

We all understand that it's impossible to navigate to a finish spot in precisely the same route with different cards and in different resolutions. Besides, such games always introduce random elements into a scene, and objects may be placed slightly differently on the same scene.

Unfortunately, websites that publish a huge number of tests do not always reveal their test methods for each game (except for those with built-in benchmarks).

So, we believe that it's better to offer a limited number of game tests, but each will be crystal clear, accurate and showing actual differences between graphics cards.