Radeon 6970: review, features, comparison with competitors and reviews

Radeon HD 6970 - AMD video card, released in December 2010. Its architecture simplified the design of stream processors and ensured their greater efficiency.

History of the model

In 2008, AMD developed and released the Radeon HD 4000 video card, codenamed R700 and branded ATI Technologies. Probably, many still use it in their PCs. The architecture of the R700 was replaced by the long-awaited R800, which paved the way for the AMD ATI Radeon HD 5000 series. The pearl of the series was the HD 5870 card, which appeared in September 2009. It consisted of 2150 million transistors and was a success, remaining a flagship uniprocessor model for about 15 months. Replaced at the end of 2010. At that time it was one of the best graphics cards in its price range, which speaks for itself.

Nvidia took about 6 months to open fire with the release of the GeForce GTX 470 and 480, and even then many believed that it did not live up to expectations. After several quick price declines and improved driver support, the card managed to make it competitive. The further appearance of the GTX 460 and the refined architecture of the GF110 helped create a viable product and open the way to the following models, namely the GTX 580 and 570.

Meanwhile, AMD offered HD 6000 from October 209, passing the baton HD 6870 as a counterbalance to the GTX 460, the price of which was then $ 240. It was an impressive model, except for the confusing name policy and constant reminders of the manufacturer that it was not a replacement for the 5870th model. Starting with this series, the ATI Technologies brand was officially finished with the goal of correlating AMD's graphics products with its computer platforms. Accordingly, the design of the logo has changed.

After a short delay, the new Radeon HD 6970 2GB was released. The graphics processor, codenamed Cayman, is made up of 2.64 billion transistors, which is 23% more than the 5870th model. Like the GTX 570 and 580, which are the upgraded versions of the GTX 470 and 480, the HD 6970 is an upgraded version of the 5870th model.

6900 series updates

The high-end architecture of the Cayman series used in the HD 6900 series is slightly different from the Barts, which debuted in the 6870 and 6850 models. It used the VLIW5 HD 5000 configuration, which includes a SIMD with 4 simple and 1 complex flow processing units. The 6900 series graphics cards differ in the VLIW4 configuration, in which the processing of the stream is organized into groups of 4 blocks with common registers. Although they all have equal opportunities, 2 of them (3 rd and 4 th) perform special functions. According to information from the company AMD, the configuration of VLIW4 provides computational power, not inferior to VLIW5, with a decrease in the area of the crystal by 10%.

Cayman processors provide a high level of parallelization compared to the Evergreen / Cypress architecture used in the HD 5800. Barts graphics processors are somewhere in the middle, as they were a step forward from Cypress in assigning individual flow controllers to each of the two SIMD blocks. Cayman chips are even more powerful thanks to the two GPE engines and the assignment of each of them to the SIMD-block. This means that the architecture now has two more efficient tessellation blocks compared to one in Barts. The video cards HD 6900, thus, the performance of tessellation can be significantly improved. According to AMD, it has increased 3 times in comparison with HD 5870. In addition, the architecture of Cayman also features a redesigned rendering engine consisting of 128 ROP Z / Stencil, 32 ROP colors, with almost double fast 16-bit integer operations and 2-4 -fold increase in the speed of 32-bit floating-point operations.

Cayman Architecture

The transition from VLIW5 to VLIW4 improved the computing capabilities of AMD: narrower SPUs are easier to use to the full, FP64 improves performance compared to FP32 by a quarter, and the saved space allows you to place additional SIMDs. But if Cayman was a serious attempt to establish itself in the graphic computing market and snatch a piece of pie from NVIDIA, then it should be done more than just adding new shaders. Therefore, AMD was struggling to improve the functionality of its graphics processor, so that it could pose a threat to the architecture of Fermi.

The main feature of Cayman is asynchronous dispatching. The term describes very precisely what the map is doing. In Fermi, NVIDIA introduced support for parallel cores, which provided the ability to calculate several code fragments simultaneously. The company AMD repeated the approach of NVIDIA, but went even further.

The limitation of Fermi's design is that although the architecture allows multiple cores to be used simultaneously, each of them must be serviced by a single stream processor. Independent applications, for example, can not issue their own code and execute it in parallel, and the GPU must contextually switch between them. Asynchronous dispatching of AMD is designed to allow independent threads and applications to generate code that would run in parallel. At least on paper this gives a significant advantage (context switching is too expensive), which should exceed Fermi's performance.

The principle of asynchronous dispatching is that the GPU hides some information about its real state from applications and code, which, in effect, results in virtualization of the resources of the GPU. After all, each piece of code assumes that it runs on its GPU with its command queue and its own virtual address space. This shifts the load on the graphics processor and drivers, but the gain is that it's better than context switching.

Asynchronous dispatching requires API support. DirectCompute is a fixed standard and this feature does not support, at least in the 11th version. Therefore, asynchronous dispatching is implemented as an extension of OpenCl.

The rest of AMD's improvements relate to memory and cache performance. The basic architecture here remains the same, but some minor changes in the computation have been made. The local data store available for each SIMD can now bypass the hierarchy of the cache and the global data store by direct reading. Cayman received the 2nd module of direct memory access, which improved the speed of reading and writing, allowing to perform simultaneously 2 operations in each direction.

Finally, the shader was slightly accelerated. Compared to Cypress, Cayman can reduce the number of operations by combining them.

Design

The length of the map is 27 cm, which is typical for modern high-performance models. For example, the length of the HD 4870 X2 is 28 cm, like the HD 5870. The dimensions of the HD 6970 are identical to the GTX 580. Like its predecessor, the AMD Radeon HD 6970 is manufactured using a 40nm process, but the company added another 486 million transistors, Which crystal size has increased by 16%. The GPU core runs at 880 MHz, which is 3.5% higher than the HD 5870, and GDDR5 also runs a little faster at 1375 MHz. The memory frequency coupled to the 256-bit bus gives the HD 6970 a theoretical throughput of 176 GB / s, which is a 14.5 percent advantage over the HD 5870.

The HD 6970 differs from the older 5870th model as well as the basic configuration. If the latter had 1600 cores, 80 TAUs and 32 blocks of rasterization, the first one received 1536 stream processors, 96 texture units and the same 32 ROPs (4% fewer cores and 20% more TAUs).

According to user feedback, the peak integer bilinear texture filtering is performed at a speed of 84.5 gigatexels / s, and the pixel filling is 28.2 Gp / s. The productivity is 2.7 Tflops. The three-dimensional scene is built at a speed of 880 million polygons / s. The card supports dual-stream HD video playback and has a built-in audio controller.

Inside the case is a printed circuit board with a GPU measuring 389 mm ² in the center. Around the processor there are 8 2-GB Hynix GDDR5 memory chips with a nominal data transfer rate of 6 Gbit / s, which is 0.5 Gbit / s above the operating frequency of the card. The use of GDDR5 at high speeds is complicated by the difficulties in creating a good memory bus, which affected this model. AMD achieved progress in achieving 5.5 Gbit / s by improving the design of printed circuit boards, but further increase in speed seems inexpedient, if only because of the design of the 256-bit bus.

Cooler

The cooling of the processor is a fairly large aluminum evaporator chamber radiator made up of 39 plates 13.5 cm long, 6.5 cm wide and 2.5 cm high. The design of the evaporator was first implemented in the AMD ATI Radeon HD5970 and borrowed from NVIDIA in the GeForce GTX 570 And 580. Finally, the radiator is cooled by a 75 x 20 mm fan that draws air from the casing and pushes it out at the back of the card.

According to user feedback, for the most part the cooler works very quietly, which is facilitated by low consumption (20 W) in idle mode. During the game, the fan, of course, accelerates, and the card under load consumes up to 250 watts. This is 33% more than the Radeon HD 5870 requires, but even with an increase in thermal load, the noise level does not rise to unacceptable values.

The radiator and fan are enclosed in a specially designed case that hides the entire video card. It is a common practice for AMD to design its most elite models. According to feedback from users, they like this design, as it provides reliable protection for the device. NVIDIA also did it in the past for its most expensive graphics cards, like the GTX 295 with two graphics processors, although in its other flagship products (for example, the GeForce GTX 580) this development has ceased to apply.

Connectivity options

In order to provide the card with sufficient power, AMD has installed 8- and 6-pin PCIe connectors. A similar solution can be found in the HD 5970 and GTX 580, since such a configuration is generally used in models that consume a lot of energy. Naturally, the HD 6970 supports CrossFire and so on a standard location you can find a pair of connectors for connecting two or more cards. Next to them is a switch that allows you to choose between 2 BIOSes. This is done to improve the reliability of the model in the event of a failure of the flash memory. In addition, you can run the GPU using the backup system, and then switch back to the primary and re-flash the damaged software. Before that, AMD did not have the support for rewritable BIOSes, but this implementation was an interesting change.

The remaining ports are located on the I / O panel. The reference video card sample is equipped with two dual DL-DVI connectors, two Mini-DisplayPort and HDMI ports. It is worth noting that the 6970 model supports a maximum resolution of up to 2560 x 1600 on 3 monitors. And with a multi-threaded hub using Mini DisplayPort 1.2 connectors, the card is capable of serving up to 6 displays. Like 5870, the back of the device is covered with a metal plate. Although there are no components requiring protection, such a decision is perceived positively by users, since it allows you to take a card without worrying about not touching sharp contacts.

Although the overall dimensions of the HD 670 are almost identical to the 5870, according to users, the 3D design of AMD is not very successful when it comes to working in CrossFire mode with 2 cards installed nearby. The case of the 5870 was slightly protruding in the center, keeping the ventilation holes from being blocked by neighboring boards. In 6970th there is no such luxury, and you can practically isolate the top card depending on how it will be installed. As a result, the temperature rises, but not to critical values. AMD company should have learned from its competitors from NVIDIA and around the fan to allocate more space that would allow it to freely "breathe". Users are strongly encouraged to install graphics cards, leaving as much distance between them as possible, if the motherboard and the chassis allow this.

Performance

According to user feedback, the Futuremark 3DMark 11 test showed that when performing extreme tests, the AMD Radeon HD 6970 is 7% faster than the GTX 570 (1821 points against 1697). If such an advantage persists in gaming tests, then this can play into the hands of the manufacturer. According to the same test, the HD 6970 is 7% slower than the GTX 580 (1962 points), 27% behind the HD 5970 (2506) and 16% more efficient than the HD 5870 (1572 points).

Dirt 2

This game has a fantastic built-in test, which very accurately measures the actual performance. Users ran Dirt 2 in DirectX 11 mode with activated 4xAA and maximum image quality settings. The Radeon 6970 on average provides 77 fps at 1920 x 1200, which makes it only 10% faster than the HD 5870. More importantly, the HD 6970 is 21% slower than its main competitor GTX 570. Other more expensive models Also fell behind. The HD 5970 was 23% faster than the 6970th model, while the GTX 580 had a 33% advantage. Dirt 2 was ruthless for the HD 6970.

F1 2010

The game with the first update has an excellent built-in test, which accurately measures the performance of the card. Users tested F1 2010 in DirectX 11 mode with 8xMSAA enabled and the best visual settings. It should be noted that they could not get CrossFire to work correctly with the latest drivers, so the HD 6970 was never able to get ahead. The F1 2010 results are not surprising, since the HD 6970 is the second fastest graphics card that was tested at 1920x1200 resolution, which is only 9% slower than the powerful GTX 580. The HD 6970 is only 5% better than the 5870th model. The results in this game are still a little non-standard, since the HD 6970 outperforms the GTX 570 with a convincing advantage of 22%.

COD: Modern Warfare 2

To evaluate the performance of the video card in this game, users used the Fraps application. Maximum game quality settings were accompanied by 4xAA and a 60-second gameplay record. As a result of testing the Radeon 6970, the refresh rate of the display was 73 fps, which is better than the GTX 480 by 1 fps. This almost corresponds to the performance of the GTX 570 (74 fps) and 16% behind the faster GTX 580. With the dual Radeon HD 5970, the 6970 is 19% slower, although it provides a frame rate that is 30% larger than the HD 5870.

COD: Black Ops

Fraps also helped to produce a comparative analysis of this single game. Users measured the frame rate for 1 minute of gameplay of the first single level (Operation 40) with maximum visual effects, including 4xAA. The results of Black Ops showed that the Radeon 6970 lags behind the GTX 570 by 5% and 13% from the GTX 580. At an average of 105 fps at 1920 x 1200 and maximum settings, there is no room for improvement anyway, the model outperforms The predecessor by 24%.

Energy consumption

Compared to the HD 5870, the 6970th model consumes 16.5% more power, but considering that it is also 24% faster, the owners find that overall this is a more productive video card. In comparison with the HD 5970, the 6970 model is 3% more economical, but on average slower by 15% and 26% in the Crysis Warehead game, which was used in stress testing. Radeon HD 6970 consumes about the same amount of energy as the GTX 570, consuming 1% more under extreme loads and 4% less at idle. Given that both video cards provide the same performance, this time, it seems, they are equally consistent with each other.

Heating temperature

According to user feedback, the HD 6970 is getting pretty hot when tested in FurMark, reaching 90 ° C, slightly above the 87 ° C HD 5870. The GeForce GTX 570, on the other hand, only heats up to 81 ° C, which makes it significantly Cooler at maximum loads.

Overclocking performance

The Catalyst Control Panel limits the maximum core frequency to 950 MHz, which the Radeon HD 6970 accepts without problems. This is quite a worthy indicator. The memory frequency can reach up to 1450 MHz, although users at the same time face stability issues and are forced to retreat to 1440 MHz. This 8 percent overclock leads to a 7.6 percent increase in performance when tested in COD: Black Ops, 7.8 percent in Crysis Warhead, and 7 percent in Battlefield Bad Company 2.

Conclusion

It seems that the HD 6900 and GTX 500 are repeating each other. With a few key differences, of course. The NVIDIA product appeared on the market in time and proved to be very effective in confronting the competing AMD models. The GeForce GTX 480 was the fastest uniprocessor video card, and the GTX 580 retained the crown, as it was beyond the power of the Radeon 6970. The price of the model corresponded to the cost of the GTX 570, and the niche of the 580 was occupied by the HD 6950. Only by the beginning of the next year AMD had a chance to return A performance advantage by releasing 2 HD 6990 series cards, code-named Antilles.

The Radeon HD 6970 is in step with the GTX 570, and the recommended retail price for this card was set at $ 369, while in the GTX 570 debuted at $ 349. Despite the fact that these are two very different models, they show the same average in many games tested by users at 1920 x 1200 resolution. For the most part, the difference is very small. The Radeon 6970 in some cases is 20% faster than the GTX 570, while in others it is about 20% slower. On energy consumption, both cards are also very close, demonstrating similar levels of operational efficiency. Therefore, the HD 6970 is a good alternative to the GTX 570. The choice between the two cards is not easy, but any solution is right, given their price and performance.

Compared to other models, such as the GTX 580, the 6970 is on average 15% slower, but costs 30% less, just like the GTX 570, which offers an even better price. The HD 6970 effectively replaces the HD 5870. Based on testing data at 1920 x 1200, the HD 6970 is on average 24% faster and only 10% more expensive. In addition, the energy efficiency of the model has also been improved. The increase in power consumption by 16% gives an additional 24% of the productivity. It is worth noting that the Radeon 6970 looks slightly better than the GTX 570 in STALKER games: Call of Pripyat and Aliens vs. Predator, in which tessellation is used. Before its advent, NVIDIA had a significant advantage here.

In general, both video cards provide an exceptional level of performance and cost, making the choice between them both easy and difficult. But any decision will be correct.