Nvidia GeForce GTX 470: Features, Overview

NVIDIA is engaged in a long enough time for detailed development of architecture in order to maintain a worthy competitiveness of the products. Thus, in the majority of cases, the company spends about two years moving from one series to the next.

The release of the GTX 470 graphics cards was delayed, as the developers had to process the architecture quite badly, some production problems arose, eventually the chip that was created turned out to be too hot during its work. Thus, the cycle remained almost exactly the same, and almost two years have passed since the release of the previous card. However, one way or another, the issued GTX 470 cards initially claimed to be called the technological leader of their time. Undoubtedly, they were like that.

GeForce 400

Traditionally, NVIDIA has simultaneously provided two models of video cards, GTX 470 and 480, but it is the first version that is more common among users today.

Experts immediately turned their attention to the number of transistors that were used in Fermi. This parameter could indirectly tell how good the new chip is, because it was twice as productive as the previous flagship of this company, and also 1.5 times more compared to the more modern Cypress, which is quite and A very good result.

Of course, this indicator directly affected the heating. In the process of creating the GTX 470, it was common to use a 40-nm process, so the difference in the thermal package with similar devices is approximately the same as the difference in the number of transistors.

Technical features

Initially, it was decided to use 512 CUDA-processors, but over time this decision was revised, resulting in the figure was slightly reduced to reduce heat dissipation, as well as increase the output of really working chips. Nevertheless, in the end, the cards were still quite expensive for the majority of users.

It was also a strange fact that the memory frequency even in the most productive model is smaller compared to the competing HD 5850 solution. The overall throughput of the GeForce GTX 470 video memory is higher, but due to the fact that it used faster chips, there was the possibility of its further overclocking .

How does she behave in games?

Both models provide DirectX 11 support. After Evergreen models began to appear on the market, there was practically nothing to test the capabilities of this device, but the situation with GeForce GTX 470 almost did not change. Games that could support the new APIs were single instances, but over time, of course, a large number of different projects were released, for which such a video card was needed. Even to this day, its use is optimal for the majority of games, and if it is impossible to play at the most modern ones at the maximum requirements, then it will maintain a medium or at least minimal stability.

As usual, the company decided to bet in favor of the future by optimizing its NVIDIA GeForce GTX 470 to work with the features that the latest DirectX provides, as well as paying special attention to GPGPU.

The architecture of the new chip

The new device used a chip with a new architecture called GF100. At first glance, there were no significant differences from its predecessor, while the design of the novelty remained scalar, so there were certain limitations. The composition of universal processors has changed insignificantly, which it was decided to combine into fairly large blocks. In the GT200 model such units were named TPC, and each of them included all the necessary components that may be needed when working with graphics, excluding blocks of rasterization. In the future, the company quickly began to produce a lot of cheaper versions of NVIDIA GeForce GTX 470, in which such units were either disabled or modified.

In each such block of the previous chip model, there were three SM (streaming multiprocessors), as well as eight texture blocks and a first level cache of 16 KB. In the block of the new device, in turn, one Raster Engine was used, as well as four SMs, each of which had its own cache memory, as well as texture units. In the same chip there are simultaneously four blocks, in the amount of providing the user 32 pixels per clock.

Streaming multiprocessors

If we consider devices at SM level, the difference becomes quite significant, since when it was released from NVIDIA GTX 470, many experts noticed that SMs are much larger and independent. In the past generation of cards there were 8 SPs, as well as 2 SFUs, but now their number and ratio was changed, as a result of which there were 4 SFUs, 32 CUDA Cores, and 16 Load / Store blocks in each device.

In each of these third generation multiprocessors, there were several texture processors, as well as shared memory providing hardware tessellation.

Due to the fact that the texture processors, as well as all the corresponding cache memory were moved inside the multiprocessors, the developers were able to achieve much more efficient operation of these devices. It was also decided to dramatically rework the internal architecture of these devices, increase the frequency of their work, plus triple the amount of second-level memory actively used in the process of texturing.

Common memory

The shared memory was merged with the L1 cache, so that the corresponding amount of memory became correctly distributed depending on which task the device is currently executing. Previously, shared memory included only 16 KB, but at the moment, depending on the type of task being performed, the device will allocate 64 KB of memory. If at the moment a lot of physical calculations are performed, then in this case 48 Kb of this memory the GTX 470 distributes to L1. If different processes access a lot of information a certain amount of time, then it is stored in the shared memory. In addition, the triple L2 cache in the GTX 470 is also useful for GPGPU. The characteristics of the device, therefore, have more and more advanced compared to competitors.

Tessellation

Tessellation is one of the key features of the Radeon HD 5000 card. The NVIDIA GeForce GTX 470 graphics card, in turn, is better geared for implementing this algorithm.

The DirectX 11 API supports certain "displacement maps" that are volume textures. Thus, each pixel in the picture of this texture will determine its height. This texture is superimposed in the game in order to make the image more detailed.

It is here that the tessellation comes in, which is equipped with GTX 470. The review of this algorithm showed that a whole mosaic is formed from one polygon. This technology is used regardless of the displacement maps, but the most impressive effect can be achieved only if they are used together. The vertices of the forming polygons will be lowered or raised to a certain height, bringing the system closer to the model that is set by the room map.

What are the advantages?

In fact, the model turns out to be absolutely identical to the standard high polygonal, being considered at the same time low polygonal, because it will give the correct glare and correctly cast the shadow, spending at the same time an order of magnitude less resources.

In other words, the NVIDIA GTX 470 characteristics are such that they can load models as quickly as possible into the memory of the card, without subsequently having to load and unload objects with other levels of detail, which was relevant for devices of the past generation. All recalculation processes are carried out directly within the card itself during various procedures, which provides much greater scalability. Depending on the presence of completely free computing power, it will be possible to approach the displacement map with a certain accuracy, which is quite important in terms of geometric performance of the GTX 470. Users feedback indicates that this technology significantly improves the overall performance of the computer during complex graphics processes.

What do users say?

Feedback from users who update their PC by installing the described card indicate that they were able to set maximum settings in all games after the upgrade. The game did not hang, no brakes were observed.

Of course, such reviews are characteristic for 2010-2012, at the moment the card no longer provides its owner with such limitless possibilities. Yet during this time, the technologies have moved far ahead, and there are already such games in which such a video card can only be played at minimum settings.

3D Vision

3D technology today has become extremely widespread, ranging from televisions to monitors. Even AMD has long been producing its own glasses, as well as the corresponding software that allows the manufacturer's video cards to output a high-quality stereoscopic image.

In GeForce GTX 470, the characteristics also provide for the possibility of using this technology, and in spite of the fact that it significantly affects the performance, the general sensations from the gameplay or watching the movie become completely different.

The entire set of tools provided by various graphics cards and APIs is quite difficult to master for ordinary users and is considered absolutely useless in the event that no one will use it. It is for this reason that NVIDIA has created a separate unit that works with modern game developers. Of course, attention has always been paid to the most successful projects, and mutually beneficial cooperation was established, because eventually the game itself became much faster and more beautiful if it was run by a user on the computer with this card.

Cooling system

To ensure adequate cooling of the map, the company developed a specialized cooler, which is one of the most powerful among the analogue reference models. The design of the device is made in the form of a standard turbine, but it has several features that definitely deserve special attention.

As the basis of the system stands a mechanical frame with a plastic casing attached to it. Through specialized thermal pads, the heat generated during the operation of memory chips and subsystems is transferred to the frame. From the GF 100 itself, heat is removed by means of five special heat pipes that contact the chip directly.

Direct contact of the heat pipes of a chip is often used in cooling systems of processors, and such a system can not be known to be more effective than the standard one. But in this case it means that when the GeForce GTX 470 was developed, the developers paid special attention to the cooling system.