GeForce 9 Series

The Geforce 9 series is a series of desktop graphics chip company Nvidia and successor of the GeForce 8 series. All graphics processors of this series support Shader Model 4.0 (SM 4.0 ) to DirectX 10, OpenGL 3.3, CUDA and PhysX with it. Since January 2009, some models of the GeForce 9 series are marketed as " GeForce 100 " for the OEM market.

  • 5.1 Measurement of power consumption of a whole system

Description

Introduced in early 2008 Geforce 9 series is largely identical with the chips of the previous series. Since two of the four graphics processors used have already been used in the Geforce 8 series, Nvidia was criticized because of the naming. The problem was solved, that the Geforce 8800 GS and 8800 GT were now being marketed under the name GeForce 9600 GSO and 9800 GT. It is generally assumed that Nvidia would react with the introduction of the Geforce 9 series on AMD's Radeon HD 3000 series.

In contrast to the older chip 8-series ( G8x ), the new GPUs and 8- 9-Series, so G92 ( b ) to G98, support for PCI Express 2.0. The Geforce 9800 series models support the hybrid power technology and can accommodate by switching off the external graphics card less power with a corresponding motherboard with Nvidia chipset under Windows Vista at idle. In contrast to AMD's Radeon HD 3000 series Nvidia still used the 65 - nm manufacturing process and introduced in the summer of 2008 to the more modern 55 - nm process. The initially formed disadvantage of the higher production costs could Nvidia compensate with better performance on the market. Criticized the omission of the support of Shader Model 4.1 (SM 4.1 ) to DirectX 10.1, and the absence of an energy -saving mode.

The first card of the GeForce 9 series, the GeForce 9600 GT was introduced on 21 February 2008. Unlike previous series presentations, these are not a high-end, but a mainstream graphics card based on the G94 GPU. In the G94 processor is in the broadest sense a halved G92 chip, which, however, still has four ROP partitions. This made it possible that Nvidia could bridge to a mainstream graphics card a 256 -bit memory interface for the first time. The GeForce 9600 GT was placed on the market against the Radeon HD 3850, which acted unrivaled at this point in the price sector of about 150 €. Despite technical disadvantages over competitors due to missing DirectX 10.1 support and the absence of an energy-saving mode, the GeForce 9600 GT was able to successfully compete in the market, which is generally attributed to the higher performance over the Radeon HD 3850. This only changed in the fall of 2008 with the introduction of the Radeon HD 4830th Only in November 2009, presented with the Nvidia Geforce GT 240 's successor, bringing the GeForce 9600 GT graphics cards for an unusually long market presence had. This was only possible with the change to the 55 - nm manufacturing process and the associated cost reduction.

On 18 March 2008, Nvidia expanded with the introduction of the GeForce 9800 GX2, the GeForce 9 series to the high-end range. The dual graphics card consisted of two individual boards, which were composed in the so -called " sandwich " design. These were practically seen two Geforce 8800 GT graphic cards, which communicate with each other via an internal SLI bridge chip and had all eight shader cluster of G92 processor. She proved in her presentation as the fastest at this time graphics card on the market, although it (eg by high anti -aliasing setting or extreme resolutions ) came in very memory-intensive applications due to the low video memory size to partially massive performance drops.

On 1 April 2008 Nvidia introduced the GeForce 9800 GTX, which is based on a full-fledged G92 processor. Nvidia tried to establish 8800 GTX with the card to the success of the Geforce. Although the GeForce 9800 GTX had both a higher computing power of stream processors and higher Texelfüllraten, she fell for memory-intensive applications behind the performance of its predecessor. As a reason for the lower memory throughput, and lower memory size apply. This meant that the Geforce 9800 GTX with only limited success selling in the market, partly because of the price premium over the minimally slower GeForce 8800 GTS 512 was disproportionately high. This only changed when Nvidia end of July 2008, the revised Geforce 9800 GTX is introduced. This used the G92 chip in the B1 revision and was therefore the first graphics card from Nvidia, which was prepared in 55 - nm manufacturing process. This reduced the manufacturing costs and enabling higher clock speeds. The new GeForce 9800 GTX placed both in terms of performance and priced between the Radeon HD 4850 and 4870th

On 29 July 2008 extended the Nvidia Geforce 9 series to the GeForce 9500 GT and 9800 GT. The Geforce 9500 GT replaced while the older Geforce 8600 GTS and was based on the G96 GPU. Collated in an upper low-end range and acts there quite successful. Some Board Nvidia's partners brought under the name GeForce 9500 GS nor a reduced-power version with DDR2 memory on the OEM market. Like the GeForce 9600 GSO, it was at the GeForce 9800 GT is a pure remake. Since only the support for the hybrid power mode was added, the GeForce 9800 GT could not achieve a performance improvement compared to its predecessor. As the competition from ATI, the Radeon HD 4850, had a higher performance, the GeForce 9800 GT could not repeat the good sales success of the GeForce 8800 GT.

The graphics processor Nvidia G98 sat first as an IGP and the Geforce M series. On the desktop market, however, the G98 Nvidia is first used on the GeForce 8400 GS. However, decided several board partners from Nvidia, the GeForce 8400 GS to bring under the name GeForce 9300 GS, GE and SE on the OEM market.

In early 2009 presented Nvidia's board partners so-called " Green Edition " of the GeForce 9600 GT and 9800 GT. The cards are operated at reduced voltage, so that the power consumption was reduced to an extent that could be dispensed with the additional power connector of the reference models. But also the clock rate must be reduced for the stable operation. Nvidia, which officially does not list the cards that supported the board partners with the selection of the necessary G92 and G94 GPUs.

GPUs

Naming

In the Geforce 9 series is a similar naming scheme as it has been used for the GeForce 6 series, are used. All graphics chips are identified by a four-digit number that generally begins with a " 9" ( for Geforce 9). The second digit then divides the family into different market segments. The abbreviation behind the model number is used to further diversification.

  • 9300/9400 / 9500 low-end
  • 9600: Mainstream
  • 9800: performance, high-end
  • No suffix - IGP variant
  • GE / GS / GSO - low powered versions (often recycling of defective part, actually higher quality chips)
  • GT - " Normal " versions ( often the products with the best value for money in the Nvidia product offering)
  • GTX - Powerful versions (name, there is only the performance segment )
  • GX2 - Most powerful versions with two GPUs on a graphics card

A " " after the model name indicates that this is a revised and usually faster version of an earlier, but identical, graphics card is - in the 9800 GTX was installed a newer revision of the processor.

Model data

Notes:

  • The specified clock rates are the established recommended by Nvidia or. However, is the final determination of the clock rates in the hands of the respective graphics cards manufacturers. Therefore, it is quite possible that there are graphics cards models or will be having different clock speeds.
  • With the stated date of the appointment of the public presentation is given, not the date of availability of the models.
  • The clock frequency of the memory is also often reported as twice as high. The reason for this is the double data rate (DDR ).

Performance

For each model, the following theoretical power data:

Notes:

  • The above performance data for the computing power of the stream processors, the pixel fill rate, the Texelfüllrate and memory bandwidth are the theoretical maximum values. The overall performance of a graphics card depends, among other things, on how well the available resources can be exploited or utilized. In addition, there are other not listed here, factors that affect the performance.
  • The specified processing power on the stream processors refers to the use of both MUL operations, which is not reached in graphics shader calculations, as further calculations must be performed. In these calculations, therefore the performance of the computing power is lower on the stream processors.

Power consumption data

The measured values ​​listed in the table refer to the pure power of graphics cards that comply with the nVidia reference design or at least form a supported nVidia's own market segment (eg, Green Edition ). In order to measure these values ​​, it requires a special measuring apparatus; depending on the employed measurement technique and given measurement conditions, including the shared program, with which the 3D load is generated, the values ​​between different equipment may vary. Therefore, here measured value ranges are specified, each representing the lowest and highest measured values ​​from different sources.

Far more common than the measurement of the consumption of the graphics card is to determine the power consumption of a whole system. To this end, a reference system is assembled, in which the various graphics cards to be installed; then finds the measurement using an energy cost meter or similar apparatus rather than directly to the outlet. However, the significance of the measurements is limited: it is not clear which consumption comes from the graphics card and what is attributable to the rest of the PC system. The difference in consumption between idle and load 3D operation of this measuring method is not only dependent on the program with which the load was produced; the utilization and efficiency of the rest of the PC system, including the power supply, motherboard and processor affect the measured difference also. Because different the tested systems are usually of one's own home PC system, the values ​​given there can not be mapped to its own system. Only data from otherwise identical systems are good (conditional) for comparison with each other. Because of this dependence overall system measurement values ​​are not listed in the local table. But since they can provide a better picture of the practical power of a concrete system with a certain graphics card, pages are listed under the links that are undertaking such measurements.

Desktop-Computer GeForce 100 Series Radeon R700 Anti-aliasing Graphics processing unit#Integrated graphics solutions Texture mapping unit Electric power
364213
de