Contents • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • Field explanations [ ] The fields in the table listed below describe the following: • Model – The marketing name for the processor, assigned by Nvidia. • Launch – Date of release for the processor. • Code name – The internal engineering codename for the processor (typically designated by an NVXY name and later GXY where X is the series number and Y is the schedule of the project for that generation). • – Fabrication process. Average feature size of components of the processor. • interface – Bus by which the graphics processor is attached to the system (typically an expansion slot, such as PCI, AGP, or PCI-Express). • – The amount of graphics memory available to the processor.

• SM Count – Number of streaming multiprocessors. • Core – The factory core clock frequency; while some manufacturers adjust clocks lower and higher, this number will always be the reference clocks used by Nvidia. • Memory clock – The factory effective memory clock frequency (while some manufacturers adjust clocks lower and higher, this number will always be the reference clocks used by Nvidia).

Now I have some problems launching simple CUDA examples. First of all I compile NVidia driver and Linux kernel. NVidia Corporation NV44 [GeForce 6200 LE] (rev a1). You can read more about this family of drivers on GeForce.com. Stock Car; Update the following. GeForce 6200 LE, GeForce 6200 TurboCache(TM), GeForce 6200SE.

All DDR/GDDR memories operate at half this frequency, except for GDDR5, which operates at one quarter of this frequency. • Core config – The layout of the graphics pipeline, in terms of functional units. Over time the number, type, and variety of functional units in the GPU core has changed significantly; before each section in the list there is an explanation as to what functional units are present in each generation of processors. In later models, shaders are integrated into a unified shader architecture, where any one shader can perform any of the functions listed. • – Maximum theoretical fillrate in textured pixels per second.

This number is generally used as a maximum throughput number for the GPU and generally, a higher fillrate corresponds to a more powerful (and faster) GPU. • Memory subsection • Bandwidth – Maximum theoretical bandwidth for the processor at factory clock with factory bus width.

GHz = 10 9 Hz. • Bus type – Type of memory bus or buses used. • Bus width – Maximum bit width of the memory bus or buses used. This will always be a factory bus width.

• API support section • – Maximum version of Direct3D fully supported. • – Maximum version of OpenGL fully supported. • Features – Added features that are not standard as a part of the two graphics libraries. Further information: • All models are made via 220 nm fabrication process • All models support 7.0 and 1.2 • All models support hardware Transform and Lighting (T&L) and Cube Environment Mapping Model Launch Core clock () Memory clock () Core config 1 Memory MOperations/s MPixels/s MTexels/s MVertices/s Size () Bandwidth (/s) Bus type Bus width () GeForce 256 SDR October 11, 1999 NV10 AGP 4× PCI 120 166 4:4:4 480 480 480 0 32 64 2.656 SDR 128 GeForce 256 DDR February 1, 2000 NV10 AGP 4× PCI 120 150 4:4:4 480 480 480 0 32 64 4.8 DDR 128 • 1:: GeForce2 series [ ].

Further information: and All models support coverage sample anti-aliasing, angle-independent anisotropic filtering, and 128-bit OpenEXR HDR. • 1:: • 2 Full G80 contains 32 texture address units and 64 texture filtering units unlike G92 which contains 64 texture address units and 64 texture filtering units • 3 To calculate the processing power, see. Further information: and All models support Coverage Sample Anti-Aliasing, Angle-Independent Anisotropic Filtering, 128-bit OpenEXR HDR • 1:: • 2 To calculate the processing power see.

Further information: and • 1:: • 2 To calculate the processing power see. Model Launch Fab () Transistors (million) Die size (mm 2) Core config 1 Clock rate Memory configuration Supported version Processing power () 2 (Watts) Comments Core () Shader () Memory () Pixel (/s) Texture (/s) Size () Bandwidth (/s) DRAM type Bus width () GeForce G 100 March 10, 2009 G98 65 210 86 PCIe 2.0 ×16 8:8:4 567 1400 500 2.15 4.3 512 8.0 DDR2 64 10.0 3.3 22.4 35 OEM products GeForce GT 120 March 10, 2009 G96b 55 314 121 32:16:8 500 1400 800 4.4 8.8 512 16.0 DDR2 128 10.0 3.3 89.6 50 OEM products GeForce GT 130 March 10, 2009 G94b 55 505 196? [ ] 48:24:12 500 1250 500 6 12 1536 24.0 DDR2 192 10.0 3.3 120 75 OEM products GeForce GT 140 March 10, 2009 G94b 55 505 196? 64:32:16 650 1625 1800 10.4 20.8 512 1024 57.6 GDDR3 256 10.0 3.3 208 105 OEM products GeForce GTS 150 March 10, 2009 G92b 55 754 260 128:64:16 738 1836 1000 11.808 47.232 1024 64.0 GDDR3 256 10.0 3.3 470 141 OEM products GeForce 200 series [ ]. Further information: and All models support Coverage Sample Anti-Aliasing, Angle-Independent Anisotropic Filtering, 240-bit OpenEXR HDR • 1:: • 2 To calculate the processing power see.

Further information: and • 1:: • 2 To calculate the processing power see. Further information: and Memory bandwidths stated in the following table refer to Nvidia reference designs. Actual bandwidth can be higher or lower depending on the maker of the graphic board. • All cards have a PCIe 2.0 ×16. • The base requirement for Vulkan 1.0 in terms of hardware features was OpenGL ES 3.1 which is a subset of OpenGL 4.3, which is supported on all Fermi and newer cards. Further information: and • 1:: • 2 To calculate the processing power see. • 3 Each SM in the GF110 contains 4 texture filtering units for every texture address unit.

The complete GF110 die contains 64 texture address units and 256 texture filtering units. Each SM in the GF114/116/118 architecture contains 8 texture filtering units for every texture address unit but has doubled both addressing and filtering units. • 4 Internally referred to as GF104B • 5 Internally referred to as GF100B • 6 Similar to previous generation, GTX 580 and most likely future GTX 570, while reflecting its improvement over GF100, still have lower rated TDP and higher power consumption, e.g. GTX580 (243W TDP) is slightly less power hungry than GTX 480 (250W TDP). This is managed by clock throttling through drivers when a dedicated power hungry application is identified that could breach card TDP. Application name changing will disable throttling and enable full power consumption, which in some cases could be close to that of GTX480. • 7 Some companies have announced that they will be offering the GTX580 with 3GB RAM.

• 9 1024 MB RAM on 192-bit bus assemble with 4 × (128 MB) + 2 × (256 MB). Further information: and • 1:: • 2 The GeForce 605 (OEM) card is a rebranded GeForce 510. • 3 The GeForce GT 610 card is a rebranded GeForce GT 520. • 4 The GeForce GT 620 (OEM) card is a rebranded GeForce GT 520.

• 5 The GeForce GT 620 card is a rebranded GeForce GT 430 (DDR3, 64-bit). • 6 The GeForce GT 630 (DDR3, 128-bit, retail) card is a rebranded GeForce GT 430 (DDR3, 128-bit). • 7 The GeForce GT 630 (GDDR5) card is a rebranded GeForce GT 440 (GDDR5). • 8 The GeForce GT 640 (OEM) card is a rebranded GeForce GT 545 (DDR3). • 9 The GeForce GT 645 (OEM) card is a rebranded GeForce GTX 560 SE. • 10 To calculate the processing power see,.

Further information: and The GeForce 700 series for desktop. The GM107-chips are -based, the GKxxx-chips.

• 1:: • 2 Max Boost depends on ASIC quality. For example, some GTX TITAN with over 80% ASIC quality can hit 1019 MHz by default, lower ASIC quality will be 1006 MHz or 993 MHz. • 3 Kepler supports some optional 11.1 features on 11_0 through the Direct3D 11.1 API, however Nvidia did not enable four non-gaming features to qualify Kepler for level 11_1. • 4 The GeForce GT 705 (OEM) is a rebranded GeForce GT 610, which itself is a rebranded GeForce GT 520.

• 5 The GeForce GT 730 (DDR3, 64-bit) is a rebranded GeForce GT 630 (Rev. • 6 The GeForce GT 730 (DDR3, 128-bit) is a rebranded GeForce GT 430 (128-bit). • 7 The GeForce GTX 740 (OEM) is a rebranded GeForce GTX 650. • 8 The GeForce GTX 760 Ti (OEM) is a rebranded GeForce GTX 670. • 9 To calculate the processing power see,. • 10 As a Kepler GPC is able to rasterize 8 pixels per clock, fully enabled GK110 GPUs (780 Ti/TITAN Black) can only output 40 pixels per clock (5 GPCs), despite 48 ROPs and all SMX units being physically present. For GTX 780 and GTX 760, multiple GPC configurations with differing pixel fillrate are possible, depending on which SMXs were disabled in the chip: 5/4 GPCs, or 4/3 GPCs, respectively.

Further information: and The GeForce 900 series for desktop. The GM20x chips are -based. • 1:: • 2 Pixel fillrate is calculated as the number of ROPs multiplied by the base core clock speed • 3 Texture fillrate is calculated as the number of TMUs multiplied by the base core clock speed. • 4 To calculate the processing power see. • 6 For accessing its memory, the GTX 970 stripes data across 7 of its 8 32-bit physical memory lanes, at 196 GB/s. The last 1/8 of its memory (0.5 GiB on a 4 GiB card) is accessed on a non-interleaved solitary 32-bit connection at 28 GB/s, one seventh the speed of the rest of the memory space. Because this smaller memory pool uses the same connection as the 7th lane to the larger main pool, it contends with accesses to the larger block reducing the effective memory bandwidth not adding to it as an independent connection could.

Model Launch Fab () Transistors (million) Die size (mm 2) Core config 1 (SM/SMP/ Streaming Multiprocessor) Clock speeds Memory Latest supported version Processing power () 4 (Watts) support 7 Release price (USD) Base core clock () Boost core clock () Max Boost 2.0 () Memory () Pixel Base Core (Base Boost) (Max Boost 2.0) (/s) 2 Texture Base Core (Base Boost) (Max Boost 2.0) (/s) 3 Size () Bandwidth (/s) Bus type Bus width () Base Core (Base Boost) (Max Boost 2.0) Base Core (Base Boost) (Max Boost 2.0) MSRP GeForce GT 945A February 2016 GM108-? 512:24:8 (3) PCIe 3.0 ×8 1072 1176? 1800 (900 MHz x 2) 8.5 (9.4) (?) 25.7 (28.2) (?) 1 2 14.4 DDR3 64 12 4.6 1.2 1.0 1097.7 (1204.2) (?) 34.3 (37.6) (?) 33 No OEM GeForce GTX 950 August 20, 2015 GM206-250 2940 227 768:48:32 (6) PCIe 3.0 ×16 1024 1188 1290+ 6612 (1653 MHz x 4) 32.7 (38.0) (41.2)+ 49.1 (57.0) (61.9)+ 2 4 105.7 GDDR5 128 1572.8 (1824.7) (1981.4)+ 49.1 (57.0) (61.9)+ 90 4-way $159 GeForce GTX 960 January 22, 2015 GM206-300 1024:64:32 (8) 1127 1178 1228 7012 (1753 Mhz x 4) 36.0 (37.6) (39.2) 72.1 (75.3) (78.5) 112.1 128 2308.0 (2412.5) (2514.9) 72.1 (75.3) (78.5) 120 $199 November 26,2015 GM204-? 5200 398 1280:80:48 (10) 924??

5012 (1253 MHz x 4) 44.3 (?) (?) 73.9 (?) (?) 3 168.2 192 2365.4 (?) (?) 73.9 (?) (?)? Further information: and • 1:: • 2 Pixel fillrate is calculated as the lowest of three numbers: number of ROPs multiplied by the base core clock speed, number of rasterizers multiplied by the number of fragments they can generate per rasterizer multiplied by the base core clock speed, and the number of streaming multiprocessors multiplied by the number of fragments per clock that they can output multiplied by the base clock rate. • 3 Texture fillrate is calculated as the number of TMUs multiplied by the base core clock speed. • 4 To calculate the processing power see. • 5 SLI HB only supports a maximum of 2-way SLI using SLI HB bridges.

• 6 As the GTX 1070 has one of the four GP104 GPCs disabled in the die, its frontend is only able to rasterize 48 pixels per clock. Analogically, the GTX 1060 features only two GPCs on its GP106 die, meaning that its frontend can only rasterize 32 pixels per clock. The remaining backend ROPs can still be used for tasks such as MSAA. • 7 The performance of FP16 is half of the performance of FP64. There is only 1 FP16x2 core for every 128 FP32 cores. Further information: The GeForce 600M series for notebooks architecture.

The processing power is obtained by multiplying shader clock speed, the number of cores, and how many instructions the cores can perform per cycle. • 1:: Model Launch Fab () Core config 1 Clock speed Memory Supported version Processing power () 2 (Watts) Notes Core () Shader () Memory () Pixel (/s) Texture (/s) Size () Bandwidth (/s) Bus type Bus width () GeForce 610M December 2011 GF119 (N13M-GE) 40 PCIe 2.0 ×16 48:8:4 900 1800 1800 3.6 7.2 1024 2048 14.4 DDR3 64 12 4.5 142.08 12 OEM. Rebadged GT 520MX GeForce GT 620M April 2012 GF117 (N13M-GS) 28 PCIe 2.0 ×16 96:16:4 625 1250 1800 2.5 10 1024 2048 14.4 28.8 DDR3 64 128 12 4.5 240 15 OEM. Die-Shrink GF108 GeForce GT 625M October 2012 GF117 (N13M-GS) 28 PCIe 2.0 ×16 96:16:4 625 1250 1800 2.5 10 1024 2048 14.4 DDR3 64 12 4.5 240 15 OEM. Die-Shrink GF108 GeForce GT 630M April 2012 GF108 (N13P-GL) GF117 40 28 PCIe 2.0 ×16 96:16:4 660 800 1320 1600 1800 4000 2.6 3.2 10.7 12.8 1024 2048 28.8 32.0 DDR3 GDDR5 128 64 12 4.5 258.0 307.2 33 GF108: OEM. Rebadged GT 540M GF117: OEM Die-Shrink GF108 GeForce GT 635M April 2012 GF106 (N12E-GE2) GF116 40 PCIe 2.0 ×16 144:24:24 675 1350 1800 16.2 16.2 2048 1536 28.8 43.2 DDR3 128 192 12 4.5 289.2 388.8 35 GF106: OEM.

Further information: The GeForce 700M series for notebooks architecture. The processing power is obtained by multiplying shader clock speed, the number of cores, and how many instructions the cores can perform per cycle. • 1:: Model Launch Fab () Core config 1 Clock speed Memory Supported version Processing power () 2 (Watts) Notes Core () Shader () Memory () Pixel (/s) Texture (/s) Size () Bandwidth (/s) Bus type Bus width () GeForce 710M January 2013 GF117 28 PCIe 2. Game Kapal Perang Terbaru Untuk Pcos. 0 ×16 96:16:4 800 1600 1800 3.2 12.8 1024 2048 14.4 DDR3 64 12 4.5 307.2 12 OEM.

About 115% of Mobile 620 & Desktop 530 [ ] GeForce GT 720M April 1, 2013 GF117 28 PCIe 2.0 ×16 96:16:4 938 1876 2000 3.8 15.0 2048 16.0 DDR3 64 12 4.5 360.19? About 130% of Mobile 625/630 & Desktop 620 [ ] GeForce GT 730M January 2013 GK208 28 PCIe 3.0 ×8 384:32:8 (2 SMX) 719 719 2000 5.8 23.0 2048 16.0 DDR3 128 12 4.5 552.2 33 Kepler, similar to Desktop GT640 GeForce GT 735M April 1, 2013 GK208 28 PCIe 3.0 ×8 384:32:8 (2 SMX) 889 889 2000 7.11 28.4 2048 16.0 DDR3 64 12 4.5 682.8? Kepler, similar to Desktop GT640 GeForce GT 740M April 1, 2013 GK208 28 PCIe 3.0 ×8 384:32:8 (2 SMX) 980 980 1800 7.84 31.4 2048 14.4 DDR3 64 12 4.5 752.6? Kepler, similar to Desktop GT640. GeForce GT 740M April 1, 2013 GK107 28 PCIe 3.0 ×16 384:32:16 (2 SMX) 810 810 1800 5000 12.96 28.8 80 DDR3 GDDR5 128 12 4.5 622.1 45 about 76% of Desktop GTX650 [ ] GeForce GT 745M April 1, 2013 GK107 28 PCIe 3.0 ×16 384:32:16 (2 SMX) 837 837 2000 5000 13.4 26.8 2048 32 80 DDR3 GDDR5 128 12 4.5 642.8 45 about 79% of Desktop GTX650 [ ] GeForce GT 750M April 1, 2013 GK107 28 PCIe 3.0 ×16 384:32:16 (2 SMX) 967 967 2000 5000 15.5 30.9 2048 32 80 DDR3 GDDR5 128 12 4.5 742.7 50 about 91% of Desktop GTX650 [ ] GeForce GT 755M?

Further information: The GeForce 800M series for notebooks architecture. The processing power is obtained by multiplying shader clock speed, the number of cores, and how many instructions the cores can perform per cycle. Further information: The GeForce 900M series for notebooks architecture. The processing power is obtained by multiplying shader clock speed, the number of cores, and how many instructions the cores can perform per cycle.

Further information: • 1::: • * NV31, NV34 and NV36 are 2×2 pipeline designs if running vertex shader, otherwise they are 4×1 pipeline designs. Further information: • 1:: Model Fab () Core clock () Shader clock () Memory clock () Core config 1 Memory Processing power () Supported version (Watts) Notes Pixel (/s) Texture (/s) Size () Bandwidth (/s) Bus type Bus width () Quadro FX 370 G84 80 PCIe ×16 360 720 1000 16:8:4 1.44 2.88 256 6.4 DDR2 64 34.56 - 10.0 3.3 35 Quadro FX 370 LP G86 80 PCIe ×16 540 1080 1000 8:8:4 2.16 4.32 256 8 DDR2 64 25.92 - 10.0 3.3 25 DMS-59 for two Single Link DVI Quadro FX 470 MCP7A-U 65 PCIe 2.0 ×16 (Integrated) 580 1400 800 (system memory) 16:8:4 2.32 4.64 Up to 256MB from system memory. Further information: • 1:: • 4 Each SM in the Fermi architecture contains 4 texture filtering units for every texture address unit. Further information: • 1:: Model Fab () Core clock () Boost clock () Memory clock () Core config 1 Memory Processing power () Supported version (Watts) Notes Pixel (/s) Texture (/s) Size () Bandwidth (/s) Bus type Bus width () Quadro P400 GP107 16 PCIe 3.0 ×16 1070 1170 1752 (7008) 256:16:16 (2 SMM) 17.1 17.1 2 32 GDDR5 64 641 20 12.1 4.5 30 Three Mini-DisplayPort 1.4 Quadro P600 GP107 16 PCIe 3.0 ×16 1354 1455 1752 (7008) 384:24:16 (3 SMM) 21.7 32.5 2 64 GDDR5 64 1195??

4.5 40 Four Mini-DisplayPort 1.4 Quadro P1000 GP107 16 PCIe 3.0 ×16 1354 1455 1752 (7008) 640:40:32 (5 SMM) 43.3 54.2 4 82 GDDR5 128 1894?? 4.5 47 Four Mini-DisplayPort 1.4 Quadro P2000 GP106 16 PCIe 3.0 ×16 1370 1470 2002 (8008) 1024:64:40 (8 SMM) 54.8 87.7 5 140 GDDR5 160 3000?? 4.5 75 Four DisplayPort 1.4 Quadro P4000 GP104 16 PCIe 3.0 ×16 1227 1480 1502 (6008) 1792:112:64 (14 SMM) 78.5 137.4 8 243 GDDR5 256 5300?? 4.5 105 Four DisplayPort 1.4 Quadro P5000 GP104-875 16 PCIe 3.0 ×16 1607 1733 1126 (9008) 2560:160:64 (20 SMM) 102.8 257.1 16 288 GDDR5X 256 ~8900 ~300 12.1 4.5 180 Four DisplayPort 1.4, One DVI-I Quadro P6000 GP102-875 16 PCIe 3.0 ×16 1417 1531 1126 (9008) 3840:240:96 (32 SMM) 136.0 340.0 24 432 GDDR5X 384 10882 (11758) ~375 12.1 4.5 250 Four DisplayPort 1.4, One DVI-I Quadro GP100 GP100 16 PCIe 3.0 x16 ~1328 ~1480 703 (1406) 3584:?:??? 16 720 HBM2 4096 ~9519 (~10608) ~5300?? 235 NVLINK support Quadro NVS [ ]. Further information: • 1::: • 2:: • * NV31, NV34 and NV36 are 2×2 pipeline designs if running vertex shader, otherwise they are 4×1 pipeline designs.

Further information: • 1 Specifications not specified by Nvidia assumed to be based on the GTX • 2 Specifications not specified by Nvidia assumed to be based on the • 3 Specifications not specified by Nvidia are assumed to be based on the • 4 With ECC on, a portion of the dedicated memory is used for ECC bits, so the available user memory is reduced by 12.5%. 4 GB total memory yields 3.5 GB of user available memory.) • 5 To calculate the processing power see,,,,. A number range specifies the minimum and maximum processing power at, respectively, the base clock and maximum boost clock. • 6 Specifications not specified by Nvidia assumed to be based on the Quadro FX 5800 • 7 GPU Boost is a default feature that increases the core clock rate while remaining under the card's predetermined power budget. Multiple boost clocks are available, but this table lists the highest clock supported by each card. • 8 Core architecture version according to the programming guide. • For the basic specifications of Tesla, refer to the GPU Computing Processor specifications.

• Because of Tesla's non-output nature, fillrate and graphics API compatibility are not applicable. Further information: Early mobile Quadro chips based on the Geforce2 Go up to Geforce Go 6. Precise reliable statistics on early mobile workstation chips appear to be scarce and conflicting between Nvidia press releases and product lineups with GPU databases. • 1::: • 2:: Model Fab () Core clock () Shader clock () Memory clock () Core config 12 Memory Processing power () Supported version (Watts) Notes Pixel (/s) Texture (/s) Size () Bandwidth (/s) Bus type Bus width () Quadro2 Go NV11GLM? AGP 4× 143 130 360 2:0:4:2 0.286 0.592 32 64 2.9 5.8 SDR DDR 128 7? 1.2 First mobile Quadro based on the Geforce2 Go, Dynamic core clock 100–143 MHz, Dynamic Voltage 1.575V, graphics core listed as 64-bit and 128-bit for DDR and SDR SDRAM types respectively according to Nvidia Quadro4 500 Go GL NV17 GLM 150 AGP 4×,8×??

2:0:4:2 0.44 0.88 64 7.0 DDR 128 7.0 1.3 Based on the Geforce4 Go, dynamic core clock 66–220 MHz, core voltage 1.35 V, uses DDR SDRAM according to Nvidia brochures Quadro4 700 Go GL NV28 A1 GLM 150 AGP 4×,8× 176 200 4:2:4:4 0.704 0.704 64 7.4? 8.1 1.3 Based on Geforce4 Go 4200, uses DDR according to Nvidia brochures Quadro FX Go 700 NV31 GLM 130 AGP 8× 295? 4:3:4:4 1.18 1.18 128 9.44? Based on the Geforce FX Go 5 (FX 56XX), uses DDR SDRAM. Quadro FX Go 1000 NV36 GLM 120 AGP 8× 295? 4:3:4:4 1.18 1.18 128 9.12?

Quadro FX Go 1400 NV41 GLM 130 PCIE 275 590 12:5:12:12? 2.20 2.2 256 18.9? Last chip designated as a Quadro FX Go, apparently uses PCIE instead of AGP 8×. Core config has been mentioned as either 8:5:8:8 or 12:5:12:12 - the latter is likely since chip is derived from Geforce Go 6800. Quadro FX (x500M) series [ ]. Further information: • 1:: Model Fab () Core clock () Shader clock () Memory clock () Core config 1 Memory Processing power () Supported version (Watts) Notes Pixel (/s) Texture (/s) Size () Bandwidth (/s) Bus type Bus width () Quadro FX 360M G86M 80 PCIe 1.0 ×16 400 800 1200 16:8:4 1.6 3.2 256 9.6 DDR2 64 38.4 10.0 3.3 17 Quadro FX 560M G73GLM 90 PCIe 1.0 ×16 500 500 1200 5:12:12:8 4 6 512 19.2 GDDR3 128 9.0c 2.1 35? Quadro FX 1600M G84M 80 PCIe 1.0 ×16 625 1250 1600 32:16:8 5 10 512 25.6 GDDR3 128 120 10.0 3.3 50?

Quadro FX 3600M G92M 65 PCIe 1.0 ×16 500 1250 1600 64:32:16 96:48:16 8 8 16 24 512 51.2 GDDR3 256 240 360 10.0 3.3 70 Quadro FX (x700M) series [ ]. Further information: • 1::: • 2:: Model Fab () Core clock () Shader clock () Memory clock () Core config 12 Memory Processing power () Supported version (Watts) Notes Pixel (/s) Texture (/s) Size () Bandwidth (/s) Bus type Bus width () Quadro NVS 110M G72M 90 PCIe 1.0 ×16 300 300 600 3:4:4:2 0.6 1.2 Up to 512 4.8 DDR 64 9.0c 2.1 10 Quadro NVS 120M G72GLM 90 PCIe 1.0 ×16 450 450 700 3:4:4:2 0.9 1.8 Up to 512 5.6 DDR2 64 9.0c 2.1 10 Quadro NVS 130M G86M 80 PCIe 2.0 ×16 400? 700 8:4:4 1.6? Up to 256 6.4? This section includes a, related reading or, but its sources remain unclear because it lacks. Please help to this section by more precise citations.

PC users looking for desktop PCs or thin and light notebook PCs have long had to compromise features and performance for value. Traditionally, the latest generation graphics hardware for gaming and multimedia applications has been out of reach for many consumers.

Now, with NVIDIA® TurboCache™ technology, PC users can get NVIDIA GeForce™ 6 Series performance and features at an affordable price. NVIDIA TurboCache is available in the NVIDIA GeForce 6200 GPU for PCI Express desktop PCs and NVIDIA GeForce Go 6200 GPU for PCI Express notebook PCs. Benefits of these GPUs include: • Support for up to 128MB and 256MB configurations • Only Microsoft® DirectX™ 9.0c Shader Model 3.0 GPUs in its class lets you play all the latest games • NVIDIA® PureVideo™technology delivers crisp, vibrant video on the PC What people are saying about the GeForce 6200 with TurboCache: 'The GeForce 6200 with TurboCache technology is the fastest sub-$100 graphics card we have ever tested.'

- PC Perspective 'This gives casual consumers the ability to see what having a 'real' graphics (processor) is like.' - Anandtech 'It brings the very same feature set found on the flagship GeForce 6800 Ultra down to a very affordable price point. The GeForce 6200 with TurboCache is also a very power efficient design that surpasses the capabilities of virtually all integrated graphics processors.' - Hot Hardware How Does it Work?

The revolutionary TurboCache technology utilizes the additional bandwidth of the PCI Express graphics bus to reach higher levels of graphics performance than traditional video memory solutions, delivering the performance and features you expect from NVIDIA graphics hardware. By allowing the graphics processing unit (GPU) to share the capacity and bandwidth of dedicated video memory and dynamically available system memory, TurboCache turbocharges performance and provides larger total graphics memory. Key features of the TurboCache architecture: • Patented hardware and software technologies that render directly to system memory • A TurboCache Manager (TCM), which dynamically allocates memory for maximum system performance • Intelligent software algorithms that maximize application performance • Bidirectional PCI Express® bandwidth, in conjunction with TurboCache architecture, which improves graphics price/performance Through its unique ability to render directly to system memory, TurboCache fundamentally redefines the price/performance of desktop and notebook PC graphics solutions. With the innovative TurboCache technology from NVIDIA, users looking for an affordable desktop or thin and light notebook solution can now experience the latest graphics and video features of the GeForce 6 Series GPUs, without compromise.