Hubbry Logo
GeForce 9 seriesGeForce 9 seriesMain
Open search
GeForce 9 series
Community hub
GeForce 9 series
logo
8 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
GeForce 9 series
GeForce 9 series
GeForce 9 series
View on Wikipedia
from Wikipedia

GeForce 9 series
An Nvidia GeForce 9800 GX2 released in 2008, the series' flagship model
Release dateFebruary 21, 2008; 18 years ago (February 21, 2008)
CodenameG9x
ArchitectureTesla
ModelsGeForce series
  • GeForce GE series
  • GeForce GS series
  • GeForce GSO series
  • GeForce GT series
  • GeForce GTX series
  • GeForce GTX+ series
  • GeForce GX2 series
Transistors210M 65 nm (G98)
  • 314M 65 nm (G96)
  • 505M 65 nm (G94)
  • 754M 65 nm (G92)
  • 754M 55 nm (G92B)
Cards
Entry-level
  • 9100
  • 9200
  • 9300 GS/GT
  • 9300 GE
  • 9400 GS/GT
Mid-range
  • 9500 GS/GT
  • 9600
High-end9800 GT/GTX/GTX+
Enthusiast9800 GX2
API support
Direct3DDirect3D 10.0
Shader Model 4.0
OpenCLOpenCL 1.1
OpenGLOpenGL 3.3
History
PredecessorGeForce 8 series
VariantGeForce 100 series
SuccessorGeForce 200 series
Support status
Unsupported

The GeForce 9 series (also known as the GeForce 9000 series) is the ninth generation of Nvidia's GeForce line of graphics processing units, the first of which was released on February 21, 2008. The products are based on an updated Tesla microarchitecture, adding PCI Express 2.0 support, improved color and z-compression, and built on a 65 nm process, later using 55 nm process to reduce power consumption and die size (GeForce 8 G8x GPUs only supported PCIe 1.1 and were built on 90 nm process or 80 nm process).

GeForce 9100 series

[edit]

Geforce 9100 G

[edit]
  • 65 nm G98 GPU
  • PCI-E x16
  • 64-bit bus
  • 4 raster operations pipelines (ROP), 8 unified shaders
  • 540 megahertz (MHz) core clock
  • 256 MB DDR2, 400 MHz memory clock
  • 1300 MHz shader clock
  • 5.1 G texels/s fill rate
  • 7.6 GB/s memory bandwidth
  • Supports DirectX 10, SM 4.0
  • OpenGL 2.1 compliance
  • Supports 1st generation PureVideo HD technology with partial VC1 decoding

GeForce 9300 series

[edit]

Geforce 9300 GS

[edit]

On May 1, 2008, the GeForce 9300 GS was officially launched.[1]

  • 80 nm G86 GPU
  • PCI-E x16
  • 64-bit bus
  • 8 ROP, 16 unified shaders
  • 450 MHz core clock
  • 512 MB DDR2, 400 MHz memory clock
  • 900 MHz shader clock
  • 3.6 Gtexels/s fill rate
  • 6.4 GB/s memory bandwidth
  • Supports DirectX 10, SM 4.0
  • OpenGL 2.1 compliance

GeForce 9400 series

[edit]

GeForce 9400 GT

[edit]

On August 27, 2008, the GeForce 9400 GT was officially launched.

  • 65 nm G96 GPU
  • 16 stream processors[2][3]
  • 550 MHz core, with a 1350 MHz unified shader clock
  • 4.4 Gtexels/s fillrate
  • 256/512/1024 MB 800 MHz DDR2 or 256 MB 1600 MHz GDDR3,[4] both with a 128-bit memory bus
  • 12.8 GB/s memory bandwidth for boards configured with DDR2 800 MHz memory
  • Supports DirectX 10, Shader Model 4.0, OpenGL 3.3, and PCI-Express 2.0
  • Supports 2nd generation PureVideo HD technology with partial VC1 decoding and HybridPower technology.[5]
  • Minimum of 300 watt power supply

GeForce 9500 series

[edit]

GeForce 9500 GT

[edit]
Gigabyte GeForce 9500 GT
BFG GeForce 9500 GT without heatsink

On July 29, 2008, the GeForce 9500 GT was officially launched.

  • 65 nm G96 GPU
  • 32 stream processors (32 CUDA cores)
  • 4 multi processors (each multi processor has 8 cores)
  • 550 MHz core, with a 1400 MHz unified shader clock
  • 8.8 Gtexels/s fillrate
  • 256/512/1024 MB 1,600 MHz GDDR3 memory or 256 MB/512 MB 1,000 MHz GDDR2 memory, both with a 128-bit memory bus
  • 25.6 GB/s memory bandwidth for boards configured with GDDR3 800 MHz memory
  • Supports DirectX 10, Shader Model 4.0, OpenGL 3.3, and PCI-Express 2.0
  • Supports 2nd generation PureVideo HD technology with partial VC1 decoding[6][unreliable source?][7]
  • Nvidia SLI-ready technology
  • DVI support

GeForce 9500 GS

[edit]

The 9500 GS is an OEM card that is based on the 9500 GT but geared towards the mainstream audience.

  • 65 nm G96 GPU
  • 32 stream processors
  • 8 ROP units
  • 550 MHz core, with a 1375 MHz defined unified shader clock
  • 8.8 Gtexels/s fillrate
  • 128/512 MB 1000 MHz DDR2 memory with a 128-bit memory bus
  • 16.0 GB/s memory bandwidth
  • Supports DirectX 10, Shader Model 4.0, OpenGL 3.3, and PCI-Express 2.0
  • Supports 2nd generation PureVideo HD technology with partial VC1 decoding
  • Nvidia SLI-ready technology
  • DVI support

GeForce 9600 series

[edit]

GeForce 9600 GT

[edit]
GeForce 9600 GT with cooler removed
Asus Geforce 9600 GT
Nvidia G94 GPU on a Geforce 9600 GT

On February 21, 2008, the GeForce 9600 GT was officially launched. It was an upgrade of 8600 GTS.

  • 65 nm G94 GPU
  • 64 CUDA cores[8]
  • 16 raster operation (ROP) units, 32 texture address (TA) / texture filter (TF) units
  • 20.8 Gtexels/s fill rate
  • 650 MHz core clock, with a 1625 MHz unified shader clock
  • 1008 MHz memory (2016 MHz datarate), 256-bit interface for 64.5 GB/s of bandwidth. (57.6 GB/s for 1800 MHz configuration)
  • 512–2048 MB of GDDR3 or DDR2 memory
  • 505M transistor count
  • DirectX 10.0, Shader Model 4.0, OpenGL 2.1, and PCI-Express 2.0[9]
  • Supports second-generation PureVideo HD technology with partial VC1 decoding
  • Is HDCP compatible, but its implementation depends on the manufacturer
  • Supports CUDA and the Quantum Effects physics processing engine
  • Almost double the performance of the previous Nvidia mid-range card, the GeForce 8600GTS

GeForce 9600 GS

[edit]
GeForce 9600GS from an HP computer; notice the SLI connector. Unlike the 9600GT, it doesn't require an external PCI-Express power supply.

The GeForce 9600GS is a Hewlett Packard OEM card. It is based on a G94a core clocked at 500 MHz. It features 768 MB of DDR2 memory on a 192-bit bus.

GeForce 9600 GSO

[edit]

The GeForce 9600 GSO was essentially a renamed 8800 GS. This tactic has been seen before in products such as the GeForce 7900 GTO to clear unsold stock when it is made obsolete by the next generation. Just like the 8800 GS, the 9600 GSO features 96 stream processors, a 550 MHz core clock with shaders clocked at 1,375 MHz, and either 384 or 768 MB of memory clocked at 800 MHz on a 192-bit memory bus. Some manufacturers have mistakenly listed some of their 768 MB models that have 96 stream processors as being based on the G94 chip, rather than the G92.[10]

GeForce 9600 GSO 512

[edit]

After clearing the old 8800 GS stock, Nvidia revised the specification with a new core, and 512 MB of memory clocked at 900 MHz on a 256-bit bus.[11] For these cards, the number of stream processors is halved to 48, with the core frequency increased to 650 MHz and the shader frequency increased to 1625 MHz. Some of these cards have 1024 megabytes of memory while still being a 512 model. The revised version is considered inferior in performance to the old version.[according to whom?]

GeForce 9600 GTX

[edit]

XFX released a 9600 GTX based on the G92 chip featuring 96 stream processors, a 580 MHz core clock, 1450 MHz shaders and 512 MB of GDDR3 running at 1400 MHz on a 256-bit bus. Other than clock speeds, it is functionally the desktop equivalent version of the 9800M GT.[12]

GeForce 9800 series

[edit]

The GeForce 9800 series contains the GX2 (dual GPU), GTX, GTX+ and GT variants.[13]

GeForce 9800 GX2

[edit]

On March 18, 2008, the GeForce 9800 GX2 was officially launched.

The GeForce 9800 GX2 has the following specifications:[14][15]

  • Dual PCBs, dual GPU design
  • about 197 W power consumption.[16]
  • Two 65nm process GPUs, with 256 total stream processors (128 per PCB).[17][unreliable source?]
  • Supports Quad SLI
  • Power of Two underclocked GeForce 8800 GTS 512 (G92) video cards in SLI Mode
  • 1 GiB (512 MiB per PCB) GDDR3 memory
  • Supports DirectX 10, Shader Model 4, OpenGL 3.3, and PCI-Express 2.0
  • Supports 2nd generation PureVideo HD technology with partial VC1 decoding
  • Outputs include two DVI ports, an HDMI output, and S/PDIF in connector on board for routing audio through the HDMI cable.[18][unreliable source?]
  • An 8-pin and a 6-pin power connector
  • Clocks (Core/Shader/Memory): 600 MHz/1500 MHz/2000 MHz[19]
  • 256-bit memory interface[19]
  • 128 GB/s memory bandwidth[19]
  • Release date: March 18, 2008
  • Launch price of $666.99[20][failed verification]

GeForce 9800 GTX

[edit]

On April 1, 2008, the GeForce 9800 GTX was officially launched.

Taken from an eVGA specification sheet:[21]

  • 128 CUDA cores
  • Clocks (Core/Shader/Memory): 675 MHz/1688 MHz/1100 MHz
  • 256-bit memory interface
  • 512 MB of GDDR3 memory
  • 70.4 GB/s memory bandwidth
  • Texture Fill Rate of 43.2 (billion/s)
  • DirectX 10, Shader Model 4.0, OpenGL 3.3, and PCI-Express 2.0
  • Supports 2nd generation PureVideo HD technology with partial VC1 decoding
  • Outputs include two DVI ports, an HDMI output (using Nvidia DVI to HDMI adapter (included)), and S/PDIF in connector on board for routing audio through the HDMI cable
  • Release date: 2008-04-01[22][unreliable source?]
  • Launch Price of $349[23][unreliable source?]

In July 2008 Nvidia released a refresh of the 9800 GTX: the 9800 GTX+ (55 nm manufacturing process). It has faster core (738 MHz) and shader (1836 MHz) clocks. Since March 2009 this design is manufactured as GeForce GTS 250.

GeForce 9800 GT

[edit]

The 9800GT is identical to an 8800GT, although some were manufactured using a 55 nm technology instead of the 65 nm technology that debuted on the 8800GT.[24] The newer (55 nm) version supports HybridPower while the 65 nm version does not.

ASUSTeK have released a 9800GT with Tri-SLI support.[25]

Taken from the Nvidia product detail page.[26]

  • 112 processor cores
  • 512–1024 MB of GDDR3 memory
  • 256-bit memory interface width
  • 600 MHz graphics clock
  • 1500 MHz processor clock
  • 900 MHz memory clock
  • 33.6 Gtexel/s texture fill rate
  • 57.6 GB/s memory bandwidth
  • Supports DirectX 10, Shader Model 4.0, OpenGL 3.3, and PCI-Express 2.0
  • Supports 2nd generation PureVideo HD technology with partial VC1 decoding

Technical Summary of Desktop G9x GPUs

[edit]
Model Year Code name Fab (nm) Transistors (Million) Die size (mm2) Bus interface Config core1 Clock rate Fillrate Memory API support (version) Processing power G FLOPs TDP (watts) Comments
Core (MHz) Shader (MHz) Memory (MHz) Pixel (GP/s) Texture (GT/s) Size (MB) Bandwidth (GB/s) Bus type Bus width (bit) DirectX OpenGL Vulkan
GeForce 9300 mGPU October 2008 MCP7A-S 65 282 162 PCIe 2.0 x16 16:8:4 450 1200 800
1333 		1.8 3.6 Up to 512 from system memory 6.4/12.8
10.664/21.328 		DDR2
DDR3 		64/128 10.0 3.3 N/a 57.6 based on 8400 GS
GeForce 9400 mGPU October 2008 MCP7A-U 65 282 162 PCIe 2.0 x16 16:8:4 580 1400 800
1333 		2.32 4.64 Up to 512 from system memory 6.4/12.8
10.664/21.328 		DDR2
DDR3 		64/128 67.2 12 based on 8400 GS
GeForce 9300 GE[27] June 2008 G98 65 ? 86 PCIe 2.0 x16 8:8:4 540 1300 500 2.16 4.3 256 8 DDR2 64 31.2 25
GeForce 9300 GS[27] June 2008 G98 65 ? 86 PCIe 2.0 x16 8:8:4 567 1400 500 2.268 4.5 512 8 DDR2 64 33.6 ??
GeForce 9400 GT August 27, 2008 G96a/b 65/55 314 144 PCIe 2.0 x16, PCI 16:8:4 550 1400 800
1600 		2.2 4.4 256, 512, 1024 12.8
25.6 		GDDR2
GDDR3 		128 67.2 50
GeForce 9500 GT July 29, 2008 G96a/b 65/55 314 144 PCIe 2.0 x16, PCI 32:16:8 550 1400 1000
1600 		4.4 8.8 256, 512, 1024 16.0
25.6 		DDR2
GDDR3 		128 134.4 50
GeForce 9600 GSO May 2008 G92 65 754 324 PCIe 2.0 x16 96:48:12 550 1375 1600 6.6 26.4 384, 768, 1536 38.4 GDDR3 192 396 84
GeForce 9600 GSO 512 October 2008 G94a/b 65/55 505 240/196? PCIe 2.0 x16 48:24:16 650 1625 1800 10.4 15.6 512 57.6 GDDR3 256 234 90
GeForce 9600 GT Green Edition 2009 G94b 55 505 196? PCIe 2.0 x16 64:32:16 600
625 		1500
1625 		1400/1800
1800 		9.6
10 		19.2
20 		512, 1024 44.8/57.6
57.6 		GDDR3 256 288
312 		59 Core Voltage 1.0V
GeForce 9600 GT February 21, 2008 G94a/b 65/55 505 240/196? PCIe 2.0 x16 64:32:16 650 1625 1800 10.4 20.8 512, 1024, 2048 57.6 GDDR3 256 312 95
GeForce 9800 GT Green Edition 2009 G92b 55 754 260 PCIe 2.0 x16 112:56:16 550 1375 1400
1600
1800 		8.8 30.8 512, 1024 44.8
51.2
57.6 		GDDR3 256 462 75 Core Voltage 1.0V
GeForce 9800 GT July 2008 G92a/b/a2 65/55/65 754 324/260/324 PCIe 2.0 x16 112:56:16 600/600/550 1500/1500/1375 1800 9.6 33.6 512, 1024 57.6 GDDR3 256 504/504/465 125/105/75 Some 65 nm cards are rebranded 8800 GT cards.

G92a2 Core Voltage 1.0V

GeForce 9800 GTX April 1, 2008 G92 65 754 324 PCIe 2.0 x16 128:64:16 675 1688 2200 10.8 43.2 512 70.4 GDDR3 256 648.192 140
GeForce 9800 GTX+ July 16, 2008 G92b 55 754 260 PCIe 2.0 x16 128:64:16 738 1836 2200 11.808 47.232 512, 1024 70.4 GDDR3 256 705.024 141
GeForce 9800 GX2 March 18, 2008 2× G92 65 2× 754 2× 324 PCIe 2.0 x16 2× 128:64:16 600 1500 2000 2× 9.6 2× 38.4 2× 512 2× 64.0 GDDR3 2× 256 2× 576 197
Model Year Codename Fab (nm) Transistors (Million) Die size (mm2) Bus interface Config core1 Core (MHz) Shader (MHz) Memory (MHz) Pixel (GP/s) Texture (GT/s) Size (MiB) Bandwidth (GB/s) Bus type Bus width (bit) DirectX OpenGL Vulkan Processing power G FLOPs TDP (watts) Comments

Features

[edit]
  • Compute Capability: 1.1 has support for Atomic functions, which are used to write thread-safe programs.
Model Features
Scalable Link Interface (SLI) PureVideo 2 with VP2,
BSP Engine, and AES128 Engine 		PureVideo 3 with VP3,
BSP Engine, and AES128 Engine
GeForce 9300 GE (G98) Yes No Yes
GeForce 9300 GS (G98)
GeForce 9400 GT Yes No
GeForce 9500 GT
GeForce 9600 GSO
GeForce 9600 GT
GeForce 9800 GT
GeForce 9800 GTX Yes
3-way
GeForce 9800 GTX+
GeForce 9800 GX2 Yes

GeForce 9M Series

[edit]

All graphical processing units in the GeForce 9M series feature:

  • Increased performance for similar power draw compared to GeForce 8M series for midrange and mid-high range notebooks
  • DirectX 10.0 and OpenGL 3.3 compatibility
  • 16X antialiasing and PCI-Express 2.0 connectivity
  • Full HD DVD / Blu-ray hardware decoding

9100M G

[edit]
  • 1 TMU per pipeline
  • 4 ROPs
  • 8 stream processors
  • 16 (v4.0) shader unified
  • 26 GigaFLOPS
  • 450 MHz core clock
  • 1100 MHz shader clock
  • Integrated RAMDAC clock at 400 MHz
  • Memory clock depends on system memory
  • Up to 256 MB shared memory, 512 MB with Turbo Cache in Windows XP
  • 64 bit memory interface (single-channel mode) / 128 bit memory interface (dual-channel mode)
  • Memory bandwidth depend on System Memory
  • 1.8 Gtexels/s texture fill rate
  • (Specification based on Acer Aspire 4530 using EVEREST Ultimate Edition Version 4.60.1500PX and TechPowerUp GPU-Z v0.4.6)[28][unreliable source?]

9200M GS

[edit]
  • 8 stream processors
  • 529 MHz core clock
  • 1300 MHz shader clock
  • 400 MHz memory clock
  • Up to 256 MB memory
  • 64-bit memory interface
  • 6.4 GB/s memory bandwidth
  • 27.1 Gpixel/s pixel fill rate
  • 4.2 Gtexel/s texture fill rate[29][unreliable source?]

9300M G

[edit]
  • 16 stream processors
  • 400 MHz core clock
  • 800 MHz shader clock
  • 600 MHz memory clock
  • Up to 512 MB memory
  • 64-bit memory interface
  • 1.8 GB/s memory bandwidth
  • 3.2 Gtexels/s texture fill rate

9300M GS

[edit]
  • 8 stream processors
  • 580 MHz core clock
  • 1450 MHz shader clock
  • 800 MHz memory clock
  • Up to 512 MB memory
  • 64-bit memory interface
  • 6.9 GB/s memory bandwidth
  • 4.6 Gtexels/s texture fill rate

9400M G

[edit]
  • 16 stream processors
  • Memory clock depends on system memory
  • 64 bit memory interface (single-channel mode) / 128 bit memory interface (dual-channel mode)
  • Memory bandwidth depends on System Memory
  • 3.6 Gtexels/s texture fill rate

9500M G

[edit]
  • 16 stream processors
  • 500 MHz core clock
  • 1250 MHz shader clock
  • 800 MHz memory clock
  • Up to 1024 MB memory
  • 128-bit memory interface
  • 25.6 GB/s memory bandwidth
  • 4.0 Gtexels/s texture fill rate

9500M GS

[edit]
  • 32 stream processors
  • 475 MHz core clock
  • 950 MHz shader clock
  • 700 MHz memory clock
  • Up to 512 MB memory
  • 128-bit memory interface
  • 22.4 GB/s memory bandwidth
  • 7.6 Gtexels/s texture fill rate

9600M GS

[edit]
  • 064A/8 core (G96)
  • 32 stream processors
  • 430 MHz core clock
  • 1075 MHz shader clock
  • 800/1600 MHz memory clock (effective)
  • Up to 1024 MB memory
  • 128-bit memory interface
  • 12.8 GB/s (with DDR2 type) or 25.6 GB/s (with GDDR3 type) memory bandwidth
  • 6.8 Gtexels/s texture fill rate
  • 103 GigaFLOPS

9600M GT

[edit]
  • 32 stream processors
  • 500 MHz core clock
  • 1250 MHz shader clock
  • 800 MHz memory clock
  • Up to 1024 MB memory
  • 128-bit memory interface
  • 25.6 GB/s memory bandwidth
  • 8.0 Gtexels/s texture fill rate

9650M GT

[edit]
  • G96 core (65/55 nm)
  • 32 stream processors
  • 550 MHz core clock
  • 1325 MHz shader clock
  • 800 MHz memory clock
  • Up to 1024 MB memory
  • 128 bit memory interface
  • 25.6 GB/s memory bandwidth
  • 8.8 Gtexels/s texture fill rate

9700M GT

[edit]
  • G96 core
  • 32 stream processors
  • 625 MHz core clock
  • 1550 MHz shader clock
  • 800 MHz memory clock
  • 128 bit memory interface
  • 25.6 GB/s memory bandwidth
  • 10.0 Gtexels/s texture fill rate
  • 148.8 GigaFLOPS

9700M GTS

[edit]
  • G94 core
  • 48 stream processors
  • 530 MHz core clock
  • 1325 MHz shader clock
  • 800 MHz memory clock
  • 256 bit memory interface
  • 51.2 GB/s memory bandwidth
  • 12.7 Gtexels/s texture fill rate
  • 190.8 GigaFLOPS

9800M GS

[edit]
  • G94 core
  • 64 stream processors
  • 530 MHz core clock
  • 1325 MHz shader clock
  • 800 MHz memory clock
  • 256 bit memory interface
  • 51.2 GB/s memory bandwidth
  • 17.0 Gtexels/s texture fill rate
  • 254 GigaFLOPS

9800M GTS

[edit]
  • G94 core
  • 64 stream processors
  • 600 MHz core clock
  • 1500 MHz shader clock
  • 800 MHz memory clock
  • 256 bit memory interface
  • 51.2 GB/s memory bandwidth
  • 19.2 Gtexels/s texture fill rate
  • 288 GigaFLOPS

9800M GT

[edit]
  • G94 core
  • 96 stream processors
  • 500 MHz core clock
  • 1250 MHz shader clock
  • 800 MHz memory clock
  • 256 bit memory interface
  • 51.2 GB/s memory bandwidth
  • 24.0 Gtexels/s texture fill rate
  • 360 GigaFLOPS

9800M GTX

[edit]
  • G92 core
  • 112 stream processors
  • 500 MHz core clock
  • 1250 MHz shader clock
  • 800 MHz memory clock
  • 256 bit memory interface
  • 51.2 GB/s memory bandwidth
  • 28.0 Gtexels/s texture fill rate
  • 420 GigaFLOPS

Technical summary

[edit]
Model Release date Codename Interface Fabrication process (nm) Core clock max (MHz) Peak fillrate Shaders Memory Texture units Raster operators Power consumption (Watts) Transistor count (Millions) Theoretical shader processing rate (GigaFLOPS)
Billion pixel/s Billion bilinear texel/s Billion bilinear FP16 texel/s Billion FP32 pixel/s CUDA cores Clock (MHz) Bandwidth max (GB/s) DRAM type Bus width (bit) Size (MB) Effective DDR clock (MHz)
GeForce 9100M G ? MCP77MH MCP79MH PCI 65 450 ? ? ? ? 8 1080 21GB/s DDR2 from RAM depend RAM configuration depend RAM configuration depend RAM configuration ? 4 ? ? 26
GeForce 9200M GS G98 65 530 ? ? ? 8 1300 GDDR2 GDDR3 256 ? ? ? ? 31
GeForce 9300M G G98 65 400 ? 3.2 ? ? 16 800 9.6 64 256 1200 (600) ? ? ? ? 38
GeForce 9300M GS G98 65 ? ? ? ? ? 16 1400 ? GDDR2 GDDR3 64 256 1400 (700) ? ? ? ? 34
GeForce 9400M G MCP79MX 65 450 ? ? ? ? 16 1100 ? 128 ? ? 12 282 54
GeForce 9500M G 65 ? ? ? ? ? 16 1250 ? GDDR2 GDDR3 128 256, 512, 1024 1600 (800) ? ? ? ? 60
GeForce 9500M GS G84 65 479 ? 7.6 ? ? 32 950 22.4 128 512 1400 (700) ? 8 ? 289 ?
GeForce 9600M GS G96 65 430 ? ? ? ? 32 1075 25.3 GDDR2 GDDR3 128 1024 1600 (800) ? ? ? 314 103
GeForce 9600M GT G96 65 500 ? ? ? ? 32 1250 ? GDDR2 GDDR3 128 256, 512, 1024 1600 (800) ? ? 23 314 120
GeForce 9650M GS 65 625 ? 10 ? ? 32 1250 25.6 128 512 1600 (800) ? ? 29 289 120
GeForce 9700M GT G96 625 ? 10 ? ? 32 1550 25.6 GDDR3 128 512 1600 (800) ? ? ? ? 148.8
GeForce 9700M GTS G94 530 ? 12.7 ? ? 48 1325 51.2 256 ? 1600 (800) ? ? ? ? 190.8
GeForce 9800M GTS G94 600 ? 19.2 ? ? 64 1500 51.2 256 512 1600 (800) ? ? ? ? 288
GeForce 9800M GT G94 500 ? 24 ? ? 96 1250 51.2 256 512 1600 (800) ? ? ? ? 360
GeForce 9800M GTX G92 500 ? 28.0 ? ? 112 1375 51.2 256 512 1600 (800) ? 75 ? ? 420

Support

[edit]

Nvidia announced that as of April 1, 2016, they would cease driver support for the GeForce 9 series.[30][31]

  • Windows XP 32-bit & Media Center Edition: version 340.52 released on July 29, 2014; Download
  • Windows XP 64-bit: version 340.52 released on July 29, 2014; Download
  • Windows Vista, 7, 8, 8.1 32-bit: version 342.01 (WHQL) released on December 14, 2016; Download
  • Windows Vista, 7, 8, 8.1 64-bit: version 342.01 (WHQL) released on December 14, 2016; Download
  • Windows 10, 32-bit: version 342.01 (WHQL) released on December 14, 2016; Download
  • Windows 10, 64-bit: version 342.01 (WHQL) released on December 14, 2016; Download

See also

[edit]

References

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

GeForce 9 series

View on Grokipedia
from Grokipedia
The GeForce 9 series is a family of graphics processing units (GPUs) developed by NVIDIA Corporation, released starting in February 2008 as the successor to the GeForce 8 series. Based on the Tesla microarchitecture with minor enhancements for efficiency, the series was manufactured using 65 nm and 55 nm process nodes, enabling lower power consumption and costs compared to prior generations while supporting DirectX 10, unified shaders, and advanced video processing via NVIDIA PureVideo HD technology. The lineup debuted with the mid-range GeForce 9600 GT on February 21, 2008, featuring 64 cores, a 65 nm G94 GPU, and a 256-bit interface for improved performance in gaming and applications. Subsequent releases included high-end models such as the GeForce 9800 GT (July 21, 2008, 112 cores on 65 nm G92) and the dual-GPU GeForce 9800 GX2 (March 18, 2008, 256 cores total), which targeted enthusiast gamers with SLI multi-GPU configurations. Mid-range options like the GeForce 9500 GT (July 29, 2008, 32 stream processors on 65 nm G96) and low-end cards such as the GeForce 9400 GT ( 2008, 16 stream processors) rounded out the discrete GPU offerings, emphasizing value and energy efficiency. A notable in the 9 series was the introduction of Hybrid SLI technology, which enabled scalable performance by combining a discrete GPU with an integrated GPU on nForce motherboards, dynamically switching for power savings during light workloads or boosting for demanding tasks. Integrated variants like the GeForce 9100 and GeForce 9200 (May 2008) further extended this capability to motherboard chipsets, supporting up to 1 GB of shared system memory and features like output for home theater PCs. Overall, the series provided a balanced price-performance ratio, with models like the 9800 GT becoming popular for their potential and quiet operation under 150W TDP limits.

Overview

Introduction

The GeForce 9 series represents the ninth generation of NVIDIA's graphics processing units (GPUs), succeeding the and preceding the . Launched starting on February 21, 2008, with the introduction of the GeForce 9600 GT as the inaugural product, the lineup marked NVIDIA's continued evolution in consumer graphics hardware during the 10 era. This series encompassed a diverse range of products built on an updated iteration of the Tesla microarchitecture, featuring enhancements such as a 65 nm process shrink for improved efficiency, PCI Express 2.0 support, and better compression techniques for color and z-buffer data. The scope included rebranded variants of select GeForce 8 models with minor tweaks for refreshed performance, alongside new low- and mid-range discrete GPUs, as well as integrated graphics solutions aimed at mainstream computing platforms. NVIDIA's branding strategy for the "9" designation emphasized a seamless generational progression, positioning the series as a bridge between established Tesla-based designs and upcoming architectural shifts. In the broader market context, the GeForce 9 series was engineered to deliver robust 10 compatibility, enabling advanced effects and high-definition gaming experiences in titles like Call of Duty 4 and Unreal Tournament 3. It targeted segments from budget-oriented users seeking affordable upgrades to high-end enthusiasts demanding multi-GPU configurations, directly competing with AMD's HD 2000 and 3000 series in a fiercely contested landscape for 10 dominance. By offering competitive pricing—such as the 9600 GT below $199— aimed to capture value-driven gamers while advancing power efficiency and video processing capabilities through technologies like HD.

Development History

NVIDIA's development of the GeForce 9 series marked a continuation and refinement of the Tesla microarchitecture initially introduced with the GeForce 8 series' G80 and G92 GPUs in 2006-2007, shifting to 65 nm production processes by late 2007 to enable smaller dies, reduced power consumption, and broader product applicability across discrete and integrated solutions. This transition emphasized scalability for mainstream and low-end markets, incorporating enhancements like PCI Express 2.0 support and improved compression techniques while maintaining unified shading capabilities. However, the series faced substantial reliability challenges, with high failure rates in 65 nm GPUs attributed to manufacturing defects, resulting in numerous customer returns and legal actions against NVIDIA. The GeForce 9 branding was first previewed at CES 2008, where NVIDIA showcased early mobile implementations such as the 9500M GS, signaling a focus on efficient, 10-compatible graphics for notebooks and desktops. The series officially launched on February 21, 2008, with the mid-range 9600 GT (G94 GPU), which NVIDIA positioned as delivering up to 116% greater performance than the 8600 GT at a sub-$200 price point, highlighting an improved performance-per-watt ratio to address growing demands for energy-efficient gaming hardware. This debut was part of NVIDIA's strategy to counter AMD's HD 3000 series, which had gained traction in 2007-2008 with competitive pricing and performance in 10 titles. Rebranding played a key role in the series' expansion, with low-end models like the 9500 GT serving as a 65 nm revision of the 8600 GT (G84 GPU), offering similar core specifications but lower power draw and costs to prolong the lifecycle of established designs without full redesigns. This approach, combined with new integrated graphics in chipsets such as the MCP7A (featuring 9400 capabilities), prioritized power efficiency for OEM systems, enabling hybrid SLI configurations and reducing thermal demands in compact builds amid AMD's push into integrated GPU markets. Development challenges included balancing performance gains with shrinking process nodes to compete effectively, as AMD's mid-2008 Radeon HD 4800 launches pressured NVIDIA on price-to-performance ratios. The rollout proceeded in phases through 2008, starting with the mainstream GeForce 9600 GT and 9650 variants in February-March, followed by high-end dual-GPU 9800 GX2 in March and single-GPU 9800 GTX in April, to methodically cover market segments. This staggered release allowed iterative refinements, culminating in the power-optimized 9800 GTX+ on January 16, 2009, based on a 55 nm G92 revision, which extended the series' viability into the next generation.

Architecture

Tesla Microarchitecture

The Tesla microarchitecture, introduced by with the and refined for the GeForce 9 series, established a unified framework for rendering, vertex processing, and general-purpose via . Building upon the initial G80 implementation, it integrated stream processors capable of executing a wide range of programs, enabling parallel workloads for both 10-compliant and high-performance compute tasks. This architecture marked a shift from fixed-function pipelines to a scalable, programmable model that supported emerging standards like Shader Model 4.0. At its core, Tesla employs a unified shader model organized around streaming multiprocessors (SMs), each containing multiple stream processors for scalar floating-point operations, along with special function units for tasks like transcendental functions. In GeForce 9 series GPUs, the number of stream processors scales from 16 in lower-end configurations to 128 in high-end ones, allowing flexible allocation across graphics pipelines and compute kernels. The architecture supports 10 for advanced effects such as geometry shaders and 3.3 for enhanced vertex array handling and uniform buffer objects, providing robust compatibility for contemporary applications. These elements are built on a Scalable Processor Array (SPA) framework, which distributes processing across texture processor clusters (TPCs) for balanced performance in rasterization and compute scenarios. Fabricated using 65 nm and 55 nm process technologies, Tesla chips in the GeForce 9 series achieved denser integration than the 90 nm G80, with examples like the G96 core produced at 65 nm to reduce power draw and heat output while maintaining core functionality. Relative to the , refinements in the GeForce 9 implementation enhanced efficiency through optimized geometry processing in the unified shaders and improved via dynamic clocking, which adjusts frequencies based on demands to optimize energy use. Integrated variants of the Tesla microarchitecture appear in motherboard solutions like the GeForce 9100, where the graphics core is embedded within NVIDIA's MCP78 and MCP79 chipsets to deliver unified processing without discrete cards, supporting hybrid graphics configurations for consumer systems.

Core Technologies

The GeForce 9 series introduced enhanced support for NVIDIA PureVideo HD technology, a hardware-accelerated video decoding solution integrated into the GPU for improved high-definition playback. This generation featured the VP3 video processing engine in select models, such as the GeForce 9600 GT and 9800 GT, enabling full hardware decoding of H.264 and partial support for VC-1 formats, which offloaded processing from the CPU to reduce system load during media playback. The VP3 engine represented an advancement over prior versions, offering better efficiency for 1080p video streams and post-processing effects like de-interlacing and noise reduction. Multi-GPU configurations were a key capability in the GeForce 9 series, with SLI technology enabling performance scaling by linking two or more compatible cards for up to 2x frame rates in supported games and applications. Introduced prominently in the 9600 and 9800 series, SLI required identical GPUs connected via a high-bandwidth bridge, supporting 9 and rendering. Complementing this, Hybrid SLI (also known as HybridPower) allowed pairing a discrete 9 GPU with an integrated graphics processor on the , dynamically switching between them to optimize power consumption during light workloads while maintaining high performance for demanding tasks. This feature was available on models like the GeForce 9800 GT and 9800 GX2, marking an early step toward power-efficient multi-GPU setups. The series marked a significant expansion in general-purpose computing on GPUs through compatibility with 1.x, NVIDIA's platform that allowed developers to leverage the GPU's processing cores for non-graphics tasks such as scientific simulations and data processing. GeForce 9 GPUs, built on the G92 and G94 chips, supported compute capability 1.1, enabling unified architecture for floating-point operations and access optimizations that were more accessible than in the preceding . This compatibility facilitated the first widespread adoption of in consumer graphics cards, with applications in fields like physics acceleration via NVIDIA . Additional rendering technologies in the GeForce 9 series included advanced modes, such as 16x Coverage Sampled Anti-Aliasing (CSAA), which improved edge smoothing in 3D graphics by using more coverage samples than traditional while maintaining comparable performance overhead. CSAA delivered image quality approaching 16x multisample AA but with efficiency gains, particularly beneficial for high-resolution gaming on models like the 9800 GT. Precursors to unified video decoding architectures were evident in the HD implementation, which laid groundwork for later dedicated decoder blocks by integrating , H.264, and support into the GPU pipeline. Driver support for the GeForce 9 series relied on the ForceWare 17x.xx series, such as versions 175.16 and 178.13, which provided WHQL certification for and early compatibility. These drivers optimized 3D rendering, SLI profiles, and features, with specific enhancements for 10 applications and reduced latency in multi-monitor setups. They also introduced better stability for workloads and integration, ensuring broad software ecosystem support during the series' lifecycle.

Desktop Processors

GeForce 9100 Series

The GeForce 9100 series consists of integrated graphics processors (IGPs) within NVIDIA's GeForce 9 lineup, utilizing the Tesla microarchitecture for entry-level desktop systems. The primary model, GeForce 9100 G, is integrated into the MCP78 chipset, marketed as the nForce 720a, and was released in the second quarter of 2008. This IGP targeted budget-oriented motherboards, particularly for home theater PCs (HTPCs), offering basic DirectX 10 compatibility and hardware-accelerated video decoding via PureVideo HD technology. Featuring 16 unified units, 8 units, and 4 render output units, the 9100 G operates at a 500 MHz core clock with a 1200 MHz clock. It shares system RAM as its source, eliminating the need for discrete VRAM and enabling compact, low-power designs suitable for everyday . Targeted applications include standard multimedia playback and light gaming at resolutions, though performance is constrained by bandwidth. Variants of the GeForce 9100 were commonly paired with the nForce 720a chipset in OEM-integrated motherboards, providing affordable all-in-one solutions for non-gaming desktops. Positioned as a successor to the earlier GeForce 7050 and 8200 IGPs, it emphasized improved efficiency and video processing for consumer electronics integration.

GeForce 9300 Series

The GeForce 9300 GS was a budget discrete graphics processing unit based on the G98 chip fabricated on a 65 nm process using NVIDIA's Tesla microarchitecture. Launched in June 2008 and priced under $100, it targeted entry-level users seeking an affordable upgrade from the GeForce 7 series without support for SLI multi-GPU configurations. The card featured 8 unified shaders with a core clock of MHz and shader clock of MHz, enabling basic computational capabilities for its class. It came with 512 MB of DDR2 on a 64-bit bus and had a power draw of under 50 W, allowing it to operate without an external power connector and fit into compact or low-power office PC builds. Primarily designed for non-gaming applications, the 9300 GS excelled in everyday tasks such as basic video playback accelerated by technology and light multimedia workloads. For gaming, it handled older titles at low settings and resolutions, providing a modest entry point for casual users but lacking the performance for contemporary games at higher details.

GeForce 9400 Series

The GeForce 9400 series introduced the low-end discrete GeForce 9400 GT graphics processing unit, aimed at enhancing efficiency for entry-level computing and multimedia applications. Launched on August 27, 2008, the 9400 GT utilizes the G96 chip fabricated on a 65 nm process node. It incorporates 16 unified shaders based on the Tesla microarchitecture, with a base core clock of 550 MHz and shader clock of 1400 MHz, though some partner models reached higher clocks up to 1100 MHz for the core. This configuration provided modest improvements in processing capabilities for its class, emphasizing power efficiency over high-end performance. Memory options for the 9400 GT included 256 MB or 512 MB of DDR2 on a 128-bit bus, delivering bandwidth up to 12.8 GB/s at an effective memory clock of 800 MHz. The card adopted a single-slot low-profile design, consuming just 50 W of power via the slot without needing external connectors, making it ideal for compact builds. Targeted at small form factor PCs and media centers, it launched at an MSRP of $59, appealing to budget-conscious users requiring basic gaming and video playback capabilities. Key features of the 9400 GT included full DirectX 10 support, offering better performance in DirectX 10 workloads compared to the preceding 9300 series due to its higher clocks and dedicated discrete implementation. It also enabled NVIDIA Hybrid SLI technology, permitting dynamic pairing with compatible integrated GPUs on supported motherboards to boost graphics output for lighter tasks. While an integrated 9400 variant appeared in 's MCP79 chipset for motherboard implementations, the series primarily centered on the discrete 9400 GT for standalone upgrades.

GeForce 9500 Series

The 9500 series introduced NVIDIA's entry into the budget discrete graphics market with 10 support, targeting users upgrading from integrated graphics or older cards for improved performance in Windows Vista-era gaming and applications. Positioned as a bridge between low-end integrated solutions and mid-range discrete GPUs, the series offered better efficiency and feature set over the 8600 lineup, particularly in performance and power consumption. The GeForce 9500 GT, released on July 29, 2008, featured NVIDIA's new single-chip G96 GPU fabricated on a 65 nm process, marking a fresh design in the Tesla microarchitecture family with 32 unified shaders (CUDA cores). It operated at a reference core clock of 550 MHz (with shader clock at 1400 MHz), though partner cards often shipped at 600-650 MHz for enhanced performance, and could reach up to 900 MHz when overclocked. Equipped with 512 MB of GDDR3 memory on a 128-bit interface (effective bandwidth of 25.6 GB/s), the card supported SLI connectivity for multi-GPU configurations, enabling scalability in compatible systems. Launched at an MSRP of approximately $99, it served as an affordable entry for 1080p gaming at medium settings in titles of the era, such as Crysis or Call of Duty 4, while maintaining a low TDP of 65 W without requiring an auxiliary power connector. The 9500 GS, also launched on July 29, 2008, was a value-oriented variant using the same G96 GPU (specifically the G96C revision) but with detuned clocks for reduced cost and power draw, typically at 550 MHz core and 504 MHz memory. Available exclusively through OEM channels like or HP systems, it paired 512 MB of DDR2 memory on a 128-bit bus, prioritizing basic 3D acceleration and video decoding over high-frame-rate gaming. This model filled the budget segment by providing a step up from integrated graphics for office PCs and light media tasks, without SLI support, and was not sold as a retail product.

GeForce 9600 Series

The GeForce 9600 series represented NVIDIA's mid-range desktop graphics offerings in the GeForce 9 lineup, emphasizing improved efficiency and performance for 1080p gaming compared to prior generations. Launched in early 2008, these GPUs utilized the Tesla microarchitecture and targeted mainstream users seeking balanced capabilities in DirectX 10-era titles without the premium cost of higher-end models. Key models included the 9600 GT as the primary performer, alongside value-oriented variants like the 9600 GS and 9600 GSO, with a rare high-end outlier in the 9600 GTX. The 9600 GT, introduced on February 21, 2008, served as the cornerstone of the series, built on the 65 nm G94 graphics processor with 64 unified shaders operating at 1620 MHz alongside a 650 MHz core clock. It featured 512 MB of GDDR3 memory clocked at 900 MHz across a 256-bit interface, delivering a of 57.6 GB/s. This configuration enabled solid 1080p gaming performance, positioning the card as a direct upgrade over the 8600 GTS while competing with AMD's HD 3850. Priced at launch between $169 and $189, it filled a mainstream market segment valued at $150–$250, succeeding the lower-tier 8800 GS in NVIDIA's portfolio. Variants expanded the lineup for budget-conscious consumers. The GeForce 9600 GS was a low-end variant using the 65 nm G94 chip with 48 unified at a 1250 MHz shader clock, 768 MB DDR2 on a 192-bit bus, and a 500 MHz core clock—offering entry-level performance without significant enhancements. In contrast, the GeForce 9600 GSO, launched in April 2008, acted as a transitional stopgap amid supply constraints for the popular GeForce 8800 GT, the existing 65 nm G92-based 8800 GS with 96 unified clocked at 1375 MHz, a 550 MHz core, and 384 MB of GDDR3 on a 192-bit bus at 800 MHz effective (some 768 MB editions used DDR2). A cut-down 9600 GSO 512 MB version used the G94 chip with 48 at 1625 MHz shader clock, a 650 MHz core, and 512 MB GDDR3 on a 256-bit bus for cost savings. The elusive GeForce 9600 GTX, a beta-level release primarily from partner in limited quantities, used a G94 chip with 64 at 1450 MHz shader clock, a 580 MHz core, and 512 MB GDDR3 on a 256-bit bus, but saw minimal distribution and was not officially endorsed by . All 9600 series models supported advanced features such as 16x for enhanced image quality, Hybrid SLI for combining discrete GPUs with integrated graphics in supported systems, and a 6-pin connector to meet demands up to 95 W. They also included initial compute capabilities for parallel processing tasks, though detailed implementation resided in broader core technologies. These attributes solidified the series' role in accessible high-definition gaming during its .
ModelChipShadersCore Clock (MHz)Shader Clock (MHz)MemoryBus WidthLaunch Date
9600 GTG94646501620512 MB GDDR3256-bitFeb 2008
9600 GSG94485001250768 MB DDR2192-bitJul 2008
9600 GSOG92965501375384 MB GDDR3192-bitApr 2008
9600 GSO 512G94486501625512 MB GDDR3256-bitMay 2008
9600 GTXG94645801450512 MB GDDR3256-bitLimited 2008

GeForce 9800 Series

The GeForce 9800 series represented NVIDIA's high-end desktop graphics offerings within the GeForce 9 lineup, built on the Tesla microarchitecture and utilizing the G92 graphics processing unit (GPU). These cards were designed for enthusiast gamers targeting resolutions from 1080p to 1440p, providing full DirectX 10 compatibility for advanced shader effects and geometry processing in contemporary titles. Launched in early 2008, the series emphasized improved power efficiency over prior generations while supporting scalable multi-GPU configurations via SLI technology. The flagship GeForce 9800 GTX, released on March 28, 2008, featured the full G92 GPU with 128 unified shading units, 64 texture mapping units (TMUs), and 16 render output units (ROPs), fabricated on a 65 nm process. It operated at a reference core clock of 675 MHz and shader clock of 1688 MHz, paired with 512 MB of GDDR3 memory clocked at 1100 MHz (2.2 GHz effective) across a 256-bit interface, delivering 70.4 GB/s of bandwidth. With a thermal design power (TDP) of 140 W and requiring two 6-pin power connectors, the 9800 GTX supported full SLI for up to three-way configurations, enabling high-frame-rate performance in demanding games at elevated settings. NVIDIA positioned it at a launch price of $299, making it a premium option for extreme HD gaming. Key variants expanded the series' appeal. The 9800 GT, a cost-reduced model launched on July 21, 2008, employed a trimmed G92 GPU with 112 units while retaining the same 64 TMUs and 16 ROPs, clocked at 600 MHz core and 1500 MHz speeds, with 512 MB GDDR3 at 900 MHz (1.8 GHz effective) on a 256-bit bus. It featured a lower 105 W TDP and a single 6-pin connector, priced at $160 for mid-range users seeking solid performance without the full overhead of the GTX. The 9800 GX2, introduced on March 18, 2008, stood out as a dual-GPU solution with two G92 chips—effectively 256 units, 128 TMUs, and 32 ROPs—each running at 600 MHz core and 1500 MHz clocks, backed by 512 MB GDDR3 per GPU (1 GB total) at 1000 MHz (2 GHz effective) per 256-bit interface. Drawing approximately 197 W with dual 6-pin connectors and SLI bridging between boards, it launched at around $600 and marked the last high-end dual-PCB design in NVIDIA's consumer lineup before the integrated dual-GPU approach in the subsequent 200 series. An overclocked variant, the 9800 GTX+, arrived later in 2008 on a 55 nm process revision, boosting core clocks to 738 MHz and shaders to 1836 MHz while maintaining 512 MB GDDR3, aimed at sustaining competitiveness against emerging rivals. These models collectively advanced the GeForce 9 high-end segment by balancing raw compute power with practical efficiency, supporting features like NVIDIA PureVideo HD for video decoding and multi-display outputs up to 2560x1600 resolution.

Desktop Technical Specifications

Chip Details

The GeForce 9 series desktop graphics processing units (GPUs) are built around several GPU dies from NVIDIA's Tesla microarchitecture, primarily fabricated on 65 nm and 80 nm processes. The lowest-end models, such as the integrated GeForce 9100, utilize the C78 die integrated into the MCP78 northbridge chipset, which combines graphics capabilities with system memory controller functions. This chip, produced on an 80 nm process by TSMC, features a die size of 127 mm² and 210 million transistors, with 16 unified shaders, 8 texture mapping units (TMUs), and 4 render output units (ROPs). Core clock speeds for the C78 typically range from 400 to 475 MHz, while memory is shared from system DDR2/DDR3 at up to 800 MHz effective. Higher-end discrete GPUs in the series shift to 65 nm fabrication for improved efficiency. The entry-level discrete models like the 9300 GS employ the G98 die, on 55 nm by UMC, with a compact 80 mm² die area and 210 million transistors. It includes 8 unified shaders, 4 TMUs, and 4 ROPs, supporting core clocks of 500 to 650 MHz and DDR2 memory at 400 to 500 MHz (up to 1000 MHz effective). The mid-range 9400 GT and 9500 GT/GS variants use the G96 die, fabricated on 65 nm by UMC, measuring 144 mm² with 314 million transistors, 32 unified shaders, 16 TMUs, and 8 ROPs. Clock ranges for G96-based cards include core speeds of 550 to 650 MHz and GDDR3/DDR2 memory from 700 to 1000 MHz effective. The performance-oriented GeForce 9600 series relies on the G94 die, produced on 65 nm by , with a 240 mm² die size and 505 million . Full configurations feature 64 unified shaders, 32 TMUs, and 16 ROPs, while lower SKUs like the 9600 GS disable units to 48 shaders, 24 TMUs, and 12 ROPs. Core clocks span 500 to 700 MHz, paired with GDDR3 memory at 700 to 1100 MHz clock (1400 to 2200 MHz effective). At the top of the lineup, the 9800 series uses the G92 die on 65 nm by , boasting a larger 324 mm² die area and 754 million , with 128 unified shaders, 64 TMUs, and 16 ROPs. The dual-GPU 9800 GX2 variant employs two G92 dies. Clock configurations for G92 range from 600 to 700 MHz core and 800 to 1100 MHz memory (up to 2200 MHz effective for GDDR3). Some later revisions, like G92b, shrank to 55 nm with a 260 mm² die while retaining counts. Rebranded 80 nm chips from the prior , such as G86 derivatives, appeared in select low-end 9300/9500 GS models, featuring 32 unified shaders, 16 TMUs, 8 ROPs, a 127 mm² die, and approximately 210 million , with core clocks of 500 to 550 MHz and DDR2 memory up to 800 MHz.
ChipUsed InProcessDie Size (mm²)Transistors (millions)Shaders / TMUs / ROPsCore Clock Range (MHz)Memory Clock Range (MHz effective)
C78 9100 (integrated)80 nm12721016 / 8 / 4400–475Shared (up to 800)
G98 9300 GS55 nm802108 / 4 / 4500–650800–1000 (DDR2)
G96 9400/9500 GT/GS65 nm14431432 / 16 / 8550–6501400–2000 (GDDR3/DDR2)
G94 9600 GT/GS65 nm24050564 / 32 / 16 (full); 48 / 24 / 12 (GS)500–7001400–2200 (GDDR3)
G92 9800 GT/GTX/GX265 nm324754128 / 64 / 16600–7001600–2200 (GDDR3)

Performance Characteristics

The GeForce 9 series desktop GPUs demonstrated modest performance improvements over their GeForce 8 predecessors, particularly in 10 workloads. The 9600 GT delivered approximately 1.2 times the performance of the GeForce 8800 GS in synthetic benchmarks like 3DMark Vantage, thanks to its unified shader architecture and higher clock speeds on the G94 core. Similarly, the GeForce 9800 GTX achieved about 1.1 times the frame rates of the GeForce 8800 GTX in 10 titles such as and Call of Duty 4, benefiting from enhanced and capabilities. Power efficiency was a notable strength of the series, with the 9800 GT at a 125 W TDP compared to the 125 W TDP of the 8800 GT, enabling better thermal management without sacrificing core performance. This efficiency stemmed from process shrinks and optimized on the G92 core, allowing for sustained boosts in setups. In gaming benchmarks at resolution with medium settings, the GeForce 9800 GTX averaged 25-40 FPS in , showcasing playable performance in demanding 10 environments while highlighting limitations in scenarios. Scalability via SLI provided 20-50% uplifts in frame rates for titles like , though low-end models like the 9500 GT exhibited diminishing returns due to bandwidth constraints in crossfire-like configurations. When compared to contemporary offerings, the 9 series held an edge in driver stability for 10 features, outperforming the HD 3850 by 10-15% in shader-heavy games, while the HD 3870 occasionally led in raw rasterization tasks by up to 20% in 9 content.

Mobile Processors

GeForce 9100M and 9200M Series

The 9100M G and 9200M GS represent the entry-level mobile graphics solutions in NVIDIA's 9 series, targeted at ultraportable laptops and netbooks requiring minimal power consumption for everyday . These GPUs emphasized efficiency over performance, supporting basic multimedia and productivity tasks while integrating seamlessly with low-power or processors of the era. Released as part of the broader 9 lineup in early 2008, they were built on NVIDIA's to address the growing demand for thin-and-light devices. The 9100M G is an integrated graphics processor embedded within NVIDIA's MCP79 mobile , designed for basic laptops with shared system memory. It features 16 shading units, 4 units, and 4 render output units, operating at a core clock of 450 MHz and a clock of 1100 MHz, with support for DirectX 10 and Shader Model 4.0. This configuration enables hardware-accelerated decoding for HD video playback via technology, making it suitable for web browsing, office applications, and light , though it lacks viability for gaming due to its limited processing power. Power draw is rated at a maximum of 12 W, allowing integration into ultralow-voltage systems without dedicated cooling. In contrast, the 9200M GS serves as a discrete low-end option, utilizing the G98 core with 8 shading units, 4 units, and 4 render output units. Clock speeds vary by implementation, typically ranging from 400 MHz to 600 MHz for the core (with a reference of 550 MHz) and up to 700 MHz for memory, paired with 256 MB of DDR2 or GDDR3 on a 64-bit bus. Like the 9100M G, it prioritizes efficiency for non-gaming workloads such as HD video playback and web-based tasks, benefiting from the same 10 compatibility. Its is around 13-15 W, enabling deployment in compact while integrated into compatible C7x0M-series chipsets for hybrid switching. These GPUs were particularly notable in the context of emerging platforms, where the 9100M G contributed to NVIDIA's efforts to enhance Atom-based systems with better video handling, though full discrete capabilities were limited to higher-tier offerings. Overall, their design focused on extending battery life in entry-level mobiles, with no emphasis on 3D acceleration beyond basic 2D operations.

GeForce 9300M and 9400M Series

The GeForce 9300M and 9400M series represented NVIDIA's entry into low-power mobile graphics solutions within the GeForce 9 lineup, targeting ultrathin laptops and integrated platforms with an emphasis on energy efficiency and basic multimedia capabilities. These GPUs were designed for office-oriented devices, enabling light 3D tasks and video playback while minimizing thermal demands in compact form factors. Released in 2008, they bridged the gap between integrated graphics and higher-end discrete options, supporting 10 and Hybrid SLI for modest performance boosts when paired with compatible integrated cores. The 9300M G and GS variants were discrete mobile GPUs built on the G86 and G98 cores, respectively, using 80 nm and 65 nm processes. The 9300M G featured 16 unified s, a core clock of 400 MHz, clock of 800 MHz, and up to 256 MB of GDDR2 or GDDR3 memory on a 64-bit bus, with a memory clock of 600 MHz. In contrast, the 9300M GS offered slightly higher clocks at 550-580 MHz core, 1400 MHz , and 256-512 MB GDDR2 memory at 700-800 MHz effective, maintaining the same 16 s for improved rasterization in entry-level scenarios. Both launched in early to mid-2008, with the G in February and GS in June, positioning them as upgrades for budget notebooks focused on everyday rather than intensive gaming. The 9400M G stood out as an integrated solution within NVIDIA's MCP79 , embedding 16 unified s directly into the for seamless operation in slim designs. It operated at a 450 MHz core clock with a 1100 MHz clock and shared system memory up to 256 MB, prioritizing low latency over dedicated bandwidth. This configuration debuted in October , particularly in Apple and models, where it enhanced for creative applications while integrating with the system's northbridge for reduced complexity. Power profiles for the series ranged from 10-25 TDP, with the 9300M models at 13 and the 9400M G at 12 W, enabling extended battery life in portable devices. They supported NVIDIA's Switchable technology, an early precursor to Optimus, which allowed dynamic toggling between the discrete GPU and integrated CPU to optimize power usage based on . Typical applications included office laptops for web browsing, document editing, and 720p video decoding via HD, with casual 720p gaming viable at low settings in titles like older strategy games.

GeForce 9500M and 9600M Series

The 9500M series, comprising the G and GS variants, represented NVIDIA's mid-range mobile graphics solutions introduced as part of the broader 9 lineup, targeting gaming laptops with improved efficiency over prior generations. The 9500M G, based on the G96M GPU fabricated on a , featured 16 unified shaders operating at up to 1250 MHz, a core clock ranging from 500 MHz, and memory speeds up to 800 MHz across a 128-bit interface supporting 256 MB to 1 GB of GDDR3. In contrast, the 9500M GS utilized the earlier G84 GPU on an 80 nm process with 32 unified shaders at 950 MHz, a 475 MHz core clock, and 700 MHz GDDR3 memory, also on a 128-bit bus with up to 512 MB capacity. These chips launched in Q3 2008, with (TDP) ratings of 20-25 W, and were compatible with MXM () form factors for easier integration into notebook designs. The 9600M series extended this mid-range segment with the GS and GT models, emphasizing enhanced gaming capabilities through higher shader counts and optional multi-GPU configurations. Built on the G94M GPU at 65 nm, the 9600M GT delivered 32 unified shaders at 1250 MHz, a 500-600 MHz core clock range, and 800 MHz GDDR3 memory on a 128-bit bus, supporting up to 1 GB of VRAM; it also enabled SLI for select high-end laptops, allowing dual-GPU setups to boost performance. The 9600M GS variant, using the G96 GPU, offered 32 unified shaders at 1075 MHz, a lower 430 MHz core clock, and similar 800 MHz memory configuration with up to 1 GB GDDR3. Both models maintained TDPs of 23-50 W depending on configuration and SLI usage, with MXM compatibility, and were released in June 2008 as part of NVIDIA's mobile 9 announcement. In terms of performance, these GPUs excelled in light to moderate gaming scenarios, such as running at medium settings with playable frame rates around 40-60 fps, thanks to their 10 support and improved shader architectures. The 9600M GT, in particular, achieved approximately 25 fps in at 1024x768 on medium settings and 50 fps in Call of Duty 4 at high settings, outperforming predecessors like the 8600M GT by up to 40% in optimized titles while balancing power efficiency for battery life in mobile environments.
VariantGPU CodenameUnified ShadersCore Clock (MHz)Shader Clock (MHz)Memory (GDDR3)TDP (W)Launch
9500M GG96M165001250256 MB-1 GB, 128-bit20-25Q3 2008
9500M GSG8432475950512 MB, 128-bit20Q3 2008
9600M GSG96324301075Up to 1 GB, 128-bit20-25June 2008
9600M GTG94M32500-6001250512 MB-1 GB, 128-bit23-50June 2008

GeForce 9700M and 9800M Series

The GeForce 9700M GT and GTS were high-performance mobile GPUs introduced as part of NVIDIA's GeForce 9 series, targeting premium gaming and workstation laptops with enhanced DirectX 10 support. The 9700M GT, based on the G96M chip fabricated on a 65 nm process, featured 32 unified shaders operating at a core clock of 625 MHz and shader clock of 1550 MHz, paired with 512 MB of GDDR3 memory on a 128-bit bus running at 800 MHz (effective 1600 MHz). Its thermal design power (TDP) was rated at 45 W, making it suitable for balanced power efficiency in larger chassis. The 9700M GTS, utilizing the G94M chip on the same process, offered improved capabilities with 48 unified shaders at a 530 MHz core clock and 1325 MHz shader clock, supported by 512 MB GDDR3 on a wider 256-bit bus at the same memory speed, and a higher TDP of 60 W for greater computational headroom. Both variants launched in July 2008, positioning them as upgrades over prior 8-series mobile GPUs for smoother performance in demanding applications. The 9800M series extended this high-end mobile lineup with models like the GS, GTS, GT, and GTX, emphasizing scalability for multi-GPU configurations and higher resolutions. The 9800M GS, derived from the G96M chip, included 32 unified at 530 MHz core and 1325 MHz clocks, with 512 MB GDDR3 on a 128-bit bus and a 60 TDP, serving as an entry point to the series launched in November 2008. The 9800M GTS, on the G94M chip, boosted to 64 unified at 600 MHz core and 1500 MHz speeds, supporting up to 1 GB GDDR3 on a 256-bit bus with a 65 TDP. Higher tiers included the 9800M GT (G92M chip, 96 , up to 700 MHz core, 512 MB to 1 GB GDDR3, 65 TDP) and 9800M GTX (G92M, 128 , up to 700 MHz core, 1 GB GDDR3 standard with some 2 GB variants, 75 TDP, peaking at 95 in SLI setups), all introduced starting July 2008. These GPUs represented the pinnacle of the 9M series, delivering substantial rasterization and for the era's mobile platforms. Key features of the 9700M and 9800M series included full support for SLI in compatible high-end configurations, such as dual 9800M GTX or GTS setups, which could yield up to 40% performance gains in optimized games depending on driver and title support. They also provided advanced capabilities, including up to 16x coverage sampling (CSAA) for reduced jagged edges without excessive performance penalties, enabling high-quality rendering at and resolutions in contemporary titles like or : . These GPUs were integrated into premium laptops, such as the M1730 equipped with 9800M GTX SLI for workstation-grade tasks and the Toshiba Qosmio X300 series featuring 9700M GTS, often commanding prices over $2,000 USD at launch due to their $300+ GPU contribution and overall build quality. Positioned for professional gamers and creators, they bridged desktop-like experiences in mobile form factors during 2008-2009, marking the peak of 65 nm mobile graphics before the shift to 40 nm architectures.
VariantChipShadersCore Clock (MHz)MemoryBus WidthTDP (W)Launch Date
9700M GTG96M32625512 MB GDDR3128-bit45Jul 2008
9700M GTSG94M48530512 MB GDDR3256-bit60Jul 2008
9800M GSG96M32530512 MB GDDR3128-bit60Nov 2008
9800M GTSG94M64600512 MB-1 GB GDDR3256-bit65Jul 2008
9800M GTG92M96625-700512 MB-1 GB GDDR3256-bit65Jul 2008
9800M GTXG92M128550-7001-2 GB GDDR3256-bit75-95Jul 2008

Mobile Technical Specifications

Chip Variants

The mobile GeForce 9 series utilizes several chip derivatives adapted from the desktop lineup, optimized for laptop constraints such as power limits and thermal management while retaining the core Tesla microarchitecture. These variants feature scaled-down configurations in shaders, clocks, and memory interfaces compared to desktop counterparts. The entry-level G98M core powers the GeForce 9200M, equipped with 8 unified shaders and fabricated on a 65 nm process at TSMC. The G96M core drives the GeForce 9500M and lower-end 9700M models (such as the GT), supporting 16 to 32 unified shaders in various configurations, also on a . Higher-performance options employ the G94M core for the 9600M series and upper 9700M variants (like the GTS), with 32 to 64 unified shaders; select implementations, including some 9800M GS models, use a 55 nm process revision for improved efficiency. The flagship GeForce 9800M lineup (GTX and GT) relies on the G92M core, offering 64 to 128 unified s, a 65 nm fabrication, and up to 754 million transistors to deliver near-desktop capabilities in a mobile package. Key adaptations from desktop chips include lower core and clocks (typically 400–625 MHz versus 550–700 MHz on desktop equivalents), built-in controllers compatible with DDR2 or DDR3 for flexible , and stricter throttling thresholds aligned with TDP limits of 12–75 W. Production emphasizes 65 nm nodes across most variants for cost-effective yields, with die sizes tailored smaller for MXM module compatibility. The integrated 9100M and 9400M are embedded within the MCP7xM family (such as MCP77MH), providing shared-memory solutions with 8 to 16 shaders directly on the GPU.

Power and Thermal Profile

The mobile GeForce 9 series GPUs featured (TDP) ratings that scaled with performance tiers to balance capability and portability. Integrated and low-end models, such as the 9100M, 9200M, 9300M GS, and 9400M, maintained low TDPs of 12-13 W, enabling their use in slim notebooks with minimal power demands. Mid-range variants like the 9500M GE and 9600M GT operated at 23-25 W, with configurable options up to 35 W for enhanced performance in mainstream laptops. High-end configurations, including the 9700M GT/GTS at 45-60 W and the 9800M GTX at 75 W (reaching 95 W in select implementations), required more substantial power supplies and were typically found in gaming-oriented systems. Efficiency improvements stemmed from the 65 nm fabrication , which permitted higher clock speeds without proportional increases in power draw compared to the 80 nm . For instance, the 9600M GT delivered roughly 1.4 unified shaders per watt based on its 32 shaders and 23 W TDP, supporting better performance in power-constrained environments. The 9800M GTX achieved greater over the GeForce 8700M equivalent, owing to refined and optimizations that enhanced output per watt during graphics-intensive tasks. NVIDIA's HybridPower further boosted runtime by dynamically switching to integrated graphics for light workloads, reducing overall system power use. Battery life varied significantly with usage, as gaming on discrete GeForce 9 GPUs accelerated drain compared to integrated Intel solutions—typically by 20-50% faster under sustained loads due to elevated GPU activity and heat generation. Low-end models minimized this impact through via chassis dissipation, preserving longer unplugged sessions for basic tasks. Mid- and high-end GPUs relied on with dedicated fans and heat pipes, while top-tier options like the 9800M GTX often incorporated vapor chambers and MXM-standard heat spreaders to handle thermal loads up to 95 W, though this necessitated larger chassis with robust airflow to prevent throttling.

Legacy

Driver Support Timeline

The GeForce 9 series debuted alongside NVIDIA ForceWare beta driver version 174.12 on February 21, 2008, providing initial support for the new GPUs including optimizations for 9 and applications, as well as NVIDIA SLI technology. WHQL support began with version 175.16 on May 13, 2008. This release marked the beginning of the R170 driver branch (later R175), which evolved through subsequent branches like R180, R190, and up to the R340 series, delivering performance enhancements, bug fixes, and compatibility updates for emerging games and operating systems over the years. Driver support progressed to the R340 , the final major lineage for the GeForce 9 series, with releases focusing on stability and compatibility for legacy hardware. The last Game Ready , version 342.01, arrived on December 14, 2016, offering optimizations for titles such as those utilizing 10 features and including updates for the nvlddmkm.sys component. Full driver support concluded in April 2016, aligning with 's policy of approximately eight years of active maintenance for consumer GPU generations, with no further updates including patches thereafter. Operating system compatibility spanned (last supported via version 340.52 in 2016) through (up to 342.01), enabling broad adoption during the transition from older to modern Windows versions, but lacking support for Windows 11. Mainline support ended in March 2014, when removed pre-Fermi GPUs—including the Tesla-based GeForce 9 series—from the mainline Unix driver, shifting them to a legacy 340.xx branch with limited updates until end of life in 2022. For macOS, drivers remained available until version 10.13 High Sierra, after which ceased web driver distribution for non-Metal-compatible GPUs like the GeForce 9 in 2018. Key updates in the later branches provided partial DirectX 11 compatibility through software emulation and hacks in select applications, though the series natively capped at DirectX 10.1 feature levels, limiting full hardware acceleration for DX11 titles. Vulkan support was absent entirely, as it debuted with Maxwell-generation GPUs in 2016. A notable 2015 driver release (341.81) included optimizations for games like , enabling playable performance on DX10-capable hardware via fallback rendering modes. NVIDIA introduced its unified driver architecture in 2014 with the R344 branch, consolidating support for newer , , and Tesla products into a single codebase for streamlined development; however, the GeForce 9 series stayed on the dedicated legacy R340 branch to maintain compatibility without benefiting from unified enhancements.

Market Impact and Discontinuation

The GeForce 9 series garnered mixed market reception upon its 2008 launch, with reviewers highlighting its efficiency improvements and performance uplifts over the while noting its largely incremental nature. For instance, the GeForce 9600 GT was lauded for delivering up to 83% better performance than the 8600 GTS in synthetic tests, offering strong value at under $200, but high-end models like the GeForce 9800 GT were described as evolutionary rather than groundbreaking in architecture or features. In terms of sales, the series performed well in the OEM segment, where mobile variants like the 9600M GT and 9700M were widely adopted in consumer systems, including Apple's lineup, contributing to NVIDIA's dominance in integrated and mobile graphics during the late . However, discrete high-end sales were hampered by the swift introduction of the in mid-2008, which overshadowed the 9800 GT and shifted consumer focus to unified shader architectures. NVIDIA's overall GPU revenue declined 29% in 2009 compared to 2008, reflecting broader market pressures amid the global . Competition intensified with AMD's launch in late 2008, sparking a mid-range where aggressively cut prices on cards like the HD 4850 and HD 4870 to capture budget market share. countered by reducing 9 pricing, but AMD's claims of superior performance in value-oriented segments led to NVIDIA ceding ground in the entry-level discrete market, with the outselling 9 equivalents in price-sensitive categories. The GeForce 9 series hardware reached end-of-life by late 2008, as retail stock rapidly depleted in favor of the , with full discontinuation from major retailers by early 2009. Driver support concluded on April 1, 2016, marking the end of the Release 340 branch, after which no further updates were provided; the series lacks compatibility with modern features like RTX or ray tracing due to its Tesla architecture limitations. In legacy terms, the series advanced the trend toward integrated graphics solutions via chips like the and 9400, influencing OEM designs for power-efficient . It also bolstered early adoption with version 1.1 support, enabling prosumer applications in general-purpose computing and laying groundwork for NVIDIA's expansion into scientific and AI workloads through accessible hardware.

References

  1. https://developer.download.nvidia.com/GPU_Programming_Guide/GPU_Programming_Guide_G80.pdf
  2. https://www.nvidia.com/docs/IO/55506/GeForce_GTX_200_GPU_Technical_Brief.pdf
  3. https://www.techpowerup.com/gpu-specs/geforce-9600-gt.c206
  4. https://www.techpowerup.com/gpu-specs/geforce-9800-gt.c635
  5. https://www.techpowerup.com/gpu-specs/geforce-9800-gx2.c208
  6. https://www.techpowerup.com/gpu-specs/?f=generation_NVIDIA%2BGeForce%2B9
  7. https://www.techpowerup.com/gpu-specs/geforce-9400-gt.c1765
  8. https://www.nvidia.com/en-us/drivers/windows/180_48/winxp_180-48_whql/
  9. https://bit-tech.net/reviews/tech/graphics/nvidia_hybrid_sli_preview/4/
  10. https://www.techpowerup.com/gpu-specs/?f=generation_NVIDIA%2BGeForce%2B9%2BIGP
  11. https://www.techpowerup.com/53358/geforce-9-series-roadmap-updated
  12. https://phys.org/news/2008-02-nvidia-reveals-next-generation-geforce-series.html
  13. https://www.tomshardware.com/picturestory/715-history-of-nvidia-gpus-2.html
  14. https://www.researchgate.net/publication/3216496_NVIDIA_Tesla_A_unified_graphics_and_computing_architecture
  15. https://arstechnica.com/gadgets/2008/09/nvidia-sued-over-notebook-gpu-failures/
  16. https://news.softpedia.com/news/CES-2008-Asus-Has-Unveiled-a-GeForce-9500M-GS-GPU-Notebook-75818.shtml
  17. https://www.guru3d.com/review/geforce-9500-gt-review/
  18. https://bit-tech.net/reviews/tech/motherboards/nvidia-mcp7a-geforce-9-series-mgpu/1/
  19. https://www.techradar.com/news/computing-components/graphics-cards/nvidia-kicks-off-geforce-9-series-with-9600-gt-244596
  20. https://sg.news.yahoo.com/2008-02-22-geforce-9-roadmap-revealed-9800-gx2-due-march-11th.html
  21. https://www.techpowerup.com/gpu-specs/geforce-9800-gtx.c237
  22. https://www.cs.cmu.edu/afs/cs/academic/class/15869-f11/www/readings/lindholm08_tesla.pdf
  23. https://www.techpowerup.com/gpu-specs/geforce-9100.c2049
  24. https://www.nvidia.com/en-us/drivers/details/5091/
  25. https://www.notebookcheck.net/GeForce-9500M-G-vs-GeForce-9500M-GE_752_1082.247598.0.html
  26. https://www.nvidia.com/en-us/drivers/windows/175_16/winvista_x86_175-16_whql/
  27. https://developer.nvidia.com/cuda-gpus
  28. https://www.nvidia.com/en-us/drivers/details/31342/
  29. https://www.techpowerup.com/gpu-specs/nvidia-c78.g610
  30. https://www.techpowerup.com/gpu-specs/geforce-9300-gs.c668
  31. https://technical.city/en/video/GeForce-8500-GT-vs-GeForce-9300-GS
  32. https://www.techpowerup.com/gpu-specs/geforce-8500-gt.c765
  33. https://videocardz.net/nvidia-geforce-8500-gt
  34. https://www.nvidia.com/en-us/drivers/geforce-8600-8500-tech-specs/
  35. https://www.techpowerup.com/gpu-specs/nvidia-g96.g149
  36. https://www.techpowerup.com/gpu-specs/geforce-9400-gt.c3159
  37. https://www.techpowerup.com/review/point-of-view-geforce-9400-gt-1-gb/
  38. https://hothardware.com/reviews/nvidia-geforce-9300-and-9400-motherboard-gpus
  39. https://pcper.com/2008/07/nvidia-geforce-9500-gt-review-budget-gaming-redux/
  40. https://hothardware.com/reviews/nvidia-geforce-9500-gt-mainstream-gpu
  41. https://www.techpowerup.com/gpu-specs/geforce-9500-gt.c3373
  42. https://technical.city/en/video/GeForce-9500-GT
  43. https://www.evga.com/products/specs/gpu.aspx?pn=DCAA3F3D-1396-49C6-AECA-174272B8B321
  44. https://www.techpowerup.com/gpu-specs/geforce-9500-gs.c3372
  45. https://www.vgamuseum.info/index.php/cpu/item/311-nvidia-geforce-9500-gs
  46. https://www.guru3d.com/review/geforce-9600-gt-512mb-6-way-shootout-review/page-2/
  47. https://www.techpowerup.com/57041/nvidia-geforce-8800-gs-rebranded-to-9600-gso
  48. https://www.techpowerup.com/gpu-specs/geforce-9600-gso.c215
  49. https://www.techpowerup.com/gpu-specs/geforce-9600-gso-512.c657
  50. https://www.evga.com/products/specs/gpu.aspx?pn=B01CE8BD-4008-454D-B7F7-FE746398ABA4
  51. https://www.techpowerup.com/gpu-specs/geforce-9800-gtx.c207
  52. https://www.nvidia.com/content/partnerforce/nvidia/us/lc_geforce_mar08_final_lowres.pdf
  53. https://www.techpowerup.com/review/zotac-geforce-9800-gx2/
  54. https://videocardz.net/gpu/nvidia-mcp78
  55. https://www.techpowerup.com/gpu-specs/nvidia-g98.g1037
  56. https://www.techpowerup.com/gpu-specs/nvidia-g94.g60
  57. https://www.techpowerup.com/gpu-specs/geforce-9600-gs-oem.c660
  58. https://www.techpowerup.com/gpu-specs/nvidia-g92.g58
  59. https://www.techpowerup.com/gpu-specs/nvidia-g92b.g94
  60. https://www.techpowerup.com/gpu-specs/nvidia-g86.g1029
  61. https://www.techpowerup.com/gpu-specs/geforce-8800-gt.c201
  62. https://www.notebookcheck.net/NVIDIA-GeForce-9100M-G.11351.0.html
  63. https://www.techpowerup.com/gpu-specs/geforce-9200m-gs.c1525
  64. https://www.techpowerup.com/gpu-specs/geforce-9100m-g-mgpu-amd.c1527
  65. https://www.notebookcheck.net/NVIDIA-GeForce-9200M-GS.9453.0.html
  66. https://www.notebookcheck.net/NVIDIA-GeForce-9300M-G.11359.0.html
  67. https://www.techpowerup.com/gpu-specs/geforce-9300m-g.c1524
  68. https://www.techpowerup.com/gpu-specs/geforce-9300m-gs.c1523
  69. https://www.notebookcheck.net/NVIDIA-GeForce-9300M-GS.9452.0.html
  70. https://www.techpowerup.com/gpu-specs/geforce-9400m-g.c1872
  71. https://www.notebookcheck.net/NVIDIA-GeForce-9400M-G.11949.0.html
  72. https://www.apple.com/newsroom/2008/10/14New-MacBook-Family-Redefines-Notebook-Design/
  73. https://www.nvidia.com/en-us/geforce/technologies/optimus/technology/
  74. https://www.techpowerup.com/gpu-specs/geforce-9500m-g.c1521
  75. https://www.notebookcheck.net/NVIDIA-GeForce-9500M-G.9451.0.html
  76. https://www.techpowerup.com/gpu-specs/geforce-9500m-gs.c1520
  77. https://www.notebookcheck.net/NVIDIA-GeForce-9600M-GT.9449.0.html
  78. https://www.techpowerup.com/gpu-specs/geforce-9600m-gt.c1518
  79. https://www.notebookcheck.net/NVIDIA-GeForce-9600M-GS.9450.0.html
  80. https://www.phoronix.com/news/NjUwMg
  81. https://www.techpowerup.com/gpu-specs/geforce-9700m-gt.c1516
  82. https://www.techpowerup.com/gpu-specs/geforce-9700m-gts.c1515
  83. https://www.notebookcheck.net/NVIDIA-GeForce-9700M-GT.9454.0.html
  84. https://www.techpowerup.com/gpu-specs/geforce-9800m-gs.c1514
  85. https://www.techpowerup.com/gpu-specs/geforce-9800m-gts.c1513
  86. https://www.techpowerup.com/gpu-specs/geforce-9800m-gt.c1512
  87. https://www.techpowerup.com/gpu-specs/geforce-9800m-gtx.c941
  88. https://www.notebookcheck.net/NVIDIA-GeForce-9700M-GTS.9899.0.html
  89. https://www.dell.com/community/en/conversations/laptops-general-locked-topics/laptop-xps-m1730-hot-by-name-hot-by-nature/647e9e59f4ccf8a8de884730
  90. https://www.notebookcheck.net/NVIDIA-GeForce-9800M-GS.12932.0.html
  91. https://www.notebookcheck.net/NVIDIA-GeForce-9800M-GTX.9919.0.html
  92. https://www.techpowerup.com/gpu-specs/geforce-9100m-g-mgpu-intel.c1526
  93. https://www.notebookcheck.net/NVIDIA-GeForce-9500M-GE-Hybrid-SLI.14029.0.html
  94. https://www.techpowerup.com/66404/geforce-9800m-9700m-offer-performance-and-energy-savings-in-a-broad-range
  95. https://www.tomshardware.com/reviews/nvidia-geforce-9600-gt%2C1780-7.html
  96. https://www.nvidia.com/en-us/drivers/windows/175_16/winxp_175-16_whql/
  97. https://www.nvidia.com/en-us/drivers/details/112596/
  98. https://nvidia.custhelp.com/app/answers/detail/a_id/3473/~/eol-windows-driver-support-for-legacy-products
  99. https://www.phoronix.com/news/MTYyODA
  100. https://nvidia.custhelp.com/app/answers/detail/a_id/3142/~/support-timeframes-for-unix-legacy-gpu-releases
  101. https://elitemacx86.com/threads/how-to-enable-nvidia-graphics-on-macos-monterey-and-later.925/
  102. https://developer.nvidia.com/vulkan-driver
  103. https://www.nvidia.com/en-us/drivers/details/89990/
  104. https://arstechnica.com/gadgets/2008/02/nvidia-9600gt-review/
  105. https://www.ebay.com/b/Apple-NVIDIA-GeForce-9600M-GT-Laptops/111422/bn_94631674
  106. https://www.sec.gov/Archives/edgar/data/1045810/000104581009000013/fy2009form10k.htm
  107. https://www.techpowerup.com/76245/nvidia-to-cut-graphics-card-prices-to-counter-amd-ati
Add your contribution
Related Hubs
User Avatar
No comments yet.