ATI Radeon HD 3830 vs NVIDIA GeForce 9800 GX2

Comparative analysis of ATI Radeon HD 3830 and NVIDIA GeForce 9800 GX2 videocards for all known characteristics in the following categories: Essentials, Technical info, Video outputs and ports, Compatibility, dimensions and requirements, API support, Memory, Technologies. Benchmark videocards performance analysis: PassMark - G3D Mark, PassMark - G2D Mark, GFXBench 4.0 - T-Rex (Frames), GFXBench 4.0 - T-Rex (Fps).

ATI Radeon HD 3830

NVIDIA GeForce 9800 GX2

Differences

Reasons to consider the ATI Radeon HD 3830

Around 25% higher pipelines: 320 vs 2x 128
A newer manufacturing process allows for a more powerful, yet cooler running videocard: 55 nm vs 65 nm
2.6x lower typical power consumption: 75 Watt vs 197 Watt
Around 66% higher memory clock speed: 1656 MHz vs 1000 MHz

Pipelines	320 vs 2x 128
Manufacturing process technology	55 nm vs 65 nm
Thermal Design Power (TDP)	75 Watt vs 197 Watt
Memory clock speed	1656 MHz vs 1000 MHz

Reasons to consider the NVIDIA GeForce 9800 GX2

2.2x more core clock speed: 1500 MHz vs 668 MHz
7.2x more texture fill rate: 76.8 billion / sec vs 10.69 GTexel / s
Around 80% better floating-point performance: 2x 384.0 gflops vs 427.5 gflops
4x more maximum memory size: 2x 512 MB vs 256 MB

Core clock speed	1500 MHz vs 668 MHz
Texture fill rate	76.8 billion / sec vs 10.69 GTexel / s
Floating-point performance	2x 384.0 gflops vs 427.5 gflops
Maximum memory size	2x 512 MB vs 256 MB

Compare benchmarks

GPU 1: ATI Radeon HD 3830
GPU 2: NVIDIA GeForce 9800 GX2

Name	ATI Radeon HD 3830	NVIDIA GeForce 9800 GX2
PassMark - G3D Mark		798
PassMark - G2D Mark		518
GFXBench 4.0 - T-Rex (Frames)		3238
GFXBench 4.0 - T-Rex (Fps)		3238

Compare specifications (specs)

	ATI Radeon HD 3830	NVIDIA GeForce 9800 GX2
Essentials
Architecture	TeraScale	Tesla
Code name	RV670	G92
Launch date	1 April 2008	18 March 2008
Place in performance rating	not rated	489
Type	Desktop	Desktop
Launch price (MSRP)		$599
Technical info
Core clock speed	668 MHz	1500 MHz
Floating-point performance	427.5 gflops	2x 384.0 gflops
Manufacturing process technology	55 nm	65 nm
Pipelines	320	2x 128
Texture fill rate	10.69 GTexel / s	76.8 billion / sec
Thermal Design Power (TDP)	75 Watt	197 Watt
Transistor count	666 million	754 million
CUDA cores		256 (128 per GPU)
Maximum GPU temperature		105 °C
Video outputs and ports
Display Connectors	No outputs	2x DVI, 1x HDMI, HDMIDual Link DVI
Audio input for HDMI		S / PDIF
HDMI
Maximum VGA resolution		2048x1536
Multi monitor support
Compatibility, dimensions and requirements
Interface	PCIe 2.0 x16	PCIe 2.0 x16
Supplementary power connectors	1x 6-pin	6-pin & 8-pin
Length		10.5" (26.7 cm)
SLI options		Quad
API support
DirectX	10.0	10.0
OpenGL	3.3	2.1
Memory
Maximum RAM amount	256 MB	2x 512 MB
Memory bandwidth	26.5 GB / s	128 (64 per GPU)
Memory bus width	128 Bit	512 Bit
Memory clock speed	1656 MHz	1000 MHz
Memory type	GDDR3	GDDR3
Technologies
3D Vision
CUDA
SLI