ATI Radeon HD 6670 vs NVIDIA GeForce 9800 GT

Theoretical performance comparison

Real-world game, 3D graphics and compute performance is dependent on several important card parameters, including pixel fillrate, texture fillrate, memory bandwidth, single-precision performance and double-precision performance. Below we explain why they are essential and which card has better specs.

Pixel fill rate (gigapixels/s)

11
8.8
6.6
4.4
2.2
0
 
 
6.4
 
9.6
 
 
Higher is better
Although it has lower graphics clock rate, the NVIDIA GeForce 9800 GT graphics card comes with better pixel fill rate, owing to twice as many Raster Operations Pipelines (ROPs). Having higher pixel fillrate allows the GPU to draw more pixels on screen and off screen, which is beneficial for some 3D effects in games, or when playing at higher display resolutions.
  - ATI Radeon HD 6670
  - NVIDIA GeForce 9800 GT

Texture fill rate (gigatexels/s)

40
32
24
16
8
0
 
 
19.2
 
33.6
 
 
Higher is better
Even though its graphics clock is slower, the GeForce 9800 GT has many more Texture Mapping Units (TMUs), which results in significantly better texture fill rate. Having better maximum texture fill rate allows the card to apply more textures and/or utilize more complex 3D effects for each texel, which improves games visual appearance.

Single Precision performance (GFLOPS)

900
720
540
360
180
0
 
 
768
 
336
 
 
Higher is better
Maximum Single Precision performance illustrates how many single-precision floating point operations the graphics card can execute per second. The performance is expressed in billions of Floating Point Operations Per Second, or GFLOPS. As a rule, the faster CUDA cores or stream processors operate at, and the more cores / processors the GPU has, the higher Single Precision performance will be. The ATI Radeon HD 6670 graphics unit is considerably faster here. Higher single-precision performance number means the graphics card will perform better in general computing applications. Since CUDA cores or stream processors are also used as vertex and geometry shaders for 3D image generation, higher performance is also beneficial to games.

Memory bandwidth (GB/s)

80
64
48
32
16
0
 
 
64
 
57.6
 
 
Higher is better
To speed up processing, the cards store 3D scene data, textures and intermediate data, used for image generation, in on-board memory. The video memory usually has much higher bandwidth than system RAM, and more bandwidth allows the GPU to run at higher screen resolutions, use larger and more detailed textures, and apply more complex 3D effects and filters. The bandwidth depends on memory type, speed, and width of memory interface. Specifically, the ATI Radeon HD 6670 comes with higher clocked memory of better kind. As such, it has higher memory bandwidth.
  - ATI Radeon HD 6670
  - NVIDIA GeForce 9800 GT

Specs comparison

All rows with different specifications or features are highlighted.

General information

Market segmentDesktop
ManufacturerATINVIDIA
ModelRadeon HD 6670GeForce 9800 GT

Architecture / Interface

Die name  
Architecture  
Fabrication process  
Bus interface  

Cores / shaders

CUDA cores  
ROPs  
Color ROPs  
Stream processors  
Pixel fill rate  
Texture units  
Texture fill rate  
Single Precision performance  

Clocks / Memory

Graphics clock800 MHz600 MHz
Processor clock 1500 MHz
Memory size1024 MB
Memory typeGDDR5GDDR3
Memory clock  
Memory interface width  
Memory bandwidth  

Other features

Maximum crossfire options  
Maximum SLI options  
Maximum power  

Better values / features are marked with green color, and worse values are in red color.


Detailed specifications:

Compare graphics cards

Type in full or partial GPU manufacturer, model name and/or part number.

More comparisons