NVIDIA GeForce GTS 450 OEM vs ATI Radeon HD 4750
We compared two Desktop platform GPUs: 512MB VRAM GeForce GTS 450 OEM and 512MB VRAM Radeon HD 4750 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce GTS 450 OEM 's Advantages
Released 1 years late
Larger VRAM bandwidth (57.73GB/s vs 51.20GB/s)
ATI Radeon HD 4750 's Advantages
496 additional rendering cores
Lower TDP (100W vs 106W)
Score
Benchmark
FP32 (float)
GeForce GTS 450 OEM
0.451 TFLOPS
Radeon HD 4750
+107%
0.934 TFLOPS
Graphics Card
Sep 2010
Release Date
Sep 2009
GeForce 400
Generation
Radeon R700
Desktop
Type
Desktop
PCIe 2.0 x16
Bus Interface
PCIe 2.0 x16
Clock Speeds
-
-
-
-
-
-
902 MHz
Memory Clock
800 MHz
Memory
512MB
Memory Size
512MB
GDDR5
Memory Type
GDDR5
128bit
Memory Bus
128bit
57.73GB/s
Bandwidth
51.20GB/s
Render Config
-
Compute Units
8
3
SM Count
-
144
Shading Units
640
24
TMUs
32
16
ROPs
16
-
-
-
-
-
-
64 KB (per SM)
L1 Cache
16 KB (per CU)
256 KB
L2 Cache
128 KB
-
-
-
Theoretical Performance
4.698 GPixel/s
Pixel Rate
11.68 GPixel/s
18.79 GTexel/s
Texture Rate
23.36 GTexel/s
-
-
-
451.0 GFLOPS
FP32 (float)
934.4 GFLOPS
37.58 GFLOPS
FP64 (double)
186.9 GFLOPS
Board Design
106W
TDP
100W
300 W
Suggested PSU
300 W
1x DVI
1x DisplayPort
1x VGA
Outputs
2x DVI
1x S-Video
1x 6-pin
Power Connectors
1x 6-pin
Graphics Processor
GF106
GPU Name
RV740
-
GPU Variant
RV740 PRO
Fermi
Architecture
TeraScale
TSMC
Foundry
TSMC
40 nm
Process Size
40 nm
1.17 billion
Transistors
0.826 billion
238 mm²
Die Size
137 mm²
Graphics Features
12 (11_0)
DirectX
10.1 (10_1)
4.6
OpenGL
3.3
1.1
OpenCL
1.1
N/A
Vulkan
N/A
2.1
CUDA
-
5.1
Shader Model
4.1
