Home GPU Comparison AMD Radeon HD 8970 OEM vs NVIDIA GeForce GTS 450 OEM

AMD Radeon HD 8970 OEM vs NVIDIA GeForce GTS 450 OEM

We compared two Desktop platform GPUs: 3GB VRAM Radeon HD 8970 OEM and 512MB VRAM GeForce GTS 450 OEM to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

AMD Radeon HD 8970 OEM 's Advantages
Released 2 years and 4 months late
More VRAM (3GB vs 512GB)
Larger VRAM bandwidth (264.0GB/s vs 57.73GB/s)
1904 additional rendering cores
NVIDIA GeForce GTS 450 OEM 's Advantages
Lower TDP (106W vs 250W)

Score

Benchmark

FP32 (float)
Radeon HD 8970 OEM +740%
3.789 TFLOPS
GeForce GTS 450 OEM
0.451 TFLOPS
VS

Graphics Card

Jan 2013
Release Date
Sep 2010
Sea Islands
Generation
GeForce 400
Desktop
Type
Desktop
PCIe 3.0 x16
Bus Interface
PCIe 2.0 x16

Clock Speeds

-
-
-
-
-
-
1375 MHz
Memory Clock
902 MHz

Memory

3GB
Memory Size
512MB
GDDR5
Memory Type
GDDR5
384bit
Memory Bus
128bit
264.0GB/s
Bandwidth
57.73GB/s

Render Config

32
Compute Units
-
-
SM Count
3
2048
Shading Units
144
128
TMUs
24
32
ROPs
16
-
-
-
-
-
-
16 KB (per CU)
L1 Cache
64 KB (per SM)
768 KB
L2 Cache
256 KB
-
-
-

Theoretical Performance

29.60 GPixel/s
Pixel Rate
4.698 GPixel/s
118.4 GTexel/s
Texture Rate
18.79 GTexel/s
-
-
-
3.789 TFLOPS
FP32 (float)
451.0 GFLOPS
947.2 GFLOPS
FP64 (double)
37.58 GFLOPS

Board Design

250W
TDP
106W
600 W
Suggested PSU
300 W
1x DVI 1x HDMI 1.4a 2x mini-DisplayPort 1.2
Outputs
1x DVI 1x DisplayPort 1x VGA
1x 6-pin + 1x 8-pin
Power Connectors
1x 6-pin

Graphics Processor

Tahiti
GPU Name
GF106
Tahiti XT (215-0821060)
GPU Variant
-
GCN 1.0
Architecture
Fermi
TSMC
Foundry
TSMC
28 nm
Process Size
40 nm
4.313 billion
Transistors
1.17 billion
352 mm²
Die Size
238 mm²

Graphics Features

12 (11_1)
DirectX
12 (11_0)
4.6
OpenGL
4.6
1.2
OpenCL
1.1
1.2
Vulkan
N/A
-
CUDA
2.1
5.1
Shader Model
5.1

Related GPU Comparisons

Related News

© 2024 - TopCPU.net   Contact Us Privacy Policy