GPU Comparison
RADEON
AMD Radeon HD 7970M
CORE STATE
Wimbledon
VRAM
2 GB
CLOCK SPEED
—
TDP
100 W
BUS WIDTH
256 bit
ARCHITECTURE
GCN 1.0
PROCESS
28 nm
LAUNCH DATE
2012
VS
GEFORCE
A10M
CORE STATE
GA102
VRAM
20 GB
CLOCK SPEED
1635 MHz
TDP
150 W
BUS WIDTH
320 bit
ARCHITECTURE
Ampere
PROCESS
8 nm
LAUNCH DATE
—
PERFORMANCE BENCHMARKS
geekbench_opencl
DETAILED SPECIFICATIONS
SPECIFICATION
HD 7970M
A10M
Core Specs
Shading Units
1,280
7,168
+460.0%
Shaders
1,280
7,168
+460.0%
TMUs
80
224
+180.0%
ROPs
32
80
+150.0%
Compute Units
20
—
SM Count
—
56
Clocks
Base Clock
—
975 MHz
Boost Clock
—
1635 MHz
GPU Clock
850 MHz
—
Memory Clock
1200 MHz
4.8 Gbps effective
1563 MHz
12.5 Gbps effective
Memory
Memory Size
2 GB
20 GB
VRAM (MB)
2,048
20,480
+900.0%
Memory Type
GDDR5
GDDR6
Memory Bus
256 bit
320 bit
Bandwidth
153.6 GB/s
500.2 GB/s
Cache
L1 Cache
16 KB (per CU)
128 KB (per SM)
L2 Cache
512 KB
6 MB
Performance
Pixel Rate
27.20 GPixel/s
130.8 GPixel/s
Texture Rate
68.00 GTexel/s
366.2 GTexel/s
FP32 (TFLOPS)
2.176 TFLOPS
23.44 TFLOPS
FP64 (TFLOPS)
136.0 GFLOPS (1:16)
732.5 GFLOPS (1:32)
FP16 (TFLOPS)
—
23.44 TFLOPS (1:1)
AI/RT
RT Cores
—
56
Tensor Cores
—
224
Power
TDP
100 W
150 W
TDP (W)
100
150
+50.0%
Suggested PSU
—
450 W
Power Connectors
None
8-pin EPS
Architecture
Architecture
GCN 1.0
Ampere
GPU Name
Wimbledon
GA102
Generation
London
(HD 7900M)
Server Ampere
(Axx)
Process Size
28 nm
8 nm
Transistors
2,800 million
28,300 million
Die Size
212 mm²
628 mm²
Foundry
TSMC
Samsung
Density
13.2M / mm²
45.1M / mm²
API Support
DirectX
12 (11_1)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.2.170
1.4
OpenCL
2.1 (1.2)
3.0
CUDA
—
8.6
Shader Model
6.5 (5.1)
6.8
Physical
Slot Width
MXM Module
Single-slot
Length
—
267 mm
10.5 inches
Height
—
112 mm
4.4 inches
Outputs
Portable Device Dependent
No outputs
Bus Interface
MXM-B (3.0)
PCIe 4.0 x16
Other
Production
End-of-life
End-of-life
Predecessor
Vancouver
Tesla Turing
Successor
Solar System
Server Ada