GPU Comparison
GEFORCE
NVIDIA PG506-232
CORE STATE
GA100
VRAM
24 GB
CLOCK SPEED
1440 MHz
TDP
165 W
BUS WIDTH
3072 bit
ARCHITECTURE
Ampere
PROCESS
7 nm
LAUNCH DATE
2021
VS
GEFORCE
Tesla M4
CORE STATE
GM206
VRAM
4 GB
CLOCK SPEED
1072 MHz
TDP
50 W
BUS WIDTH
128 bit
ARCHITECTURE
Maxwell 2.0
PROCESS
28 nm
LAUNCH DATE
2015
PERFORMANCE BENCHMARKS
geekbench_opencl
DETAILED SPECIFICATIONS
SPECIFICATION
PG506-232
Tesla M4
Core Specs
Shading Units
3,584
1,024
-71.4%
Shaders
3,584
1,024
-71.4%
TMUs
224
64
-71.4%
ROPs
96
32
-66.7%
SM Count
56
—
Clocks
Base Clock
930 MHz
872 MHz
Boost Clock
1440 MHz
1072 MHz
Memory Clock
1215 MHz
2.4 Gbps effective
1375 MHz
5.5 Gbps effective
Memory
Memory Size
24 GB
4 GB
VRAM (MB)
24,576
4,096
-83.3%
Memory Type
HBM2
GDDR5
Memory Bus
3072 bit
128 bit
Bandwidth
933.1 GB/s
88.00 GB/s
Cache
L1 Cache
192 KB (per SM)
48 KB (per SMM)
L2 Cache
24 MB
1024 KB
Performance
Pixel Rate
138.2 GPixel/s
34.30 GPixel/s
Texture Rate
322.6 GTexel/s
68.61 GTexel/s
FP32 (TFLOPS)
10.32 TFLOPS
2.195 TFLOPS
FP64 (TFLOPS)
5.161 TFLOPS (1:2)
68.61 GFLOPS (1:32)
FP16 (TFLOPS)
10.32 TFLOPS (1:1)
—
AI/RT
Tensor Cores
224
—
Power
TDP
165 W
50 W
TDP (W)
165
50
-69.7%
Suggested PSU
450 W
250 W
Power Connectors
8-pin EPS
—
Architecture
Architecture
Ampere
Maxwell 2.0
GPU Name
GA100
GM206
Generation
Server Ampere
(Axx)
Tesla Maxwell
(Mxx)
Process Size
7 nm
28 nm
Transistors
54,200 million
2,940 million
Die Size
826 mm²
228 mm²
Foundry
TSMC
TSMC
Density
65.6M / mm²
12.9M / mm²
API Support
DirectX
—
12 (12_1)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
8.0
5.2
Shader Model
—
6.8
Physical
Slot Width
Dual-slot
Single-slot
Length
267 mm
10.5 inches
—
Height
112 mm
4.4 inches
—
Outputs
No outputs
No outputs
Bus Interface
PCIe 4.0 x16
PCIe 3.0 x16
Other
Production
End-of-life
End-of-life
Predecessor
Tesla Turing
Tesla Kepler
Successor
Server Ada
Tesla Pascal