GPU Comparison
RADEON
AMD Instinct MI100
CORE STATE
Arcturus
VRAM
32 GB
CLOCK SPEED
1502 MHz
TDP
300 W
BUS WIDTH
4096 bit
ARCHITECTURE
CDNA 1.0
PROCESS
7 nm
LAUNCH DATE
2020
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
Instinct MI100
Tesla M4
Core Specs
Shading Units
7,680
1,024
-86.7%
Shaders
7,680
1,024
-86.7%
TMUs
480
64
-86.7%
ROPs
64
32
-50.0%
Compute Units
120
—
Clocks
Base Clock
1000 MHz
872 MHz
Boost Clock
1502 MHz
1072 MHz
Memory Clock
1200 MHz 2.4 Gbps effective
1375 MHz
5.5 Gbps effective
Memory
Memory Size
32 GB
4 GB
VRAM (MB)
32,768
4,096
-87.5%
Memory Type
HBM2
GDDR5
Memory Bus
4096 bit
128 bit
Bandwidth
1.23 TB/s
88.00 GB/s
Cache
L1 Cache
16 KB (per CU)
48 KB (per SMM)
L2 Cache
8 MB
1024 KB
Performance
Pixel Rate
96.13 GPixel/s
34.30 GPixel/s
Texture Rate
721.0 GTexel/s
68.61 GTexel/s
FP32 (TFLOPS)
23.07 TFLOPS
2.195 TFLOPS
FP64 (TFLOPS)
11.54 TFLOPS (1:2)
68.61 GFLOPS (1:32)
FP16 (TFLOPS)
46.14 TFLOPS (2:1)
—
Power
TDP
300 W
50 W
TDP (W)
300
50
-83.3%
Suggested PSU
700 W
250 W
Power Connectors
2x 8-pin
—
Architecture
Architecture
CDNA 1.0
Maxwell 2.0
GPU Name
Arcturus
GM206
Generation
Instinct (MIx)
Tesla Maxwell
(Mxx)
Process Size
7 nm
28 nm
Transistors
25,600 million
2,940 million
Die Size
750 mm²
228 mm²
Foundry
TSMC
TSMC
Density
34.1M / mm²
12.9M / mm²
API Support
DirectX
—
12 (12_1)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
2.1
3.0
CUDA
—
5.2
Shader Model
—
6.8
Physical
Slot Width
Dual-slot
Single-slot
Length
267 mm 10.5 inches
—
Height
111 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
Radeon Instinct
Tesla Kepler
Successor
—
Tesla Pascal