GPU Comparison
GPU
Intel Arc A570M
CORE STATE
DG2-256
VRAM
8 GB
CLOCK SPEED
1300 MHz
TDP
75 W
BUS WIDTH
128 bit
ARCHITECTURE
Xe-HPG
PROCESS
6 nm
LAUNCH DATE
2023
VS
GEFORCE
H100 CNX
CORE STATE
GH100
VRAM
80 GB
CLOCK SPEED
1845 MHz
TDP
350 W
BUS WIDTH
5120 bit
ARCHITECTURE
Hopper
PROCESS
5 nm
LAUNCH DATE
2023
PERFORMANCE BENCHMARKS
geekbench_opencl
DETAILED SPECIFICATIONS
SPECIFICATION
A570M
H100 CNX
Core Specs
Shading Units
2,048
14,592
+612.5%
Shaders
2,048
14,592
+612.5%
TMUs
128
456
+256.3%
ROPs
64
24
-62.5%
SM Count
—
114
Execution Units
256
—
Clocks
Base Clock
900 MHz
690 MHz
Boost Clock
1300 MHz
1845 MHz
Memory Clock
1750 MHz
14 Gbps effective
1593 MHz
3.2 Gbps effective
Memory
Memory Size
8 GB
80 GB
VRAM (MB)
8,192
81,920
+900.0%
Memory Type
GDDR6
HBM2e
Memory Bus
128 bit
5120 bit
Bandwidth
224.0 GB/s
2.04 TB/s
Cache
L1 Cache
—
256 KB (per SM)
L2 Cache
8 MB
50 MB
Performance
Pixel Rate
83.20 GPixel/s
44.28 GPixel/s
Texture Rate
166.4 GTexel/s
841.3 GTexel/s
FP32 (TFLOPS)
5.325 TFLOPS
53.84 TFLOPS
FP64 (TFLOPS)
—
26.92 TFLOPS (1:2)
FP16 (TFLOPS)
10.65 TFLOPS (2:1)
215.4 TFLOPS (4:1)
AI/RT
RT Cores
16
—
Tensor Cores
—
456
XMX Cores
256
—
Power
TDP
75 W
350 W
TDP (W)
75
350
+366.7%
Suggested PSU
—
750 W
Power Connectors
—
8-pin EPS
Architecture
Architecture
Xe-HPG
Hopper
GPU Name
DG2-256
GH100
Generation
Alchemist
(Arc 5 Mobile)
Server Hopper
(Hxx)
Process Size
6 nm
5 nm
Transistors
11,500 million
80,000 million
Die Size
269 mm²
814 mm²
Foundry
TSMC
TSMC
Density
42.8M / mm²
98.3M / mm²
API Support
DirectX
12 Ultimate (12_2)
—
OpenGL
4.6
—
Vulkan
1.4
—
OpenCL
3.0
3.0
CUDA
—
9.0
Shader Model
6.6
—
Physical
Slot Width
IGP
Dual-slot
Length
—
267 mm
10.5 inches
Height
—
111 mm
4.4 inches
Outputs
Portable Device Dependent
No outputs
Bus Interface
PCIe 4.0 x8
PCIe 5.0 x16
Other
Production
Active
Active
Predecessor
—
Server Ada
Successor
—
Server Blackwell