GPU Comparison
GPU
Intel Arc A580
CORE STATE
DG2-512
VRAM
8 GB
CLOCK SPEED
2000 MHz
TDP
175 W
BUS WIDTH
256 bit
ARCHITECTURE
Xe-HPG
PROCESS
6 nm
LAUNCH DATE
2023
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
3dmark_3dmark_steel_nomad_dx12
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
A580
A10M
Core Specs
Shading Units
3,072
7,168
+133.3%
Shaders
3,072
7,168
+133.3%
TMUs
192
224
+16.7%
ROPs
96
80
-16.7%
SM Count
—
56
Execution Units
384
—
Clocks
Base Clock
1700 MHz
975 MHz
Boost Clock
2000 MHz
1635 MHz
Memory Clock
2000 MHz
16 Gbps effective
1563 MHz
12.5 Gbps effective
Memory
Memory Size
8 GB
20 GB
VRAM (MB)
8,192
20,480
+150.0%
Memory Type
GDDR6
GDDR6
Memory Bus
256 bit
320 bit
Bandwidth
512.0 GB/s
500.2 GB/s
Cache
L1 Cache
—
128 KB (per SM)
L2 Cache
8 MB
6 MB
Performance
Pixel Rate
192.0 GPixel/s
130.8 GPixel/s
Texture Rate
384.0 GTexel/s
366.2 GTexel/s
FP32 (TFLOPS)
12.29 TFLOPS
23.44 TFLOPS
FP64 (TFLOPS)
1.536 TFLOPS (1:8)
732.5 GFLOPS (1:32)
FP16 (TFLOPS)
24.58 TFLOPS (2:1)
23.44 TFLOPS (1:1)
AI/RT
RT Cores
24
56
+133.3%
Tensor Cores
—
224
XMX Cores
384
—
Power
TDP
175 W
150 W
TDP (W)
175
150
-14.3%
Suggested PSU
450 W
450 W
Power Connectors
2x 8-pin
8-pin EPS
Architecture
Architecture
Xe-HPG
Ampere
GPU Name
DG2-512
GA102
Generation
Alchemist
(Arc 5)
Server Ampere
(Axx)
Process Size
6 nm
8 nm
Transistors
21,700 million
28,300 million
Die Size
406 mm²
628 mm²
Foundry
TSMC
Samsung
Density
53.4M / mm²
45.1M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
3.0
3.0
CUDA
—
8.6
Shader Model
6.6
6.8
Physical
Slot Width
Dual-slot
Single-slot
Length
—
267 mm
10.5 inches
Height
—
112 mm
4.4 inches
Outputs
1x HDMI 2.13x DisplayPort 2.0
No outputs
Bus Interface
PCIe 4.0 x16
PCIe 4.0 x16
Other
Production
Active
End-of-life
Predecessor
Xe Graphics
Tesla Turing
Successor
Battlemage
Server Ada