GPU Comparison
RADEON
AMD Radeon Pro Vega 16
CORE STATE
Vega 12
VRAM
4 GB
CLOCK SPEED
1190 MHz
TDP
75 W
BUS WIDTH
1024 bit
ARCHITECTURE
GCN 5.0
PROCESS
14 nm
LAUNCH DATE
2018
VS
RADEON
Radeon PRO W6400
CORE STATE
Navi 24
VRAM
4 GB
CLOCK SPEED
2321 MHz
TDP
50 W
BUS WIDTH
64 bit
ARCHITECTURE
RDNA 2.0
PROCESS
6 nm
LAUNCH DATE
2022
PERFORMANCE BENCHMARKS
geekbench_metal
geekbench_vulkan
geekbench_opencl
DETAILED SPECIFICATIONS
SPECIFICATION
Pro Vega 16
PRO W6400
Core Specs
Shading Units
1,024
768
-25.0%
Shaders
1,024
768
-25.0%
TMUs
64
48
-25.0%
ROPs
32
32
0.0%
Compute Units
16
12
-25.0%
Clocks
Base Clock
815 MHz
2039 MHz
Boost Clock
1190 MHz
2321 MHz
Memory Clock
1200 MHz
2.4 Gbps effective
2000 MHz
16 Gbps effective
Memory
Memory Size
4 GB
4 GB
VRAM (MB)
4,096
4,096
0.0%
Memory Type
HBM2
GDDR6
Memory Bus
1024 bit
64 bit
Bandwidth
307.2 GB/s
128.0 GB/s
Cache
L1 Cache
16 KB (per CU)
128 KB per Array
L2 Cache
1024 KB
1024 KB
L3 Cache
—
8 MB
L0 Cache
—
32 KB per WGP
Performance
Pixel Rate
38.08 GPixel/s
74.27 GPixel/s
Texture Rate
76.16 GTexel/s
111.4 GTexel/s
FP32 (TFLOPS)
2.437 TFLOPS
3.565 TFLOPS
FP64 (TFLOPS)
152.3 GFLOPS (1:16)
222.8 GFLOPS (1:16)
FP16 (TFLOPS)
4.874 TFLOPS (2:1)
7.130 TFLOPS (2:1)
AI/RT
RT Cores
—
12
Power
TDP
75 W
50 W
TDP (W)
75
50
-33.3%
Suggested PSU
—
250 W
Power Connectors
—
None
Architecture
Architecture
GCN 5.0
RDNA 2.0
GPU Name
Vega 12
Navi 24
Generation
Radeon Pro Mac
(Vega Series)
Radeon Pro Navi
(Navi II Series)
Process Size
14 nm
6 nm
Transistors
—
5,400 million
Die Size
—
107 mm²
Foundry
GlobalFoundries
TSMC
Density
—
50.5M / mm²
API Support
DirectX
12 (12_1)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.3
1.4
OpenCL
2.1
2.2
Shader Model
6.0
6.8
Physical
Slot Width
IGP
Single-slot
Outputs
Portable Device Dependent
2x DisplayPort 1.4a
Bus Interface
PCIe 3.0 x16
PCIe 4.0 x4
Other
Production
End-of-life
End-of-life
Predecessor
—
Radeon Pro Vega