GPU Comparison
RADEON
AMD Radeon Pro Vega II
CORE STATE
Vega 20
VRAM
32 GB
CLOCK SPEED
1720 MHz
TDP
475 W
BUS WIDTH
4096 bit
ARCHITECTURE
GCN 5.1
PROCESS
7 nm
LAUNCH DATE
2019
VS
GEFORCE
GeForce RTX 4090 D
CORE STATE
AD102
VRAM
24 GB
CLOCK SPEED
2520 MHz
TDP
425 W
BUS WIDTH
384 bit
ARCHITECTURE
Ada Lovelace
PROCESS
5 nm
LAUNCH DATE
2023
PERFORMANCE BENCHMARKS
geekbench_metal
geekbench_opencl
geekbench_vulkan
3dmark_3dmark_steel_nomad_dx12
DETAILED SPECIFICATIONS
SPECIFICATION
Pro Vega II
RTX 4090 D
Core Specs
Shading Units
4,096
14,592
+256.3%
Shaders
4,096
14,592
+256.3%
TMUs
256
456
+78.1%
ROPs
64
176
+175.0%
Compute Units
64
—
SM Count
—
114
Clocks
Base Clock
1574 MHz
2280 MHz
Boost Clock
1720 MHz
2520 MHz
Memory Clock
806 MHz
1612 Mbps effective
1313 MHz
21 Gbps effective
Memory
Memory Size
32 GB
24 GB
VRAM (MB)
32,768
24,576
-25.0%
Memory Type
HBM2
GDDR6X
Memory Bus
4096 bit
384 bit
Bandwidth
825.3 GB/s
1.01 TB/s
Cache
L1 Cache
16 KB (per CU)
128 KB (per SM)
L2 Cache
4 MB
72 MB
Performance
Pixel Rate
110.1 GPixel/s
443.5 GPixel/s
Texture Rate
440.3 GTexel/s
1,149.1 GTexel/s
FP32 (TFLOPS)
14.09 TFLOPS
73.54 TFLOPS
FP64 (TFLOPS)
7.045 TFLOPS (1:2)
1,149.1 GFLOPS (1:64)
FP16 (TFLOPS)
28.18 TFLOPS (2:1)
73.54 TFLOPS (1:1)
AI/RT
RT Cores
—
114
Tensor Cores
—
456
Power
TDP
475 W
425 W
TDP (W)
475
425
-10.5%
Suggested PSU
850 W
800 W
Power Connectors
—
1x 16-pin
Architecture
Architecture
GCN 5.1
Ada Lovelace
GPU Name
Vega 20
AD102
Generation
Radeon Pro Mac
(Vega Series)
GeForce 40
Process Size
7 nm
5 nm
Transistors
13,230 million
76,300 million
Die Size
331 mm²
609 mm²
Foundry
TSMC
TSMC
Density
40.0M / mm²
125.3M / mm²
API Support
DirectX
12 (12_1)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.3
1.4
OpenCL
2.1
3.0
CUDA
—
8.9
Shader Model
6.7
6.8
Physical
Slot Width
Quad-slot
Triple-slot
Length
—
304 mm
12 inches
Height
—
137 mm
5.4 inches
Outputs
1x HDMI 2.0b4x Thunderbolt
1x HDMI 2.13x DisplayPort 1.4a
Bus Interface
Apple MPX
PCIe 4.0 x16
Other
Launch Price
2,199 USD
1,599 USD
Production
End-of-life
End-of-life
Predecessor
—
GeForce 30
Successor
—
GeForce 50