GPU Comparison
GEFORCE
NVIDIA CMP 50HX
CORE STATE
TU102
VRAM
10 GB
CLOCK SPEED
1545 MHz
TDP
250 W
BUS WIDTH
320 bit
ARCHITECTURE
Turing
PROCESS
12 nm
LAUNCH DATE
2021
VS
GEFORCE
L40
CORE STATE
AD102
VRAM
48 GB
CLOCK SPEED
2490 MHz
TDP
300 W
BUS WIDTH
384 bit
ARCHITECTURE
Ada Lovelace
PROCESS
5 nm
LAUNCH DATE
2022
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
CMP 50HX
L40
Core Specs
Shading Units
3,584
18,176
+407.1%
Shaders
3,584
18,176
+407.1%
TMUs
192
568
+195.8%
ROPs
80
192
+140.0%
SM Count
56
142
+153.6%
Clocks
Base Clock
1350 MHz
735 MHz
Boost Clock
1545 MHz
2490 MHz
Memory Clock
1750 MHz
14 Gbps effective
2250 MHz
18 Gbps effective
Memory
Memory Size
10 GB
48 GB
VRAM (MB)
10,240
49,152
+380.0%
Memory Type
GDDR6
GDDR6
Memory Bus
320 bit
384 bit
Bandwidth
560.0 GB/s
864.0 GB/s
Cache
L1 Cache
64 KB (per SM)
128 KB (per SM)
L2 Cache
5 MB
96 MB
Performance
Pixel Rate
123.6 GPixel/s
478.1 GPixel/s
Texture Rate
296.6 GTexel/s
1,414.3 GTexel/s
FP32 (TFLOPS)
11.07 TFLOPS
90.52 TFLOPS
FP64 (TFLOPS)
346.1 GFLOPS (1:32)
1,414.3 GFLOPS (1:64)
FP16 (TFLOPS)
22.15 TFLOPS (2:1)
90.52 TFLOPS (1:1)
AI/RT
RT Cores
56
142
+153.6%
Tensor Cores
448
568
+26.8%
Power
TDP
250 W
300 W
TDP (W)
250
300
+20.0%
Suggested PSU
600 W
700 W
Power Connectors
2x 8-pin
1x 16-pin
Architecture
Architecture
Turing
Ada Lovelace
GPU Name
TU102
AD102
Generation
Mining GPUs
Server Ada
(Lxx)
Process Size
12 nm
5 nm
Transistors
18,600 million
76,300 million
Die Size
754 mm²
609 mm²
Foundry
TSMC
TSMC
Density
24.7M / mm²
125.3M / 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
7.5
8.9
Shader Model
6.8
6.8
Physical
Slot Width
Dual-slot
Dual-slot
Length
267 mm
10.5 inches
267 mm
10.5 inches
Height
116 mm
4.6 inches
111 mm
4.4 inches
Outputs
No outputs
4x DisplayPort 1.4a
Bus Interface
PCIe 1.0 x4
PCIe 4.0 x16
Other
Production
End-of-life
End-of-life
Predecessor
—
Server Ampere
Successor
—
Server Hopper