Report an Error

NVIDIA CMP 50HX

Graphics Processor
TU102
Cores
4352
TMUs
272
ROPs
88
Memory Size
10 GB
Memory Type
GDDR6
Bus Width
320 bit
This product is not released yet.
Data on this page may change in the future.
The CMP 50HX will be a professional graphics card by NVIDIA, that is expected to launch in 2021. Built on the 12 nm process, and based on the TU102 graphics processor, in its TU102-100-A1 variant, the card supports DirectX 12 Ultimate. The TU102 graphics processor is a large chip with a die area of 754 mm² and 18,600 million transistors. Unlike the fully unlocked TITAN RTX, which uses the same GPU but has all 4608 shaders enabled, NVIDIA has disabled some shading units on the CMP 50HX to reach the product's target shader count. It features 4352 shading units, 272 texture mapping units, and 88 ROPs. Also included are 544 tensor cores which help improve the speed of machine learning applications. The card also has 68 raytracing acceleration cores. NVIDIA has paired 10 GB GDDR6 memory with the CMP 50HX, which are connected using a 320-bit memory interface. The GPU is operating at a frequency of 1350 MHz, which can be boosted up to 1545 MHz, memory is running at 1750 MHz (14 Gbps effective).
Being a dual-slot card, the NVIDIA CMP 50HX draws power from 2x 8-pin power connectors, with power draw rated at 250 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. CMP 50HX is connected to the rest of the system using a PCI-Express 3.0 x16 interface. The card's dimensions are 267 mm x 116 mm x 35 mm, and it features a dual-slot cooling solution.

Graphics Processor

GPU Name
TU102
GPU Variant
TU102-100-A1
Architecture
Turing
Foundry
TSMC
Process Size
12 nm
Transistors
18,600 million
Die Size
754 mm²

Graphics Card

Release Date
2021
Generation
Mining GPUs
Production
Unreleased
Bus Interface
PCIe 3.0 x16

Relative Performance

1%
GeForce 210
1%
GeForce 9400 GT
1%
Radeon HD 4550
2%
Radeon HD 5450
2%
Radeon HD 6450
2%
GeForce GT 520
2%
GeForce GT 220
3%
GeForce GT 430
3%
Radeon HD 5570
4%
Radeon HD 4670
4%
GeForce GT 440
4%
GeForce GT 240
4%
GeForce 9600 GT
5%
Radeon HD 5670
5%
GeForce GT 640
5%
GeForce 9800 GT
5%
Radeon HD 6670
6%
Radeon HD 4830
6%
Radeon HD 4770
6%
Radeon HD 4850
6%
GeForce GTS 250
7%
GeForce GTS 450
7%
Radeon HD 5750
8%
Radeon HD 7750
8%
Radeon HD 5770
8%
GeForce GTX 260
8%
Radeon HD 4870
8%
Radeon Vega 8
8%
GeForce GTX 550 Ti
8%
GeForce GTX 650
9%
Radeon HD 5830
9%
Radeon HD 6790
9%
Radeon HD 4890
10%
GeForce GTX 460
10%
Radeon RX Vega 11
10%
GeForce GT 1030
10%
GeForce GTX 275
10%
Radeon HD 7770 GHz Edition
10%
GeForce GTX 280
10%
GeForce GTX 465
11%
Radeon HD 6850
11%
GeForce GTX 285
12%
Radeon HD 5850
12%
GeForce GTX 650 Ti
12%
Radeon HD 7790
13%
Radeon RX 550
13%
GeForce GTX 470
13%
Radeon HD 6870
14%
Radeon HD 5870
14%
GeForce GTX 560 Ti
14%
Radeon HD 4870 X2
15%
GeForce GTX 750 Ti
15%
Radeon HD 6950
15%
GeForce GTX 295
15%
GeForce GTX 650 Ti Boost
16%
Radeon HD 7850
16%
GeForce GTX 480
17%
Radeon HD 6970
17%
Radeon R7 265
17%
Radeon RX 460
17%
Radeon R7 370
17%
GeForce GTX 660
17%
GeForce GTX 570
18%
Radeon RX 560
19%
GeForce GTX 950
19%
Radeon HD 7870 GHz Edition
19%
Radeon R9 270X
19%
GeForce GTX 660 Ti
20%
GeForce GTX 580
20%
GeForce GTX 1050
20%
Radeon HD 7950
20%
Radeon HD 5970
21%
GeForce GTX 760
22%
GeForce GTX 670
23%
GeForce GTX 960
23%
Radeon R9 380
24%
Radeon R9 285
24%
Radeon HD 7970
24%
GeForce GTX 680
25%
GeForce GTX 1050 Ti
25%
GeForce GTX 770
26%
Radeon HD 6990
27%
Radeon R9 280X
27%
GeForce GTX 590
27%
Radeon HD 7970 GHz Edition
30%
GeForce GTX 780
31%
GeForce GTX 1650
33%
Radeon RX 470
33%
Radeon R9 290
35%
Radeon RX 570
35%
Radeon R9 390
36%
Radeon R9 290X
36%
GeForce GTX 970
37%
GeForce GTX TITAN
37%
GeForce GTX 780 Ti
38%
Radeon R9 390X
38%
Radeon HD 7990
38%
Radeon RX 480
38%
GeForce GTX 690
39%
GeForce GTX 1060 6 GB
40%
Radeon RX 5500 OEM
40%
Radeon RX 580
41%
Radeon RX 5500 XT
41%
GeForce GTX 980
42%
GeForce GTX 1650 SUPER
42%
Radeon R9 FURY
44%
Radeon RX 590
46%
Radeon R9 295X2
46%
GeForce GTX 1660
47%
Radeon R9 FURY X
47%
GeForce GTX 980 Ti
49%
GeForce GTX TITAN X
52%
GeForce GTX 1660 SUPER
53%
GeForce GTX 1070
53%
GeForce GTX 1660 Ti
56%
Radeon RX Vega 56
60%
GeForce GTX 1070 Ti
61%
Radeon RX 5600 XT
62%
Radeon RX Vega 64
62%
GeForce GTX 1080
62%
GeForce RTX 2060
65%
Radeon RX 5700
70%
GeForce RTX 2060 SUPER
73%
GeForce RTX 2070
74%
GeForce RTX 3060
74%
Radeon RX 5700 XT
75%
Radeon VII
77%
TITAN X Pascal
79%
GeForce GTX 1080 Ti
80%
GeForce RTX 2070 SUPER
85%
GeForce RTX 2080
91%
GeForce RTX 2080 SUPER
92%
GeForce RTX 3060 Ti
97%
Radeon RX 6700 XT
100%
GeForce RTX 2080 Ti
100%
CMP 50HX
101%
GeForce RTX 3070
112%
Radeon RX 6800
129%
Radeon RX 6800 XT
131%
GeForce RTX 3080
137%
Radeon RX 6900 XT
145%
GeForce RTX 3090
Based on TPU review data: "Performance Summary" at 1920x1080, 4K for 2080 Ti and faster.
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

Base Clock
1350 MHz
Boost Clock
1545 MHz
Memory Clock
1750 MHz
14 Gbps effective

Memory

Memory Size
10 GB
Memory Type
GDDR6
Memory Bus
320 bit
Bandwidth
560.0 GB/s

Render Config

Shading Units
4352
TMUs
272
ROPs
88
SM Count
68
Tensor Cores
544
RT Cores
68
L1 Cache
64 KB (per SM)
L2 Cache
5 MB

Theoretical Performance

Pixel Rate
136.0 GPixel/s
Texture Rate
420.2 GTexel/s
FP16 (half) performance
26.90 TFLOPS (2:1)
FP32 (float) performance
13.45 TFLOPS
FP64 (double) performance
420.2 GFLOPS (1:32)

Board Design

Slot Width
Dual-slot
Length
267 mm
10.5 inches
Width
116 mm
4.6 inches
Height
35 mm
1.4 inches
TDP
250 W
Suggested PSU
600 W
Outputs
No outputs
Power Connectors
2x 8-pin
Board Number
PG150 SKU 100

Graphics Features

DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
1.2
Vulkan
1.2
CUDA
7.5
Shader Model
6.5