CMP 50HX: specs and benchmarks

Summary

NVIDIA started CMP 50HX sales 24 June 2021. This is a Turing architecture desktop card based on 12 nm manufacturing process and primarily aimed at designers. 10 GB of GDDR6 memory clocked at 1.75 GHz are supplied, and together with 320 Bit memory interface this creates a bandwidth of 560.0 GB/s.

Compatibility-wise, this is a dual-slot graphics card attached via PCIe 3.0 x16 interface. Its manufacturer default version has a length of 267 mm. Two 8-pin power connectors are required, and power consumption is at 250 Watt.

Primary details

Some basic facts about CMP 50HX: architecture, market segment, release date etc.

Place in the ranking	not rated
Place by popularity	not in top-100
Architecture	Turing (2018−2022)
GPU code name	TU102
Market segment	Workstation
Release date	24 June 2021 (3 years ago)

Detailed specifications

CMP 50HX's specs such as number of shaders, GPU base clock, manufacturing process, texturing and calculation speed. These parameters indirectly speak of CMP 50HX's performance, but for precise assessment you have to consider its benchmark and gaming test results.

Pipelines / CUDA cores	3584	of 21760 (GeForce RTX 5090)
Core clock speed	1350 MHz	of 2670 MHz (Arc B580)
Boost clock speed	1545 MHz	of 3599 MHz (Radeon RX 7990 XTX)
Number of transistors	18,600 million	of 208,000 million (B200 SXM 192 GB)
Manufacturing process technology	12 nm	of 3 nm (Arc Graphics 140V)
Power consumption (TDP)	250 Watt	of 2400 Watt (Data Center GPU Max Subsystem)
Texture fill rate	296.6	of 2,554 (Radeon Instinct MI300X)
Floating-point processing power	11.07 TFLOPS	of 109.7 (GeForce RTX 5090)
ROPs	80	of 192 (Radeon RX 7900 XTX)
TMUs	192	of 1280 (Data Center GPU Max NEXT)
Tensor Cores	448	of 1216 (Radeon Instinct MI300X)
Ray Tracing Cores	56	of 170 (GeForce RTX 5090)

Form factor & compatibility

This section provides details about the physical dimensions of CMP 50HX and its compatibility with other computer components. This information is useful when selecting a computer configuration or upgrading an existing one. For desktop graphics cards, it includes details about the interface and bus (for motherboard compatibility) and additional power connectors (for power supply compatibility).

Interface	PCIe 3.0 x16
Length	267 mm
Width	2-slot
Supplementary power connectors	2x 8-pin

VRAM capacity and type

Parameters of memory installed on CMP 50HX: its type, size, bus, clock and resulting bandwidth. Note that GPUs integrated into processors have no dedicated memory and use a shared part of system RAM instead.

Memory type	GDDR6
Maximum RAM amount	10 GB	of 294912 (Radeon Instinct MI325X)
Memory bus width	320 Bit	of 8192 Bit (Radeon Instinct MI250X)
Memory clock speed	1750 MHz	of 20000 (RTX 5000 Ada Generation Mobile)
Memory bandwidth	560.0 GB/s	of 5,171 GB/s (Radeon Instinct MI300X)

Connectivity and outputs

Types and number of video connectors present on CMP 50HX. As a rule, this section is relevant only for desktop reference graphics cards, since for notebook ones the availability of certain video outputs depends on the laptop model, while non-reference desktop models can (though not necessarily will) bear a different set of video ports.

Display Connectors

No outputs

API compatibility

APIs supported by CMP 50HX, sometimes including their particular versions.

DirectX	12 Ultimate (12_2)
Shader Model	6.6
OpenGL	4.6
OpenCL	3.0
Vulkan	1.2
CUDA	7.5

Benchmark performance

Synthetic benchmark performance of CMP 50HX. The combined score is measured on a 0-100 point scale.

---

---