Radeon Steam Deck 8CU vs CMP 50HX

Primary details

GPU architecture, market segment, value for money and other general parameters compared.

Place in the ranking	not rated	485
Place by popularity	not in top-100	not in top-100
Architecture	Turing (2018−2022)	RDNA 2 (2022−2023)
GPU code name	TU102	RDNA 2 Sephiroth
Market segment	Workstation	Laptop
Release date	24 June 2021 (4 years ago)	9 November 2023 (2 years ago)

Detailed specifications

General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Note that power consumption of some graphics cards can well exceed their nominal TDP, especially when overclocked.

Pipelines / CUDA cores	3584	512
Core clock speed	1350 MHz	no data
Boost clock speed	1545 MHz	1900 MHz
Number of transistors	18,600 million	no data
Manufacturing process technology	12 nm	6 nm
Power consumption (TDP)	250 Watt	no data
Texture fill rate	296.6	no data
Floating-point processing power	11.07 TFLOPS	no data
ROPs	80	no data
TMUs	192	no data
Tensor Cores	448	no data
Ray Tracing Cores	56	no data
L1 Cache	3.5 MB	512 KB
L2 Cache	5 MB	1 MB

Form factor & compatibility

Information on compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).

Interface	PCIe 1.0 x4	no data
Length	267 mm	no data
Width	2-slot	no data
Supplementary power connectors	2x 8-pin	no data

VRAM capacity and type

Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. Integrated GPUs have no dedicated video RAM and use a shared part of system RAM.

Memory type	GDDR6	no data
Maximum RAM amount	10 GB	no data
Memory bus width	320 Bit	no data
Memory clock speed	1750 MHz	no data
Memory bandwidth	560.0 GB/s	no data
Shared memory	-	+

Connectivity and outputs

This section shows the types and number of video connectors on each GPU. The data applies specifically to desktop reference models (for example, NVIDIA’s Founders Edition). OEM partners often modify both the number and types of ports. On notebook GPUs, video‐output options are determined by the laptop’s design rather than the graphics chip itself.

Display Connectors

No outputs

no data

API and SDK support

List of supported 3D and general-purpose computing APIs, including their specific versions.

DirectX	12 Ultimate (12_2)	12_1
Shader Model	6.8	no data
OpenGL	4.6	no data
OpenCL	3.0	no data
Vulkan	1.3	-
CUDA	7.5	-
DLSS	+	-

Pros & cons summary

---

Recency	24 June 2021	9 November 2023
Chip lithography	12 nm	6 nm