Radeon Instinct MI300 vs CMP 50HX
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | not rated | not rated |
| Place by popularity | not in top-100 | not in top-100 |
| Architecture | Turing (2018−2022) | CDNA 3.0 (2023−2024) |
| GPU code name | TU102 | Aqua Vanjaram |
| Market segment | Workstation | Workstation |
| Release date | 24 June 2021 (4 years ago) | 4 January 2023 (3 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 | 14080 |
| Core clock speed | 1350 MHz | 1000 MHz |
| Boost clock speed | 1545 MHz | 1700 MHz |
| Number of transistors | 18,600 million | 153,000 million |
| Manufacturing process technology | 12 nm | 5 nm |
| Power consumption (TDP) | 250 Watt | 600 Watt |
| Texture fill rate | 296.6 | 1,496 |
| Floating-point processing power | 11.07 TFLOPS | 47.87 TFLOPS |
| ROPs | 80 | no data |
| TMUs | 192 | 880 |
| Tensor Cores | 448 | 880 |
| Ray Tracing Cores | 56 | no data |
| L1 Cache | 3.5 MB | 3.4 MB |
| L2 Cache | 5 MB | 16 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 | PCIe 5.0 x16 |
| Length | 267 mm | 267 mm |
| Width | 2-slot | 2-slot |
| Supplementary power connectors | 2x 8-pin | 2x 8-pin |
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 | HBM3 |
| Maximum RAM amount | 10 GB | 128 GB |
| Memory bus width | 320 Bit | 8192 Bit |
| Memory clock speed | 1750 MHz | 1600 MHz |
| Memory bandwidth | 560.0 GB/s | 3,277 GB/s |
| Resizable BAR | - | + |
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 outputs |
API and SDK support
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 12 Ultimate (12_2) | N/A |
| Shader Model | 6.8 | N/A |
| OpenGL | 4.6 | N/A |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.3 | N/A |
| CUDA | 7.5 | - |
| DLSS | + | + |
Pros & cons summary
| Recency | 24 June 2021 | 4 January 2023 |
| Maximum RAM amount | 10 GB | 128 GB |
| Chip lithography | 12 nm | 5 nm |
| Power consumption (TDP) | 250 Watt | 600 Watt |
CMP 50HX has 140% lower power consumption.
Instinct MI300, on the other hand, has an age advantage of 1 year, a 1180% higher maximum VRAM amount, and a 140% more advanced lithography process.
We couldn't decide between CMP 50HX and Radeon Instinct MI300. We've got no test results to judge.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
