Radeon Instinct MI300 vs Tesla S870
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 | Tesla (2006−2010) | CDNA 3.0 (2023−2024) |
| GPU code name | G80 | Aqua Vanjaram |
| Market segment | Workstation | Workstation |
| Release date | 2 May 2007 (18 years ago) | 4 January 2023 (3 years ago) |
| Launch price (MSRP) | $11,999 | no data |
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 | 128 ×4 | 14080 |
| Core clock speed | 600 MHz | 1000 MHz |
| Boost clock speed | no data | 1700 MHz |
| Number of transistors | 681 million | 153,000 million |
| Manufacturing process technology | 90 nm | 5 nm |
| Power consumption (TDP) | 800 Watt | 600 Watt |
| Texture fill rate | 38.40 ×4 | 1,496 |
| Floating-point processing power | 0.3456 TFLOPS ×4 | 47.87 TFLOPS |
| ROPs | 24 ×4 | no data |
| TMUs | 32 ×4 | 880 |
| Tensor Cores | no data | 880 |
| L1 Cache | no data | 3.4 MB |
| L2 Cache | 96 KB | 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 x16 | PCIe 5.0 x16 |
| Length | no data | 267 mm |
| Width | 2-slot | 2-slot |
| Supplementary power connectors | no data | 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 | GDDR3 | HBM3 |
| Maximum RAM amount | 1536 MB ×4 | 128 GB |
| Memory bus width | 384 Bit ×4 | 8192 Bit |
| Memory clock speed | 800 MHz | 1600 MHz |
| Memory bandwidth | 76.8 GB/s ×4 | 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 | 11.1 (10_0) | N/A |
| Shader Model | 4.0 | N/A |
| OpenGL | 3.3 | N/A |
| OpenCL | 1.1 | 3.0 |
| Vulkan | N/A | N/A |
| CUDA | + | - |
| DLSS | - | + |
Pros & cons summary
| Recency | 2 May 2007 | 4 January 2023 |
| Maximum RAM amount | 1536 MB | 128 GB |
| Chip lithography | 90 nm | 5 nm |
| Power consumption (TDP) | 800 Watt | 600 Watt |
Instinct MI300 has an age advantage of 15 years, a 8433% higher maximum VRAM amount, a 1700% more advanced lithography process, and 33% lower power consumption.
We couldn't decide between Tesla S870 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.
