Radeon Instinct MI60 vs Tesla T4
Aggregate performance score
We've compared Tesla T4 and Radeon Instinct MI60, covering specs and all relevant benchmarks.
MI60 outperforms T4 by a moderate 18% based on our aggregate benchmark results.
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 264 | 229 |
| Place by popularity | not in top-100 | not in top-100 |
| Power efficiency | 26.05 | 7.14 |
| Architecture | Turing (2018−2022) | GCN 5.1 (2018−2022) |
| GPU code name | TU104 | Vega 20 |
| Market segment | Workstation | Workstation |
| Release date | 13 September 2018 (7 years ago) | 18 November 2018 (7 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. Real power consumption of some graphics cards can well exceed their nominal TDP, especially if overclocked.
| Pipelines / CUDA cores | 2560 | 4096 |
| Core clock speed | 585 MHz | 1200 MHz |
| Boost clock speed | 1590 MHz | 1800 MHz |
| Number of transistors | 13,600 million | 13,230 million |
| Manufacturing process technology | 12 nm | 7 nm |
| Power consumption (TDP) | 70 Watt | 300 Watt |
| Texture fill rate | 254.4 | 460.8 |
| Floating-point processing power | 8.141 TFLOPS | 14.75 TFLOPS |
| ROPs | 64 | 64 |
| TMUs | 160 | 256 |
| Tensor Cores | 320 | no data |
| Ray Tracing Cores | 40 | no data |
| L1 Cache | 2.5 MB | 1 MB |
| L2 Cache | 4 MB | 4 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 3.0 x16 | PCIe 4.0 x16 |
| Length | 168 mm | 267 mm |
| Width | 1-slot | 2-slot |
| Supplementary power connectors | None | 1x 6-pin + 1x 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 | HBM2 |
| Maximum RAM amount | 16 GB | 32 GB |
| Memory bus width | 256 Bit | 4096 Bit |
| Memory clock speed | 1250 MHz | 1000 MHz |
| Memory bandwidth | 320.0 GB/s | 1,024 GB/s |
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_1) | 12 (12_1) |
| Shader Model | 6.5 | 6.4 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 1.2 | 2.0 |
| Vulkan | 1.2.131 | 1.2.131 |
| CUDA | 7.5 | - |
| DLSS | + | - |
Synthetic benchmarks
Non-gaming benchmark results comparison. The combined score is measured on a 0-100 point scale.
Combined synthetic benchmark score
This is our combined benchmark score.
Passmark
This is the most ubiquitous GPU benchmark. It gives the graphics card a thorough evaluation under various types of load, providing four separate benchmarks for Direct3D versions 9, 10, 11 and 12 (the last being done in 4K resolution if possible), and few more tests engaging DirectCompute capabilities.
GeekBench 5 OpenCL
Geekbench 5 is a widespread graphics card benchmark combined from 11 different test scenarios. All these scenarios rely on direct usage of GPU's processing power, no 3D rendering is involved. This variation uses OpenCL API by Khronos Group.
Gaming performance
Let's see how good the compared graphics cards are for gaming. Particular gaming benchmark results are measured in FPS.
Pros & cons summary
| Performance score | 23.68 | 27.83 |
| Recency | 13 September 2018 | 18 November 2018 |
| Maximum RAM amount | 16 GB | 32 GB |
| Chip lithography | 12 nm | 7 nm |
| Power consumption (TDP) | 70 Watt | 300 Watt |
Tesla T4 has 329% lower power consumption.
Instinct MI60, on the other hand, has a 18% higher aggregate performance score, an age advantage of 2 months, a 100% higher maximum VRAM amount, and a 71% more advanced lithography process.
The Radeon Instinct MI60 is our recommended choice as it beats the Tesla T4 in performance tests.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
