Home / GPU Comparison / AMD Radeon Instinct MI308X or NVIDIA A800 PCIe 80 GB: What's better?

AMD Radeon Instinct MI308X

vs

NVIDIA A800 PCIe 80 GB

GPU Comparison Result

Below are the results of a comparison of AMD Radeon Instinct MI308X and NVIDIA A800 PCIe 80 GB video cards based on key performance characteristics, as well as power consumption and much more.

Advantages

AMD Radeon Instinct MI308X

Higher Boost Clock: 2100 MHz (2100 MHz vs 1410MHz)
Larger Memory Size: 192GB (192GB vs 80GB)
More Shading Units: 19456 (19456 vs 6912)
Newer Launch Date: December 2023 (December 2023 vs November 2022)

NVIDIA A800 PCIe 80 GB

Higher Bandwidth: 2039 GB/s (10.3TB/s vs 2039 GB/s)

Basic

AMD

Label Name

NVIDIA

December 2023

Launch Date

November 2022

Desktop

Platform

Professional

Radeon Instinct MI308X

Model Name

A800 PCIe 80 GB

Radeon Instinct

Generation

Ampere

1000 MHz

Base Clock

1065MHz

2100 MHz

Boost Clock

1410MHz

PCIe 5.0 x16

Bus Interface

PCIe 4.0 x16

153 billion

Transistors

54,200 million

304

Compute Units

1216

Tensor Cores

Tensor Cores are specialized processing units designed specifically for deep learning, providing higher training and inference performance compared to FP32 training. They enable rapid computations in areas such as computer vision, natural language processing, speech recognition, text-to-speech conversion, and personalized recommendations. The two most notable applications of Tensor Cores are DLSS (Deep Learning Super Sampling) and AI Denoiser for noise reduction.

432

1216

TMUs

Texture Mapping Units (TMUs) serve as components of the GPU, which are capable of rotating, scaling, and distorting binary images, and then placing them as textures onto any plane of a given 3D model. This process is called texture mapping.

432

TSMC

Foundry

TSMC

5 nm

Process Size

7 nm

CDNA 3.0

Architecture

Ampere

Memory Specifications

192GB

Memory Size

80GB

HBM3

Memory Type

HBM2e

8192bit

Memory Bus

The memory bus width refers to the number of bits of data that the video memory can transfer within a single clock cycle. The larger the bus width, the greater the amount of data that can be transmitted instantaneously, making it one of the crucial parameters of video memory. The memory bandwidth is calculated as: Memory Bandwidth = Memory Frequency x Memory Bus Width / 8. Therefore, when the memory frequencies are similar, the memory bus width will determine the size of the memory bandwidth.

5120bit

2525 MHz

Memory Clock

1593MHz

10.3TB/s

Bandwidth

Memory bandwidth refers to the data transfer rate between the graphics chip and the video memory. It is measured in bytes per second, and the formula to calculate it is: memory bandwidth = working frequency × memory bus width / 8 bits.

2039 GB/s

Display and Media

No outputs

Outputs

No outputs

Theoretical Performance

Pixel Rate

Pixel fill rate refers to the number of pixels a graphics processing unit (GPU) can render per second, measured in MPixels/s (million pixels per second) or GPixels/s (billion pixels per second). It is the most commonly used metric to evaluate the pixel processing performance of a graphics card.

225.6 GPixel/s

2554 GTexel/s

Texture Rate

Texture fill rate refers to the number of texture map elements (texels) that a GPU can map to pixels in a single second.

609.1 GTexel/s

653.7 TFLOPS

FP16 (half)

An important metric for measuring GPU performance is floating-point computing capability. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable. Single-precision floating-point numbers (32-bit) are used for common multimedia and graphics processing tasks, while double-precision floating-point numbers (64-bit) are required for scientific computing that demands a wide numeric range and high accuracy.

77.97 TFLOPS

81.72 TFLOPS

FP64 (double)

An important metric for measuring GPU performance is floating-point computing capability. Double-precision floating-point numbers (64-bit) are required for scientific computing that demands a wide numeric range and high accuracy, while single-precision floating-point numbers (32-bit) are used for common multimedia and graphics processing tasks. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable.

9.746 TFLOPS

83.354 TFLOPS

FP32 (float)

An important metric for measuring GPU performance is floating-point computing capability. Single-precision floating-point numbers (32-bit) are used for common multimedia and graphics processing tasks, while double-precision floating-point numbers (64-bit) are required for scientific computing that demands a wide numeric range and high accuracy. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable.

19.88 TFLOPS

Miscellaneous

SM Count

Multiple Streaming Processors (SPs), along with other resources, form a Streaming Multiprocessor (SM), which is also referred to as a GPU's major core. These additional resources include components such as warp schedulers, registers, and shared memory. The SM can be considered the heart of the GPU, similar to a CPU core, with registers and shared memory being scarce resources within the SM.

108

19456

Shading Units

The most fundamental processing unit is the Streaming Processor (SP), where specific instructions and tasks are executed. GPUs perform parallel computing, which means multiple SPs work simultaneously to process tasks.

6912

16 KB (per CU)

L1 Cache

192 KB (per SM)

16 MB

L2 Cache

80MB

750W

TDP

250W

3.0

OpenCL Version

3.0

CUDA

8.0

None

Power Connectors

8-pin EPS

ROPs

The Raster Operations Pipeline (ROPs) is primarily responsible for handling lighting and reflection calculations in games, as well as managing effects like anti-aliasing (AA), high resolution, smoke, and fire. The more demanding the anti-aliasing and lighting effects in a game, the higher the performance requirements for the ROPs; otherwise, it may result in a sharp drop in frame rate.

160

1150 W

Suggested PSU

600W