Report an Error

AMD Aqua Vanjaram

Default
Aqua Vanjaram
AMD's Aqua Vanjaram GPU uses the CDNA 3.0 architecture and is made using a 5 nm production process at TSMC. With a die size of 1017 mm² and a transistor count of 153,000 million it is a very big chip. Aqua Vanjaram does not support DirectX. For GPU compute applications, OpenCL version 3.0 can be used. It features 19456 shading units, 1216 texture mapping units and 0 ROPs. Also included are 1216 tensor cores which help improve the speed of machine learning applications.
Further reading: CDNA Architecture

Graphics Processor

Released
Jan 4th, 2023
GPU Name
Aqua Vanjaram
Architecture
CDNA 3.0
Foundry
TSMC
Process Size
5 nm
MCD Process
6 nm
Process Type
N5 / N6 FinFET
Transistors
153,000 million
GCD Transistors
129,600 million
MCD Transistors
2,050 million x8
Density
150.4M / mm²
GCD Density
179.0M / mm²
MCD Density
55.4M / mm²
Die Size
1017 mm²
GCD Die Size
181 mm² x4
MCD Die Size
36.6 mm² x8
Package
MCM

Graphics Features

DirectX
N/A
OpenGL
N/A
OpenCL
3.0
Vulkan
N/A
Shader Model
N/A
GC
9.4.0
Shader ISA
GFX9 (gfx940)
VCN
?

Render Config

Shading Units
19456
TMUs
1216
ROPs
0
Compute Units
304
FP16 Units
65536 x4
FP64 Units
4096 x4
Tensor Cores
1216
L1 Cache
16 KB per CU
L2 Cache
16384 KB
Max. TDP
750 W

All CDNA 3.0 GPUs

  • AMD Aqua Vanjaram

AMD GPU Architecture History

Graphics cards using the AMD Aqua Vanjaram GPU

Name Chip Memory Shaders TMUs ROPs Base Clock Boost Clock Memory Clock
AMD Radeon Instinct MI300
128 GB 14080 880 0 1000 MHz 1700 MHz 1600 MHz
AMD Radeon Instinct MI300X
192 GB 19456 1216 0 1000 MHz 2100 MHz 2525 MHz

Aqua Vanjaram GPU Notes

Graphics/Compute: 9.4.0
Shader ISA: GFX9 (gfx940)
Video Core Next: ?
May 21st, 2024 13:32 EDT change timezone

Latest GPU Drivers

New Forum Posts

Popular Reviews

Controversial News Posts