Report an Error

AMD Aldebaran

Default
Aldebaran
Block Diagram
Block Diagram
Memory Diagram
Memory Diagram
AMD's Aldebaran GPU uses the CDNA 2.0 architecture and is made using a 6 nm production process at TSMC. With a die size of 724 mm² and a transistor count of 58,200 million it is a very big chip. Aldebaran does not support DirectX. For GPU compute applications, OpenCL version 3.0 can be used. It features 7040 x2 shading units, 440 x2 texture mapping units and 0 ROPs.
Further reading: CDNA Architecture

Graphics Processor

Released
Nov 8th, 2021
GPU Name
Aldebaran
Architecture
CDNA 2.0
Foundry
TSMC
Process Size
6 nm
Transistors
58,200 million
Density
80.4M / mm²
Die Size
724 mm²
Memory Size
64 GB x2
Memory Type
HBM2e
Memory Bus
4096 Bit x2
Bandwidth
1638 GB/s x2

Graphics Features

DirectX
N/A
OpenGL
N/A
OpenCL
3.0
Vulkan
N/A
Shader Model
N/A
Compute
9.4.2
Shader ISA
GFX9 (gfx90a)
VCN
2.6.0
SDMA
4.4.0
MP0
13.0.2

Render Config

Shading Units
7040 x2
TMUs
440 x2
ROPs
0
Compute Units
110 x2
FP16 Units
65536 x2
FP64 Units
4096 x2
L1 Cache
16 KB per CU
L2 Cache
8192 KB x2
L2 Cache
16384 KB
Max. TDP
500 W

All CDNA 2.0 GPUs

  • AMD Aldebaran

AMD GPU Architecture History

Graphics cards using the AMD Aldebaran GPU

Name Chip Memory Shaders TMUs ROPs Base Clock Boost Clock Memory Clock
Aldebaran XT 128 GB 14080 880 0 1000 MHz 1700 MHz 1600 MHz
Aldebaran 128 GB 13312 832 0 1000 MHz 1700 MHz 1600 MHz
Aldebaran 64 GB 6656 416 0 1000 MHz 1700 MHz 1600 MHz
Aldebaran 64 GB 6656 416 0 1000 MHz 1700 MHz 1600 MHz

Aldebaran GPU Notes

IP Versions
Graphics/Compute: 9.4.2
Shader ISA: GFX9 (gfx90a)
Video Core Next: 2.6.0
Data Fabric: 3.6.2
Hortonworks Data Platform: 4.4.0
System DMA: 4.4.0
XGMI: 6.1.0
Unified Memory Controller: 6.7.0
NorthBridge I/O: 7.4.4
Thermal Sensors: 13.0.2
Platform Security Processor: 13.0.2
System Management Unit: 13.0.2
May 21st, 2024 22:12 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts