GeekRadar / The comparison desk
COMPARE.
GPU
ASUS

ASUS ROG MATRIX RTX 4090 PLATINUM

Graphics processor

Video memory
24 GB*
Memory bus
384 bit*
Rated memory bandwidth
1,010 GB/s*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture
Ada Lovelace*
Process node
5 nm*
GPU cores
16,384*
Graphics interface
PCIe 4.0 x16*
Reported chip die area
609 mm²*
Transistors
76,300 million*
GPU L2 cache
72 MB*
Graphics processor
AD102*
Texture mapping units
512*
Raster output units
176*
Streaming multiprocessors
128*
Tensor cores
512*
Ray tracing cores
128*
Chip foundry
TSMC*
Card or module format
Quad-slot*
Declared GPU TDP
450 W*
Suggested system PSU
850 W*
External power connectors
1x 16-pin*
DirectX / feature level
12 Ultimate (12_2)*
OpenGL
4.6*
OpenCL
3.0*
Vulkan
1.4*
CUDA compute capability
8.9*
Shader model
6.9*

Memory & storage

Memory options
24*
Video memory
24 GB*
Video memory type
GDDR6X*
Memory bus
384 bit*
Rated memory bandwidth
1,010 GB/s*

Power specifications

Thermal design power
450 W*

System configuration

Detailed specifications

Specifications 1 · 30 fields
proc transistors
76,300 million*
proc die size
609* mm²
cores
16,384*
tensor cores
512*
rt cores
128*
sm count
128*
base clock
2,235*
boost clock
2,670*
memory clock
1,313*
ram
24*
name
ASUS ROG MATRIX RTX 4090 PLATINUM*
ram type
GDDR6X*
bus width
384*
bus interface
PCIe 4.0 x16*
ram bandwidth
1.01 TB/s*
fp16 performance
87.49*
fp32 performance
87.49*
fp64 performance
1.367*
tdp
450* W
suggested psu
850* W
slot width
Quad-slot*
power connectors
1x 16-pin*
api cuda
8.9*
is retail board
Yes*
gpi value
33.1*
manufacturer
ASUS*
arch name
Ada Lovelace*
card release date
2022-09-20*
proc foundry
TSMC*
proc process size
5* nm
Specifications 2 · 39 fields
CUDA compute capability
8.9*
DirectX / feature level
12 Ultimate (12_2)*
OpenCL
3.0*
OpenGL
4.6*
Shader model
6.9*
Vulkan
1.4*
Architecture
Ada Lovelace*
Graphics interface
PCIe 4.0 x16*
Memory bus
384* bit
Reported shader processors
16,384*
Reported theoretical FP16 peak
87.49* TFLOPS
Scope and conditions
Arithmetic density
Not stated*
Reported theoretical FP32 peak
87.49* TFLOPS
Scope and conditions
Arithmetic density
Not stated*
Reported theoretical FP64 peak
1.367* TFLOPS
Scope and conditions
Arithmetic density
Not stated*
Declared L1 cache configuration
128 KB (per SM)*
Declared GPU L2 cache
72 MB*
Reported display outputs
Active
2x HDMI 2.1, 3x DisplayPort 1.4a*
Struck-out declarations
  • 1x HDMI 2.1, 3x DisplayPort 1.4a*
Theoretical pixel fill rate
469.9 GPixel/s*
External power connectors
1x 16-pin*
Reported chip die area
609 mm²*
Chip foundry
TSMC*
Process node declaration
5 nm*
Transistors
76,300 million*
Graphics processor
AD102*
Reported video memory
24* GB
Rated memory bandwidth
1.01 TB/s*
Video memory type
GDDR6X*
Raster output units
176*
Ray tracing cores
128*
Reported shading units
16,384*
Card or module format
Quad-slot*
Streaming multiprocessors
128*
Suggested system PSU
850 W*
Declared GPU TDP
450 W*
Tensor cores
512*
Theoretical texture fill rate
1,367.0 GTexel/s*
Texture mapping units
512*
GPU cores
16,384*
Video memory
24* GB
Video memory type
GDDR6X*

* Provisional data