NVIDIA GeForce RTX 4090

NVIDIA GeForce RTX 4090
$1,599.00
Brand: NVIDIA
Category: GPU Benchmarks
  • Graphics Processor: AD102
  • TMUs: 512
  • Memory Type: GDDR6X
  • Bus Width: 384 bit
  • ROPs: 176

SpecsByte Rating

Official Specs Rating: "Expert evaluation based on authentic manufacturer specifications and detailed product datasheets."

8
  • Gaming 8 / 10
  • Memory and Bandwidth 9 / 10
  • Power Consumption 7 / 10
  • Performance 9 / 10
  • Temperature and Noise 6 / 10
  • Value for money 9 / 10

NVIDIA GeForce RTX 4090 Specs

NVIDIA released the GeForce RTX 4090, a graphics card designed for enthusiasts, on September 20, 2022. In its AD102-300-A1 form, the card supports DirectX 12 Ultimate and is based on the AD102 graphics processor and built using the 5 nm technology. This guarantees that GeForce RTX 4090 will run all contemporary games. The DirectX 12 Ultimate capability also ensures that future video games will support hardware raytracing, variable-rate shading, and other features.

The NVIDIA GeForce RTX 4090 is also a triple-slot card that uses a single 16-pin power connection to draw power, with a maximum power draw rating of 450 W. Three DisplayPort 1.4a and one HDMI 2.1 display output are available. Through a PCI-Express 4.0 x16 interface, the GeForce RTX 4090 is linked to the rest of the system. It has a triple-slot cooling solution and measures 304 x 137 x 61 mm. At debut, it cost 1,599 US dollars.

Additionally, there are 512 tensor cores, which aid in accelerating machine learning applications. There are 128 raytracing acceleration cores on the GPU as well. A 384-bit memory interface connects the GeForce RTX 4090 to 24 GB of GDDR6X memory that NVIDIA has coupled with it. While the GPU operates at 2235 MHz (with the potential to be increased to 2520 MHz), the RAM operates at 1313 MHz (effectively 21 Gbps).

Specifications

BASIC GPU INFO

GPU name AD102
Manufacturer NVIDIA
Released date Sep 20th, 2022
Launch price $1,599
Foundry TSMC
Purpose gaming
GPU codename AD102-300-A1
Architecture Ada Lovelace
Generation GeForce 40
Production Available
Successor GeForce 50
Predecessor GeForce 30

BOARD DESIGN

TDP 450 W
Slot width Triple-slot
Power connectors 1x 16-pin
Length 304 mm 12 inches
Height 61 mm 2.4 inches
Width 137 mm 5.4 inches
Outputs 1x HDMI 2.1 and 3x DisplayPort 1.4a
Board number PG139 SKU 330
Weight N/A
PSU 850 W
Bus interface PCIe 4.0 x16

GRAPHICS PROCESSING UNIT

Base clock 2235 MHz
Boost clock 2520 MHz
Transistors 76,300 million
Process size 5 nm
Process type N/A
Die size 609 mm²
MCD Die density N/A
Density 125.3M / mm²

GPU MEMORY

Memory type GDDR6X
Memory size 24 GB
Memory bus 384 bit
Bandwidth 1.01 TB/s
Memory clock 1313 MHz 21 Gbps more effective

GRAPHICS FEATURES (API)

Vulkan 1.3
DirectX 12 Ultimate (12_2)
OpenGL 4.6
OpenCL 3.0
Display port N/A
Shader model 6.8
Ray tracing N/A
CUDA 8.9

CONFIGURATION

ROPs 176
Shading units 16384
TMUs 512
L1 cache 128 KB (per SM)
L2 cache 72 MB

GPU PERFORMANCE

Pixel rate 443.5 GPixel/s
Texture rate 1,290 GTexel/s
FP16 (half) 82.58 TFLOPS (1:1)
FP32 (float) 82.58 TFLOPS
FP64 (double) 1,290 GFLOPS (1:64)

FULL VIDEO REVIEW

Share your thoughts

  • Be the first to add a Review

    Please post a user review only if you have / had this product.

Disclaimer Note

We can not guarantee that the information on this page is 100% correct.

Back to top button