GPUs

ASUS GeForce RTX 3090 Turbo Full Specs

10,496 Shaders
1.7GHz
24GB GDDR6X936.2GB/s
··
35.58 TFLOPS
Form Factor
PCIe Card
TDP
350W
Power Connectors
2x 8-Pin

Clock Speed
···
Peak OPS
1.14 POPSINT4 Tensor Sparse
Tensor FP16-16
142.3 TFLOPS
FP16-16 Tensor Sparse
284.7 TFLOPS
Tensor FP16-32
71.16 TFLOPS
FP16-32 Tensor Sparse
142.3 TFLOPS
BF16
35.58 TFLOPS
Tensor BF16
71.16 TFLOPS
BF16 Tensor Sparse
142.3 TFLOPS
Tensor TF32
35.58 TFLOPS
FP32
35.58 TFLOPS
FP64
556 GFLOPS
Tensor INT4
569.3 TOPS
Tensor INT8
284.7 TOPS
Ray
53.66 TOPS
Pixel Rate
189.8 GPixel/s
Texture Rate
556 GTexel/s

Shaders
10,496 Shaders
TMUs
328 TMUs
ROPs
112 ROPs
Tensor Cores
328 T-Cores
RT Cores
82 RT-Cores
SMs
82 SMs

Base Clock
1.4GHz
Boost Clock
1.7GHz

L2 Cache
6MB shared

24GB GDDR6X
Memory Bus
384-bit
Memory Speed
19.5GT/s
Memory Bandwidth
936.2GB/s
ECC
No

TDP
350W
Max Temp
93°C Max

Multi-Monitor
4
HDCP
HDCP 2.3

3x DisplayPort 1.4
1x HDMI 2.1

Encoder Model
NVENC 7
Codec
AVC (H.264)
HEVC (H.265)

Decoder Model
NVDEC 5
Codec
MPEG-1
MPEG-2
MPEG-4
VC-1
VP8
VP9
AVC (H.264)
HEVC (H.265)
AV1

Form Factor
PCIe Card
PCIe
2.7-Slots
Height
138 mm (5.43")
Width
313 mm (12.32")
Depth
55 mm (2.17")
Cooling
Blower
1x Fan
Power Connectors
2x 8-Pin

Manufacturer
Chip Designer
Architecture
Family
Branding
GeForce RTX 2018 branding
Codename
NV172
Chip Variant
GA102-200-KD-A1
Market Segment
Desktop
Release Date
Sep 24, 2020

Foundry
Samsung
Fabrication Node
8N
Die Size
628mm²
Transistor Count
28.3 Billion
Transistor Density
45.04 MTr/mm²