AMD Radeon Pro Duo vs NVIDIA GeForce RTX 3060 Ti GDDR6X Full Specs

Summary

8,192 Shaders 1GHz	4,864 Shaders 1.67GHz
8GB HBM1.02TB/s	8GB GDDR6X608.3GB/s
Peak·FP32·FP16 16.38 TFLOPS	Peak·FP32·FP16 16.2 TFLOPS
Form Factor PCIe Card	Form Factor PCIe Card
TDP 350W	TDP 200W
Power Connectors - 3x 8-Pin	Power Connectors 1x 12-Pin -

Highlights

Peak OPS

Peak OPS FP32 FP16 Tensor FP16-16 FP16-16 Tensor Sparse Tensor FP16-32 FP16-32 Tensor Sparse FP64 BF16 Tensor BF16 BF16 Tensor Sparse Tensor TF32 Tensor INT8 Tensor INT4 Memory Bandwidth TDP

Radeon Pro Duo16.38 TFLOPSFP32 (2x 8.19 TFLOPS)

x1

GeForce RTX 3060 Ti GDDR6X518.3 TOPSINT4 Tensor Sparse

x31.64

Show Top Score

Theoretical Performance

Clock Speed1000MHz UC·Boost·OC	Clock Speed1665MHz UC·Base·Boost·OC
Peak OPS 16.38 TFLOPSFP32	Peak OPS 518.3 TOPSINT4 Tensor Sparse
-	Tensor FP16-16 64.79 TFLOPS FP16-16 Tensor Sparse 129.6 TFLOPS Tensor FP16-32 32.39 TFLOPS FP16-32 Tensor Sparse 64.79 TFLOPS
-	BF16 16.2 TFLOPS Tensor BF16 32.39 TFLOPS BF16 Tensor Sparse 64.79 TFLOPS
Tensor TF32 -	Tensor TF32 16.2 TFLOPS
FP32 16.38 TFLOPS	FP32 16.2 TFLOPS
FP64 1.02 TFLOPS	FP64 253.1 GFLOPS
Tensor INT4 -	Tensor INT4 259.2 TOPS
-	Tensor INT8 129.6 TOPS
Ray -	Ray 24.43 TOPS
Pixel Rate 64 GPixel/s	Pixel Rate 133.2 GPixel/s
Texture Rate 256 GTexel/s	Texture Rate 253.1 GTexel/s

Core Config

Shaders 8,192 Shaders	Shaders 4,864 Shaders
TMUs 512 TMUs	TMUs 152 TMUs
ROPs 128 ROPs	ROPs 80 ROPs
Tensor Cores -	Tensor Cores 152 T-Cores
RT Cores -	RT Cores 38 RT-Cores
CUs 128 CUs	SMs 38 SMs

Clocks

Base Clock -	Base Clock 1.41GHz
Boost Clock 1GHz	Boost Clock 1.67GHz

Cache

L2 Cache 2MB shared

L2 Cache 4.1MB shared

Memory

8GB HBM	8GB GDDR6X
Memory Bus 4096-bit	Memory Bus 256-bit
Memory Speed 1GT/s	Memory Speed 19GT/s
Memory Bandwidth 1.02TB/s	Memory Bandwidth 608.3GB/s
ECC No	ECC No

Power & Thermals

TDP 350W

TDP 200W

Video Output

Multi-Monitor 4	Multi-Monitor 3
HDCP HDCP 1.4	HDCP HDCP 2.3

Ports

-	3x DisplayPort 1.4
3x DisplayPort 1.2	-
-	1x HDMI 2.1
1x HDMI 1.4	-

Encoder

Encoder Model VCE 3.0

Encoder Model NVENC 7

Decoder

Decoder Model UVD 6.0

Decoder Model NVDEC 5

Card

Form Factor PCIe Card	Form Factor PCIe Card
PCIe 2-Slots	PCIe -
Height 111 mm (4.37") Width 277 mm (10.91") Depth 40 mm (1.57")	-
Cooling Liquid 120mm Radiator 1x Fan	Cooling Passive -
Power Connectors 3x 8-Pin	Power Connectors 1x 12-Pin

Chip

Manufacturer AMD	Manufacturer NVIDIA
Chip Designer AMD	Chip Designer NVIDIA
Architecture GCN 3	Architecture Ampere
Family Radeon R 300	Family GeForce 30
Branding	Branding
Codename Fiji	Codename NV174
Chip Variant Fiji Pro	Chip Variant GA104-150-A1
Market Segment Desktop	Market Segment Desktop
Release Date Apr 26, 2016	Release Date Oct 27, 2022

Fabrication

Foundry TSMC	Foundry Samsung
Fabrication Node 28nm	Fabrication Node 8N
Die Size 596mm²	Die Size 393mm²
Transistor Count 8.9 Billion	Transistor Count 17.4 Billion
Transistor Density 14.93 MTr/mm²	Transistor Density 44.33 MTr/mm²

Images

No images available

No images available

Competitors

AMDRadeon Pro Duo vs GPU AMDRadeon RX 9070 XT AMDRadeon Pro Duo vs GPU NVIDIAGeForce RTX 5070 AMDRadeon Pro Duo vs GPU AMDRadeon RX 7800 XT Show 12 more

NVIDIAGeForce RTX 3060 Ti GDDR6X vs GPU AMDRadeon RX 7600 NVIDIAGeForce RTX 3060 Ti GDDR6X vs GPU AMDRadeon RX 7700 NVIDIAGeForce RTX 3060 Ti GDDR6X vs GPU AMDRadeon RX 7700 XT Show 12 more