NVIDIA GA10B 28SM vs Intel UHD 770 1650MHz Full Specs

Summary

1,792 Shaders 930MHz	256 Shaders 1.65GHz
Shared Memory204.8GB/sLPDDR5	Shared Memory89.6GB/s DDR5·DDR4
Peak·FP32·FP16 3.33 TFLOPS	Peak·FP32·FP16 844.8 GFLOPS
Form Factor iGPU	Form Factor iGPU
TDP Shared	TDP Shared

Highlights

Peak OPS

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

GA10B 28SM213.3 TOPSINT4 Tensor Sparse

x126.25

UHD 770 1650MHz1.69 TFLOPSFP16

x1

Show Top Score

Theoretical Performance

Clock Speed930MHz UC·Boost·OC	Clock Speed1650MHz UC·Base·Boost·OC
Peak OPS 213.3 TOPSINT4 Tensor Sparse	Peak OPS 1.69 TFLOPSFP16
Tensor FP16-16 26.66 TFLOPS FP16-16 Tensor Sparse 53.33 TFLOPS Tensor FP16-32 26.66 TFLOPS FP16-32 Tensor Sparse 53.33 TFLOPS	-
Tensor BF16 26.66 TFLOPS BF16 Tensor Sparse 53.33 TFLOPS	-
Tensor TF32 13.33 TFLOPS	Tensor TF32 -
FP32 3.33 TFLOPS	FP32 844.8 GFLOPS
-	FP64 211.2 GFLOPS
Tensor INT4 106.7 TOPS	Tensor INT4 -
Tensor INT8 53.33 TOPS	-
Pixel Rate 22.3 GPixel/s	Pixel Rate 13.2 GPixel/s
Texture Rate 13 GTexel/s	Texture Rate 26.4 GTexel/s

Core Config

Shaders 1,792 Shaders	Shaders 256 Shaders
TMUs 14 TMUs	TMUs 16 TMUs
ROPs 24 ROPs	ROPs 8 ROPs
Tensor Cores 56 T-Cores	Tensor Cores -
SMs 14 SMs	EUs 32 EUs

Clocks

Base Clock -	Base Clock 300MHz
Boost Clock 930MHz	Boost Clock 1.65GHz

Cache

L2 Cache 4.1MB shared

L2 Cache -

Memory

Shared MemoryLPDDR5	Shared Memory DDR5·DDR4
Memory Bus 256-bit	Memory Bus 128-bit
Memory Speed 6.4GT/s	Memory Speed 5.6GT/s
Memory Bandwidth 204.8GB/s	Memory Bandwidth 89.6GB/s
ECC No	ECC No

Video Output

Multi-Monitor -

Multi-Monitor 4

Encoder

Encoder Model NVENC 7

Encoder Model Quick Sync Video 8

Decoder

Decoder Model NVDEC 5

Decoder Model Quick Sync Video 8

Card

Form Factor iGPU	Form Factor iGPU
Cooling Open-Air	Cooling Passive

Chip

Manufacturer NVIDIA	Manufacturer Intel
Chip Designer NVIDIA	Chip Designer Intel
Architecture Ampere	Architecture Gen 12
Family Jetson	Family Gen 12
Branding	Branding
Codename Hercules	Codename Xe LP
Chip Variant -	Chip Variant GT1
Market Segment Embedded	Market Segment Desktop
Release Date Oct 1, 2022	Release Date Jan 4, 2022

Fabrication

Foundry Samsung	Foundry Intel
Fabrication Node 8N	Fabrication Node Intel 7
Die Size -	Die Size 257mm²
Transistor Count 17 Billion	Transistor Count 18.5 Billion
Transistor Density -	Transistor Density 72 MTr/mm²

Images

No images available

No images available

Used In

CPUs AGX Orin 32GB

CPUs Core i9-13900 Core i9-13900KS Core i9-13900T Show 10 more

Competitors

GPU NVIDIAGA10B 28SM vs GPU QualcommAdreno 840 GPU NVIDIAGA10B 28SM vs GPU Apple10-Core M5 GPU NVIDIAGA10B 28SM vs GPU Apple10-Core M5-FL Show 12 more

IntelUHD 770 1650MHz vs GPU IntelGraphics 2C 2300MHz