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

Summary

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

UHD 770 1650MHz1.69 TFLOPSFP16

x1

GA10B 28SM213.3 TOPSINT4 Tensor Sparse

x126.25

Show Top Score

Theoretical Performance

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

Core Config

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

Clocks

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

Cache

L2 Cache -

L2 Cache 4.1MB shared

Memory

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

Video Output

Multi-Monitor 4

Multi-Monitor -

Encoder

Encoder Model Quick Sync Video 8

Encoder Model NVENC 7

Decoder

Decoder Model Quick Sync Video 8

Decoder Model NVDEC 5

Card

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

Chip

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

Fabrication

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

Images

No images available

No images available

Used In

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

CPUs AGX Orin 32GB

Competitors

IntelUHD 770 1650MHz vs GPU IntelGraphics 2C 2300MHz IntelUHD 770 1650MHz vs GPU AMDRadeon 840M 2900MHz

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