AMD Latte vs NVIDIA Tegra X1+ Full Specs

Summary

320 Shaders 550MHz	256 Shaders 1.27GHz
2GB DDR312.8GB/s placeholder	Shared Memory34.1GB/s LPDDR4X·LPDDR4
Peak·FP32·FP16 352 GFLOPS	Peak·FP32·FP16 648.7 GFLOPS
Form Factor Soldered	Form Factor iGPU
TDP 33W	TDP Shared

Theoretical Performance

Clock Speed550MHz UC·Boost·OC	Clock Speed1267MHz UC·Boost·OC
Peak OPS 352 GFLOPSFP32	Peak OPS 1.3 TFLOPSFP16
FP32 352 GFLOPS	FP32 648.7 GFLOPS
-	FP64 20.27 GFLOPS
Pixel Rate 4.4 GPixel/s	Pixel Rate 20.3 GPixel/s
Texture Rate 8.8 GTexel/s	Texture Rate 20.3 GTexel/s

Core Config

Shaders 320 Shaders	Shaders 256 Shaders
TMUs 16 TMUs	TMUs 16 TMUs
ROPs 8 ROPs	ROPs 16 ROPs
CUs 4 CUs	SMs 2 SMs

Clocks

Boost Clock 550MHz

Boost Clock 1.27GHz

Cache

L2 Cache -

L2 Cache 256KB shared

Memory

2GB DDR3 placeholder	Shared Memory LPDDR4X·LPDDR4
Memory Bus 64-bit	Memory Bus 64-bit
Memory Speed 1.6GT/s	Memory Speed 4.3GT/s
Memory Bandwidth 12.8GB/s	Memory Bandwidth 34.1GB/s
ECC No	ECC No
EDRAM 32MBSize 1024BitBus Width 70.4GB/sBandwidth	EDRAM -

Power & Thermals

TDP 33W

TDP Shared

Video Output

Multi-Monitor 1

Multi-Monitor -

Encoder

-	Encoder Model NVENC 3
Codec -	Codec AVC (H.264) HEVC (H.265)

Decoder

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

Card

Form Factor Soldered

Form Factor iGPU

Chip

Manufacturer AMD	Manufacturer NVIDIA
Chip Designer AMD	Chip Designer NVIDIA
Architecture TeraScale 2	Architecture Maxwell 2
Family Console	Family Tegra
Branding	Branding
Codename Latte	Codename Mariko
Market Segment Console	Market Segment Tablet
Release Date Nov 18, 2012	Release Date May 1, 2019

Fabrication

Foundry Renesas	Foundry TSMC
Fabrication Node 40nm	Fabrication Node 16FFC
Die Size 146mm²	Die Size 100mm²
Transistor Count 880 Million	Transistor Count 2 Billion
Transistor Density 6.03 MTr/mm²	Transistor Density 20 MTr/mm²

Images