Performance x64: Cache Blocking (Matrix Blocking) @WhatsACreel
Performance x64: Cache Blocking (Matrix Blocking)  @WhatsACreel
Uploaded August 2017 | Updated November 2026, 13 minutes ago
In this video we'll start out talking about cache lines. After that we look at a technique called blocking. This is where we split a large problem into smaller sections, with the hopes that the data fits into cache.

In reality, without my screen recorder going, blocking improves performance by about 20%. This 20% is in addition to performance increase from the previous video. At the end of this vid, the product is about 12% of the original.

Support What's a Creel? on Patreon:
patreon.com/whatsacreel

Or support by purchasing Official merchandise:
teespring.com/stores/whats-a-creel-3

FaceBook:
facebook.com/whatsacreel

Music Channel:
youtube.com/channel/UC2nWRGqPL6O-Lm8mexWhISQ

Another channel with random things:
youtube.com/channel/UCK7bPumWNzWRnb3N8LCzKsA
Performance x64: Cache Blocking (Matrix Blocking)C++ Bit FieldsModern x64 Assembly 13: Control Structures (If statements, do while, while and for loops)IEEE 754: 32 Bit Single Precision FloatsObject Oriented Programming is a Dirty Rotten Low Down Trick: A Look at How C++ WorksLearn Python Programming 2: Arithmetic OperatorsAVX512 (2 of 3): Programming AVX512 in 3 Different WaysDirect3D 11 2: Swap ChainFor Loop PuzzlesIntroduction to Function Pointers in C++: What are they and how are they used?Linux x64 Assembly Tutorial 1: Getting a Value from NASM into G++AES Encryption 1: Intro and Outline
Creel |

Performance x64: Cache Blocking (Matrix Blocking)

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER