Daniel Lemire's blog

Splitting a long string in lines efficiently

Processors are getting wider

Debugging C++ with Visual Studio Code under macOS

Predictable memory accesses are much faster

Why do we even need SIMD instructions ?

Innovation starts with consumers, not academia

Rebels on campus

Dot product on misaligned data

Why measuring productivity is hard

Memory-level parallelism :: Apple M2 vs Apple M4

Just say no to broken JSON

Base64 for compression

From code reuse to the impact of generative AI

Discover C++26’s compile-time reflection

Life expectancy of men in Canadian provinces

Don’t argue: build

Metcalfe’s Law against Brooks’ Law

Let us bury the linear model of innovation

Fast character classification with z3

Is the job market for software developers collapsing?

Models and science

Dividing an array into fair sized chunks

Returning several values from a function in C++ (C++23 edition)

Producing useful commands on the go using C++ and AI

C++20 concepts for nicer compiler errors

Streamlined iteration: exploring keys and values in C++20

Detect control characters, quotes and backslashes efficiently using ‘SWAR’

How can really smart people appear totally incompetent?

How helpful is AI?

Faster shuffling in Go with batching

Mixing ARM NEON with SVE code for fun and profit

Unsigned comparisons using signed types

Speeding up C++ code with template lambdas

An overview of parallel programming (Go edition)

How fast can you open 1000 files?

AVX-512 gotcha: avoid compressing words to memory with AMD Zen 4 processors

Thread-safe memory copy

Programmer time and the pitfalls of wasteful work

Regular expressions can blow up!

Checking whether an ARM NEON register is zero

The ivory tower’s drift: how academia’s preference for theory over empiricism fuels scientific stagnation

JavaScript hashing speed comparison: MD5 versus SHA-256