241
C vs. C#: racing VLC's Gaussian blur
A real VLC video filter, ported line by line to C# and raced against the original C inside VLC 4, then optimized on both sides, with byte-identical output throughout.
242
Building a web server in aarch64 assembly to give my life (a lack of) meaning
a static http server in aarch64 assembly
243
How I reverse engineered a commercial spatial audio effect
Measuring a commercial spatial audio processor, extracting its stereo impulse responses, and rebuilding the effect as a convolution filter on Linux.
244
How to Scale a System from 0 to 10M+ Users
Scaling is a complex topic, but after working at big tech on services handling millions of requests and scaling my own startup (AlgoMaster.io) from scratch, I’ve realized that most systems evolve through a surprisingly similar set of stages as they grow.
245
Tracking down a 25% Regression on LLVM RISC-V
Similar to the previous post, this post covers my analysis of a benchmark on RISC-V targets. Unlike the previous post, I was able to land a patch to eliminate the performance gap to GCC (for this benchmark)!
246
Microcode inside the Intel 8087 floating-point chip: register exchange
In 1980, Intel introduced the 8087 floating-point chip, a co-processor that made floating-point operations up to 100 times faster. This chip...
247
Run Windows 2000 on a DEC Alpha with a new es40 fork
Published: 05-07-2026 20:33 | Author: Remy van Elst | Text only version of this article
248
Running Doom on Our Custom CPU and Going Viral
I get kidnapped by the MIG while adding DDR3 to our CPU
249
The Lost Atomic Update on Loongson CPU
杰哥的{教学，运维，编程，调板子}小笔记
250
Microcode in Intel's 8087 floating-point chip: the scale instruction
In the 1970s, floating-point arithmetic was a mess. Computer manufacturers had a dozen incompatible arithmetic standards. Moreover, floating...
251
Linux latency measurements and compositor tuning
Linux latency on the desktop and in games compared to Windows. Network streaming with Sunshine and Moonlight. Tuning the KWin compositor for lower latencies.
252
Why is the first C++ (m)allocation always 72 KB?
Why is my first C++ (m)allocation always 72 KB?? It’s the ’emergency pool’ for exceptions, here’s how I uncovered it using GDB and digging through the source code of libstdc++, glibc and Valgrind!
253
Zero-copy protobuf and ConnectRPC for Rust
As part of my work at Anthropic, I open sourced two Rust crates that fill a gap in the RPC ecosystem: buffa, a pure-Rust Protocol Buffers…
254
A few CPU hardware bugs
Catherine (Whitequark)’s recent observations on poorly-engineered firmware reminded me of a few mistakes I’ve seen in vendors’ CPUs; some unimportant and others surprisingly bad. Since I’ve never seen these widely discussed, here’s some discussion and links to supporting evidence to make them more widely known, since…
255
Windows XP 2002 for the Itanium: Unbridled rage
This is exciting news, there is a fork of Qemu that has workable Itanium Merced emulation, and it’s good enough to run Windows XP/2003! Enter Malte Kuhlmann ‘s Qemu fork of syunnPC ‘s AI Itanium infused Qemu fork .
256
Columnar Storage Is Normalization
Something I didn't understand for a while is that the process of turning row-oriented data into column-oriented data isn't a totally bespoke, foreign concept...
257
AMD Based FreeBSD Desktop Reloaded
My previous AMD Based FreeBSD Desktop is still up and running – it will now be my kids computer … so I needed a new one … but not a bigger one :) I spent way too much time finding…
258
Linux Internals: How /proc/self/mem writes to unwritable memory (2021)
Introduction An obscure quirk of the /proc/*/mem pseudofile is its “punch through” semantics. Writes performed through this file will succeed even if the destination virtual memory is marked unwritable. In fact, this behavior is intentional and actively used by projects such as the Julia JIT compiler and rr debugger…
259
Zigzag Decoding with AVX-512
I’ve been working on speeding up AVX-512 vertex decoding in meshoptimizer recently; in the process I stumbled upon two optimizations that I did not end up using but I thought they might be fun to write about! The optimizations that actually made it in require some higher level background / explanations that will have…
260
AI errno(2) values
errno(2) values for the AI age
261
MacOS Extensions Pack for the Apple Network Server v1.0
By now regular readers of this blog are quite familiar with my affection for the 1996 Apple Network Server , Apple's Unix-through-and-thro...
262
The internet discovers TLA+. Now what?
A practical introduction to TLA+, why it matters for agentic coding, and how AI could take formal verification from models to machine-checked proofs and ultimately to verified software.
263
264
Gzip decompression in 250 lines of Rust
i wanted to have a deeper grasp of how compression works, so i wrote a gzip decompressor from scratch. the result is about 250 lines of rust that can decompress gzip from a file or stdin.
265
An ARM Homelab Server, or a Minisforum MS-R1 Review
I’ve always wanted an ARM server in my homelab. But earlier, I either had to use an underpowered ARM system, or use Asahi which not only requires expensive Mac hardware but also slowed down in the past few years.
266
Engineering for Bounded Cognition
The mind that builds software is far smaller than the software it builds, and almost everything good in engineering is a way of living with that gap.
267
AVX-512: First Impressions on Performance and Programmability
A simple, whitespace theme for academics. Based on [*folio](https://github.com/bogoli/-folio) design.
268
269
Virtualisation on Apple Silicon Macs is different
Virtualising macOS, Linux and Windows on Intel Macs has been relatively straightforward, and device support left to the developer. That won’t work for Apple silicon Macs. This explains what h…
270
State of Homelab 2026
Evolution of a self-hosting hobby: from clueless to amateur.