TinyComputers.io

Latest Hardware Reviews

Pine64 Board Comparison: RockPro64 vs Quartz64-B

 · 5 min read

Comprehensive performance comparison between Pine64's RockPro64 (RK3399) running FreeBSD and Quartz64-B (RK3566) running Debian Linux. Detailed benchmarks covering CPU performance, memory bandwidth, storage I/O, and network throughput reveal the RockPro64's superiority in write performance and multi-core tasks, while the Quartz64-B excels in memory bandwidth and read operations. Analysis includes ideal use cases for each board, from build servers to containerized workloads.
Recent
benchmarks Rust Compilation Performance Benchmark Report  · 7 min read Hardware Reviews, Single Board Computers Raspberry Pi Compute Module 5 Review: Performance Analysis and CM4-Compatible Ecosystem Comparison  · 15 min read Hardware Reviews AMD AI Max+ 395 System Review: A Comprehensive Analysis  · 12 min read RK3588 Orange Pi 5 Max Review  · 11 min read

Sponsors

More Articles

ballistics, machine learning

Transfer Learning for Transonic Drag Prediction: A Two-Stage Approach Using Ogive Geometry Inference

 · 10 min read

Accurate prediction of transonic drag coefficients remains one of the most challenging problems in external ballistics, with traditional fixed-percentage methods achieving error rates exceeding 100% at Mach 1.0. This paper presents a novel two-stage machine learning approach that leverages transfer learning to predict transonic drag behavior without requiring direct ogive radius measurements. By training an intermediate model on 648 bullets with known ogive geometry, we enable accurate ogive radius prediction from basic bullet parameters (weight, caliber). This predicted geometry then enhances transonic drag prediction for 272 bullets with Doppler-derived drag curves. Our approach achieves an R2 score of 0.311 with 26.7% mean absolute error, representing a 77% reduction in prediction error compared to industry-standard fixed BC degradation methods. The model has been successfully deployed in production, processing over 3,000 trajectory calculations daily with sub-20ms latency overhead.

Review of *The Well-Grounded Rubyist, Third Edition* by David A. Black and Joseph Leo III

 · 21 min read

*The Well-Grounded Rubyist, Third Edition* by David A. Black and Joseph Leo III stands out as the definitive Ruby programming text, transforming beginners into skilled practitioners through its innovative three-part structure and spiral learning methodology that revisits concepts with increasing depth. Unlike typical programming tutorials, the book teaches not just Ruby syntax but Ruby thinking, combining practical code examples with deep explanations of why the language works as it does, making it valuable for everyone from novices to experienced developers seeking mastery. The book's exceptional pedagogical approach (featuring progressive complexity, real-world applications, and comprehensive coverage from basic syntax to advanced metaprogramming) has made it the gold standard for Ruby education in universities, bootcamps, and the professional development community.
Programming & Software Development

A Critical Analysis of "Tiny C Projects" by Dan Gookin

 · 10 min read

"Tiny C Projects" by Dan Gookin offers intermediate C programmers a hands-on approach to skill development through 15 practical command-line utility projects, from simple greeting programs to AI-powered tic-tac-toe. While the book excels at teaching through incremental project development with clear explanations and immediately useful results, it falls short by avoiding modern C standards (C11/C17/C23), excluding GUI and network programming, and limiting itself to text-mode applications. Gookin's accessible writing style and "start small and grow" methodology make complex topics digestible, but the book's narrow focus and prerequisite requirements may disappoint beginners seeking an introduction or experienced developers looking for contemporary C practices. Best suited for programmers with basic C knowledge wanting practical project experience, the book delivers solid value for its target audience (7/10) but serves as limited general C instruction (5/10), making it a worthwhile but incomplete resource that benefits from supplementation with more comprehensive texts.

Topics

Retro Computing AI & Inference Hardware Projects Philosophy Economics Systems Programming

Reading Series

8+ posts Jevons Paradox & AI Economics What happens when cognitive output gets cheaper. Demand expansion, the biological ceiling, and the AI vampire. 3 parts Heidegger & Technology Enframing, standing reserve, and what happens when AI treats specifications as raw material for code generation. 8+ posts Z80 RetroShield Running a real Z80 on modern hardware. CP/M, Forth, Zork, custom PCBs, and a dual-processor design. 3 parts Sampo CPU & Compilers A custom RISC CPU from ISA design through FPGA implementation to an LLVM backend. 6 posts Tesla P40 Home Lab Four eBay GPUs, 96GB of VRAM, and every workload we can throw at them. LLMs, TTS, video generation. 3 parts The Lattice Language A programming language where mutability is a phase transition. Novel type system, bytecode VM, 100 opcodes.
View All Articles →
Written by Alex Jokela Software engineer by trade, tinkerer by nature, single-board computer hoarder by choice. More about me →