My Favorite Dev Talks of 39C3
The 39th Chaos Communication Congress (39C3) delivered an information-rich engineering feast as usual.
Once again, over a hundred talks, several days of intense content, as well as tens of thousands of developers and hackers. As usual for the Computer Chaos Club (CCC), there was a focus on reverse engineering, systems hacking and software concepts.
This post is about some fascinating talks from the 39C3.
Asahi Linux – Porting Linux to Apple Silicon
This multi-year engineering odyssey features the reverse engineering of undocumented Apple GPU/PMU blocks, the reconstruction of the boot chain from scratch, PCIe quirks, IOMMU traps, power management bring-up and the upstreaming of thousands of kernel patches. A masterclass in porting Linux to a proprietary SoC architecture!
Breaking Architecture Barriers: Running x86 Games & Apps on ARM
A deep dive into binary translation, JIT compilation, and ABI mismatch handling. Expect syscall translation tables, SIMD emulation strategies, performance benchmarks, and real‑world gaming workloads. Essential for anyone working on cross‑ISA compatibility.
Bluetooth Headphone Jacking – A Key to Your Phone
A practical exploitation talk showing how Bluetooth audio devices can be hijacked to access sensitive phone functions. Includes RF sniffing, packet injection, protocol fuzzing, and attack surface mapping.
Liberating Bluetooth on the ESP32
Firmware‑level hacking of the ESP32’s Bluetooth subsystem. Topics include undocumented registers, custom firmware injection, BLE stack patching, and performance measurements. A must‑watch for embedded devs.
Hacking Washing Machines
One of the most surprisingly technical talks of the Congress. Featuring MCU reverse engineering, UART probing, safety bypasses, and more. A reminder that every appliance is based on a computer — and thus may be hacked.
The Art of Text Rendering
A visual and algorithmic journey into glyph rasterization, subpixel rendering, font hinting, and GPU acceleration paths. Perfect for graphics engineers and UI/UX devs who want to understand the math behind pixels.
AI‑Generated Content in Wikipedia – A Tale of Caution
A data‑driven analysis of LLM‑generated misinformation, detection heuristics, and statistical anomaly detection.