Undocumented Power Macintosh G3 Easter Egg Discovered After 27 Years

2025-06-24

A developer accidentally stumbled upon an undocumented easter egg hidden within the ROM of the original Power Macintosh G3. The egg is a JPEG image featuring the team who worked on the Mac models. By analyzing the SCSI Manager code in the ROM, the developer discovered the trigger: formatting the RAM disk after startup and typing 'secret ROM image' into the format dialog. This creates a JPEG file named 'The Team' on the RAM disk, revealing the team photo. This find might be one of the last easter eggs on Macs before their reported banning in 1997, adding a layer of mystery to Mac history.

Read more
Hardware

Hacking a Dummy Plug's EDID with a Raspberry Pi

2025-06-15

The author cleverly used a Raspberry Pi and I2C tools to modify the EDID information of a cheap HDMI dummy plug. By reading and writing new EDID data, the dummy plug was disguised as a 1080p capture device, solving the incompatibility problem with 4K resolution. The whole process details the steps and reminds readers to be careful to avoid damaging the equipment.

Read more
Hardware

Basilisk II Black Screen Bug: A Twelve-Year Debugging Odyssey

2025-05-17

Twelve years ago, a black screen bug in the Basilisk II emulator plagued users on newer Windows versions. The author, through meticulous debugging, traced the issue to a subtle memory allocation detail: the order of two memory allocations in Windows caused the emulator's ROM to fail to load. The solution involved modifying the memory allocation code, resolving this long-standing problem and showcasing debugging skills and deep OS understanding.

Read more
Development

Resurrecting a Lost Piece of Apple History: The Performa 550's Secret Recovery Partition

2025-03-16

While rescuing data from a failing hard drive in an old Apple Performa 550, the author uncovered a hidden recovery partition containing a fascinating piece of Apple's software history. This partition, designed to boot in case of system failure, allowed users to reinstall the OS. A three-year quest involving online appeals culminated in finding a pristine hard drive, revealing the partition's mechanics and leading to the sharing of its image. This compelling story highlights the thrill of tech archeology and software preservation.

Read more

Resurrecting 30-Year-Old Apple SCSI Hard Drives: The Rubber-Decay Data Recovery

2025-03-02

This post details the author's experience restoring 1990s Apple-branded Quantum and Conner SCSI hard drives. These drives commonly suffer from a failure mode where they spin up and immediately stop. By opening the drives, the author discovered the root cause: aging rubber bumpers causing the read/write head to stick. Two methods—manually moving the head and using Kapton tape to hold it in place—were successfully employed to recover data. The article also shares interesting details about how data is physically stored on these drives and serves as a reminder to regularly back up important data.

Read more

The Undocumented 68030 Instruction That Saved the Mac Classic II

2025-01-25

While debugging a Macintosh Classic II in MAME, the author discovered a bus error causing a 'Sad Mac' in 32-bit addressing mode. Reverse engineering revealed an out-of-bounds jump in the ROM, leading the CPU to execute an undocumented 68030 instruction. Surprisingly, this instruction corrected the value of register A1, preventing a crash and allowing the Classic II to boot successfully. The author validated this by repairing a vintage Classic II and modifying its ROM. This case highlights the power of emulators in uncovering hardware quirks and showcases how a software bug was inadvertently fixed by hardware.

Read more
Hardware

Chumby 8 Kernel Upgrade: Solving the 100% CPU Usage Mystery

2025-01-13

While upgrading the Linux kernel of his Chumby 8 device to version 6.x, the author encountered a persistent 100% CPU usage problem. Through time-reversal debugging, kernel profiling, and a deep dive into the `/proc/stat` file, the root cause was traced to a timing issue in the kernel code that reads the PXA168 hardware timer register, resulting in inaccurate idle time counting. The author fixed this problem and contributed the solution back to the main Linux kernel.

Read more
Development Hardware Debugging