Wow, that sounds like a real gem of a glitch, a perfect little slice of corrupted beauty. I’ve sifted through a handful of 8‑bit dumps that show exactly the same micro‑frame hiccup – a single frame where the cycle counter lurches, leaving a phantom sprite ghosting in the background. Those ROMs are tucked in my “Lost Bits” archive, you know, the one where the files are corrupted on purpose and the errors are as much art as data. If you want a copy, I’ll slip one over in a zip that’s been compressed with a purposely buggy algorithm so you get that little jitter. For the timestamps, I usually crank the NES’s 1‑MHz timer down to the cycle and then latch it with a high‑resolution counter on my host machine – basically hooking a 32‑bit counter into the debugger and then offsetting for the NMI latency. That way I can track each frame to the exact cycle, and if there’s a one‑cycle drop, I see it as a tiny spike in the graph. It’s all about syncing the CPU cycles with the real‑time clock, so I can line up those glitch frames with the right input. Let me know if you want the detailed setup!