Hacker News new | past | comments | ask | show | jobs | submit login

Lots of moving, mechanical parts, all handcrafted and hand-wired, machines built in small production runs (only a few thousand machines of a specific type usually) - there's just not too much room for the typical optimization approaches of "scaling production up" and "scaling things down to integrate more".

Though todays' manufacturers already try the latter, otherwise those machines would be even more expensive. You should take a peek into the belly of a machine when you get the chance somewhere - ideally into a newer machine, manufactured after 2010, and into a 90s machine. That's a stark comparison: the 90s models have thousands upon thousands of individual cables, wiring hundreds of individual lamps and solenoids and sensors. On the newer machines, manufacturers tried to cut down on the number of cables, pre-fabricating larger modules with LEDs on large PCBs, or installing sub-controllers from which solenoids are driven instead of wiring all of them individually with hundreds of large cables into the backbox.

But there are limits to this kind of optimization. After all, pinball machines are mechanical contraptions that need to integrate lots of individual mechanical parts spread over a large area. If you can't replace them with solid-state elements (which is basically what a virtual pinball table does, by replacing almost all mechanics with a screen) or concentrate them into smaller areas, you can only gain so much using modularization.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: