I worked as a show and ride control engineer for Disney in the early 1980’s, and witnessed several spectacular (though non-lethal) accidents.
[Get a 10% discount on ARM TechCon 2012 conference passes by using promo code EDIT. Click here to learn about the show and register.]
The first took place at Epcot center. An electrician was using a three foot long open-ended aluminum wrench to tighten a connection, and accidentally dropped it across a live 480V line inside the Kodak pavilion.
50,000 square feet of interior space lit up like the sun while the wrench vaporized in the arc. Everyone froze, not wanting to be the first one to find out if electric current was going to use their body as a ground. The electrician had to go home, as his pants were a smoking mess.
A few of us close to him had to change our shorts, too.
The next incident occurred at WED Imagineering, where the R&D group had built a huge neon tube chamber with ten pairs of switchable electrodes to simulate lightning arcing in the chamber.
Everything had worked flawlessly in Glendale California – however, a few weeks after it was installed in Florida, disaster struck.
The high voltage switching control box was built out of plywood, and after a few weeks in the Florida humidity, the high voltage started arcing around the plywood box to the steel support beam it was mounted on.
Unfortunately, that same steel beam also mounted the only circuit breaker between the box and the main Florida Power transformer (oops!). Florida power was not too happy about shutting down their power grid, and somewhere between 10 and 20,000 homes went dark for an hour in Orlando while power was cycled.
In space, no one can hear you scream, but that was not the case on Space Mountain in Tokyo Disneyland. The thrill ride was designed to handle twice the number of riders per hour of prior Space Mountain rides.
Since the track was the same length, the Oriental Land Company had decided that the way to run more people thru per hour was to speed up the ride.
After days of meetings trying to get them to reconsider their folly, Disney gave up, presented them with a bill for re-engineering the ride for the higher speeds, and OLC signed a (rather large) check for NRE. New high speed rollers, higher strength truck materials, and a host of other changes were part of the redesign effort.
After the ride was built and given safety clearance, the executives from Oriental Land Company were invited to try out the new high speed ride. With the work lights on, the big green go switch was pressed. When the ride came back into the station, most of the eighteen executives had thrown up, and all but two could barely stand.
“Too fast” was about all they could say (with the lights out, it would have been even worse…). The ride was slowed down to a more “sedate” pace for the grand opening in 1983.
Last, but certainly not least, has to do with the importance of computer equipment at theme parks.
A typical show control computer in 1980’s was a Data General S250 – with 19 inch rack mount 10MB hard drive, UPS, and show interface the beast weighed over 1000 lbs.
Instead of using a forklift to remove it from the truck, the Japanese decided to use bamboo rollers and manpower.
Unfortunately, they severely underestimated the manpower needed…. $300,000 worth of computer and equipment rolled off the side of the truck, hit a retaining wall, then fell ten feet onto a concrete pad below the retaining wall. The good news was that two spare show control units had been shipped, so Tokyo Disneyland opened on time, 100% operational.
-- “Dave Riness has done everything from mining pixie dust for Disney, to working on sensitive KH-11 Keyhole satellites, as well as Mattel toys to the first 900 DPI 24 bit color image scanner for the original Macintosh desktop publishing market. Dave keeps a spotting scope handy for that next chance encounter with Sasquatch (or a wookie).”
Unfortunately, they severely underestimated the manpower needed…. $300,000 worth of computer and equipment rolled off the side of the truck, hit a retaining wall, at http://mihaela-aioanei.ro/
then fell ten feet onto a concrete pad below the retaining wall. The good news was that two spare show control units had been shipped, so Tokyo Disneyland opened on time, 100% operational.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.