Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!bloom-beacon!mit-eddie!uw-beaver!tektronix!tekig5!tekig4!brianr From: brianr@tekig4.TEK.COM (Brian Rhodefer) Newsgroups: comp.sys.amiga Subject: Re: Joysticks Keywords: rugged Message-ID: <2997@tekig4.TEK.COM> Date: 5 Jul 88 21:31:44 GMT References: <890@esunix.UUCP> <9808@g.ms.uky.edu> <339@dcdwest.UUCP> Reply-To: brianr@tekig4.UUCP (Brian Rhodefer) Organization: Tektronix, Inc., Beaverton, OR. Lines: 65 It amazes me how badly the human inputs for computer games are implemented. The remarks of one recent poster notwithstanding, this is true for commercial arcade games as much as (or more than) for home games. The better the game, at least for the skill&action variety, the more exuberantly its operator(s) slam its controls around; yet, the controls seem always to involve the human's frenzied muscles pushing two bits of metal together. "Pushing"? Say, rather, "*Mashing*". Go to any arcade, and you'll see at least one or two machines down because their controls have been beaten into junk. Though the Wico joysticks seem more rugged in construction, if you open one, you'll see that its handle is essentially one end of a lever whose other end presses some leaf-spring contacts together. The only saving grace is that after the switches close, BOTH leaves flex, and that the massive plastic chassis limits control motion before the switch leaves are permanently damaged. Oveall, I wouldn't expect a Wico controller to be any less reliable than, say, an open-frame telephone relay. A company I worked for as a summer student sold a `joyswitch' used as the controller for the cargo-loading system of 747s, that I consider to be the ultimate mechanical joystick design. With 4 independent axes of motion (right/left, up/down, in/out, and clockwise/counterclockwise), it's overkill for a game, but its implementation of the basic up/down, right/left function was brilliant: the force that actuated the switching element was decoupled from the force that moved the control handle. Here's a cutaway view drawing showing one axis of this control (The sliding yoke is actually cross-shaped). The key point is that the switching is done by cam action where the direction of travel of the microswitch feeler is rotated 90 degrees away from the direction of human-applied force. The travel of the microswitch is rigidly defined by the slope & size of the sliding yoke cam surface. It may have been a bit bulky, but I've never seen its equal (passive switches only here; no fair using optosensors) for ruggedness. ______ / \ \ knob / | | ~ ~ ~ ~ ------------ _______ | | ______ ----------- | | |.......| | | |......| | | | Micro | |.......| | | |......| | Micro | | Switch | |.......| | | |......| | Switch | | | |_______| | | |______| | | ---------#-- ____________ | | ___________ --#-------- U / Sliding | | | | \ U ------/ Yoke | | | | \-------- WMWM| <-> | | | | |WMWM <- springs -------------------- | | --------------------- --------- | | -------- |.......| | | |......| |.......| \ / |......| |.......| \ / |......| |.......| \/ |...... ....... \ ** /.... . ** Ball joint ..... --------- ....... If this design isn't patented, it sure ought to be used in commercial games. Divulgingly, Brian Rhodefer ...!tektronix!tekig4!brianr