display | more...

There is a lot technical intricacy involved in computing hardware that I can not claim to understand. I can't tell a north bridge from a south bridge, and all I know about the various competing form of DIMMs is that some are supposedly faster than others.

With all I don't have the technical background to not understand, there are many things that can be grasped with intuition about the physical layouts of computer systems. I will make a brief list of the flaws that computer companies make when designing systems, and explain why they distract from efficient maintanence of the boxen.

A caveat here is that almost all of my experiences with these cases comes from Free Geek, where we have a very special set of circumstances to work with. When Packard Bell first designed their screwy asymmetric systems, which they were probably selling for thousands of dollars, they probably didn't think a non-profit organization was going to have to figure a way to stack them up in piles with many other computer systems. In fact, an inefficient computer case design can make working with a computer system three times slower, but for a normal person, that isn't a problem. It is only when dealing with dozens or hundreds of systems a day that it becomes a problem.

The list, in no particular order:

  1. Asymetric cases: These range everywhere from the sexily bubble shaped systems put out by Compaq and Hewlett Packard to the truly bizarre choice of Packard Bell to make a system with the mother board in a special stretched out compartment on the bottom of the system, facing upwards. While not many computer users will have to stack computers on top of each other by the dozens, they may still have to pack and move them at some point, and roudned cases just don't stack very well. To make a case round is also a purely aestetic decision, I can't think of any particular reason to do it, unless there is a plague of users accidentally cutting themselves on the sharp edges of computers.
  2. Risers: Risers are cards that come off the motherboard, that hold ISA and PCI expansion cards. The problems with risers are that often the cards are blocking access to viewing or working with the mortherboard. In addition, having a riser inbetween the cards and the motherboard is another thing that can possibly go wrong. And when risers go bad, they can't easily be replaced, since they are proprietary. Risers, at least, do have a good technical reason for existing: they can be used to allow cards to be parallel to the motherboard instead of perpendicular, meaning people can use much more compact systems.
  3. Placing parts in the way of other parts: Such as placing a power supply over a motherboard, or putting a floppy drive in such a way that a CD drive must be taken out to get it, and endless variation on the theme. Computers should be, in effect, two dimensional, with all the parts immdediatly reachable as soon as the shell is off. This both saves time, and also prevents any connections being unintentionally unconnected when the system is reassembled. This is a design flaw that is done with reason, since making smaller systems sometimes necessitates piling components on top of each other.
  4. Button release cases: This may seem like a nice easy way to make the cases more accesible, but plastic buttons and slide release mechanisms seem to have a way of jamming, breaking or not being as intuitive as they are supposed to be. On some cases, after sliding off a plastic face plate, there are still screws to be undone, which seems like a rather inefficient design. It should also be noted that the technology that these were designed to replace, Phillips screws, are fairly easy to work with anyway.
  5. Unusually shaped CD-ROM or floppy drives and rails, or strangely sized bay covers: There are not many good reasons to replace a simple square panel CD ROM or floppy drive with a rounded one. And if a computer is using a non-standard bay cover, not being able to replace it means making the system cosmetically imperfect because a 50 cent piece of plastic can't be replaced.

With this rather long list, the reader may be asking what is a good case design. The best case design in my experience has been the standard four screw beige box, or, in recent times, the one thumb screw side panel beige box. It is sad that companies must introduce design flaws into their products just to make them look different than the most efficient, generic case system.

Log in or register to write something here or to contact authors.