Awl --
OK, so ahm still on my CAD/laptop jihad....
But here, I think, is the nitty-gritty:
How does one effectively *assess* the suitability of sundry CAD programs for a given use or context ? What standard benchmarks can one apply?
These are obvious for cars, stereos, proly even pyooters, but mebbe not AS obvious for pyooters. And even for something like cars, context itself plays a big role: 1/4 mile times for an off-road vehicle would be much less relevant than for the yupster's sports car.
For pyooters, one test I sort of recall reading about was how fast a processor/chipset could compile digits of pi, e, etc, or how fast it could sort very long random lists into an alphabetized state, etc. Or some other intense number-crunching task or multiple sorting of complicated data, or perhaps manipulating 3d forms, fractals, etc.
This wouldn't be a bad idea to do on a new pyooter, to see if in fact it is slowing down over time, iow, an initial benchmark..
What pyooter tests might be available to the DIYer?
Similar for CAD, but no doubt much more involved:
What standard "builds" might one use, to assess a program's efficiency, usability, suitability, etc?
One immediate problem is to first *define* the actual context itself. It seems this is not as clear as it might be, altho certainly cad for "boat building" would seem to be different than punch-and-die cad.
It might be useful to concoct a spectrum of these tests, so that hybrid-type programs could in fact be evaluated in a number of areas for a "usability score", averaged over different areas.
I was thinking more along the lines of how a novice might download the free trials that virtually all programs offer, and perhaps be able to quickly assess (relatively) how usable the program is. I'm thinking of a standard array of a variety of parts, with perhaps a "time to completion" of each part or project in a given program. But then, this would presume a certain amount of training in the program, unless the test was indeed of the learning curve itself..
Quite the dicey proposition, this program evaluation thing, it seems, esp. for novices. Right now, it seems that one has to rely on the opinion of people who do similar work, and hope they are right.
I have the following sneaking suspicion:
In the days of DOS, you had really different programs in a given genre. For example, in word processors, XyWrite was simply the king of the effing hill, a program that even today will run circles around modern word processors -- graphics notwithstanding, of course. Modern Wordprocessors = Microsoft Word, no??
But, the Windows platform seemed to turn everything into a clone. You seen one word processor, you seen'em all. Even Windows XyWrite sucked!!
I wonder if fundamentally all this goddamm CAD isn't just the same shit, but made so effing complicated ito rules, sequences, etc. that the end user can't tell that they are all the same.
I'm not saying they ARE all the same, just wondering if maybe there isn't some wool being pulled over our eyes.... again.... After all, what is a solid but just an array of g-d points in x, y, and z?? Just how effing complicated can/should that be???