Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Re: Non Professionals doing Engineering

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Christopher-
I agree with the concept of checking the input and results.  However,
others seem to be suggesting that you have no business running a
computer analysis unless you can predict the outcome ahead of time.  

I do not subscribe to that line of thinking.

And just for the record, computers do make mistakes.  It is true that a
computer error is much less likely now than in the 1960's.  But I
remember the days of silicon failure and faulty chips / bad transistors
/ tubes actually burning out making errors.  Those were the good old
days when you could actually truly blame errors on the computer.

I must be getting old if I am the only one who can remember this :)

Lynn



Christopher Wright wrote:
> 
> >Computers don't make mistakes, programmers and
> >users do. You may be different, but I have found that I am not very
> >good at checking my input for concept, accuracy, applicability, and typing.
> An honest man. I've heard a lot of lectures about convergence checks and
> checking over source code, but the nastiest bites I've ever gotten were
> my own (usually) simple blunders. Like the punch press
> episode--everything looked great except my client thought the
> displacements didn't jibe with his measurements. Displacement and stress
> plots looked fine, equilibrium checked, but it turned out my elastic
> modulus was input as 3E6, I'd mixed up 30E6 and 3E7 when I entered them.
> Turned out to be no big deal--the client said the difference looked off
> by almost exactly 10, and all he really cared about were weld loads
> anyway, don't bother re-running.
> 
> So I'd quibble with the 'Do the analysis and then it's over,' position.
> It's only begun when the numbers come out. That's the time you start
> asking all the questions you'd ask if your client presented you with
> results sent him by your competition.
> 
> Christopher Wright P.E.