View Single Post
  #5   Report Post  
Posted to sci.electronics.design,alt.binaries.schematics.electronic,sci.electronics.basics
Don Y[_2_] Don Y[_2_] is offline
external usenet poster
 
Posts: 68
Default Time to Upgrade ?:-}

On 8/2/2015 9:16 AM, Jim Thompson wrote:
I think it's time I upgraded my 'Spice' machine... my present machine
is as follows... no laughter please... I've successfully done at least
at least 20 chip designs on this machine. What modern equivalent
should I replace it with?


A lot of that depends on how you work. I have friends who are perpetually
upgrading -- trying to eek out the last epsilon of performance, never
considering the time spent in the upgrade process (reinstallling software
and then reconfiguring it for the various "options"/preferences you had);
nor the "losses" that come with it (e.g., peripherals and applications that
no longer work).

But, if you look at their work process, they sit and *watch* their
machine, waiting for it to cough up a result. So, in their minds,
every increase in performance (if not counteracted by inefficiencies in
software "upgrades") is a net improvement.

OTOH, I prefer to wait a bit for each action I expect from my machines.
This gives me time to reorganize my thoughts: what will I do *when*
the machine is finished? what is my next priority? how will I verify
that the machine has done what I expected of it? etc.

Likewise, if the time involved is "more than a cup of tea", I can move
to another machine (or chore) and make some progress there. No need
to sit and wait for a machine to do the job it *will* perform.

So, the trick is finding the right amount of "wait" -- too little and
you can't get started on something else; too much and you risk the
task taking too long for your schedule, etc.

In the 80's, I had a pair of 25MHz 386's. It would take a full 24 hours
to render some of my 3D CAD models. I'd turn off the monitor (save
power) and put a note on the keyboard: "Do not turn off" (lest I
forget in a moment of distraction). Then, move to the other machine
and keep working on mode models, or a schematic, or a layout, or some
software, or assembling a prototype, or ordering components, or office
supplies, etc. Always *something* that could be done in the time waiting
(without it feeling like you're "waiting")