PSX underground peek

[Jeroen Hoppenbrouwers]

Here a quote from a recent post on the alpha team forum, one of many, many, many.

3. When pulling F19 (C27) Hydraulic SYS FAULT 1 and 2 lights go off (so is #3 but this is ok). Their Test Ground pin (5B) should be connected to K05 relay [C122 (F20)] and not K07.

Questions about the colour of the leaves of the trees during Autumn deferred. First things first.

[Will]

Wow, I'd love to hear about how you design and test something like this.  I'm assuming the alpha testers are looking at PSX as a black box, right?  So there's a widget called a "Test Ground Pin" that they can see, and other widgets that look like wires, showing connections to something that looks like a K05 relay?  Or are they under the hood, looking at logic, code, and metaphor?

[delcom]

Good luck guys, and keep up the good work. Don't forget to have a beer now and then.

[Jeroen Hoppenbrouwers]

Looking at flight deck lights and browsing through 1 m² large electrical diagrams. Popping a CB, set up various flight deck switches, push the IND LTS TEST switch, and seeing what looks weird. Then find out which wire on the diagram could have created this problem, and suggesting a fix.

Hey, if you simulate the real thing, you test it like the real thing.   :-)

[Qavion]

Good luck guys, and keep up the good work. Don't forget to have a beer now and then.

Looking at the wiring schematics through the bottom of a beer glass provides that helpful extra bit of magnification

Jeroen's quote is purely related to official wiring schematics (with a few modifications added... because of errors and omissions found in the real diagrams)

(Delcom, ref SSM 33-11-71)

Cheers.
Q>

[Joe Clark]

Wow! One of the reasons why we pay so much for such a class A product! Thanks Jeroen for the post and Hardy and the alpha team for all of your hard work!

[Phil Bunch]

Is there really very much difference between programming and verifying a realistic simulator vs design/testing the real airliner's avionics and other systems?  

----------------------

At a MUCH more modest systems test level, a laser printer R&D project team I worked on created what they called a "regression test" subteam who (as far as I could understand it) tested every combination of switches and software-based parameter settings to see if the printer should be released to customers.   This was with the real system, AFTER a similar test had been passed in a software logic test.  The tests took several weeks since there were so many options on this medical imaging laser printer and its interfaces.

I assume real airliners have to undergo similar exhaustive analysis and testing, during circuit and software design and ultimately with the real airliner.  It would be interesting to learn a bit more about how they accomplish such a thing.  Now that I think about it, can one really test everything with a real airliner since pushing certain buttons or circuit breakers with everything powered up would do great harm.  E.g., pushing the fire extinguisher buttons with a running engine is best not done unless you have a real fire!  Same with the fuel jettison and many other things, but they have to be tested in use before flight.  

Also, the number of combinations of switch and relay settings and software parameter settings seems much too large to exhaustively test with an airliner...hmmm...how many combinations and permutations of such settings and parameters are there for a 747-400?!?!

[Mariano]

Hey, if you simulate the real thing, you test it like the real thing.   :-)

Nailed it! This is why PS1 is still alive and well after thirteen years, and why PSX is going to blow all of us away. I believe the hard-core PS1 fan base will be stunned by PSX, even if we are accustomed to Hardy's perfectionism.

Thanks again to Hardy and the alpha team for all you are doing.

Mariano