Backup Computer went into Deco

[oldschoolto]

Do you expect all computers give identical info on the same dive?

That's how they (Engineers) use computers to do things ... Have 3 computers working the same thing and if there is a problem, they look at each other and throw out the one that doesn't match..

Jim...

 

[Centrals]

That's how they (Engineers) use computers to do things ... Have 3 computers working the same thing and if there is a problem, they look at each other and throw out the one that doesn't match..

Jim...

You are not suggesting that 3 dive computers from three different manufacturers will give identical info on everything eg ndl?

 

[oldschoolto]

You are not suggesting that 3 dive computers from three different manufacturers will give identical info on everything eg ndl?

NO... Three of the same computers,,,

 

[BenjaminF]

If you really want the engineering approach, use an even number (2 or more) of identical dive computers. Group them into sets of 2 computers. Compare each set and discard any sets which don't match (for a given accuracy, of course). Take the average of all valid sets and use that as your NDL/whatever. Or just use the first valid set. Both methods work and are used for certain fields. Now I think I'll start off my day by drinking something REALLY strong in an attempt to forget that I thought about work at home...

 

[Centrals]

NO... Three of the same computers,,,

They might still give slightly different reading eg depth.
My two identical Uwatec Bottom timers rarely give identical reading on depth.

 

[Dave Bevan]

It was a joke...

 

[dmaziuk]

NO... Three of the same computers,,,

Whose components came from exact same batches, all 3 assembled on the same line in the same hour of the same day, all subjected to exact same conditions on your exact same dives. And their failures will be absolutely textbook-independent.

If you really want the engineering approach, use an even number (2 or more) of identical dive computers.

FVO "even" = "odd".

 

[stuartv]

NO... Three of the same computers,,,

I would use 3 computers from 3 different manufacturers, all that implement the same algorithm. 3 of the same computer mean 3 computers that all have all the same bugs - plus any individual manufacturing defects.

I use 2 different computers now. They generally agree (very close). Without a 3rd as a tie-breaker, I will just have to rely on the fuzzy logic between my ears to decide which one to believe when/if they ever do have a significant disagreement.

 

[TrimixToo]

I would use 3 computers from 3 different manufacturers, all that implement the same algorithm. 3 of the same computer mean 3 computers that all have all the same bugs - plus any individual manufacturing defects.

I use 2 different computers now. They generally agree (very close). Without a 3rd as a tie-breaker, I will just have to rely on the fuzzy logic between my ears to decide which one to believe when/if they ever do have a significant disagreement.

I was thinking the same thing, and I also use two different ones. This was happenstance, because the thought of having the same bug on two different ones wasn't anything I thought about until someone else brought it up.

John

 

[oldschoolto]

I'm just pointing out how complex computers are paired to over come failures... A good example is aircraft fly by wire systems.. 3 to 4 computers all talking to one another and not listening to the one that doesn't match the others.. I tend to trust my computer between my ears... BUT, BUT... I learned in flight training... ALWAYS TRUST YOUR INSTRUMENTS.. lol... So , I'm back to square one..

Jim...