Dev.SN ♥ developers
Sections
Dev.SN
ed.goforth writes:
"Gears of war: When mechanical analog computers ruled the waves over at Ars is a quick recap of the history of naval targeting & fire control systems and the transition from analog to digital. In short, you'll never believe just how good those old designs were, and how much work it's taken to make GPS-based guidance be as accurate."
This discussion has been archived.
No new comments can be posted.
A Brief History of Analog Targeting Systems
|
Log In/Create an Account
| Top
| 1000 moderator points
| 14 comments
| Search Discussion
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
I'm not a robot like you. I don't like having disks crammed into me... unless they're Oreos, and then only in the mouth.
-- Fry
(Score: 5, Insightful) by egcagrac0 on Tuesday March 18 2014, @11:21AM
It's accurate in exactly the same way a slide rule or a vernier caliper is accurate.
It's a bit more impressive, since it has more inputs and outputs.
Digital computers are great, if you're dealing with data and operations that nicely break down into 0 and 1. Analog computers don't suffer from rounding errors in the same way, but can be a touch more difficult to get precise readouts from.
(Score: -1, Flamebait) by Anonymous Coward on Tuesday March 18 2014, @02:20PM
from TFA
"Because they use physical rather than digital inputs and outputs, they can represent curves and other geometric elements of calculations with an infinite level of resolution "
No, they can't, for the same reason that a photograph doesn't have an infinite level of detail just because it's "analog"
(Score: 4, Informative) by egcagrac0 on Tuesday March 18 2014, @06:55PM
The limit of a photograph's resolution is the grain size of the silver halide (or dye cloud, if color). Back when I was doing film photography, we were typically unable to do better than about 35 megapixels from a 35mm double frame (the "common" 35mm still format, 24x36mm) - with a few exceptions for things like TechPan.
Accuracy and precision are different. With an analog scale readout, I may know very well that it's about halfway between 24 and 25, but because the scratches on the bar are fairly close together, it can be difficult to determine if that's 24.47, 24.48, or 24.49.
Fortunately, if the explosion is big enough, "close enough" on the order of "about 24.5" works just fine. (As in horseshoes, with a 2700lb high-explosive shell, close counts.)
As for the accuracy of a mechanical analog computer, that limit is tied to the mating surfaces of the cogs, pinions, cams and followers - if they're machined accurately, if they're not particularly worn, if the backlash is controlled, that accuracy is very high indeed. Further, there's no interstage rounding error - the rounding errors really only show up at the output stage.