continued fraction program

I once had a coworker in the USAF who had been in on that missile prgram that discovered that the earth is kinda pear-shaped - a little oblate in the south, and a little prolate in the north.

What happened was, they calculated a trajectory, launched a missile, and it always landed short and to the north of the target. From that, they determined that the Earth has a bulge.

Isn't one of those inch/metric things an ezact ratio? Either the 39.37 or the 2.54?

Cheers! Rich

It's been pointed out that the inch has been defined as being exactly

Given the non-repeating value, it's caused some discussion in gEDA/PCB about changing our internal units to metric to avoid rounding errors. It would mean our base unit is *nanometers* though, to get the inch resolution (1E-5) we want, and limit boards to 2 meters per edge with the default types.

Ah, a NIST reference:

"Standard Inch. The value for the inch, derived from the value for the yard effective July 1, 1959, is exactly equivalent to 25.4 millimeters."

I thought most of the shape issues had been calculated and observed as far back as the 18th Century. One experiment was to measure out known straight line due South in the Andes, and compare lattitude with expected. Same expedition brought back things like rubber from the Amazon rain forests.

See this for confusion:

Jon

Why would that confuse things? If the yard/inch are defined as some fraction of a meter, if the meter changes, the yard/inch change with it.

I wasn't talking about the ratio that you wrote about. You just reminded me that recently a meter was redefined as the length traveled by light in vacuum in 1/299782458 seconds, so the unit is determined in terms of another fundamental unit -- time. That's a change from prior definitions based upon atomic transitions. Sorry I wasn't clear. There have been a lot of redefinitions as tools improved rapidly, of late.

Jon

I don't get it. What's so confusing about it? Though not great prose, it seems clear enough

See my other recent response.

Jon

A base unit of 1/12.7E6 inch would give integer representation of 10 micro inches and 10nm with a bit more than +/-4 meters range from a 32bit int.

Doesn't that work out better all round?

I.e. a 2nm unit. Yeah, but who has boards bigger than 2 meters?

... snip 110 useless lines ...

Did this exquisite reply really require quoting 110 useless lines?

It's just a confusing web page.

Say, are we sure that the Universe is expanding, or are we shrinking, so that the rest of the Universe only _looks_ like it's expanding? ;-)

Cheers! Rich

That could very well be the case - maybe what was significant at the time is that they did it in rockets, and/or that the "discovery" was reported in Time magazine, in about a 6-page spread with color glossy artists' conceptions and stuff. :-)

Thanks! Rich

Seriosly. Are there people using gEDA (or any other CAD program) to layout boards bigger then 2m?

And if they are, what software do they currently use? If one can afford to build 2m PCBs the costs of the layout software is probably negligible anyway.

Some years ago Delorme took their Street Atlas USA software and split it up into various versions (basic/deluxe/commercial/etc.)... one of the compalints of people with the new version was that the cheap version (but not cheaper than the old version) refused to print to anything larger than 11x17. Most annoying to cripple software like that...

If they are, they can rebuild PCB on a 64-bit platform too ;-)

(if anyone complains that they need a PCB bigger than the galaxy, let them buy a 128-bit platform ;)

Consider relativity, as expounded by Einstein about 1905.

My computer has things called "scroll bars."

John