Revolutionary Science

There are two reasons I can think of that might explain why Nasa uses Julian day, though I don't know if either one is correct. One is that it's simple and ubiquitous - there are libraries for it in every computer language. The other is that it goes back before 1970, probably very useful if you're using old astronomical data to extrapolate planetary and moon positions. Anyway, there are ton's of Julian day libraries out there.But the Julian day libraries have no problem with end of year wraparound - it's just a conversion, and the conversion routines were completely debugged eons ago in computer years. Julian day is just a simple count, and the counter isn't really going "365, 366, 367..." It's going something like (I haven't calculated correct numbers for the end of this year) "2454102, 2454103, 2454104..."Where'd you see the article, maybe I'll take a look at it.--Percy

not Julian date, Julian Day. I will see if I can find it. Found oneand another oneEdited by jar, : No reason given.Edited by jar, : No reason given.

---

Aslan is not a Tame Lion

The question is what made your computer a dog - your {ability\inability} to do the kind of computing whatever work, or the addition of "ringtones" to so many programs as upgrades?It seems to me that software expands to fill the available ram*Hz so that computers are always run at max, even though the average user doesn't even use spellcheck.It is the rare person that takes advantage of all the bells and whistles of any modern program ... imh(ysa)o.I'm betting that the drive for additional computer ability will be making games more and more realistic and have nothing intentional to do with ability to do {jobs\work\analysis\etc}Enjoy.

---

Join the effort to unravel {AIDS/HIV} {Protenes} and {Cancer} with Team EvC! (click)

---

we are limited in our ability to understand
by our ability to understand

​

RebelAAmericanOZen[Deist
... to learn ... to think ... to live ... to laugh ...
to share.

Well Jar, having worked at KSC, I can tell you that most of the software and hardware involved was designed and implemented in the 70's. In 1997 I was working on a replacement for the shuttle launch control system. The system being replaced had 32k of memory, could run 8 programs at a time with 2k stack and 2k for the program per program.

Who said "Julian date"? Not me.Here's the reason for the problem from one of the articles you found: 'Nuf said! Julian day approaches dominated software date/time libraries 30 years ago, it's a very simple approach, and to fit in tiny memories I imagine NASA chose Julian day and had to develop custom software that skipped details like end-of-year rollover.Nonetheless, it was surprising to me to realize that this means no space shuttle mission has ever celebrated New Year's Day in space, so I checked a list of missions at Wikipedia. Nope, never happened before.--Percy

The reason I mentioned Julian Date as opposed to Julian Day was this sentence. The later example is Julian Date and if they were using that then the year change would not be a problem. But the article describes the former, an actual Julian Day. Julian Day seems to be what they were using.And as I pointed out in Message 45, if compactness was an issue, using a Mission Day counter would be even smaller and would have additional advantages not found using Julian Day.

---

Aslan is not a Tame Lion

Huh? Sorry, don't follow. Julian day is a count of the number of days since I don't know when (looking it up, at Wikipedia it says since Jan 1, 4713 BC), not since the beginning of the year. I imagine the "365, 366, 367..." stuff in the articles is just a simplification for popular consumption. Sure, if the code is only for counting mission days, but I doubt it's as simple as the article makes it sound - software is rarely simple, even from 30 years ago. Navigational issues seem to be the primary concern, so it isn't just an issue of counting mission days.--Percy

If it were a true Julian Date then the problem of year end simply wouldn't come up. They claim that there is a problem.Why would a true Julian Date as opposed to Julian Day (which refers to the day of the year) have a problem with end of year?

---

Aslan is not a Tame Lion

I can't even guess the root causes of the slowdown, but each security update from Microsoft lessened the performance. When I last turned the machine on (it's now used as a door stop), it took five minutes to log in. Positioning the cursor over a javascript dynamic menu would take it a few seconds to generate the menu. It was a 1.4 GHz machine, should have done fine, but it didn't. It wasn't worth fiddling with (reinstall operating system, retune, etc.), it's from 2000 or 2001 - I phased it out. The processor and graphical demands of work related activity can be extremely substantial. PCs are now being used to do the analysis for chip design that was once done by mainframes. PCs are also being used for 3D modeling, animation, etc, which places significant demands upon graphic performance. The demand for more power does not come from just the game-playing side of the fence.--Percy

in response to GDR's question about the impact of doubling the speed of light: The big question is: "What else changes?" The electro-magnetic coupling constant give by alpha = e2/hc determines the strength of eletonic and electromagnetic forces. Notice the 'c', speed of light in the denominator. Double the speed of light and the sun and earth would not even exist, let alone orbit, unless one of the other constants (e is the charge of the electron and h is planks constant) changes to compensate.The speed of light shows up in alot of other contexts that would be impacted by its change. These are all part of the strong evidentiary base for the accdeptance of the constancy of the speed of light.It needs to be recognized that the 'speed of light' actually has nothing to do with light per se; It is a property of space-time. The nature of this property is that anything with zero mass and non-zero energy will appear to move at a certain fixed speed: the 'speed of light'. If it were ever found that photons have non-zero mass then light would not move at the speed of light.

Two second choices that come to mind are Ludwig Boltzman, whose pioneering work in statistical mechanics was roundly rejected (except in America, which didn't count for much at that time) and Albert Einstein. Einstein's Nobel prize in 1921 was awarded for his work on the photoelectric effect and the awarding committee made an uncharactoristic point of stating that it was not for his theory of special relativity, which still had very limited acceptance at that time.

But not for the average home computer, not for the average business work computer.I've done major 3D modeling on a PC for several years now, and do not see much growth in necessary power.For the average home computer games are going to be the more extreme usage of most machines -- imh(ysa)o. It would be interesting to uninstall windoz and install linux on that machine and see what the performance difference was.Or a previous version of windoz ...Edited by RAZD, : added end.

---

Join the effort to unravel {AIDS/HIV} {Protenes} and {Cancer} with Team EvC! (click)

---

we are limited in our ability to understand
by our ability to understand

​

RebelAAmericanOZen[Deist
... to learn ... to think ... to live ... to laugh ...
to share.

You're right. The applications I mentioned were engineering/scientific. One person's mountain is another's anthill. That's great if you're satisfied with the power you have. You're actually reinforcing what Bob said in the conference room after his talk, that computer power will soon exceed demand, if it hasn't already. Linux is much lighter on it's feet, so the getinto/getoutof times are quicker, but for big tasks the Linux scheduler isn't significantly better than Windows. But the home applications that run on Linux (I need things like Quicken, etc.) are not very numerous.--Percy