Full-time on Celestia!

[Don. Edwards]

Hey Guys,

I think it might be a bit early to say Chris has abandoned us again. I think the real way we will know that he has made the choice to stop doig any work on Celestia will be the day the shatters.net server goes offline and never comes back up. I for one hope that never happens.

So just give him a little wiggle room here. He just made a big choice to end a long term job. Its the summer time and he might have taken off to a nice tropical island for a break, I would if I had the money and the time.

Of course it would have been nice for him to drop us a line saying that he was taking off for a while, but its his baby and its his choice as to what to do. It is no different than when I go quiet for a while and don't post any projects or even make a comment here in the forum. Sometimes I have to take a break. So lets wait till Autumn and than see if he is back coding and if he isn't than we can start to worry.

Don. Edwards

[Fenerit]

I agree. Whereas the codeproject it's stopped, better to work on the catalog's improvements.

I definitely don't think Chis would want to abandon Celestia at this time. Celestia is really important for him. This I know for sure. I also know of a pretty important reason for his distraction from Celestia.

Bye Fridger

Said as it is, seem a love question... mmmm...

[fsgregs]

OK Fridger, if the big star data bases are inaccurate, why cannot they be made more correct? 100,000 stars is simply too sparce in Celestia space. Is this something that can be updated via some kind of automated script that can produce a more accurate database, as you have so meticulously done with your galaxy database?


Frank

[LordFerret]

2-cents

I tend to agree with Don's view, and I think the answer to all this speculation is right within Chris' original post.

Next week will be my last at NVIDIA--it's been a great company to work for, but there are other projects that I want to pursue. One of the big reasons I'm leaving is to spend more time working on Celestia. I'm really looking forward to having the opportunity to spend full days improving Celestia. At this point, I'm not sure how long it will be possible to work on it full-time, but at the very least it will be several months.

--Chris

Perhaps this disappearing act of his was part of his plan... time away... a chance to step back and re-think direction?

[ANDREA]

Well, I'm an incorrigible romantic, so... couldn't it be a matter of love?
Bye

Andrea

[t00fri]

OK Fridger, if the big star data bases are inaccurate, why cannot they be made more correct? 100,000 stars is simply too sparce in Celestia space. Is this something that can be updated via some kind of automated script that can produce a more accurate database, as you have so meticulously done with your galaxy database?


Frank

Frank,

it's solid distance information that is lacking in case of (fainter) stars being further away from us!

How about an instructive little calculation in your astronomy class, finding out what the largest distance of a star can be when the typical uncertainties on (trigonometric) parallax measurements (milli-arcsecond!) become ~20% of the values of the parallax itself ...? Acceptance of up to 20% trigonometric parallax errors includes about 50 000 HIP stars.



Besides measuring the tiny parallax angles, for stars there are no other reliable sources of their distances IN GENERAL. This is quite unlike galaxies, where we have about 10 different methods to estimate their distances.

The only other source of distance information for stars comes from the so-called "photometric parallax" meaning estimates of the distance modulus

m - M = -5 + 5 log(d)

Within reasonable uncertainties all this is properly implemented in form of a weighted mean in the 100 000 Hipparcos stars.
Only by applying UNREASONABLE extrapolations and thus with UNREASONABLE uncertainties one reaches ~ 1 or 2 million stars this way.

Celestia being a 3d visualization framework does NEED accurate distance information for its stars. Some users might mainly care for an attractive "background glittering" from huge numbers of stars, but this is not conform with Celestia's design goal.

Bye Fridger

[Spaceman Spiff]

Frank,

your wish for millions of stars with accurate location shall be granted (hopefully) by ESA's Gaia mission, sometime by 2020: http://sci.esa.int/science-e/www/area/index.cfm?fareaid=26. That web site says Gaia will locate one billion stars, which is really only 0.25% of the whole Milky Way population*.

Expanding on Fridger's mention of distance modulus, the trouble with the method is that populations of stars with the same surface temperature, thus spectrum or colour index (such as B-V), naturally have a spread of luminosity of a factor of two. The cause is varying age and metallicity.

Out of the extra two million stars calatogue, Celestia may show two stars being 5,000 light years away from us and 10 light years from each other, but the truth could be that they are 500 light years apart.

100,000 stars is the best we have. As Fridger suggests, it may be a very good idea to explain in classes how uncertain we are about star distances.

Spiff.

* The web site says one billion stars is 1% of the Milky Way because it assumes the average star mass is 1 solar mass. It's more like 0.25 solar masses, so the popuation is likely to be nearer 400,000,000,000 stars. Petabyte server for futureproofing Celestia, anyone?

[Spaceman Spiff]

Well, I'm an incorrigible romantic

Ah, the Italians... .

Spiff.

[ANDREA]

Well, I'm an incorrigible romantic

Ah, the Italians... .
Spiff.

Thank you!

Andrea

[Fenerit]

Fridger, just to learn more: epistemologically speak, this "uncertainty" is the same both for the professional astronomers with the best affidavits as for the Celestians or not, that is, we are we that did'nt have a method to traslate accurate extants data (thing that seem false as you have mathematically shown) or astronomers too, fails in this method; because in this case nothing we can add with accuracy and no differences exists among we and the astronomers.

[t00fri]

Fridger, just to learn more: epistemologically speak, this "uncertainty" is the same both for the professional astronomers with the best affidavits as for the Celestians or not, that is, we are we that did'nt have a method to traslate accurate extants data (thing that seem false as you have mathematically shown) or astronomers too, fails in this method; because in this case nothing we can add with accuracy and no differences exists among we and the astronomers.

Fenerit,

sorry here I have some problems from your English post in understanding /precisely/ what you mean.

Guessing at this point: yes the problems I mentioned are the same for professional astronomers and Celestia. However astronomers might often not require precise distances of the stars, depending on the problems under study. In Celestia we DO need the distances since Celestia lives in 3 space dimensions.

Bye Fridger

[Fenerit]

Fridger, just to learn more: epistemologically speak, this "uncertainty" is the same both for the professional astronomers with the best affidavits as for the Celestians or not, that is, we are we that did'nt have a method to traslate accurate extants data (thing that seem false as you have mathematically shown) or astronomers too, fails in this method; because in this case nothing we can add with accuracy and no differences exists among we and the astronomers.

Fenerit,

sorry here I have some problems from your English post in understanding /precisely/ what you mean.

Guessing at this point: yes the problems I mentioned are the same for professional astronomers and Celestia. However astronomers might often not require precise distances of the stars, depending on the problems under study. In Celestia we DO need the distances since Celestia lives in 3 space dimensions.

Bye Fridger

Fridger, I apologize for the unclarity (was matter of philosophy of science). I understood. The main problem is that the astronomers lacking to fournish the star's distances to the web community, so that everyone can make own personal use... Oh, well...

[selden]

Fenrit,

Yes, astronomers do provide distance information on the Web. The problem is that the distances to most stars are not well known: there is no accurate information which can be provided.

The "spectrographic distances" of many stars have been calculated, but those distances can be very unreliable. The procedure for calculating them has to take into account many parameters which are not well known. Pascal Hartmann's Web pages at http://pascal.hartman.free.fr/newcat.html describe how the spectrographic distances were calculated for the stars in the "2 million star database".

[Fenerit]

Ok, Selden... Although I do not know how a negative parallax can be more reliable for the astronomers than the Celestians... Very strange. The formers should account for the real space, the latters for a virtual one; but whereas the latters wish the real, the formers give the virtual.

[t00fri]

Fridger, I apologize for the unclarity (was matter of philosophy of science). I understood. The main problem is that the astronomers lacking to fournish the star's distances to the web community, so that everyone can make own personal use... Oh, well...

Fenerit,

just to reiterate: The Celestia community has access to the SAME astronomical data catalogs the scientific astronomer community has access to!

So we are all in the same "boat".

Parallax-based distance measurements are HARD and the "reach" in distance is simply limited due to measuring uncertainties of these tiny angles.

If one is daring enough, of course, one may try various "empirical" recipees that at least furnish a result with the dimension of length for fainter, more distant stars . But the associated uncertainties are OUT OF CONTROL! The results are often grossly wrong. So indeed, one is stuck with the 100 000 HIP stars, if one requires <=20% error on the star distance!

It should be easy to understand why the possibilities of distance determination are significantly richer in case of galaxies.

Just to name 3 methods that in general have NO counterparts for individual stars :

a) Cepheids, i.e. so-called standard candles. Either a given star is a Cepheid or it is not. In case of galaxies the chances are high to locate a Cepheid star among the billions of their stars!
I assume most of you know the crucial point about Cepheids, else -> click the net.

b) The "Surface Brightness Fluctuation (SBF)" distance method. It is associated with exploiting contrast fluctuations of extended objects (i.e. galaxies), hence cannot apply to pointlike stars.

c) "Tully-Fisher" distances: Here the rotation speed of galaxies is exploited, which again is inapplicable in case of pointlike stars.

d) Hubble distances: Since this method is based on the Hubble expansion rate of the Universe, it requires cosmological distance scales and corresponding redshifts. Hence it's out for stars in our Milky Way.

e)...

These examples should illustrate clearly (also to Frank!) why the situation of distance determination is qualitatively different for stars and galaxies!

Bye Fridger

[selden]

Fenrit,

Where have you seen a negative parallax?

I have a vague recollecion of something like that being seen in a database where the errors in the measurement were larger than the value being measured.

[Fenerit]

Ok, Fridger. Sincerely, if the things stay so, better leave the star catalog's improvement apart. I do not see how to escape from this tangle. Perhaps to return to consider what happened to Chris is less binding.

[Fenerit]

Fenrit,

Where have you seen a negative parallax?

I have a vague recollecion of something like that being seen in a database where the errors in the measurement were larger than the value being measured.

I'm rely upon the words found here:

http://pascal.hartman.free.fr/newcat.html

where is said: "Well, the problem was to extract some "good" stars out of the 2454555 Tycho stars of the ASCC catalog where parallax values are often unusable, either because they are negative, or lesser than their corresponding parallax errors values."

I do not have seen the true and proper catalog.

[t00fri]

Although I do not know how a negative parallax can be more reliable for the astronomers than the Celestians... Very strange. The formers should account for the real space, the latters for a virtual one; but whereas the latters wish the real, the formers give the virtual.

Sorry, all this is just wrong....

Negative parallax values simply indicate that the measurement is entirely incorrect/unreliable. They have NO physical meaning!

Incidentally, an imaginary parallax has a much closer relation to virtual spaces . (negative parallaxes are just wrong)

Suppose we measure the (real, positive) parallax of stars in a Universe with Euclidean space-time geometry (metric tensor g_mu_nu ={1,1,1,1} instead of a Minkowskian space-time, where we live in ( metric g_mu_nu ={1,-1,-1,-1} ) . Then a real parallax angle in this Euclidean space would continue to become an imaginary (sqrt(-1)) parallax in our physical space-time...

But I am afraid you can't buy anything for this fact...

Bye Fridger

[Fenerit]

Thank you Fridger for the delucidations. Just the last question then I leave off to annoying you in these matters. Did the ours computers, at the present, can run Celestia with the mass calcolus in terms of general relativity theory?