Ultimate space simulation software

 
User avatar
FastFourierTransform
Pioneer
Pioneer
Topic Author
Posts: 356
Joined: 17 Nov 2016 15:09

The Gaia Mission

02 Oct 2020 07:32

Hi, whats the difference between these and what I downloaded from Phunnie?  I actually tried both his 30 million and 15 million star add-ons and while I found the 30 million add-on workable, there was some graphics flickering from time to time so I switched to the 15 million star add-on which is working fine.
This thread is not about the addon to SE but about what the addon is based on, a.k.a. the Gaia mission. Phunnie could at best base its addon on the available data at the time, which is Gaia DR2. Here we are talking about Gaia DR3, more stars, better astrometry and other things I've explained in the previous post.
 
A-L-E-X
Galaxy Architect
Galaxy Architect
Posts: 3498
Joined: 06 Mar 2017 20:19

The Gaia Mission

07 Oct 2020 15:14

Thanks, this explains the different sizes of the add-ons, as I have them all saved.  The original DR1 5 million star add on was 400 MB, while the DR2 5 million star add on was less than 200 MB.
 
User avatar
FastFourierTransform
Pioneer
Pioneer
Topic Author
Posts: 356
Joined: 17 Nov 2016 15:09

The Gaia Mission

08 Oct 2020 06:15

Thanks, this explains the different sizes of the add-ons, as I have them all saved.  The original DR1 5 million star add on was 400 MB, while the DR2 5 million star add on was less than 200 MB.
This is not possible. Let me clear some possible misunderstandings (If there are any):
Gaia DR1 (data release number 1) was a catalog of 1.4 billion stars (exactly 1,142,679,769). But the first release lacked enough observations to perform accurate stellar parallaxes by its own (thus distances). Thanks to the "old" Tycho-2 and Hipparcos catalogs we could establish some constraints on the new Gaia DR1 data so for the stars appearing in these catalogs we were able to yield a catalog of parallaxes putting everything into the mix. From the 1.4 billion stars in DR1 we got only 2 million (exactly 2,057,050) in the TGAS (Tycho-Gaia Astrometric Solution), which is the 0.18% of the catalog. Thus, at most Phunnie could have made a 2 million star addon for DR1 (not 5 million).

In Gaia DR2 (the current most recent data release, until December 3) contains 1.7 billion (exactly 1,692,919,135) stars. Now with lots of parallaxes. But Phunnie based its addon not in all of the DR2 parallax measurements but in a secondary catalog created by Bailer-Jones et al which estimated distances using a statistical analysis on the parallaxes. The Bailer-Jones distance estimates include 1.3 billion stars (exactly 1,331,909,727) of the Gaia DR2 (78.7% of the catalog). Phunnie did a wonderful job selecting only the best estimates and that's why the addon is at most of 30 million stars (which is just the 2.25% of what Gaia DR2 yielded in "distance measurements").

Phunnie could have made a larger addon with currently available data since Gaia DR2 is huge! but some artifacts might appear in SpaceEngine (due to large uncertainties and biases for the farthest stars), making the gameplay seem unnatural. And also because SpaceEngine is currently uncapable of handling that many catalogued objects without crashing.

I think you are confusing the first realease of Phunnie's addon for SpaceEngine with other later enhancements he did. These were all based on the real data of Gaia second data release (DR2), not in DR1 or any other. Now we are expecting for December the third data release (DR3), which will contain more stars. But that is probably not going to change the addon since it is limited by SpaceEngine capacity to the point it can't even give you the entire DR2 now. Thus the improvements for SpaceEngine will come in the form of better parallax measurements, and more precise distances (a 20% better estimates in many cases).
 
A-L-E-X
Galaxy Architect
Galaxy Architect
Posts: 3498
Joined: 06 Mar 2017 20:19

The Gaia Mission

08 Oct 2020 06:44

Thanks, this explains the different sizes of the add-ons, as I have them all saved.  The original DR1 5 million star add on was 400 MB, while the DR2 5 million star add on was less than 200 MB.
This is not possible. Let me clear some possible misunderstandings (If there are any):
Gaia DR1 (data release number 1) was a catalog of 1.4 billion stars (exactly 1,142,679,769). But the first release lacked enough observations to perform accurate stellar parallaxes by its own (thus distances). Thanks to the "old" Tycho-2 and Hipparcos catalogs we could establish some constraints on the new Gaia DR1 data so for the stars appearing in these catalogs we were able to yield a catalog of parallaxes putting everything into the mix. From the 1.4 billion stars in DR1 we got only 2 million (exactly 2,057,050) in the TGAS (Tycho-Gaia Astrometric Solution), which is the 0.18% of the catalog. Thus, at most Phunnie could have made a 2 million star addon for DR1 (not 5 million).

In Gaia DR2 (the current most recent data release, until December 3) contains 1.7 billion (exactly 1,692,919,135) stars. Now with lots of parallaxes. But Phunnie based its addon not in all of the DR2 parallax measurements but in a secondary catalog created by Bailer-Jones et al which estimated distances using a statistical analysis on the parallaxes. The Bailer-Jones distance estimates include 1.3 billion stars (exactly 1,331,909,727) of the Gaia DR2 (78.7% of the catalog). Phunnie did a wonderful job selecting only the best estimates and that's why the addon is at most of 30 million stars (which is just the 2.25% of what Gaia DR2 yielded in "distance measurements").

Phunnie could have made a larger addon with currently available data since Gaia DR2 is huge! but some artifacts might appear in SpaceEngine (due to large uncertainties and biases for the farthest stars), making the gameplay seem unnatural. And also because SpaceEngine is currently uncapable of handling that many catalogued objects without crashing.

I think you are confusing the first realease of Phunnie's addon for SpaceEngine with other later enhancements he did. These were all based on the real data of Gaia second data release (DR2), not in DR1 or any other. Now we are expecting for December the third data release (DR3), which will contain more stars. But that is probably not going to change the addon since it is limited by SpaceEngine capacity to the point it can't even give you the entire DR2 now. Thus the improvements for SpaceEngine will come in the form of better parallax measurements, and more precise distances (a 20% better estimates in many cases).
Thanks- I wonder how large the files will be?
This is what I was referencing.....
Extreme-5Mstars20LALL.pak  size 406,526  downloaded 12/8/18
vs
5M-1L.pak  size 155,154  downloaded 9/25/20
also, despite the large file size, the earlier file loaded fairly quickly with SE (load time 1 min).
now I use the 14M-14mag.pak file size 395,741 downloaded 9/25/20
That takes about 3 min to load at startup.

I used that 30 million star pak that you referenced for awhile, but it took about 8 min to load and although the program didn't crash for me, the window flickered quite a bit.  It was 30M-15mag.pak  size 794,645  downloaded 8/3/20
 
User avatar
FastFourierTransform
Pioneer
Pioneer
Topic Author
Posts: 356
Joined: 17 Nov 2016 15:09

Re: The Gaia Mission

20 May 2022 05:37

So, in just 24 days (13th of June, 2022) we will have the new data release from Gaia (this is called DR3). There will be no updates in astrometry (parallax and proper motions) since Gaia EDR3 (which was released in December 2020), but the new release will present low-resolution spectra for the first time in this mission, which means an alternative way of validating parallaxes by doing spectroscopic parallax. The good thing in regards to SpaceEngine is that for the first time, we will have kind of reliable spectral types for millions of stars.

In any case, since the extension of the mission has been approved we expect to see an important increase in Gaia parallaxes for DR4 and subsequent, wich means better distance estimates and a good 3D map of the Solar neighbourhood.
 
julianstout
Observer
Observer
Posts: 1
Joined: 24 Mar 2019 16:53

Re: The Gaia Mission

06 Jun 2022 16:20

Will the addon get updated? and if so, how long would it take?
 
User avatar
Mosfet
Star Engineer
Star Engineer
Posts: 1770
Joined: 24 Oct 2016 11:34
Location: Italy
Contact:

Re: The Gaia Mission

07 Jun 2022 04:01

Will the addon get updated? and if so, how long would it take?
Assuming you're referring to Phunnie's Gaia addons available in the Steam Workshop, again, this thread is not dedicated to them. The correct thread for discussing addons is this: GAIA to SpaceEngine catalogs
Nobody can answer to these questions but the authors. Unfortunately the authors of the addons seem to be busy with other things at the moment.
"Time is illusion. Lunchtime doubly so". Douglas N. Adams
| My mods: http://forum.spaceengine.org/viewtopic.php?f=3&t=80 | My specs: Asus x555ub - cpu i5-6200u, ram 12gb, gpu nvidia geforce 940m 2gb vram |
 
User avatar
SpaceEngineer
Author of SpaceEngine
Author of SpaceEngine
Posts: 1125
Joined: 17 May 2016 22:16
Location: Saint-Petersburg
Contact:

Re: The Gaia Mission

09 Jun 2022 14:06

I still didn't start work on a proper support of GAIA data. The main problem is that SE require spectral class for every star, but GAIA data doesn't have this. It is possible to change engine so that it will use only observable data to generate procedural systems (i.e. luminosity and B-V). The other thing is implementing a different type of octree for catalog stars, probably extending the procedural stars octree by replacing generation function by disk data loading function. This is not that hard, but will kill out existing star database (including exoplanets), all star names (HIP, Kepler etc) must be cross-matched with GAIA again. Also new octree will require some pre-processig of the GAIA data (probably be a special command line tool). Only stars with known parallax must be included to SE catalog, which will reduce the database size. Other stars with no parallax still can be included, if SE (or preprocessing tool) will be able to calculate distance, for example from apparent magnitude and spectrum. These extre stars will have bad distance accuracy, but may be released as addon.
 
A-L-E-X
Galaxy Architect
Galaxy Architect
Posts: 3498
Joined: 06 Mar 2017 20:19

Re: The Gaia Mission

16 Jul 2022 05:26

So, in just 24 days (13th of June, 2022) we will have the new data release from Gaia (this is called DR3). There will be no updates in astrometry (parallax and proper motions) since Gaia EDR3 (which was released in December 2020), but the new release will present low-resolution spectra for the first time in this mission, which means an alternative way of validating parallaxes by doing spectroscopic parallax. The good thing in regards to SpaceEngine is that for the first time, we will have kind of reliable spectral types for millions of stars.

In any case, since the extension of the mission has been approved we expect to see an important increase in Gaia parallaxes for DR4 and subsequent, wich means better distance estimates and a good 3D map of the Solar neighbourhood.
Really excited about reliable spectral types!  I'm going to once again advocate for the presence of a FOV dynamic H-R diagram within SE that shows the position of the star you are pointing to on the H-R diagram (and vice versa if you point to a dot in H-R it would show what star it corresponds to in the FOV.)
 
TrigMan59
Observer
Observer
Posts: 4
Joined: 27 Jun 2022 20:32

Re: The Gaia Mission

31 Jul 2022 16:34

I hope this hasn’t already been mentioned in the thread (I couldn’t find it) but was the satellite mission named Gaia in reference to the last book in the Issac Asimov Foundation series Foundation and Earth which talks about Gaia? Just wondering…..
 
A-L-E-X
Galaxy Architect
Galaxy Architect
Posts: 3498
Joined: 06 Mar 2017 20:19

Re: The Gaia Mission

01 Aug 2022 04:06

I hope this hasn’t already been mentioned in the thread (I couldn’t find it) but was the satellite mission named Gaia in reference to the last book in the Issac Asimov Foundation series Foundation and Earth which talks about Gaia? Just wondering…..
Love that series and the last two were my favorite because of Gaia... I hope all the Foundation planets get into Space Engine one day!

Who is online

Users browsing this forum: No registered users and 18 guests