JackDole wrote:

Phunnie wrote:

Source of the post Updated to V2.9. All details in changelog.

Can you publish the new program you used to create the new catalogs? And the parameters necessary for the Gaia query?

I'll need to make some changes to the code so it takes into consideration the new HR diagram based algorithm I've implemented and convert from the basic gaia ADQL query I've been using so far to SE.

Phunnie wrote:

A-L-E-X wrote:

Phunnie wrote:

You could, it might take a while to load because SE will have to delete a substantial amount of duplicate stars. I'll keep that idea in mind and I might make other paks in the future when I'll have time. I do really want to make a pak with giants and above only, as well as some based on apparent magnitude and perhaps even cross sections of the milky way to hopefully get some sort of structure.

Thats understandable however on the Gaia wiki page I saw that they were also talking about discovering thousands of exoplanets and it would be useful to have sunlike stars and even red dwarfs included because planets discovered around those stars could be habitable.  They also mentioned quasars, and it would be nice to have a full catalog of quasars in SE.

I agree, having quasars would be nice. However, I have no idea what to query in the ADQL search to get the data for galaxies. Once I figure that out, I'll add galaxy paks, or include them in each version.

I wonder how it will be once Gaia starts including exoplanets.  Right now we have to add systems one at a time, it would be great if there was a way for the program to update its list of exoplanets automatically and give each a default landscape (but also randomized on smaller scales for some variety) based on type.

Just wanted to thank you Phunnie, the 59 spectral class errors in the 5M file are now gone!  I tried to run the 2M and 5M files together but SE would not load.  With the 5M file, SE takes about 1 min to load and uses up about 2.3 GB of RAM.  It also uses 50% of my GPU while the vanilla SE used about 30%.

The only errors left now are those duplicate errors, which there was 1 from the intergalactic stars pak, 4 from the white dwarfs pak and 3 from the 5M pak.  You said the program self-corrects those though so those shouldn't be an issue?

Well I got super side-tracked and ended up deciding to cut a flat cross section of the plane of the milky way to maybe get a glimpse of where the nearest dust "clouds" are.

and well while doing that I realized that the milky way is actually crooked

so I corrected it:

Note: It's still not 100% perfect.

Fix attached below:

I also managed to pull a 28M stars query lol.


and converted it to a SE .csv catalog that's 2.3GB in size, then took a 20pc thick cross section of those 28M stars on the plane of the MW giving some interesting results:

750k stars in that tiny cross section from a abridged (28M) catalog, Imagine the entire catalog (over 1B). Ouch.

Thanks Phunnie, do I need to replace the files I downloaded lol (800k, 2M and 5M) or are those the same as before?  I downloaded them on the same day I made that post thanking you- 12/8 or 12/9 I think.

I see this in the changelog- also there is no 28M star file.  You can go higher with stars than 5M, Starry Night has 17M stars and runs fine as long as all the stars aren't being viewed all at once lol.  What is that Localgroup.sc file supposed to do?  Since I am using the 5M file from 12/8 (2.9) do I need to use that too or are all those stars already included in the 5M pak?

Changelog
V2.91 - Small hotfix for missing clusters in 3M and 4M. 09/12/2018
V2.9 - Changed the algorithm of the way spectral classes are assigned to classes using the HR diagram as a model. Now has more than just the V type class (Main Sequence). Includes Ia (luminous supergiant, Ib(supergiant), II(bright giant), III (giant) and V (main sequence). IV (subgiant) line is ambiguous so I did not add them. Spectral subclass (1-9) are still randomized (does not affect anything). 08/12/2018

A-L-E-X wrote:

Thanks Phunnie, do I need to replace the files I downloaded lol (800k, 2M and 5M) or are those the same as before?  I downloaded them on the same day I made that post thanking you- 12/8 or 12/9 I think.

I see this in the changelog- also there is no 28M star file.  You can go higher with stars than 5M, Starry Night has 17M stars and runs fine as long as all the stars aren't being viewed all at once lol.  What is that Localgroup.sc file supposed to do?  Since I am using the 5M file from 12/8 (2.9) do I need to use that too or are all those stars already included in the 5M pak?

Changelog
V2.91 - Small hotfix for missing clusters in 3M and 4M. 09/12/2018
V2.9 - Changed the algorithm of the way spectral classes are assigned to classes using the HR diagram as a model. Now has more than just the V type class (Main Sequence). Includes Ia (luminous supergiant, Ib(supergiant), II(bright giant), III (giant) and V (main sequence). IV (subgiant) line is ambiguous so I did not add them. Spectral subclass (1-9) are still randomized (does not affect anything). 08/12/2018

No i haven't made any changes. Also really, you can go over 5M? I've always crashed upon startup past 5M. Not just me either, I ran test on other systems and it seemed SE just couldn't support more than 5M stars in one pak. Also localgroup file is a fix for galaxies. The file should go in addons/catalogs/galaxies or data/catalogs/galaxies.

Phunnie wrote:

A-L-E-X wrote:

Thanks Phunnie, do I need to replace the files I downloaded lol (800k, 2M and 5M) or are those the same as before?  I downloaded them on the same day I made that post thanking you- 12/8 or 12/9 I think.

I see this in the changelog- also there is no 28M star file.  You can go higher with stars than 5M, Starry Night has 17M stars and runs fine as long as all the stars aren't being viewed all at once lol.  What is that Localgroup.sc file supposed to do?  Since I am using the 5M file from 12/8 (2.9) do I need to use that too or are all those stars already included in the 5M pak?

Changelog
V2.91 - Small hotfix for missing clusters in 3M and 4M. 09/12/2018
V2.9 - Changed the algorithm of the way spectral classes are assigned to classes using the HR diagram as a model. Now has more than just the V type class (Main Sequence). Includes Ia (luminous supergiant, Ib(supergiant), II(bright giant), III (giant) and V (main sequence). IV (subgiant) line is ambiguous so I did not add them. Spectral subclass (1-9) are still randomized (does not affect anything). 08/12/2018

No i haven't made any changes. Also really, you can go over 5M? I've always crashed upon startup past 5M. Not just me either, I ran test on other systems and it seemed SE just couldn't support more than 5M stars in one pak. Also localgroup file is a fix for galaxies. The file should go in addons/catalogs/galaxies or data/catalogs/galaxies.

Thanks I will put in the galaxies folder, does it include stars or just images for galaxies?  Does it alter the numbering scheme for procedural stars in galaxies as far as you know? (I thought all galaxy add ons did that.)
I downloaded the three different ones to test them and I installed the 2M+5M pak for awhile and it did load the first time but every subsequent time I loaded it, it would hang on start up so I just use the 5M pak now. 

Still haven't figured out how to remove duplicates. However, with the most recent 64-bit update, I have sucessfully ran a 49.3 million stars pak, albeit with a lot of suffering. I also got a 10M, 30M and 78M (the entire catalog of stars with radius values) that I will upload as addons sometime in the future. I also managed to pull out the LMC and SMC, but the radius of the stars have been randomized (because they are missing) and the distance set to an arbitrary number somewhat close to what it should be.

bonus downscaled milky way pano with 49M

You shouldn't randomize radius if data is missing. Just leave the field blank, SE will calculate radius itself.

That's pretty neat

Phunnie wrote:

Still haven't figured out how to remove duplicates. However, with the most recent 64-bit update, I have sucessfully ran a 49.3 million stars pak, albeit with a lot of suffering. I also got a 10M, 30M and 78M (the entire catalog of stars with radius values) that I will upload as addons sometime in the future. I also managed to pull out the LMC and SMC, but the radius of the stars have been randomized (because they are missing) and the distance set to an arbitrary number somewhat close to what it should be.

► Show Spoiler

bonus downscaled milky way pano with 49M

► Show Spoiler

Any update?, I believe that this one it's for 0.9.9.0 now, or still can be for 0.9.8.0? This may make my pc explode lol

I have his biggest catalog installed for .990 and it actually works much better there no lag.  I have the 5M catalog installed, dont know what this 49M thing is lol.  

Phunnie wrote:

Still haven't figured out how to remove duplicates. However, with the most recent 64-bit update, I have sucessfully ran a 49.3 million stars pak, albeit with a lot of suffering. I also got a 10M, 30M and 78M (the entire catalog of stars with radius values) that I will upload as addons sometime in the future. I also managed to pull out the LMC and SMC, but the radius of the stars have been randomized (because they are missing) and the distance set to an arbitrary number somewhat close to what it should be.

► Show Spoiler

bonus downscaled milky way pano with 49M

► Show Spoiler

Nothing new to install since last Dec right?  I am fine so far with the v2.9 5M star file you uploaded back then lol.

The 5M version works fine on my PC. Incredible amount of stars. Thank you!

So it's been a while since any work has been done on this, but I would like to announce that duplicate stars are no more!

Before:
The duped stars are clearly visible where it looks like there are rays of stars pointing at earth


After:


Now this doesn't 100% fix all duped stars (more like 99.999% of them) as I only used HIP cross-referenced to GAIA.

I will update it soonish after I redo everything and add a couple extra even more extreme versions.

GIF:
https://gfycat.com/disastrousaltruisticballoonfish