A Forking Commit Thread.

The place to discuss creating, porting and modifying Celestia's source code.
pirogronian
Developer
Posts: 177
Joined: 05.01.2018
Age: 33
With us: 1 year 8 months
Location: Wrocław
Contact:

Post #41by pirogronian » 11.04.2019, 17:27

onetwothree wrote:Celestia doesn't support start orbits.
Doesn't it? I always thought it does! Star may orbit other star or barycenter. Only DSOs doesn't have any trajectories...

onetwothree
Developer
Posts: 279
Joined: 22.09.2018
With us: 11 months 24 days

Post #42by onetwothree » 11.04.2019, 17:48

pirogronian wrote:Doesn't it? I always thought it does! Star may orbit other star or barycenter. Only DSOs doesn't have any trajectories...

Barycenters and other stars - maybe, but that's all. Center/Goto the Barnard's star and speed up time. You'll see that star field doesn't change, while BS is very fast, its period is 180 years only.

pirogronian
Developer
Posts: 177
Joined: 05.01.2018
Age: 33
With us: 1 year 8 months
Location: Wrocław
Contact:

Post #43by pirogronian » 11.04.2019, 19:53

Janus wrote:Should I see about making a DP star database?
And if so, would you mind if I included the RA, Dec & Dist, as well as proper motion in it.
I ask because part of my tweaks use RaDecDist directly, and storing it in the star DB is faster than computing it.

Do You mean database file? Why not.
According to motion, I have a plan for making all objects "trajectorable" like planet bodies currently are. However I have no idea, how to include RaDecDisc coordinates in this mechanism...

Topic author
Janus
Posts: 405
Joined: 13.08.2016
With us: 3 years 1 month

Post #44by Janus » 12.04.2019, 04:08

@pirogronian

I have asterisms working finally.
Kind of kludgy, but it works.

My next step is make the text entries into DP.
I have not looked to see they are or not, but a quick compare to 5445 indicated no.

A question, is the DP resolution actually being used yet?
I ask because the asterisms usable data is still float, as are boundaries.


Janus.

Avatar
Lafuente_Astronomy
Moderator
Posts: 352
Joined: 04.08.2018
Age: 21
With us: 1 year 1 month
Location: Cebu City, Cebu Province, Philippines
Contact:

Post #45by Lafuente_Astronomy » 12.04.2019, 04:22

Janus wrote:@pirogronian

I have asterisms working finally.
Kind of kludgy, but it works.

My next step is make the text entries into DP.
I have not looked to see they are or not, but a quick compare to 5445 indicated no.

A question, is the DP resolution actually being used yet?
I ask because the asterisms usable data is still float, as are boundaries.


Janus.

Janus, would the things you mentioned go into your next fork and commit? Just curious

onetwothree wrote:Celestia doesn't support start orbits.

Well, I think it may well be about time it should. So that we can observe the paths they take relative from Earth's view over millions and billions of years. Same with Precession.

If current star motion esitmates are correct, Alpha Centauri and Beta Centauri would have a conjunction around 6200 AD, making it an optical star, and I'd love to see that in Celestia if I would go all the way to that specific era in Celestia's time feature.

Oh and one more thing: Is there a Black Hole model file that can be used for making Black Holes in Celestia? I'd love to see how M 87's Black Hole would look like in Celestia. Thanks in advance
Official Administrator of the Celestia Discord Server.
Invite: https://discordapp.com/invite/WEWDcJh
If you don't have a Discord account, register here: https://discordapp.com/register
Have a blessed day.

Topic author
Janus
Posts: 405
Joined: 13.08.2016
With us: 3 years 1 month

Post #46by Janus » 12.04.2019, 04:33

@Lafuente_Astronomy

The DP star position is not my work.
I grabbed the source from the double fork on github.
What I did was make the asterisms work again.
It is something I have been working toward over time, but not made any real progress.

My stuff is current with main branch release {5466} right now, and I am playing with the whole DP thing.
Once I have the text entries going into DP.
I will be making a DP star database and package to use it.
When I get all of that working, I will bring my tweaks over to it.


Janus.

Avatar
Lafuente_Astronomy
Moderator
Posts: 352
Joined: 04.08.2018
Age: 21
With us: 1 year 1 month
Location: Cebu City, Cebu Province, Philippines
Contact:

Post #47by Lafuente_Astronomy » 12.04.2019, 04:45

Janus wrote:@Lafuente_Astronomy

The DP star position is not my work.
I grabbed the source from the double fork on github.
What I did was make the asterisms work again.
It is something I have been working toward over time, but not made any real progress.

My stuff is current with main branch release {5466} right now, and I am playing with the whole DP thing.
Once I have the text entries going into DP.
I will be making a DP star database and package to use it.
When I get all of that working, I will bring my tweaks over to it.


Janus.

Alright, I'll wait for your new update of Commit 5466.

However, it still causes letters which form the spelling of the Greek letters to form into Greek letters themseves. So, it may still need work. Just a heads up.

Other than that, do what you think is best.
Official Administrator of the Celestia Discord Server.
Invite: https://discordapp.com/invite/WEWDcJh
If you don't have a Discord account, register here: https://discordapp.com/register
Have a blessed day.

Topic author
Janus
Posts: 405
Joined: 13.08.2016
With us: 3 years 1 month

Post #48by Janus » 12.04.2019, 05:13

@Lafuente_Astronomy

For my part I would be happier if I could disable/remove all the greek letters.
They drive me bonkers for my part, and I may develop a tweak/fork that filters them out if I can ever wrap my head around strings/internationalism.
I do not understand why they were added in the first place, but this is not my project, not my choice.
I prefer ascii 20h-7Fh, it is so much simpler and more direct.

On another note, my 5466 is up in my forking thread, and in My Downloads though that is a complete source package for VS2015, but it does contain the precompiled exes.

The DP version is not ready for prime time, it will be a while.


Janus.

Avatar
Lafuente_Astronomy
Moderator
Posts: 352
Joined: 04.08.2018
Age: 21
With us: 1 year 1 month
Location: Cebu City, Cebu Province, Philippines
Contact:

Post #49by Lafuente_Astronomy » 12.04.2019, 07:34

Janus wrote:@Lafuente_Astronomy

For my part I would be happier if I could disable/remove all the greek letters.
They drive me bonkers for my part, and I may develop a tweak/fork that filters them out if I can ever wrap my head around strings/internationalism.
I do not understand why they were added in the first place, but this is not my project, not my choice.
I prefer ascii 20h-7Fh, it is so much simpler and more direct.

On another note, my 5466 is up in my forking thread, and in My Downloads though that is a complete source package for VS2015, but it does contain the precompiled exes.

The DP version is not ready for prime time, it will be a while.

Alright. I can wait

As for the Greek Letters, I think you should confer with the devs and reach a settlement regarding them. Only then will the problem probably be fixed

Also, will the 100 GLY star field limit be reinstated in your future forks and commits or not?
Official Administrator of the Celestia Discord Server.
Invite: https://discordapp.com/invite/WEWDcJh
If you don't have a Discord account, register here: https://discordapp.com/register
Have a blessed day.

pirogronian
Developer
Posts: 177
Joined: 05.01.2018
Age: 33
With us: 1 year 8 months
Location: Wrocław
Contact:

Post #50by pirogronian » 12.04.2019, 10:53

Janus wrote:I have asterisms working finally.
Kind of kludgy, but it works.

Do you mean it didnt work? But I though my last work with Greeks solved all bugs?

And did my code solve problem with stars distribution? Problem mentioned in first post in this topic? Anyone knows? Without it it's not worth to continue rewriting things into doubles, I think.

Topic author
Janus
Posts: 405
Joined: 13.08.2016
With us: 3 years 1 month

Post #51by Janus » 12.04.2019, 13:25

@pirogronian

The kludgy I was referring to is how I got asterisms working.
I had to switch asterisms back to float, then convert star position chains from double back to float during prepare.
I am having a lot of trouble following most of the render path, but I think I am making some progress.
Now if customers would quit distracting me with paying jobs I could buckle down and study, you know, do something really interesting.

As for the greek letters, them being shown is my problem.
I realize I may be in the minority, but if I never saw them in a stellar name again, I would be happy.
The fully spelled out names are faster and easier for me to read.

@Lafuente_Astronomy

I will be posting a DP version once I have a DP database to go with it.
Until then however, I am studying how these work.

http://orbitsimulator.com/gravitySimulatorCloud/properMotionHome.html

http://t.nomoto.org/HippLiner/index-e.html

To get an idea of how to do the math for proper motion.
My intent at this time is work out stellar speed into a movement vector, then add it to the database.
The star class would have a base position, a proper motion movement, and a current position.
Any call to setting the date, or date manipulation would trigger a star position recalculation.
The 100Gly will be enabled when I put it all up.
I also need to verify that DP is used for star position instead of SP, which I cannot be sure of yet.

Text entries into DP positions is my current step.


Janus.

Avatar
Lafuente_Astronomy
Moderator
Posts: 352
Joined: 04.08.2018
Age: 21
With us: 1 year 1 month
Location: Cebu City, Cebu Province, Philippines
Contact:

Post #52by Lafuente_Astronomy » 12.04.2019, 14:14

Janus wrote:@Lafuente_Astronomy

I will be posting a DP version once I have a DP database to go with it.
Until then however, I am studying how these work.

http://orbitsimulator.com/gravitySimulatorCloud/properMotionHome.html

http://t.nomoto.org/HippLiner/index-e.html

To get an idea of how to do the math for proper motion.
My intent at this time is work out stellar speed into a movement vector, then add it to the database.
The star class would have a base position, a proper motion movement, and a current position.
Any call to setting the date, or date manipulation would trigger a star position recalculation.
The 100Gly will be enabled when I put it all up.
I also need to verify that DP is used for star position instead of SP, which I cannot be sure of yet.

Text entries into DP positions is my current step.


Janus.

Alright. Take your time and do what you think is best. Don't let your job and your clients distract you from the fact that you're making a great contribution to Celestia's development. More power to you
Official Administrator of the Celestia Discord Server.
Invite: https://discordapp.com/invite/WEWDcJh
If you don't have a Discord account, register here: https://discordapp.com/register
Have a blessed day.

onetwothree
Developer
Posts: 279
Joined: 22.09.2018
With us: 11 months 24 days

Post #53by onetwothree » 12.04.2019, 17:51

Janus wrote:My plan is to use the stars position in RaDec{Dist} as a base, then add or subtract the proper motion as relevant.
It is a crude method, but it should be enough to answer my question(s).

I believe RaDec is only useful when looking from Earth only. If an observer changes its position then it's better to have xyz set of coordinates.

pirogronian wrote:According to motion, I have a plan for making all objects "trajectorable" like planet bodies currently are.

Do you have enough knowledge in celestial mechanics for such changes?

pirogronian
Developer
Posts: 177
Joined: 05.01.2018
Age: 33
With us: 1 year 8 months
Location: Wrocław
Contact:

Post #54by pirogronian » 12.04.2019, 17:55

onetwothree wrote:pirogronian wrote:
According to motion, I have a plan for making all objects "trajectorable" like planet bodies currently are.


Do you have enough knowledge in celestial mechanics for such changes?

As I mentioned, there is already working code for Body class. It's matter of adaptation.

onetwothree
Developer
Posts: 279
Joined: 22.09.2018
With us: 11 months 24 days

Post #55by onetwothree » 12.04.2019, 18:58

pirogronian wrote:As I mentioned, there is already working code for Body class. It's matter of adaptation.

Your "already working code" is just a 1% of work. And "matter of adaptation" is another 99%.

pirogronian
Developer
Posts: 177
Joined: 05.01.2018
Age: 33
With us: 1 year 8 months
Location: Wrocław
Contact:

Post #56by pirogronian » 12.04.2019, 19:33

onetwothree wrote:Your "already working code" is just a 1% of work. And "matter of adaptation" is another 99%.

Hmm, maybe. Without try I won't find out it for sure.
Update: I think the really hard thing will be an octree. It already will be hard to make it work with dynamic database.

Topic author
Janus
Posts: 405
Joined: 13.08.2016
With us: 3 years 1 month

Post #57by Janus » 13.04.2019, 00:29

Speaking entirely for myself.

The entire reason I am going with proper motion adding is to avoid that 99% adaptation.
I am not a real astronomer, mathematician, just a tinker, yet even i know how much work doing it right is.
Computing their proper orbit requires computing ours first for an approximation.
The process is quite iterative, and time consuming.

On the other hand, proper motion will handle a few degrees of galactic rotation quite easily, at ~95% of the final accuracy.
Since a Galactic year is ~240My, ~1,5Degree per Million years.
Handling plus or minus three degrees of galactic rotation covers plus or minus 2My.
If you accept some slop, that can be expanded to 6 degrees, or plus or minus 4My.
We might need temporally sensitive textures for earth if we go that far.

The KT boundary occurred ~95 degrees ago for comparison.
Here for anyone that cares.

You are not going to go past those few degrees on a desktop simulator.
At least those of us not running Thread ripper 32 core monsters for desktops that is, and not likely then.
Not that I wouldn't, but I have better uses for that amount of money.
Or putting it in simpler terms, that sort of simulator would eat any bitcoin mining operation for breakfast without even burping.

Using simple proper motion on the other hand, ours included, is just simple addition.
Every star has a speed, and a direction it is going.
As long as we restrict ourselves to the equivalent of the horizon {From a galactic center perspective.}, we can consider those paths straight.
Add the offsets in microlightyears restricting yourself to the scale where secants/chords occur where the arc looks straight.

The database currently stores positions in microlight years with 7 digits accuracy.
The move to DP will raise that accuracy to 16 digits.
Or going from a minimum current position error rate of ~30LS per LY, to ~30ns per LY.

Hence using proper motion addition or subtraction is much better for our needs..
If crossing a year boundary, recalculate star positions, then mark asterisms unprepared.
While it is not "Exact", it is plenty close enough, and will also enable fairly accurate star charts from other systems.


Janus.

Avatar
Lafuente_Astronomy
Moderator
Posts: 352
Joined: 04.08.2018
Age: 21
With us: 1 year 1 month
Location: Cebu City, Cebu Province, Philippines
Contact:

Post #58by Lafuente_Astronomy » 13.04.2019, 08:07

FYI, I merged Janus' post, entitled "A Forking Commit Thread." with with Kochav Israel's post, entitled "Why does 1 GY celestia do this? Is this because star distances are still laid somewhat according to parallax?", because I think they somewhat have the same questions and contents. If you want them separate, I can split them.

But I think splitting it won't be necessary, at least judging by the contents of the 2 posts being about improvements of Janus' forks and commits, and its relation to the Dev team's developments of certain parts of Celestia 1.7.0. Have a blessed day everyone
Official Administrator of the Celestia Discord Server.
Invite: https://discordapp.com/invite/WEWDcJh
If you don't have a Discord account, register here: https://discordapp.com/register
Have a blessed day.

Topic author
Janus
Posts: 405
Joined: 13.08.2016
With us: 3 years 1 month

Post #59by Janus » 13.04.2019, 08:31

I have no opinion on the merge.

However, during testing I noticed a pattern to texture issues.
I am going to make a tweak/fork with greek lettering disabled.
No substitutions, no changes in display or storage either one.

Would anyone be willing to do some testing with </greek> to see how it does with addons?


Janus.

Avatar
Lafuente_Astronomy
Moderator
Posts: 352
Joined: 04.08.2018
Age: 21
With us: 1 year 1 month
Location: Cebu City, Cebu Province, Philippines
Contact:

Post #60by Lafuente_Astronomy » 13.04.2019, 09:15

Janus wrote:I have no opinion on the merge.

However, during testing I noticed a pattern to texture issues.
I am going to make a tweak/fork with greek lettering disabled.
No substitutions, no changes in display or storage either one.

Would anyone be willing to do some testing with </greek> to see how it does with addons?


Janus.

Not sure how that works. Could you explain how? But I'll try regardless
Official Administrator of the Celestia Discord Server.
Invite: https://discordapp.com/invite/WEWDcJh
If you don't have a Discord account, register here: https://discordapp.com/register
Have a blessed day.


Return to “Development”

Who is online

Users browsing this forum: 33 guests