The Feature Requests Collecting Thread

[Apollonius]

Request:

An "export equirectangular image" function would be really nice, with a few (2:1) sizes to pick from, to be used in a QTVR panorama. I know that Celestia is already far better than a QTVR file, but those are nice to share with people as an example view from your addon. You could even make multi-node tours with hotspots, to really show it off.

I've made one QTVR from Celestia screenshots, but it took forever and it's still not perfect.

Seems like a cool feature and one that wouldn't be too hard to add. Do you have a link to information about the types of projections required by QTVR?

--Chris

I'm still looking for a good link, but from what I know right now, you would need either 1 equirectangular projection with an aspect ratio of 2:1 (width to height), or 6 square cube facets; one for up, down, left, right, front, and back. You can also make cylindrical panoramas, but since it's outer space in Celestia, with no horizon line, it seems a little unnecessary.

Cubic (spherical) panoramas use space more efficiently (less data, better resolution, and less distortion), but exporting 1 equirectangular image might be much easier to implement (?) Also, quality doesn't matter so much if the resolution of the equirectangular image is high enough. People could still convert it to 6 cube faces to save data space if they wanted to, losing little or no resolution.

Within the cubic format are other variations like 'T-strip', 'strip', and 'series', but those are often program-specific, and all are based on the same 6 facets anyway. For the sake of ease though, putting all 6 facets into one image might be a good thing. Again, if people wanted or needed to, they could extract the facets into 6 separate square images.

[Teto]

Sorry if the request has been already done:

- When a planet is selected, we can, by right-click, select the moons.
- But when a moon is selected, we can't 'go back'. And sometimes the moon is far away from planet, so we must type the name of the planet to select it again...
- It would be great to have, for each object from satellite/object/something to star, the parent and the child(s) selectable by right-click.

Teto.
Great software, anyway.

You can reselect the planet by typing it's number. If it's the 5th planet hit the 5 key. etc.

Thanks, Reiko. I didn't remember this feature.
But I don't like it anyway. I mean, you can go in one direction by mouse, and by keyboard for the other one? Well... And I continue to say that having the possibility to select the parent of an object/planet/star or its children(s) by right-click would be a useful feature for the user.

Teto.

[Andy74]

Well, at least by keyboard you can always select the parent object:
Just press the back-space key.
E. g. when Moon is selected, after pressing back-space once, earth is selected. If you continue pressing back-space you select sun, then solar system barycenter and finally you are in the mode when no object is selected.
At least this works for me in a WinXP system with celestia 1.6.0 or a current SVN build.

Regards
Andy

[Teto]

Thanks @Andy74.
Next time I'll read the manual more precisely... Sorry. But too much keyboard's actions to learn...

I'd like having 2 improvements, but before asking, I'll read all I can about the features of the soft...

Edit 05/nov:
Ok, I've read many documents, so one of my questions is out of date: Having volumetric effects for nebulae. It's not easy to do (I understood nothing about the format CMOD, but I have a little brain, so... ) but it's possible.
The second one is about other galaxies:
- With version 1.6, the distance limitation does not longer exist, but we have a problem of accuracy, that's ok for me.
- But the galaxies are represented by 'limited' models: we can't go inside. And when we're flying around, it looks like a small object, not a big galaxy. Few stars visible, and so on.
- If I'm wrong, don't hesitate to snap me...
- I really want to go inside other galaxies.
- I know there's a memory problem. We can't have billions of galaxies with their trillion stars, or something. But is it possible to flush the memory of the galaxy we just left, and load the new one?
- Having, for each type of galaxy, a 'standard' stars.dat. If we go inside a galaxy with no special file for the stars, we load the one for this type of galaxy.
- I know that these improvements are not really scientific because for many galaxies we don't have data on stars inside. But I think it could be great for the immersion inside this great soft.

That my point,
Teto.

[PlutonianEmpire]

Can't remember if this has been posted before, but...

One thing I would like to see is the ability to switch between distance measurements in light years and distance measurements in parsecs, or a checkbox that we can check if we want to include distance measuring in parsecs. Having to do some math to figure out the distance in light years can be pretty frustrating for some people, including me. Plus, I'm offline during the week and can only come on on the weekends, so I don't have the 24/7 ability to look up the distance on the internet.

[Reiko]

Thanks @Andy74.
Next time I'll read the manual more precisely... Sorry. But too much keyboard's actions to learn...

I'd like having 2 improvements, but before asking, I'll read all I can about the features of the soft...

Edit 05/nov:
Ok, I've read many documents, so one of my questions is out of date: Having volumetric effects for nebulae. It's not easy to do (I understood nothing about the format CMOD, but I have a little brain, so... ) but it's possible.
The second one is about other galaxies:
- With version 1.6, the distance limitation does not longer exist, but we have a problem of accuracy, that's ok for me.
- But the galaxies are represented by 'limited' models: we can't go inside. And when we're flying around, it looks like a small object, not a big galaxy. Few stars visible, and so on.
- If I'm wrong, don't hesitate to snap me...
- I really want to go inside other galaxies.
- I know there's a memory problem. We can't have billions of galaxies with their trillion stars, or something. But is it possible to flush the memory of the galaxy we just left, and load the new one?
- Having, for each type of galaxy, a 'standard' stars.dat. If we go inside a galaxy with no special file for the stars, we load the one for this type of galaxy.
- I know that these improvements are not really scientific because for many galaxies we don't have data on stars inside. But I think it could be great for the immersion inside this great soft.

That my point,
Teto.

You can have many stars in other galaxies just like in the Milky Way and not have to flush your memory of the Milky Way stars.

[duds26]

Sorry if the request has been already done:

- When a planet is selected, we can, by right-click, select the moons.
- But when a moon is selected, we can't 'go back'. And sometimes the moon is far away from planet, so we must type the name of the planet to select it again...
- It would be great to have, for each object from satellite/object/something to star, the parent and the child(s) selectable by right-click.

Teto.
Great software, anyway.

You can reselect the planet by typing it's number. If it's the 5th planet hit the 5 key. etc.

Thanks, Reiko. I didn't remember this feature.
But I don't like it anyway. I mean, you can go in one direction by mouse, and by keyboard for the other one? Well... And I continue to say that having the possibility to select the parent of an object/planet/star or its children(s) by right-click would be a useful feature for the user.

Teto.

Would definitely/certainly be useful to right-click and selecting the parent of the current selected object.
vote +1.

[duds26]

Feature request: OpenGL-based UI


It offers the following advantages:
- Cross platform Win/Mac/Linux (OpenGL ES 2 is available on a lot of mobile devices: http://en.wikipedia.org/wiki/OpenGL_ES, OpenGl ES 2 corresponds loosely to OpenGL 2 )
- only need opengl 2.0 (or maybe even only OpenGL ES 2.0 subset needed )
- fully customizable (ui elements done as textured polygons please)
- high performance, efficiency
- unified ui, familiarity (a consistent ui on all platforms)
- very flexible, not tied to a specific toolkit (and/or it's look and feel, other limitations)

Examples of ui's that use OpenGL:
There are examples of UI's in the wild done only in OpenGL.
e.g. Blender has an OpenGL-based one.

There are several tool kits based on OpenGL:
http://en.wikipedia.org/wiki/List_of_widget_toolkits#Based_on_OpenGL

Both QT can make use of OpenGL to render itself.
GTK is planning an OpenGL-backend/hardware acceleration.

There is even a sourceforge project offering a blender style opengl ui library:
http://sourceforge.net/projects/libblendgui/

Celestia would only need these things.
menu bar, button, radio button, selection box, text area, drop down boxes, spinner

Looks much but isn't such a big deal.
Those elements will be very powerful and flexible if done correctly.

[VikingTechJPL]

BRING BACK AN OPTION TO KEEP THE MAGENTA SPHERE TURNED ON WHEN YOU'RE IN INTERSTELLAR SPACE

When out in interstellar space in 1.4.1, I always liked the ability to render the Magenta Sphere of Constellation Borders. (I think some of you call it the "Death Star.") Yes, of course, the Borders are no longer valid when you leave the domain of our Sun. But the 10-light-year-radius sphere serves as a very useful "reference beacon" when you're hopping around stars less than about 200 light years from Earth. It helps to give you a sense of where you are.

Moreover, when I added the Ecliptic and Zodiac to 1.4.1's Constellation Borders, the Magenta Sphere becomes even more useful by suggesting your orientation in space. (This addition also works with 1.6.0, but only as long as Constellation Borders are on.)

Instead of 1.6.0's solution of having the Magenta Sphere disappear when we distance ourselves from Earth, could we have the option of displaying it? Out in instellar space, it's a very effective "celestial signpost" that helps take a lot of confusion out of the dark voids.

Thanks,
VikingTechJPL

[VikingTechJPL]

Draw the Part of a Planet's Orbit Above the Ecliptic in a Lighter Color Than the Part Below the Ecliptic

When the orbits of planets are displayed, can the part of an orbit above the Ecliptic be drawn in a lighter color than the part below? I used this technique in a rather simple astronomy program that I wrote about fifteen years ago, and it's amazing how much it helps you visualize 3-D orbits on a 2-D monitor. It immediately tells you where the nodes of the orbits lie, and emphasizes how much each planet's orbit is tilted relative to the Ecliptic. Even when you look down on an orbit from directly above, it conveys the orbits' tilt.

Similarly, for planetary moons, could the parts of a moon's orbit above the parent-planet's equator be drawn lighter than the parts below? This gives a good sense of an orbit's frame relative to the planet's equatorial frame.

It would also be useful to be able to toggle perihelia and aphelia (or, more generally, any periapsides and apoapsides) on and off, though that is fairly easy to implement in a script.

Taken together, all of these things go a long way toward conveying an intuitive sense of a body's orientation and orbital elements.

--VikingTechJPL

[Hungry4info]

This would only work in our solar system.

How, for example, would you define the ecliptic in multiple-planet systems for which inclinations are known (Gliese 876 b, c, PSR B1257+12 B, C)?

[VikingTechJPL]

Re: Draw the Part of a Planet's Orbit Above the Ecliptic in a Lighter Color Than the Part Below the Ecliptic

Greetings, Hungry4Info,

Yes, I'd favor this feature being confined to our own Solar System at this point for a number of reasons:

1) the orbital planes of so many exoplanets are unknown, and
2) astronomy's favored status of the Ecliptic is only the result of our happening to be here on Earth. Everything is relative, and things are going to get complicated when people start growing up on Mars or, worse, on a moon like Triton. Can you imagine the conversations?
Astronaut 1: "Are you talking about Earth's Vernal Equinox?"
Astronaut 2: "No, I'm talking about Mars' Vernal Equinox!"
Astronaut 3: "I thought you were talking about Triton's Vernal Equinox!"
Astronaut 2: "Does Triton have an Equinox? Does it even HAVE a Vernal?"

Regarding the orbital planes of exoplanets, I like to visit rho1 Cnc and watch the many transits that you can see from its outer planets. But I have to remind myself that this is probably not accurate as, in our own Solar System, no two planets share the same orbital plane.

Here are a few examples of what I'm suggesting:



Incidentally, the Vernal Equinox would be essentially toward the upper left corner in both pics (moreso in the bottom one).

As you can see, the dual-color orbit lines instantly convey a sense of depth and some important information:
1) that the "upper" (most ecliptic northward) parts of all of the planets' orbits lie within a limited area of the sky; the average of their "upper-most points" seems to lie roughly in the direction of the Fall Equinox; (I never knew this before I plotted the orbits this way);
2) that Mercury's nodes are quite near its perihelion and aphelion;
3) that dwarf-planet Pluto's nodes are NOT near its perihelion and aphelion; (I wrote this WAYYY back in the days of EGA 16-color graphics, when Pluto was still a full-fledged planet; EGA graphics were around probably before many Celestia-users were born);
4) in the Moon's case, the dual-color orbit lines tell whether the lunar-orbit nodes are pointing toward the Sun (when that happens we look for eclipses, of course);
5) in the Moon's case too, this would show how its orbital tilt rotates around the Earth.

Anyway, these are just suggestions. As some anonymous writer once said, "There is no shortage of ideas, only the time to implement them all." Ain't that the truth.

Regards,
--VikingTechJPL

[Hungry4info]

So what side of the orbits of extrasolar planets would you make brighter? And why?

Even extrasolar planets aside, what if someone wants to make a fictional system around a star? How does Celestia choose the ecliptic for those systems?

I'm not sure how this would work anywhere than in our Solar System.

[VikingTechJPL]

Hungry,

Your last post seems to suggest that you don't think I'm agreeing with you. I do agree that drawing orbits this way and using the Ecliptic as the reference plane really only makes good sense in our own Solar System.

If you're asking what should be done at other star systems—well, your guess is as good as anyone's. Using 1) the parent star's equatorial plane, or 2) the average plane of the star's known planets' orbits both present problems.

In the first case, what happens if we find planets around a double star whose components have non-aprallel axes? Is that even possible? If so, which star would you use? Or do you use the plane of the binary's orbit?

In the second case, every time we discover a star's new planet we'd have to recalculate the reference plane for the star. GROAN, what a pain that might be. Wouldn't want to have do that.

Though it sounds like too much work for the Development Team, what would really be awesome would be to have the ability to draw the dual-color orbits relative to the orbit of any other body in the same system. And then cycle through them. Ah, so many ideas, so little time!

You sound like you do a lot in Celestia with exoplanets. What would your suggestion be?

VikingTechJPL

PS What about using the plane of the first "habitable" planet found in a system? Uh oh, but what happens if there IS NO habitable planet? I'd better let you "exoplanet enthusiasts" make that decision.

[Hungry4info]

I apologise. I did not mean to imply I thought you were disagreeing, I just wasn't sure your previous post addressed the extrasolar systems issue fully.

I like the idea of the ecliptic being defined as the average of the inclinations, yet one can see with your graphics that it is not so (While we're on the subject, what is the inclination of Earth's orbit relative to the solar equator? I can't seem to find it anywhere). And yeah, finding new planets in known systems would mess that up quite a bit.

For the binary star system question, I'm assuming you're asking about circumbinary planets like those at HW Vir and QS Vir as opposed to the multitude of wide-binary systems in which one component is known to host a planetary companion. I honestly have no idea. It may even be that the inclination of a coplanar planetary system would be different than the inclination of the close binary itself (such that an observer on such a planet would not observe star-star transits throughout much of the year).

What would your suggestion be?

As the ecliptic is defined for Earth's orbit, I don't think it's something that can easily be defined for other planetary systems, due to complications you pointed out. The ecliptic seems to be a very Sol System concept.

But it doesn't feel right, making something apply only for one planetary system. It might damage the Universal feel of Celestia, making one system (ours) unique like that.

While I really like your idea of shading the orbit fraction below the ecliptic, I would think it would be more consistent to not do so.

I'm only one mind. I am curious to hear what others think.

[VikingTechJPL]

Re: Draw the Part of a Planet's Orbit Above the Ecliptic in a Lighter Color Than the Part Below the Ecliptic

REVISED: Re: For OUR Solar System, Draw the Part of a Planet's Orbit Above the Ecliptic in a Lighter Color Than the Part Below the Ecliptic

Hungry,

I found two references to the Sun's axial inclination being 7.25?, i.e. it's Equator is tilted by that amount to the Ecliptic.

Furthering our lively deliberation, I'm a little curious about why you don't think our own Solar Sytem should be treated differently from other star systems, since so much of astronomy is Solar-System-centric and based on the Earth-Sun relationship: the equinoxes, solstices etc. And the IAU defines even the phases of the Moon by the relative "Ecliptic" longitudes of the Sun and Moon. So, I disagree with you (respectfully, of course, an in the good spirit of improving Celestia); I don't think its "Universal Feel" is damaged because Celestia shows some traits or detail for our Solar System and not other star systems. For one thing, what we "really know" about other star systems is pretty thin and subject to much revision. And our own Solar System is already "privileged" by having is own set of Constellation Borders.

Anyway, being very intested in Solar System dynamics, I obviously favor the dual-color orbit feature, though I do think it should be able to be toggled on and off.

Thanks for the spirited debate. I too am eager to hear if others are interested in such a feature, and if the Development Team thinks it's worth implementing it.

Regards, and see you out at the exoplanets,
--VikingTechJPL

[chris]

While Celestia feature requests are appreciated--especially detailed ones with images--this thread is not the right place to put them. Start a new topic with 'Feature Suggestion' in the title instead, for example, "Feature Request: Show Celestial Coordinates for Objects" This way, all discussion related to the particular feature is located together in a single topic instead of intermingled with discussion about a dozen other unrelated features.

--Chris

[VikingTechJPL]

Point well taken, Chris. I'll certainly do that next time.

Is there any way to break up this current long thread into its component topics? Is that something an administrator has to do?

--Gary

[Hungry4info]

I found two references to the Sun's axial inclination being 7.25?, i.e. it's Equator is tilted by that amount to the Ecliptic.

Ah, I appreciate that. I hadn't thought to look for the sun's axial tilt, but searched for "Earth's orbital inclination" instead.

I'm a little curious about why you don't think our own Solar Sytem should be treated differently from other star systems, since so much of astronomy is Solar-System-centric and based on the Earth-Sun relationship: the equinoxes, solstices etc.

Long ago, there was just Earth and the sky. The stars were nothing but dots in the heavens. Nothing nearly as special as our sun. Yet now with a more advanced perspective, we know they are stars, and our sun is but one voice in a choir, so to speak.

I bought a 3D Space sim, I forgot what it was called, but it had several thousand galaxies in it (Deep Space Explorer?). Around Sol, you could visit a few of the closest stars (galacticaly speaking of course), but they would not show up like the sun. Regardless of how close you got to, say, Alpha Centauri, it was just a dot. The sun was rendered fully as a sphere. When I found Celestia (and OMG it was free too!), I instantly fell in love with it because I could go to the other stars and they would show up as spheres. Suddenly, they seemed like actual places, not just dots.

Not showing the Solar System as unique is very important to me because I don't think it's unique at all. Since 1995, we've been more and more able to put our solar system in context with extrasolar planets. Ours is not the only solar system, but one in a multitude of systems that we're beginning to find. This is a perspective I think is best represented by our solar system just being another star circled by just another set of planets.

You're certainly right in that we know very little about extrasolar planetary systems. I liken this to us not knowing anything about stars other than our own back in the day. Eventually, we will know much more.

[quote=VikingTechJPL"]I don't think its "Universal Feel" is damaged because Celestia shows some traits or detail for our Solar System and not other star systems. ... And our own Solar System is already "privileged" by having is own set of Constellation Borders.[/quote] While I never use Celestia's constellation features (save for the occasional use of Celestia as a sky chart), I admit you have a very good point here.

I obviously favor the dual-color orbit feature, though I do think it should be able to be toggled on and off.

I, too, would support it if it could be toggled on and off (that seems to be the fix for any differences in how features should be implemented haha).