1.5.0 prerelease 2

[Johaen]

For the record, what's HDR? High Definition Rendering?

I'm not sure if your machine can run this program, but if it can, it will give you a good idea of what HDR does. Basically bright things look really bright, dark things look really dark. Also, the brightness of objects can change, similar to how bright the bathroom light is when you wake up in the middle of the night until your eyes have a chance to adjust.

Real-Time High Dynamic Range Image-Based Lighting (rthdribl)

[Johaen]

For the record, what's HDR? High Definition Rendering?

Also, read this article.

High dynamic range rendering

[buggs_moran]

I say : remove completely this glare attempt. It's really a bad effect. It doesn't feel realistic at all. We never see something like this on pictures.

Don't we?

To be clear, I'm not arguing against changing or eliminating the halos on planets; I'm just challenging you to reconsider your views on what is and is not realistic.

--Chris

But that shot was taken from Earth in Arizona. The glow we see there is mostly a by-product of our atmosphere. I am sure we can all say that on a clear night we have all seen how sharp the planets are in contrast to the stars. And, I'll agree with Fridger on the glow I see through my 6" Newtonian, but again, that is through our atmosphere and is a direct result of seeing and clarity of the sky.

Above the atmosphere I would expect that our eyes would still perceive some glow due to circular aperture diffraction in our eyes (it's the same thing that causes rings in telescope pics). But I think the "glow" should definitely drop off more quickly. The Airy disk that we should see with our eyes would never be so big as the glow that is currently presented. In fact, with my limited knowledge and a little digging, the best size of a disk I could come up with was somewhere between .45 and 1.1 arcminutes in size... With my screen resolution at 1280 pixels and my FOV at 1x (~22 deg) I wouldn't even detect a hint of an Airy disk (with my eyes)...

Of course, realistically, with my glasses off, all stellar objects have a glow around them, so "reality" is in the eyes of the beholder. My vote would be to remove the glow, or at least significantly diminish it.

edit
----------------------------------
I did just read one paper that gave the resolution of the human eye at a high of 3.9 arc minutes. You might be able to model a slight glow 3 pixels wide at 1x resolution in Celestia . However, if you dropped your FOV to 7 deg (3x) you could have a glow (Airy disk) of 12 pixels wide... So perhaps there is some argument to have some glow increase at low FOVs...

[chris]

I say : remove completely this glare attempt. It's really a bad effect. It doesn't feel realistic at all. We never see something like this on pictures.

Don't we?

To be clear, I'm not arguing against changing or eliminating the halos on planets; I'm just challenging you to reconsider your views on what is and is not realistic.

--Chris

But that shot was taken from Earth in Arizona. The glow we see there is mostly a by-product of our atmosphere. I am sure we can all say that on a clear night we have all seen how sharp the planets are in contrast to the stars. And, I'll agree with Fridger on the glow I see through my 6" Newtonian, but again, that is through our atmosphere and is a direct result of seeing and clarity of the sky.

The glow in that image is not due to the atmosphere. It's caused by light reflecting and scattering within the optics of the telescope. You can see similar effects in Hubble images:

http://heritage.stsci.edu/2003/21/big.html

Also, planets aren't any sharper than stars. They do appear more stable because their apparent size is larger and are thus less affected by atmospheric distortions.

Above the atmosphere I would expect that our eyes would still perceive some glow due to circular aperture diffraction in our eyes (it's the same thing that causes rings in telescope pics). But I think the "glow" should definitely drop off more quickly. The Airy disk that we should see with our eyes would never be so big as the glow that is currently presented. In fact, with my limited knowledge and a little digging, the best size of a disk I could come up with was somewhere between .45 and 1.1 arcminutes in size... With my screen resolution at 1280 pixels and my FOV at 1x (~22 deg) I wouldn't even detect a hint of an Airy disk (with my eyes)...

Of course, realistically, with my glasses off, all stellar objects have a glow around them, so "reality" is in the eyes of the beholder. My vote would be to remove the glow, or at least significantly diminish it.

edit
----------------------------------
I did just read one paper that gave the resolution of the human eye at a high of 3.9 arc minutes. You might be able to model a slight glow 3 pixels wide at 1x resolution in Celestia . However, if you dropped your FOV to 7 deg (3x) you could have a glow (Airy disk) of 12 pixels wide... So perhaps there is some argument to have some glow increase at low FOVs...

The paper by Spencer, Shirley, et al that I linked to earlier gives empirically derived point spread functions for the human eye, and the glare due to light scattering in the eye is clearly spread over a much wider angle than 3.9 arcminutes. However, this glare is much dinner than the unscattered component and will only be perceptible if it's from a source much brighter than the background.

--Chris

[chris]

In the CVS version I've disabled halos around planets. I also fixed a long-existing bug in point stars mode--now the planets with an apparent size of a pixel or less will also be rendered as points.

--Chris

[Cham]

Chris,

what about all the other objects around the sun ? I had a strong halo around some spacecraft too.

[chris]

Chris,

what about all the other objects around the sun ? I had a strong halo around some spacecraft too.

I removed the glare effect from all nonstellar objects.

--Chris

[Cham]

Good !

And what about the star glare changing abruptly with distance ?

[chris]

Good !

And what about the star glare changing abruptly with distance ?

I can't reproduce that on my system--stars fade smoothly to points. I haven't had a chance to see if it happens on my PowerBook yet. Hopefully it does, otherwise it's going to be hell to figure out what's going on.

--Chris

[Cham]

IMO, you should also put back an atmosphere on the stars. Having a sharp border isn't good.

[PlutonianEmpire]

Now now now, cham. be a good kitty and go play and let Master be.

[Cham]

Now now now, cham. be a good kitty and go play

MEOOWW! FSCHHHTT ! FSSCHHHHTT! FCHHHT!

[PlutonianEmpire]

Now now now, cham. be a good kitty and go play

MEOOWW! FSCHHHTT ! FSSCHHHHTT! FCHHHT!

Hehe.

[buggs_moran]

The glow in that image is not due to the atmosphere. It's caused by light reflecting and scattering within the optics of the telescope. You can see similar effects in Hubble images:

Okay, agreed over a time lapse shot, but our eyes don't work on time lapse per se. if I remember correctly we see at approx 25 frames per second so the intake of light and therefore the diffusion (scattering) of it is limited, right?

Also, planets aren't any sharper than stars. They do appear more stable because their apparent size is larger and are thus less affected by atmospheric distortions.

Understood. I meant that they "look" sharper to our bare eyes than the diffuse versions through telescopes and cameras.

[neo albireo]

In the CVS version I've disabled halos around planets. I also fixed a long-existing bug in point stars mode--now the planets with an apparent size of a pixel or less will also be rendered as points.

--Chris

Now they have too few halo

It was just not good with the halo that strong, but now I miss some halo already - I have the feeling now that the moons are not bright enough compared with the stars.
Maybe some compromise until HDRR? It's really not easy...

Bug: When leaving the earth, it first is a disk getting smaller. Then, at 0.07 AU, both earth and moon disappear totally. Then, with 2.2 AU, they disappear again as points. What happens in between?

Also, there is still the cloud shadow bug causing Celestia to close. It only happens with a big cloud texture. However, as cloud shadows make only sense with rather detailed cloud maps, I would really like this problem to be solved.

[AlexChan]

I find a problem too,
when lunar eclipse, the moon just only have 1 side of night side light, not whole moon...
Could you fix this bug?

[PlutonianEmpire]

IMO, you should also put back an atmosphere on the stars. Having a sharp border isn't good.

Hmmm... now that I think about it, how 'bout instead of giving back atmosphere to stars, we enable anti-aliasing for them? Just a though.

[Reiko]

Will this install over the 1.4.1 version? I would like to take a look at this but I don't want to lose the older version just yet.

[rthorvald]

Wow!
I am playing with the new atmosphere ssc possibilities. It is beautiful! What an insanely cool improvement!
- rthorvald

[selden]

Alex,

Are you running Celestia on a Mac? You've never said.

Exactly when and from what viewpoint do you see the problem? Do you have "ambient light" turned on, perhaps?

The moon goes completely black (as it should) when I look at it in the Windows version of Celestia v1.5.0pre2 on the night of 2007 March 3.

As seen from Greenwich, in the OpenGL 2.0 render path, the Moon starts darkening at about 20:40 UTC (i.e. first contact), is completely dark at about 22:53, starts getting light at about 23:48, and is fully lit at about 01:51 on the 4th.

In the Basic render path, the corresponding times are about 21:27, 23:04, 23:38 and 01:28.

According to http://sunearth.gsfc.nasa.gov/eclipse/lunar.html
totality is supposed to last 1h14m. Both renderpaths show it as less than an hour, so Celestia isn't getting it quite right. Yet.

Someone else will have to verify this on a Mac.