General Question Lighting Effects Regarding Earth's Terminator

[DrSolidRocketMotor]

Although I have been using (and enjoying) Orbiter for almost a full year, this is a phenomenon that I have just recently noticed, and I'm interested in any feedback from others who have flown spacecraft in similar situations.

I have seen, as I'm sure many of you have, countless NASA images of the Earth taken from orbit, as well as on the way to the moon. When viewing images (as well as motion picture and video images) that display the earth's terminator (the line that separates daytime from nighttime, for those of you unfamiliar with this term), there is always a clear, sharp boundary between day and night (given the slight scattering effect of the earth's atmosphere).

Recently, while experimenting with the Space Shuttle and elliptical orbits, I achieved an apogee (apoapsis) that was high enough to see most of the planet in a single glance. When I was about to cross from day into night, I was struck by how the terminator did not look like the one I had seen in actual photos. The line of demarcation seemed to have little curves and dips, most notably near the horizon line. When I used the 'Page Up' key to raise my viewpoint even higher to the point where the entire Earth was visible on my monitor, the effect seemed even more pronounced.

You would probably not notice this while in low Earth orbit, which may be why I haven't read any comments about it. I'm just wondering if this is something that occurs with all the planets in the simulation, or if it is the way the program maps lighting effects to the planetary texture maps.

Any comments will be appreciated!

 

[Ripley]

Are you using any external graphics client (D3D9, D3D11), or the inline standard (D3D7) engine?

 

[asbjos]

Could you please post a picture of your screen? I have problems understanding what you are seeing. A real-life reference photo could be nice too.

 

[DrSolidRocketMotor]

Are you using any external graphics client (D3D9, D3D11), or the inline standard (D3D7) engine?

Since I don't know what an external graphics client is, I would assume that I'm using the inline standard (D3D7) engine.

---------- Post added at 03:08 PM ---------- Previous post was at 03:04 PM ----------

Could you please post a picture of your screen? I have problems understanding what you are seeing. A real-life reference photo could be nice too.

Thanks for your response -- please give me a few days to create some screenshots that I believe will more clearly illustrate the phenomenon I'm talking about. I'm also going to look online for some photos to provide a baseline reference, since the only photos I have right now are in books, and I don't have a scanner.

 

[worir1]

Since I don't know what an external graphics client is, I would assume that I'm using the inline standard (D3D7) engine.

Could we have a screen shot please?

 

[HiPotOk1978]

I think I understand what he is talking about so let me take a shot.

Since cameras are not as good with light and dark in a scene as our eyes are, the terminator in pictures is bright and dark, there is no fade to black. On the Earth, your start to see dawn approaching and get very little light from the sun. In orbiter you can see the dawn's early light from orbit. Now since I have yet to go to space (here's hoping I win the lottery and can give the Russian space agency like 70 million USD to send me up there to take a look for myself), I can only guess that you will be able to see dawns light on the surface of the Earth like you can see in orbiter....

In short, I believe that Orbiter's rendering of the terminator even with the default engine is very close as to what it will look like looking out the window with my human eyes. Hope that clears it up

 

[DrSolidRocketMotor]

Could you please post a picture of your screen? I have problems understanding what you are seeing. A real-life reference photo could be nice too.

Hello everyone -- sorry to keep all of you waiting....it's been a busy week. Well, I've got some screenshots and some real photos for everyone to look at. I'm either going to provide some additional insight, or perhaps confuse the issue even further! I think I've got a better understanding of the issue now, but if anyone disagrees with me, please don't hesitate to speak!

Basically, I think the terminator in the simulation looks the way it does because I have all of the planetary effects parameters turned on. In particular, I think turning on Horizon Haze and Specular Water Reflections creates reflections within the "atmosphere" that can alter the appearance of the terminator, depending on the visual angle to the orbiter.

As a baseline reference, here are a few real photos of the earth from orbit --

I think the first image, showing a group of four pictures, compares favorably to the perspective I used when shooting screenshots of the simulation. In all of the photos, the terminator is either a straight line (when viewed from directly above), or follows a smooth curve (when viewed off center).

Now, look at these two screenshots made from an external view, with all of the planetary effects parameters turned on --


Hopefully, everyone can see what I mean. If you mentally connect the two "horns" (the points where the terminator meets the horizon) with a smooth line (allowing for the earth's curvature), you can see dark areas that seem to cross over the terminator line.

Compare the above images to the following, which is a screenshot made with all of the planetary effects parameters turned OFF --

Although this is a darker image due to the lack of land masses, I think it illustrates effectively how the terminator is rendered with all effects turned off -- although I think the simulation displays a wider terminator (I'm talking about west to east width) than you would see from actual orbit -- compare to baseline real photos.

Now, while I'm still standing on my soapbox -- nitpicking -- I would love to hear any thoughts on the following topic -- has anyone noticed how the position of the terminator as depicted in the space shuttle's earth map MFD seems to be slightly off from what you see in an external view?

I present the following pair of images for your consideration. The images were taken virtually at the same moment, so as to insure that the perspectives were accurate. In my discussion, you may need a pocket atlas to identify the geographical areas that I'm talking about.


Although the MFD map indicates that the position of the terminator is such that you should be able to see all the way to the southern tip of South America, the external view has the southern end of the continent already in darkness, as if the terminator's position was slightly west of the position shown on the MFD.

Since I have reached my limit for including images in this post, I cannot show you any more examples. I'm just curious to know if anyone else has noticed this phenomenon, and if it varies with the earth's orbital position and/or inclination.

Well, that's it. I don't mean to sound too nit-picky, but when I'm using this magnificent simulation, I just want it to be as perfect as humanly possible.

My humble thanks to all of you for listening and contributing!

 

[orb]

Are you using 16 bpp color depth (it looks like it on gradients)? If so, try if it improves when you switch to 32 bpp color depth.

 

[statickid]

I blew up the contrast on your screen shot, and as you can see, the MFD draws the line where the actual darkness starts. You can see that orbiter is rendering this as a true black by how much the background is brightened. I'm guessing you had ambient light set to zero. There is no image information below where orbiter is determining the terminator to be. I drew a light grey line showing approximately where the black starts, by comparing it to Tierra del Fuego and Costa San Blas, I'd say the MFD is pretty accurate

 

[DrSolidRocketMotor]

Are you using 16 bpp color depth (it looks like it on gradients)? If so, try if it improves when you switch to 32 bpp color depth.

Thanks for your suggestion -- believe it or not, my video card is already set to 32 bbp. If I may ask -- what kind of monitor are you using? A member of my family gave me an older LCD display, so I switched to that from my earlier CRT tube monitor -- although that really shouldn't affect producing screenshots other than image dimensions.

 

[statickid]

There may have been a loss of quality when you saved the screen shot as a jpg. you can check by looking at your own screen shot and seeing if the gradient is of lower quality than when you are actually playing orbiter, or compare it to a screenshot saved in a different format like png or high quality jpg

 

[DrSolidRocketMotor]

I blew up the contrast on your screen shot, and as you can see, the MFD draws the line where the actual darkness starts. You can see that orbiter is rendering this as a true black by how much the background is brightened. I'm guessing you had ambient light set to zero. There is no image information below where orbiter is determining the terminator to be. I drew a light grey line showing approximately where the black starts, by comparing it to Tierra del Fuego and Costa San Blas, I'd say the MFD is pretty accurate

Yes, you're correct -- I had ambient light set to zero. As for your results -- amazing! I did not even think to tweak the monitor's settings. The problem is -- I use a variety of 3D software, so I try to find a happy medium in my monitor's adjustments. But you certainly have shown me what to try next -- many thanks!

 

[worir1]

In parameters what is your ambient lighting setting. I noticed this causes a blur in the terminator. Set it to 0 and see if it helps

---------- Post added at 22:23 ---------- Previous post was at 22:21 ----------

Sorry didnt see that post above. Please ignore my post

 

[orb]

If I may ask -- what kind of monitor are you using?

Well, I use CRT (until they replace LCD with something better for the same price, or until it will break), But monitor doesn't have anything with taking screenshots (unless you take the photo of the screen ). Maybe it's the high level of lossy compression. :shrug: I simply compared your screenshot to what I see in Orbiter and such gradients are much smoother in mine.

 

[Urwumpe]

Well, I use CRT (until they replace LCD with something better for the same price, or until it will break), But monitor doesn't have anything with taking screenshots (unless you take the photo of the screen ). Maybe it's the high level of lossy compression. :shrug: I simply compared your screenshot to what I see in Orbiter and such gradients are much smoother in mine.

here a good 23" TFT costs already less than a 17" CRT of 2010. Maybe different market. The ambient light factor is likely the cause there, orbiter has a rather simple calculation by default. Maybe the other clients are more accurate by the better shader model available.

 

[FordPrefect]

The thing pointed out here is the (inferior) lighting model used in Orbiter's inline client. I believe it uses the Lambert model to calculate the illumination for the planetary surfaces.

SpaceEngine for example uses the more sophisticated/more realistic Lommel-Seeliger Lighting model.
You can see that clearly in those three unedited screenshots, SpaceEngine is just pure beauty!

 

[DrSolidRocketMotor]

here a good 23" TFT costs already less than a 17" CRT of 2010. Maybe different market. The ambient light factor is likely the cause there, orbiter has a rather simple calculation by default. Maybe the other clients are more accurate by the better shader model available.

Thanks for your input! -- I have both an old SONY Trinitron Multiscan 200ES, and a widescreen monitor made by somebody called Pandigital. They were both hand-me-downs from a family member who was upgrading to something state-of-the-art, so I couldn't complain about the price. I decided to use the Pandigital since it was a widescreen monitor, but after much tweaking in the past day or so, I think I've come to the conclusion that the Pandigital has a very restrictive range in terms of brightness and contrast. I'm using a Radeon video card with enough memory to support 32 bit, but, all-in-all, I think the SONY had a much better picture. With the Pandigital, after testing in various modes, I see "banding" happening depending upon the distance from Earth, which is something I used to see a long time ago before dedicated video cards were even around! I think this is probably responsible for creating a terminator that looked misplaced -- the monitor could just not handle the gradient in the image.

I'm also wondering if the culprit could be the software I'm using to perform screen captures -- I'm using a program called MGrab -- do you have any experience with it, or could you suggest something better?

Thanks again!

---------- Post added at 03:44 PM ---------- Previous post was at 03:30 PM ----------

The thing pointed out here is the (inferior) lighting model used in Orbiter's inline client. I believe it uses the Lambert model to calculate the illumination for the planetary surfaces.

SpaceEngine for example uses the more sophisticated/more realistic Lommel-Seeliger Lighting model.
You can see that clearly in those three unedited screenshots, SpaceEngine is just pure beauty!

I agree with you completely -- in fact, if you hadn't told me, I would have almost thought that they were actual photos!

Since I don't know the first thing about using external graphics clients, any guidance you would be willing to provide would be greatly appreciated! No need to go into great detail -- just point me in the right direction. Thanks again!

 

[hedzup456]

I believe Space Engine is a seperate sim, not a GC.
~Hedz

 

[DrSolidRocketMotor]

I believe Space Engine is a seperate sim, not a GC.
~Hedz

Yes -- you are correct. I should have checked out the website first. Although now that I have seen it, I realize that my current system is slightly underpowered to run it.

Thanks!

 

[C3PO]

But monitor doesn't have anything with taking screenshots (unless you take the photo of the screen ).

Oh yes it does. If you have a bad monitor, and tweak the settings so that it looks OK on that, a screengrab can look very different on other people's systems.