Apparent diameter of Moon from Earth incorrect

[Size_Mick]

I know a lot of you will read this and just think, "pfft not this again" and write me off as some kind of moron. But I wouldn't post this just to draw flames. I am totally serious when I say that I think the apparent diameter of the moon in Celestia (as viewed from Earth) is only about half the size of the way it looks in the real night sky. I had brought this point up before only to be told that it was all just an illusion. However, I was not convinced of this and have since been looking through the night sky, waiting for the moon to be in some position where its size could be compared by distance from certain stars, etc. and I have since become absolutely convinced that its apparent diameter is wrong in Celestia.

My latest check is against two stars, Algieba and Regulus. Tonight the moon was very close to both of these stars, one being on the lower left and the other on the lower right of the moon from my point of view. As soon as I noticed this I straightaway went inside and called up the same view in Celestia, and sure enough, it wasn't the same. The moon was a paltry diameter in comparison to the diameter of the real thing outside, when measured against how much distance it covered between the two stars.

Now, I hope I don't have a bunch of people replying to tell me how wrong I am, when it's easy enough to try it yourself and see. What I'd rather get in way of reply is that someone is going to actually check it out for themselves before telling me that there's no way I could be right.

Either way, I won't bring up the issue again. I just urge the developers or anyone who's heavy into this Celestia stuff to try the same experiment and see what I'm talking about.

[chris]

My latest check is against two stars, Algieba and Regulus. Tonight the moon was very close to both of these stars, one being on the lower left and the other on the lower right of the moon from my point of view. As soon as I noticed this I straightaway went inside and called up the same view in Celestia, and sure enough, it wasn't the same.

You must have been looking at different stars in the sky. Algieba (Gamma Leo) and Regulus are over eight degrees apart, whereas the Moon is only about half a degree in diameter. There's no way that the Moon could have been as close as you're implying to both star simultaneously.

--Chris

[Size_Mick]

You can figure out which stars they are. They formed a pretty even triangle with the moon around the time I saw them. I live in South Florida, where there is a great deal of haze and city lights, so these two stars obviously stand out from the rest. That is how I arrived at the conclusion that these stars were what I saw in Celestia also. I lowered the magnitude limit until basically just those stars were visible, as in real life.

After agonizing minutes of right-clicking to get these stars selected (any way to make this easier with the mouse that I'm not aware of, like holding down a magic key and dragging a box around the star I want to select?) to get their names, that was what I came up with.

I'll send you the ss in an email so you can see what I'm talking about and judge for yourself.

[granthutchison]

You're reporting a rather vague observation, since you don't really know the names of the stars involved, and can't be precise about the distances. To get this right, we need to subject Celestia to a very precise test, and those of us who've been disagreeing with you are doing so because we know Celestia can pass such tests easily - we have tested it against observation, and it works. But this is going to be my last post on this topic, too - if this doesn't convince you, there's no point in further discussion .

Go to Fort Myers, Florida (Lat 26.65, Lon -81.85). Set the time and date to around 08:10 17 July 2003 (UTC, which is equivalent to 04:10 EDT). Centre and track Mars, and then watch for the next few minutes while the Moon slides across and partially (but never completely) hides the planet.
Now go to this site, a space.com article which talks about that Mars occultation, and look at the map. Fort Myers lies in a very narrow band which was the only place from which partial occultations like the one you just simulated could be seen. Even a minute error in the size of the Moon would have prevented Celestia from simulating the event with accuracy. The Moon is the right size.

Grant

[maxim]

Well, I watched the other thread and smiled about the discussion on the psychology of moon sizes. Last fullmoon when I stood outside, watching the sky I remembered that thread and did a measuring -

- you now: straight arm, thumbs up and pinpointing the moon

and it was indeed about half the diameter of my thumb. Then I went inside and did the same with celestia on a 19', 1024x789, default FOV,
and it was less than a quarter of my thumb - even pressing my thumb against the glass didn't help much. Then I took a ruler and did the same, with the same result. So the default appearing of the moon in celestia seems to be a quarter of the real one - for what reason ever.

maxim (on a cold night )

[galileo]

the best way to do this is to measure it against the stars as Size_Mick was doing. I've got another astronomy program which is deadly accurate for this type of dealy.. it has an option to "enlarge moon" which makes it look like it does out my window.. turning this off, it shrinks to about a quarter of that.

i'm very interested in this.. I'll see what kind of experiments I can concoct and get back to y'all with the results. It'll probabl be a few days before I can see an actual moon outside, though.. (darn clouds)

edit: also remember when trying to get an accurate spacial estimation of the size of the moon, close one eye.. the illusion of its largeness almost immediately disappears.

[maxim]

...it has an option to "enlarge moon" which makes it look like it does out my window.. turning this off, it shrinks to about a quarter of that.

Does your program explain why it has this option? Sounds that this seems to be a common 'problem'.

maxim

[granthutchison]

So the default appearing of the moon in celestia seems to be a quarter of the real one - for what reason ever.

All this tells you is that the default FOV doesn't match the apparent vertical size of your monitor at your arm's length. Zoom in until the Moon fills the screen vertically, and you'll see that the FOV comes down to half a degree, which is correct.
Your observation doesn't have any relevance to Size_Mick's point, in which he maintains that the Moon in Celestia is the wrong size relative to the stars.

Grant

[galileo]

yeah the program has the option when viewing at large FOVs (the default FOV is 100˚) but restores the size when you zoom in. it does this to make it easier to see the moon phases, and this way the moon looks more "correct" at this large FOV. remember that you're looking at a very large FOV given that you're looking through a monitor probably a good quarter-half metre from your face. the actual FOV you should be using is probably closer to 25˚ or so, but then it would be too restrictive.

try this.. make a frame roughly the size of your monitor, hold it the distance of your monitor away from your face and take note of the spacial points of various stars.. try first to fit different constellations in it, etc.. outside.. then set celestia up so that the view is mimicked (centre the constellation you used for reference, and zoom in until it fits the same way)... you should now have what FOV you're actually looking at, and should know why the moon looked so small before.

i have verified time and time again through observation with binoculars and telescopes whose FOVs and magnifications were well known.. the moon is roughly half a degree worth of sky, regardless of how it may appear to the unaided eye.

[maxim]

Zoom in until the Moon fills the screen vertically, and you'll see that the FOV comes down to half a degree, which is correct.
Your observation doesn't have any relevance to Size_Mick's point, in which he maintains that the Moon in Celestia is the wrong size relative to the stars

Well, every technical aspect in celestia is obviously correct. It's the correct diameter, distance from earth, eclipse shadows look exact as the real ones - as you stated. I don't think it's a problem of inaccurathy (is this correct spelling???) inside celestia.

All this tells you is that the default FOV doesn't match the apparent vertical size of your monitor at your arm's length.

To the arm's length: that doesn't matter as it is a relative measurement. It only has to be the same distance from eye to pin for both measurings, and that is a simple way to ensure that - could also be a stick(or a geographer's equipment)

To the monitor: From certain (different) distances every monitor (even moviescreens) occupies the same FOV. If you show a picture with the same FOV then, all have the same size from your viewpoint. It should be possible to approach closer and have the shown object appear bigger than in reality. If you can't get any closer, and the object is still smaller than the real one, your FOV is obviously too wide.

any counter-arguments?

maxim

[granthutchison]

any counter-arguments?

A person with a shorter arm than yours will (presumably) have a smaller thumb. So if they perform your experiment out of doors, they'll also find that the Moon is half the width of their thumb (because their smaller thumb is proportionally closer to their eye, and so subtends the same angular diameter).
But when they come indoors and press their outstretch thumb to your screen, they'll be standing closer to it (because of their shorter arm), so it will subtend a greater FOV for them. Their smaller thumb will cover less of the screen, so they'll find the Moon is larger in proportion to their thumb than you did. So they'll report a larger Celestia Moon than you found.
As I said - it depends on your screen and your arms.

Grant

[maxim]

But when they come indoors and press their outstretch thumb to your screen, they'll be standing closer to it (because of their shorter arm), so it will subtend a greater FOV for them. Their smaller thumb will cover less of the screen, so they'll find the Moon is larger in proportion to their thumb than you did. So they'll report a larger Celestia Moon than you found.

Hmmm... so you think it's a problem of the projection plane?

When both stand away one meter should they both report the same size?
I started with one meter distance - I only came closer to force the moon to appear big enought - which didn't work.

maxim

[granthutchison]

Hmmm... so you think it's a problem of the projection plane?

When you compare the real Moon to your outstretched thumb, you're effectively projecting the Moon on to a plane the same distance from your eye as your thumb. So what you need to do is go out and find out how far from your eye your thumb has to be so that it looks a little more than four times the width of the Moon (to match what you found in Celestia when you placed your thumb on the screen). That distance is how close you have to sit to your screen for the Moon to be half a degree across at your eye (the correct apparent diameter).
Trouble is, when you're that close to the screen, your brain knows you're focussing on something nearby, and so it knows how tiny the Celestia Moon really is - it feels small. With the real Moon, your eyes are focussed on infinity and so you judge it to be much bigger.

Grant

[t00fri]

This issue is really getting a bit boring, since the people who do not believe the Celestia moon size apparently have problems to produce /precisely reproducable/ references.

Below, I have compared the moon today at 22:39 UT from Hamburg: on the lhs, Celestia 1.3.1CVS and on the rhs. Xephem 3.6. I have worked for 11 1/2 years together with Elwood Downey on Xephem to get things right!:lol: Note that the (bluish) star right below the moon is 42 Leo and the one to the left HD 89906.

Does anybody see a discrepancy?

Bye Fridger

[Size_Mick]

...and the ones I saw in the sky are definitely Algieba and Regulus. If anyone cares to host a pic, I'll gladly email a screenshot to them.

If anyone cares to give me tips on how to use either a home video cam or a digital cam to take a decent picture for comparison, I'll do my best to repeat this experiment properly.

P.S. I will speculate on one theoretical possibility about this all being an optical illusion. I don't believe it, but to me it seems the most likely candidate of the "optical illusion" theories. The apparent size of the real life moon is magnified (to our naked eyes) because of the brightness of the sunlight reflected off of it. This would explain why it looks one way to the naked eye but another in a photograph or on an accurate computer simulation. The only problem I have with this theory is that the moon doesn't simply look larger in real life (compared to distances between nearby stars), it looks *much* larger, like 2x larger.

[galileo]

a funny thing, that.
i can host pics for ya. e-mail 'em to galileo at mikeeheler.ca

man, i totally forgot about xephem. definitely the best program to use for comparisons.. that's one intense app.

one big problem with celestia is it doesn't really have a proper ground-based mode.. i would love to see this in future versions.. it's undoubtedly the most beautiful space app i've seen, now if i can actually use it for something scientific...

[Size_Mick]

try this.. make a frame roughly the size of your monitor, hold it the distance of your monitor away from your face and take note of the spacial points of various stars.. try first to fit different constellations in it, etc.. outside.. then set celestia up so that the view is mimicked (centre the constellation you used for reference, and zoom in until it fits the same way)... you should now have what FOV you're actually looking at, and should know why the moon looked so small before.

i have verified time and time again through observation with binoculars and telescopes whose FOVs and magnifications were well known.. the moon is roughly half a degree worth of sky, regardless of how it may appear to the unaided eye.

Yeah I thought of that already. It may help to refer to my earlier thread in the Development section:

http://www.shatters.net/forum/viewtopic.php?t=3858

BTW guys, thanks for all the replies thus far. You really have me convinced -- until I step outside again. Then I see the moon in relation to the stars and I think, naaaaah they're still wrong! Not to sound like a hard-headed person, but I'm really gonna need direct photographic evidence to really be satisfied. BTW I just found out that I *do* have access to a reflecting telescope (a cheap one but good enough to see Saturn's rings so I'm told), so if anyone can tell me what I need to do to get a good picture of the moon and some background stars, I'll get to work on it. Please note I only have access to a digital cam and a home video cam -- though I'm hoping there's something I can do with long exposure on the digital cam. Perhaps a filter on the telescope that will cut down the moon's light so that the stars can be picked up better?

[Kendrix]

I've just checked the moon (since it's in conjunction with Jupiter tonight) and there is no problem at all !!!

I have taken a shot of the real moon and a screenshot in Celestia.

Since my picture is in a quite bad quality to measure exactly the diameter of the moon the ratio don't match exactly but nearly !

So I can't see the pb !!!

If you want the photos just email me !

kendrix@wanadoo.fr

I think that the major problem is that some people doesn't understand what is the FOV in every 3D application !

Draw two lines : from your eye to the top of your screen and from your eye to the bottom of your screen.

The angle between these 2 lines is the FOV you should set in your 3D application to match the "reality"...

But well, granthutchinson has already explained it perfectly !

[Size_Mick]

Well, I'm sort of embarrassed to say this, but in a way I'm glad to say it. Tonight the moon was very close to Jupiter, and using my trusty bic lighter and a long armreach I was able to approximate how many moons away Jupiter was. I checked it in Celestia and it was correct.

Therefore I will make one further question/request and drop the subject:

Is there a way to control the brightness of reflected light on planets? Perhaps that way you could possibly simulate the same effect that I believe must be causing the moon to sometimes look bigger in real life, to the naked eye. I'm not sure if I am making this idea clear, though: it isn't about how white/bright the moon gets in Celestia, but more about the glow that sort of surrounds it. I'm guessing an atmospheric haze might be partly to blame for this. If there isn't a feature like this, can this be added in for a visually appealing but entirely optional effect?

Thanks again to everyone who has shown so much patience with me!

[jestr]

Could it not also be magnified by the atmosphere in real life,but not in Celestia,Jestr