As I've already shown (with illustrations) the moon has sufficient surface pattern to clue your brain in that it is spherical, yet they don't work, because the overwhelming effect is that the moon is flat because of its remarkably uniform surface brightness, at least when it is full, due to its highly Lambertian nature. Steopsis is irrelevent, there have been many pictures posted in this thread that allow your brain to discern their shape, and none of them involve steropsis. No object at the distance of the moon would show any steropsis, but a much less Lambertian heavenly body would show limb darkening every when full and appear to be spherical.

Shading from a single light source isn't as powerful a cue as you're making it out to be, at least not the component that comes from a non lambertian surface with a single light source. Texture and patterning is more important that the Lambertian nature of the surface. The research I posted earlier backs that up.

You (or someone else) complained before about a side on image of the moon, that looked very spherical. Well, here's your plain white sphere again with frontal lighting, in a high quality image that can't benefit from banding. Also a patterned sphere for comparison with absolutely no shading.





They both say "3D", but in different ways and neither are particularly strong. To me the first one looks slightly more "real" (because we're used to shading IRL), but the second one leaps out of the page more. If I put a tiny bit of shading on the second surface (I don't know how), it would look a truckload more 3D than either the pure shaded one or the pure patterned one. As you can see, there are multiple contributors to 3D appearance, and without other other cues shading alone isn't terribly strong (as the research I posted earlier backs up). Now compare this with the purely Lambertian surfaces I posted earlier:



Clearly, Lambertian or lack of is not the major differentiator here.

Your lambertian blocks are side lit, Phil. Like I said, light them from the front and tell me what they would look like.

Also, the first sphere above is obviously spherical, and the second sort of looks like a fisheye view of bacteria, really.

This doesn't look particularly "more" spherical than the first one to me:

We must process 3D imagery differently or look for different cues. The second one looks far more spherical than either of the two I posted, to me.





Anyway, thanks for humoring this conversation. It must be differences in perception shared by me and Stu but not others in this thread.

Phil -

I work with spectralon every day. I'm immensely familiar with it. I have LabSphere (the only producers of Spectralon) on my speed dial.

I would like to point out that your pictures have no spheres in them. They also aren't illuminated by a point source.

In order for THE MOON to look flat, it needs to be Lambertian.
In order for A SPHERE ILLUMINATED FROM A SINGLE POINT to look flat, it needs to be Lambertian.

If you want to compare apples and oranges, go ahead, keep doing it. If you want to compare apples and apples, you'll realize that the moon looks flat because it's Lambertian.

Josh

Actually, in order for ANY non plane surface to look flat, it needs to be Lambertian. It also requires all of the other conditions I listed, there's nothing special about Lambertian-ess that makes it different from the other conditions. You're confusing necessary, sufficient and contributing.

In order for a uniform sphere (or something dang close) that is illuminated by a point source to appear flat, it needs to be lambertian. Also, it being lambertian is sufficient.

You can talk about planes and cylinders and objects illuminated by something other than a point source all you want. However, for a uniform sphere illuminated by a point source, a lambertian surface is both necessary and sufficient.

If you disagree that it's sufficient, give me an example of a uniform sphere illuminated by a point source that is lambertian that does NOT appear flat.

If/when you can't do that, please just admit that being Lambertian is sufficient. It seems that you've already agreed that it's necessary.

Josh

yeah i play all sports fine, and reasonably well.

The moon is NOT a lambertian reflector
A lambertian spere does NOT look like a disk when it is illuminated from the direction of the observer, but rather shows darkening to the edges giving a visiual cue of depth.
The moon is a rough surface and THAT is the reason why it looks like disk
Light reflection off from rough surface is totally different than from smooth lambertian surfaces.
You can easily do the experiment yourself by the way:
Shoot a smooth round diffuse surface in near darkness with a flash on your camera (paper wrapped around a roll)
Now do the same with a rough round surface (white towel wrapped around a roll)

I find that even through a pair of 7x magnification binoculars, the moon looks spherical. To the naked eye, it looks like a disk. That may be because even with the binoculars, you can see some shadowing from craters and mountains that give some perspective, whereas with the naked eye, you can't

I know this is an ancient thread. But this is probably the only topic I could find regarding this question.

J-dub is wrong! Moon is not Lambertian

From wikipedia:

https://en.wikipedia.org/wiki/Lamber..._and_radiators

Which says:

Haha 7 years later, much before my time in SMP to remember this.

Moon doesnt appear flat if one is careful. Look at craters. Near the edges the craters look like ellipses, which is only possible if the local surface is at an angle other than 90 deg, exactly as you expect in a curved surface.



The reason the moon, moreover that detail, looks a bit flat is because it doesnt have prominent features ie the largest mountains are only 1/1000 of the radius which is impossible to detect at such detail. The same is true for earth disk when you look it from the moon or even when you look down on earth from 30-50km height (eg a space balloon or a spy plane or space shuttle when looking directly down not the sides) at the ground features (it appears very flat and mountains no longer make a big difference in the ground topography in terms of relative depth).

Look here what i mean (the volcanic mountains are visible but not as severely prominent as you can imagine they must look from lower heights)



or even the great Himalayas



Furthermore the moon is observed with zoom lens to fill the screen. You have the same effect you have when you observe a person up close vs with a zoom lens say 300mm. If a person is 3 meters away from you their top of their head is at a distance of 3m+0.1m and the nose is say at 3m-0.04m if the eyes is the 3m distance say. So various features appear at variations of the order of 1/30 from the avg distance. If you hold an object eg an apple at 20cm away from your eyes this relative distances variation may be even smaller like 4cm to 20 cm or 1 in 5. If you take the apple or the human a distance that is 10 times larger away than the above example and look at them with zoom lens they will appear flatter too. The relative distance variation will become similar to what it is for the moon. For the moon for example the various features/points in the edge of the disk are only at best 1700km (moon radius) further away than the front part that is about 384000 km away from us the observer. So you have a 1700 vs 384000 relative distance depth of 1/225.

The objects observed at such distance eg with telephoto lens will appear to have a shallow depth of field now. They will appear to be closer to each other in relative terms than they would if observed from nearby.

Eg see here;

vs

The issue is how it looks to the naked eye - like a disk. I am fairly sure that everyone is aware that we now know it is spherical because we can see that it is (not using the naked eye).

I believe that the LACK of stereopsis on the moon, in a world where pretty much everything else we see is subject to stereopsis, contributes to our tendency to perceive it as flat.

But mostly it's the lambertian-ness.

164 posts and 0 chance of "Because you aren't standing on it."

wtf you been, Borodog?

Does this account for the fact that you're also seeing more surface area per square unit of arc at the limb as compared with the center?

is there also an explanation why the earth looks flat?

From a ground perspective you are not tall enough. But if you are a good observer you can note that on a calm ocean the top portion of a large boat (in the old days the mast and sails) will appear first on the horizon. Make a logical inference from there.............

Speak for yourself.