arrgh, the second paragraph should read “…consistent with a photo taken during the day when the moon is very low in the sky.”
Well, the way I see the moon the illumination direction is about 10-15 degrees above the horizon, not high in the sky. That is is consistent with a mostly full moon which can be near but above the horizon while the sun is in a similar position opposite.
Two reasons for low contrast - no distinct shadows - on the mountain. It appears distant because of atmospheric hazing. That’s why distant mountains looks less distinct and also bluish. The sun is also fairly low in the sky giving that soft “golden hour” light so desirable for portraits. The subject is illuminated by diffuse sky as much as by direct sun.
I hold my position that this may be a completely unmanipulated shot. I will attempt one similar at my earliest opportunity.
Oh, that’s the shot. I’m almost certain (99.94%) that it’s a composite. There’s no period of twilight or dawn where the exposure of the moon and an unlit mountain landscape would be close enough to yield detail in both. And the ND graduated filter is pretty much straight out, as the exposure of the sky is completely even, with no tell-tale signs of a graduated density filter being used.
I cannot think of anyway to take this picture and have it perfectly exposed. I mean, the mountain is in the shadow. The moon looks like it’s probably exposed one stop (maybe two stops) over the “sunny 16” rule for it. That puts us, at best, at 1/125 sec at f/8 at ISO 100. No way you’re getting any detail out of a shadowed mountain at that exposure.
What you’re both missing is that cloud patterns are clearly visible in the picture that do not cross the face of the moon, even thought there seem to be clouds in the dark of the moon (the unlit portion of its face). The moon appears to be in front of the clouds, therefore the shot must (IMO) be fake.
Are we looking at the same picture? I’m referring to the link originally posted by Colpholon. I don’t see any clouds over any part of the moon. In any event it may be perfectly normal to see faint clouds in front of the unlit the moon but not see them in the lit portions because in the case of this digital photo the bright portions are in saturation.
“Shadow” is a much more complex thing than you may realize. At noonday sun the difference between direct sun and the umbra shadow (the darkest part) is extremely high and difficult to photograph. Don 't forget that “shadow” doesn’t mean absolute dark, it means illuminated by light scattered over the sky and other nearby objects. When the sun is near the horizon things are much different as direct sun is filtered and scattered through a much greater thickness of atmosphere. This means that the ratio of illumination from full sun to shadow is much lower and the edges of shadows are very indistinct if they can be made out at all. Add to this that the mountains are really, really far away so the color blends into the blue sky which itself is an indicator that the sun is still providing a lot of light.
Don’t take my word for it. Do this demonstration for yourself. Take a familiar object out in the noonday sun, say a ball. Notice the very sharp demarkation between light and shadow. Observe the same object when the sun is close to the horizon. The difference between bright and shadow is much less and the transition is much less distinct. I’ll take some test shots to show you what I mean but it may not be until Friday.
The human eye is a really poor light meter becase it auto adjusts so quickly. Dusk light is dimmer than noonday sun but not as much as it appears bcause you don’t see such high contast ratios.
We are looking at the same picture. And I stand by my conclusion that I think it’s most likely a manipulated image.
I would expect a picture similar to this, or this given the apparent lighting conditions of that wallpaper photo.
I’m not going to go so far as to say it’s impossible. I’ll take my 99.94% figure back and settle on 95%. But it looks like a typical double exposure to me, and I’d be extremely surprised if it wasn’t. I really would expect for the foreground to be much more in silhouette at that exposure or the moon to be completely blown out.
pulykamell, those photos were taken under vastly different conditions from the photo we are discussing as both show a completely full moon which indicates it is almost directly opposite the sun. That means when the moon is up the sun is already near or below the horizon which means terrestrial objects are much darker. The OP image we are discussing shows the moon in a gibbous phase which means both it and the sun can be quite a bit above the horizon at the same time.
I reiterate that I cannot conclusively say it isn’t a manipulated photo but that I don’t think it is. That photo can only be taken at specific times. I’m taking a stab and saying it’s a morning photo rather than evening which, if I am reading the chart correctly, can only be taken during the waning gibbous phase and then only during a brief time window in the morning. Too early and the mountain will be too dark, too late and it will be more ovbiously sunlit.
http://www.enchantedlearning.com/subjects/astronomy/moon/Phases.shtml
Look at the exposure table here. It basically agrees with my rule of thumb (1/60 or 1/125 @ f/16 @ ISO 100). For a gibbous moon, it suggests you open up a stop. So we have 1/60 @ f/11 @ ISO 100. Now, barring any substantial atmostpheric haze, the moon is not going to be darker than this. It physically can’t. It’s lit by a constant, steady light source. We don’t really have to guess too much about the moon exposure because we can confidently say what it is.
Now, an object in deep shade (such as the mountains in question) will be at least 3, more likely 4, f-stops darker than the moon. I can’t see any way otherwise. The only time the moon and foreground will have the absolute same exposure is in full sun, or possibly even half sun. This is clearly not the case here. I do not see how it would be possible for both foreground and background to be properly exposed. It physically does not make sense. The only hesitation I have is atmospheric conditions that would stop down the moon. That’s the only reason why I won’t say with 100% certainty it’s a composite.
Well we haven’t come to fisticuffs yet but I think people are taking bets.
1 - The moon is lit by a constant source but it can vary in apparent brightness depending on position in the sky. You can readily observe this with the sun. When directly overhead the path through the atmosphere is shortest and it is too bright to look at safely with the naked eye. When the sun is just above the horizon the path through the atmosphere is much longer and the apparent brightnes of the sun is dim enough to look at directly. Please feel free to correct me if you think the same does not apply to the moon. Remember, your eyes suck as a light meter so the apparent brightness of the mooon is hard to detect without a narrow angle spot meter.
2 - I’m not sure how your are applying the chart. It appears it is for taking photos of the moon not of things illuminated by it. It’s also a bit misleading as the chart appears to give a rule for average exposure but remember the lit part of the moon doesn’t vary in brightness with phase.
The mountains however are illuminated by the sun which is still in the sky as indicated by the direction of the lit portion of the moon. The mountain doesn’t appear directly lit with clearly defined shadows as most light is coming from the sky which is diffused and very little from the virtual point source of the sun. This is the effect photographers get from using an umbrella. The same exposure value looks dimmer than a shot taken with a small flash reflector because you don’t have such distinct shadows to compare the bright part to.
3 - The moon and mountain are not in the same exposure range. The moon is near overexposure as most of the white parts are at saturation. The mountain is significantly darker than the sky which is itself much darker than the moon.
“Proper exposure” is not a single number from a meter. If one metered on the mountain the entire moon would be in overexposure with no detail. If the meter reading was taken of the moon the mountain would appear as a dark silhouett against the sky. Whoever took this photo was careful to make sure that both moon and mountain were within the exposure latitde of whatever medium he was using but just barely. Ansel Adams formalized this process and called it the zone system.
So far the arguments against this being an unmanipulated photo seem to be from comparisons to photos taken under different conditions. As I said before this image requires very specific conditions but I will try to make a similar shot as soon as I can. Unfortunately I’m working in Central Ohio now and there seems to be a lack of snow capped peaks on the horizon.
No fisticuffs. This is definitely a friendly dispute of two professional opinions. 
Yes. And this is the only reason I hesitate is because of the effects of the atmosphere. Yes, I know your eyes suck as a light meter (hence the “squinting” trick for evaluating scene contrast). That’s also why I’m stating exposures clearly and precisely. Like I said, I’m not 100% sure for this reason, and this reason only.
Yes. What’s the question? Of course I’m applying it to the correct exposure for the moon. I don’t care what the moon is illuminating. Most of the light on the mountain is presumably from the early sunrise, not the moon, so its effects are negligible. We have a well-exposed moon in our picture, so we need to know what the exposure for the moon would be. Or am I misunderstanding you?
We don’t disagree here.
It’s one stop overexposed, if I had to guess. A “properly” exposed moon (as referenced by those charts) would not be that white.
I agree on all of this, too. Hence my use of “proper” exposure in quotes most of the time.
All I’m saying is that I cannot imagine a lighting situation which would make that photo possible without manipulation. I don’t think there’s any way the moon and mountain could possibly be in the same exposure latitude to produce that picture. Like I said, I could be wrong. But I don’t think I am.
Are you still talking about this picture? Can anyone identify this mountain? I ask because, if the image is not manipulated, the moon appears to be near its maximum elevation in the sky (because the moon’s north-to-south axis, as indicated by the terminator, is nearly vertical) but it appears to be only a few moon-diameters (1 or 2 degrees) above the horizon. This would only be possible at high latitudes (and only high northern latitudes, since the moon’s north pole is up), so unless the mountain is McKinley or some other far-northern mountain I don’t see how the picture could be real. But unfortunately I have no idea what mountain this is.
That was my guess too but I don’t think that is relavent to the exposure issue.
pulykamell if you follow that chart for shooting the moon you’ll overexpose for anything but a full moon shot and you won’t see any detail. This is why photographers use things like spot and incident meters and 18% gray cards rather than relying on the averaging meter in a camera.
I used that chart because it gives you the benefit of a doubt and allows a longer exposure for the moon. (Thereby helping you get more detail in the landscape and bringing it into our exposure latitude.) Theoretically, the exposure times should be the same, whether full, gibbous, cresent or whatnot. And that exposure time would be the sunny-16 (plus one stop overexposure) rule. And, yes, I know how to use a gray card, spot meter, incident meter, color meter, metering off the grass, meter off the hand plus one, etc… I’ve shot chromes the bulk of my professional photo life, so, needless to say, I can expose film properly.
I was looking at the moon earlier today. I noticed that when some wisps of cirrus clouds went over it, it had the effect of an ND filter—stopping down the moon a couple stops— and the presence of the clouds were not really noticable. In that case, and that case only, I could see the possibility of getting the shot in the desktop wallpaper.
Ya know, Padeye, the more I look at it, the more I’m starting to convince myself that you might be right. There seems to be enough atmospheric haze to make the photo possible. Thanks. You’ve completely made me doubt my instincts. 
I think you are missing Omphaloskeptic’s point - he seems to be saying that because the terminator is almost vertical, the moon ,ust be nearly at its highest point, despite being so low in the sky. But I don’t know enough about the mechanics of the moon’s orbit and its appearance at various stages to know whether that is true.
And Padeye, you may well be right about the haze - the mountain is very hazy, and the moon of course is behind a LOT greater thickness of haze. I’ve taken photos where even the sun is not overexposed, near the horizon, due to thick haze, while still capturing land detail.
Good capture if it’s real, nonetheless. I wonder what Microsoft pay for wallpaper shots
Great pictures! My urge-to-travel fever just went up a few degrees.
I finally had a chance to sit down and download some photos and I have a sample that will support the MS Windows Ascent wallpaper being a real, uncomposited photograph. The photo I took is quite different for several reasons as I did not have a similar location or sufficently long telephoto lens to make the moon appear as large. The shot was taken in a single exposure and not manipulated outside the camera aside from converting the raw NEF file into a JPG and resizing it for screen display. Original image was 3008x200 pixels with the resized version at 602x400 with no cropping. The detail shot of the moon is cropped with no resizing.
Photo was taken on Mar 27, 2005 at 6:23AM MST with a Nikon D100 and 24-85mm F3.5-4.5 G lens at 85mm (equvalent to a 127.5mm lens on a 35mm film camera) Shot was taken at 1/320 second, f9.0 ISO, 800. Camera setting were for high sharpness, color mode sRGB and a custom contrast curve (Rob Galbraith’s “wedding” curve IIRC) with all other settings neutral, auto or off.
Even with the low magnification of the moon there is quite a bit of detail on its surface. I set exposure so that none of the moon gave a blown highlight warning when I reviewed the images and basically had to wait until the sun was high enough to illuminate terrestrial objects.
If I had been shooting with a 400mm+ lens the moon would have appeard as big in the frame as in the original image without cropping and of course detail would be five times better and noise less apparent.
http://www.padeye.net/welovedamoon.jpg
http://www.padeye.net/moondetail.jpg