Considering the Digital Photos of Screenshots used in the Reviews of Home Theater Projectors
Here are the key points for you to consider:
Limits of a dSLR, in accurately capturing scenes viewed on home projectors.
First, and foremost: These photos of scenes are not, by any means a perfect, or near perfect way to compare projectors. They exist in the projector reviews, to complement the commentary, rather than the other way around.
The limits of my digital SLR (currently, I'm using the relatively new Olympus E510), or any dSLR, are such that what you see on the projected screen is far superior to the images seen in the reviews.
- Digital cameras, even higher end digital SLRs (dSLR), cannot begin to capture the full dynamic range of the images projected on the screen by a good home theater projector.
- Take a normally exposed image of a balanced (plenty of bright and dark areas) scene, and a dSLR will lose the darker shadow detail, making dark objects easily visible on the screen look black, and blurring into the dark background. Near whites - bright areas will also get "blown out", turning near whites to white, and losing color saturation in bright colors as they are "overexposted" relative to what the camera can cope with.
For that reason, to look at how well a projector deals with shadow detail, and black levels, you will see a number of images in each review, where the scene is intentionally overexposed, anywhere from about 1 to 3 f-stops. This completely overexposes the mid and brighter parts of the scene, but reveals the black levels and shadow detail lost by the camera at a normal exposure.
And, yes, I could also underexpose to show you what highlight details are getting lost (by the dSLR), along with color saturation, on very bright areas, but, simply put, I do not.
Color accuracy of the photos in the projector reviews
First, even at best, there is going to be a slight shift in the colors, relative to what is on the screen. Despite the best efforts so far, to "tune" my dSLR (with it's color temperature controls, etc.), there is always some color shift. Typically, I find (on my display) that images come out a little more redish than the image on the screen. (That's an oversimplification.)
You will also see huge amounts of color balance shift in those dark, and intentionally widely overexposed images I use for shadow detail and black level comparison. Some might appear very red, others shift towards green. For these images I'm normally doing time exposures from 1.5 to 10 seconds, and they are very, very dark scenes. At these low light levels, and when overexposed significantly, the camera definitely tends to make the color shifts far more dramatic than what you see on the projected image. Two images of the same scene, might find one very redish, another heavy on the green, a third, fairly well balanced. Viewing the projected images though, might show only slight differences. Why my cameras accentuate this on very dark, highly overexposed images at long time exposures, I can't properly explain, but the shifts can be dramatic, while they are subtle on the original projected scene, between any two projectors.
Losses in image accuracy, due to the process of taking, posting, and viewing the images on typical computer monitors
First, the contrast ratios and dynamic of these home theater projectors, is much greater than your display monitor on your desktop or computer monitor.
Second, I don't try to adjust anything, to compensate. You are looking at images that have not been modified, other than to crop them, and resize them. I don't "fix" shifts in colors or contrast, to better match the screen. Why? Many would say that would be cheating. So, instead, what you see, is what the camera captured, further effected by the limits of your computer display.
Third, I bracket the photos. I take as many as 8 different exposures of the same scene, and choose the one that I think best represents (given the limitations), what I saw on the screen. And again, I point out that I don't modify the photos. There are rare exceptions, but I note those changes when they occur.
Your computer monitor is not up to the task. It lacks the dynamic range and contrast, but also is not perfect in color balance. Your monitor might be strong on reds, weak on greens, it might oversaturate the image (making it look like there is too much color), which is very typical.
As such, if images seem oversaturated - too much color intensity, try dialing back the color saturation on your monitor, so that the image no longer looks oversaturated.
In a perfect world, your computer monitor would have a 10,000:1 contrast ratio (not 400:1 to 2000:1), and it would be color calibrated to D65 (6500K color temperature). That simply isn't the case.
Software comes into play, as well. Two different browswers, are quite likely to produce slightly different colors, saturation, and other aspects of the image. Look at one of these images, in your browser, try saving it, and then view it in another program (say Photoshop, or a simpler viewer), and other programs. You are likely to see significant differences depending on software as well.
I can tell you, from my own experience, that an image definitely looks different on my MacBook Pro (which I use for everything), when I'm cropping and resizing in Photoshop, than when I look at the same image, in other programs, and definitely when I look at the same image fed back to me from the website, when viewing the final review.
So, the bottom line is that these photos, while useful, and not accurate enough, by themselves to really help you determine which projector is best. They definitely do, though,reveal some useful ideas about the picture quality, when viewed in context.
I could say more, but you get the idea.