White screens: BOC vs C&S Ultra vs RS-MM-LL vs Sherwin Willams ProClassic

Long story short, I painted my BOC screen with RS-MaxxMudd-LL. I also made panels of C&S Ultra from Hometheatershack, the Sherwin Williams ProClassic recommended in the Projector Central article, and unpainted BOC to compare them directly.

I guess I shouldn't really be surprised by the results. In order from least light to most light, they go: BOC, C&S-U, RS-MM-LL, SW. And by "light" I mean both brighter whites and lighter blacks. The closest of the four are RS-MM-LL and SW, with the Sherwin Williams edging it out a bit in lightness. The SW is actually not quite 100% dry yet so I'll check it again in a few days.

Given the fact that presumably someone wanting a screen of these shades is probably after brightness, I give the nod for now to Sherwin Williams. I should point out that though none of these had both brighter whites AND darker blacks than any of the others, for me there is more apparent difference between a brighter white and a dimmer white than between one shade of black and a slightly lighter one. Not only that, but the brighter whites made that panel's corresponding blacks seem darker in comparison. In other words, as far as perception goes, I saw far less difference between the blacks of the four than between the whites, so for me at least, going up the scale in lightness actually improved perceived contrast. On many scenes, for example a hallway of the Nostromo in Alien, a scene from the Road, or heavily detailed scenes like the royal court in the sacrificial scene in Apocalypto, I had to really look hard to see where one panel ended and another began; while on brighter scenes, like snowboarding in Art of Flight, the difference was immediately and significantly apparent.

Color was noticeably different on all four on a blank white screen, but in every movie scene I put up, any color differences were for all practical purposes undetectable to my eye. The guys at HTS seem to be the only ones who bother to actually test/measure this kind of thing, and I believe C&S and BOC are both very neutral with SW and then RS-MM-LL coming in behind.

Sherwin Williams is off the shelf, costs $20 a quart, and supposedly rolls on like a dream if you don't want to bother spraying.

The RS-MM-LL costs about $80 at a minimum, although this gives you a much greater quantity. It is not an open the can and go solution, due to requiring 4 ingredients and more particular quantities.

C&S Ultra is around $32 for a quart or something like $52 iirc for two quarts, assuming you get the BASICS silver with a 40-50% off coupon (that goes for my stated cost of RS-MM-LL as well). Only two ingredients and simple to measure though, just equal amounts of each of the two ingredients. Note: I might be mis-remembering and the BASICS might have been $2.50 with the coupon rather than $5. If so, it would be $22 for a quart and $32 for two quarts. I can't remember if the regular Michael's price is $10 or $5 per 4oz tube.

BOC is almost always on sale at Joann's and there are almost always half off coupons as well, so usually work out to $7 a yard or less (54" high).

BOC requires a frame which is around $20 of material at Home Depot. The total costs to build my frame for a 126" x 54" 2.35 screen and get the BOC were around $40.

The paint solutions can be applied directly to a good wall or another material. If your wall is in rough shape and you want to paint on the ideal surface, a sheet of Sintra, that will cost you; a 5x10 sheet to almost-but-not-quite match the dimensions of my BOC screen would run you $95-100. I went with spraying and personally found it an enormous chore. The time and effort to dropcloth off the area so the paint dust doesn't get all over the place, the cleaning of the gun between every coat, and the (for me) difficulty of getting a good smooth result had me wishing I had rolled or just stuck with BOC, to be honest. Not to mention $80+ for the sprayer, mask, dropcloths, and other supplies. I should add that I'm not handy at all, though.

Personally I'm not sure why BOC gets kind of short shrift as a "passable starter screen". It seems to me that it has a niche just like any other solution regardless of cost, that it fits right into the spectrum, having darker blacks than the brightest white and white+metallic screens, but better whites than the grey screens. Add in the extremely low cost and the ease of putting a screen of pretty much any size together, and I don't see why seemingly everybody rushes to either a grey screen or a paint solution. I would imagine there'd be many projectors, screen sizes, etc for which BOC would be the sweet spot of white/black performance, just like any other solution.

I'm sticking with the RS-MM-LL because it's brighter than my original BOC -- important on a 10.5 ft wide 2.35 screen with a Mits HC4000 -- and like I said, the hassle and expense of painting is enough that I don't want to repaint it. When I move next year, if I have a good wall available and the same projector, I'll most likely paint the wall with the Sherwin-Williams. My new screen is significantly brighter than my original BOC in bright scenes, as I said, but honestly I wouldn't have known it if I didn't have a BOC panel to compare it directly to.

Thanks to MississippiMan and Harpmaker at HTS for all the help and suggestions.
Edited by curttard - 12/9/12 at 10:56pm

I began calibrating contrast and brightness but the settings did not change, and honestly, why would they? The point at which I can distinguish between my projector's black and blacker-than-black did not change just because one screen was a bit brighter than another, and the same went for white. I did not bother with color/greyscale because I trust the HTS crew's measurements as far as accuracy, for those looking for the most accurate reference color. For myself, in actual scene content, the differences were negligible or unnoticeable. Really I was just trying to compare white and black performance.

I will put up a couple simple GIFs later. They will be simple white/black comparisons, cycling through the four screens, cropped to show only the panel. Each picture was taken in RAW mode with identical full manual settings including white balance.

You are correct, it is difficult or impossible to take one photo that captures the full range of what's happening on the screen; either the blacks or whites or both will be clipped. This is why I always find it funny when people post a picture of a scene and say "look at my inky blacks" or whatever. You can get pitch black and blinding white in a photo of almost any projector on any screen.

So, I took two exposures of each panel's ANSI checkerboard. One slightly over-exposed, to make sure it would pick up the difference in the blacks between the panels, and one slightly under-exposed, to pick up the variation in the whites.

Two screen materials/paints in one shot can be illustrative of relative differences, for sure. It potentially suffers as I said, in that the camera can't capture the full range -- so the blacks of two solutions might appear the same, because both of them are clipping the camera, while the whites might show more difference because the shot is exposed for the whites. This would give the impression that one has better whites with the same black level, thus making it appear to have a greater contrast range than it actually does -- because while its whites were brighter, so were its darks, but they are still dark enough to clip the camera and thus appear identical to the darks of the other..

Also, I think many projectors are probably not uniform across the image -- mine certainly is not. Thus a panel in one position might have very different bright/dark properties than the same panel in another position.

Finally, a scene from a movie makes a poor source to compare blacks and whites in a single shot with multiple panels, as of course there is no guarantee that any two areas of light or dark are even *supposed* to be the same in the area of one panel vs another.

It's not hard to avoid these issues, which is what I did with my shots. First, it's a simple ANSI checkerboard -- you know that everything you see is supposed to be pure white or pure black. Second, I take two shots, exposing one to get more range in the darks, the other to get more range in the whites. Third, they are shots taken from a fixed camera tripod position with each material/paint in the same position -- so if the projector is brighter in one spot than another, this is reflected in the shots of ALL the tests. And finally, shots are taken in RAW mode -- thus no JPG processing or compression -- and with full manual settings, to make sure each panel gets the exact same treatment from the camera.
Edited by curttard - 12/10/12 at 5:44pm

I should mention that I was reminded on HTS that I forgot about Glidden Diamond 450 Titanium White, which is what I will actually use in the future most likely -- brighter than the rest and measured very neutral iirc.
Edited by curttard - 12/10/12 at 7:36pm

As promised, here is the first image: This is 100% white, taken at 1/4 sec exposure, ISO 80, f 4.0.



The SW is only a bit brighter than the RS-MM-LL; the difference between the others is more noticeable.

Measuring a large square in the center of that with Photshop's luminosity histogram (I would have done the whole frame but the BOC wrinkles would have thrown it off), we get median luminosities of:

SW: 186
RS: 182
CS: 173
BOC: 162

And the same thing for black at 15sec exposure. I should probably do this one over at a greater exposure to get more differentiation, but you get the idea:



SW: 19
RS: 18
CS: 17
BOC: 16

Edit: The black one shows almost no differentiation between them in my browser while the gif looked correct in Photoshop, I'll have to look into that.
Edited by curttard - 12/10/12 at 11:00pm

I think if you posted an ANSI checkerboard with no obvious hotspots or that type of thing that could affect how light one area was as opposed to another, it would be pretty good for showing the relative black/white performance of two things in the same shot, sure. Ideally you would do two shots, one underexposed (for the whites) and one overexposed (for the blacks) but otherwise identical. Even point and shoots should let you do that.

I say "ideally" two shots but I think it might be necessary. I don't know that you can get a checkerboard (at least on a white screen) where one or the other of black or white isn't clipping the camera. I know I couldn't with my S95.

Also ideally I guess we should post a link to the RAW image, or at least the full-size original JPG (i.e. not one re-uploaded to Photobucket etc), because the gifs above show me that something, either the browser or Photobucket or resizing, has changed the values. I just copied one of the gifs above and brought it into Photoshop alongside the original file I uploaded here, and they are significantly different.

Edit: I should add that there would still be the issue of the projector not being perfectly uniform across its image. I think the best apples-to-apples comparison possible is comparing the exact same area of the projected image with the exact same manual camera settings.
Edited by curttard - 12/11/12 at 8:16am

They have white/white as well. White/ecru could work if the white side is the side you want to use. You want to use the less textured side.

I'm withdrawing my recommendation of Sherwin-Williams ProClassic. The hotspotting on my test panel is horrendous. I don't know why I have this issue when I haven't heard of anyone else having it, but all I can go by is what I'm seeing myself.

I even rolled a few coats on, and while I think the overall texture is actually improved versus my sprayed coats, the hotspotting is at least as bad if not worse.

The test panel is 1.5' wide by about 3.5' long. I didn't shrink my projected image down or anything.

It was hotspotting before after a week. These rolled coats are new.

Yeah. Just lucky I guess! I've never heard of anyone else having hotspotting with the SW and certainly loads of people are using it or have tried it.