What causes less of a hot spot on a silver screen?

[Steve Guttag]

Sure...the spot shown here was on a curved screen, however. Note, it is 21-22fL in the middle (and all of the way from edge to edge) but top to bottom, it drops to nearly half...around 12fL. So, if it was as flat screen, it would have been MUCH worse...more like 1/4 or less at the edges compared to the center.

This is the PSA on a Flat-Silver screen that is about 2/3rds as the one shown above (this one is about 26 feet across:

FlatSilverCapture.JPG

So, you can see the light really sucks at the sides. The uniformity issue will be due to where the projector is relative to screen center and that the meter is roughly in the middle of the auditorium. So, even though the light out the projector is uniform, such is not the case in the theatre. Where you sit relative to the projector and screen play a big role in how even the light is.

[Alan Gouger]

Sure...the spot shown here was on a curved screen, however. Note, it is 21-22fL in the middle (and all of the way from edge to edge) but top to bottom, it drops to nearly half...around 12fL.

Stewart Film Screen calls that saddling, caused by too much side tension when lacing the screen.

[Steve Guttag]

You are misinterpreting what Stewart was telling you. Around here, we would would call it "bellying" but it is the same thing. That does refer to when too much side tension is placed on a curved screen (it wants to pull the screen flat but the frame is holding it top and bottom). This screen didn't have too much side tension. We normally specify a lose-lace on the sides for such screens to avoid the belly. Just enough tension to keep the edges from flapping/wrinkling. Then spring it top/bottom (or bungie, if it is a real flexible screen). The PSA I showed for the the curved screen is not the results of bellying. That is the predicted results of a well-curved screen. As noted, it is a bit too curved or the center would be the brightest part. For a silver screen, the light distribution is pretty remarkable, none-the-less.

Now Stewart is a company that can produce a real Matte-White screen. They look incredible. A Stewart Snomatte100, microperf is tough to beat. A Stewart Ultramatte130 is the highest gain screen one can use and get SMPTE light levels side-to-side. Just be prepared to pay 4-5 times the price of the typical cinema brands. They ARE better...I don't know about 4-5 times better. I tend to use them in screening rooms.

[Harold Hallikainen]

On a non-gain lambertian screen, isn't there always fall off from the center to the edges? I think the projector has to be infinitely far away to make it perfectly uniform.

I read about the variable sized perforations when I was researching stuff for the SMPTE screen luminance group. Clever! But, it seems that today we could do that electronically. Can't the video be passed through a variable gain block where the gain varies with the XY position of the pixel? This could drop the center brightness. Of course, if the hot spot moves with viewer position, that doesn't help (and can make things worse).

Harold

[Mark Gulbrandsen]

The original D-150 screens had variable perforations AND variable lenticels from the center out to the left and right. I don't know if they kept that in the later installations though.

[Steve Guttag]

Any time you have a screen gain, yes, you will get fall off. Even on a matte-white screen, as you noted, you'll never get that perfectly collimated light plus nobody is making the screen material THAT perfect. As I recall, matte-white is defined by a piece of ceramic tile stored at NIST or some such thing. That probably isn't such a great screen surface either.

SMPTE standard even call for a fall off. Truth be told 14fL (48cd/m²) everywhere is ALSO outside of SMPTE spec (unless they changed it while I wasn't looking). RP-98 has it at 75% - 90% while review rooms were to be "at least 80%" of center. This is reinforced in ANSI/SMPTE 196 I never understood the upper limit for theatres. If you can hit the reference in the center and hold it at or near 100% on the sides/corners, more power to you.

As for pre-dimming the middle to compensate for a hot spot, that won't work unless your projector is perfectly on center line and the observer is also perfectly on center line. You've now introduced one fix but caused another problem. The goal is to have the greatest coverage over the most about of seats. Pre-dimmnig doesn't take into consideration where the person may be sitting.

This image shows more of where I was standing when taking the picture than where the projector was. If you pre-dim it in the middle, you'd exaggerate the hot spot on the right:

Screen Shot 2021-07-17 at 11.36.25 AM.png

The answer is...DON'T USE SILVER SCREENS!!!!! Furthermore, don't use higher gain screens, nothing above 1.3 unless you are willing to curve the screen. And if you do that, be willing to accept the geometric distortions and the cropping that comes along with that (or use curved lower making board to conform to the barrel distortion.

[Sean McKinnon]

I HATE silver screens. I actually prefer to to go the other way when possible (Steve and I have disagreed on whether this provided provides better appearing contrast) I really like the results I get from less than unity gain screens and a brighter projector. It also helps in smaller rooms where you can’t get the brightness low enough. I have gotten very good results with this method and actually had a consultant tell me yesterday how impressed he was with the image quality on a 4K NEC laser on a 0.8 gain screen. I have used Xpand 3D very successfully on a MDI .7 gain screen 26’ wide with a Barco 23B so if you’re willing to spend the money you CAN have the best of both worlds.

[Steve Guttag]

They problem with most grey screens (below 1.0 gain) is that they have a 1.3 or so gain characteristic (they are not uniform) and, if the room is dark enough, they don't bring anything to the table. Grey screens are for high-ambient light situations where one uses the screen to lower the effect of the higher ambient light from reflecting off of it. Then use a brighter projector. There is no magic there. More likely, what they interpreted as higher contrast was the the screen's directivity. Using a 1.3 gain white screen would give you that too. It has a bit of "pop" to it without calling attention to a hotspot.

This is what I find as "typical" of a grey screen:

Screen Shot 2021-07-22 at 12.29.55 AM.png

It is anything but uniform but that is what gives the illusion of higher contrast. You could do the same thing by using a 1.3 gain screen but get a bit more light for your effort. Now, if the theatre has bright walls and there is a reason that the ambient light levels may be up, that could be a different story. Stewart does have a grey screen that has a matte-white like distribution (essentially 180-degree viewing angle) but that is Stewart. Theirs is just about the only Matte-White screen that can deliver.

Screen Shot 2021-07-22 at 12.45.26 AM.png
In my top-end screening rooms, they get a Stewart Snomatte 100 with microperfs. They look phenomenal but are a lot more money than the typical cinema screen.

[Marcel Birgelen]

In my top-end screening rooms, they get a Stewart Snomatte 100 with microperfs. They look phenomenal but are a lot more money than the typical cinema screen.

Since we'll be constructing a new screening room later this year, I've been looking at some screen options with microperf. While I'm mostly happy with our current material, which has a gain of 0.95, under certain circumstances, like a total white image on screen and you know what you're looking for, you still can see the perforations in the first row.

While the pitch between the perforations is the primary factor for this, there are other factors like the size of the perforations and I guess even the surface of the material itself has some effect on the end-result. Manufacturers never give you reliable data on this, so that's why I'm interested in your experience. What's the closest viewing distance before the perforations become visible on this screen material with microperfs?

[Steve Guttag]

Something like 2m-3m They go invisible pretty quick. I'm sure Stewart can provide the details you want on their Microperfs.

[Marcel Birgelen]

Yeah, that's about the indication I get from most manufacturers, including the manufacturer of our current screen material. I don't think there are any absolute numbers about this, because it's one of those things that also depends on your personal vision. In our latest design, the front of the first row will be about 3m away from the screen. I guess we'll end up ordering some sample patches to make sure the pattern is invisible at that distance.

[Max Biela]

It also depends a lot on the brightness of the projected image. I could see perfs on my Harkness miniperf screen at 21fl from about 4m that vanished when viewed at 14fl.

[Sean McKinnon]

A lot of the smaller size rooms that seem to be all the rage these days have trouble getting light levels low enough to begin with so getting that "pop" without the extra perceived brightness is a win in my book!

[Steve Guttag]

Put in a High Contrast lens. Lowers the light, increases the contrast ratio, improves the picture. With lasers going down to 20% in many projectors and 850W xenon for xenon based, there shouldn't be a problem of getting low enough light.

Again, if the room has high ambient light, that is another story.

[Tony Bandiera Jr]

When I worked at a So Cal amusement park which had a 3D attraction, we had a roll-down screen that seemed to work ok, it was a long and rather narrow space. Almost zero projection angle and both machines were pretty close to center (we only used one at a time, with first a Potts Alpha then later Christie ELF platter doing the handling.)

Because of the 3D lens, the polarized glasses and the narrow room, we didn't seem to have a hot spot per se, but man the light really fell off to those on the sides.

To answer the OP's question, what causes less of a hot spot on a silver screen is, don't use a silver screen.