AVS Forum

AVS Forum (http://www.avsforum.com/forum/)
-   Screens (http://www.avsforum.com/forum/23-screens/)
-   -   PRACTICAL properties of curvrd screen? (http://www.avsforum.com/forum/23-screens/19168-practical-properties-curvrd-screen.html)

Jeff Smith 10-11-2000 02:31 PM

Forget, for a monent, the numbers... how does it look? For simplicity sake, lets assume a good CRT ( they need the gain the most), lets say a 9" CRT on a screen 107x60". Lets also limit the curve to the horizontal only ( no toroidal's).

1) must you be sitting in the center to get the proper effect? Does the curve limit your viewing angle, and if so, how much does off angle viewing suffer (remember the awful brightness variations on the toroidal Kloss Novabeam screens as you move even a few feet up-down-sideways!) ie. if 90% viewing is dead on centered,8% of the time someone sits off center but still within the screen width 12' back, and 2% of the time intoxicated but fanatic fans are all over the room and floor looking at all angles at the super bowl...WHAT DOES IT LOOK LIKE?
Could someonr with a curved srceen just walk all around and tell us dreamers what it looks like from MANY angles...please?

2) Would you get and advantage by curving your existing gain screen (ie. 1.3), or would you have to go up in gain (ie 2.0) and curve to help the problems associated with the higher gain?

3) How much curve? I have always assumed you would take a string centered on the projector and make the screen distance equal all across...True or more complicated?

In case it isnt obvious, I'm a CRT fan who would love to go digital, but have yet to see a digital projector with as good a picture, but REALLY want to go to a 107x60 minimum screen. I'm looking at a 9" but worry about it with a 1.3 flat screen. I would feel better with a 2.0 and would love it if the curve helped the situation, but I wouldn't do it if the curve really screwed up and gave me much "Kloss toroid brightness variation). It sure would be nice to have someone with roughly the same specs give me advice.

Thanks

Jeff



------------------
Jeff Smith

KFung 10-12-2000 09:04 PM

1) No you don't have to sit at the centre to get the proper effect. The curve of the screen is meant to optimize viewing for EVERYONE in the room. Just the name Novebeam indicates that the screen probably had monstorous gain numbers and a toroidal shape was necessary to get any kind of decent picture.

2) Curving your screen will probably not help a 1.3 gain screen, as you probably get pretty good uniformity already. That was the point after all, to give you a bit of a light boost without hotspotting, it doesn't need a curve.

3) No that's not complicated enough! http://www.avsforum.com/ubb/smile.gif If are looking to reduce hotspotting on your high gain screen you need to find the combination of curves that redirects to most light to all the members of the audience and maintains uniformity over the screen area for all the viewers. A challenging task to say the least! What the people at TORUS do is ray-trace the light from the screen and optimize it for everyone.

As for horizontal curves only, it's better than nothing on a high gain screen. Since your vertical height is relatively small there should be less hotspotting, so you could reduce hotspotting to almost nothing with just a horizontal curve.

What you should know is that gain will fall off the farther you get off axis, what TORUS does is find the angles that give a visually uniform brightness (I think that's about 10% of brightness?) this angle can be used to define a cone with the pointed end on any point of the screen. The center line of the cone would be perpendicular (90 degrees) to the surface of the screen. What you would do is find a curvature that would make the cone cover all the audience members from every point on the screen. In reality you only have to find the optimum values for a few points on the screen and derive the curve from there (usually the corners...).

Hope that helps,

Kam Fung

CINERAMAX 10-12-2000 09:15 PM

Oh I don't know.

I have seen 23 foot wide curved screen shot with a Barco light cannon and instead of being a Hot Spot it was more like a Hot vertical oval. Again the gain was 2.

CINERAMAX 10-12-2000 09:21 PM

Jeff my advice is to go to A Vutec 54 x 96 curved screen but supply them with a Stewart 2.5 to 2.8 white screen sheet. They will make it for you given sufficient lead times. The dynamic range of the image should look as good as double stacks or better in ambient light situations.

Just seat a little closer.

This is the poorman's TORUS and a steal at less than 4 grand.

Any ways that is the way I will end up going with the Cineramax leaving out the perforations.


KBK 10-12-2000 09:32 PM

Pardon my ignorance here but....

If you are using current projectors, and their standard focussing methods, then the curvature's characteristics are defined in the mathematical parameters and limitations of interactions between the main and outer lens focussing parameters.

------------------
---Place Signature Here---

Jeff Smith 10-13-2000 06:38 AM

So is there any downside to a curved screen with CRT, other than cost? How about off axis viewing,is it any worse (if you are sitting fairly far off to the side) than with a flat screen, or do I read KFung correctly in that even that area is better?

Peter, have you heard of any slightly larger screens from Vutec or others?

thanks

------------------
Jeff Smith

KFung 10-13-2000 08:11 PM

Yes, KBK is right, if your screen has too much curvature it will defocus at the edges. Most projectors are designed to project a focused image onto a flat surface, ironically a cheaper lens that focused onto a hemisphere would have been better suited to a toroidal screen. In any case, it would be necessary to determine the distance from a flat surface that a projector's lens will still retain its focus (sort of like depth of field for cameras). You would need to find this and make sure your curved screen was not deeper that the projectors ability to focus. I believe someone may have checked the depth of focus for a D-ILA with an ISCO and found it to be 6" or so...I've never seen the numbers for a curved screen, so I don't know if this is enough.

Jeff,

Yes, if done correctly a curved screen with produce a better image for everyone, especially the people at the corners! (traditionally where the image of a flat high-gain screen suffers most) Make no mistake a TORUS-designed screen is not cheap, I believe the number they gave me when I asked was something like $3-5k for engineering services and $8-9k for the screen! They said they charged ~$150/hr for consulting and [I assume] it would take a significant amount of time to do the calculations for a toroidal screen. A horizontally curved screen is significantly cheaper and their consulting fee is something like $1,500 max. (the result is not as good though...)

Check out www.sigmadesigngroup.com (they design the TORUS CCS and Stewart makes them)

Regards,

Kam Fung


[This message has been edited by KFung (edited 10-15-2000).]


All times are GMT -7. The time now is 07:36 AM.

Powered by vBulletin® Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
User Alert System provided by Advanced User Tagging (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.