AVS Forum banner
Status
Not open for further replies.
1 - 20 of 30 Posts

·
Registered
Joined
·
1,108 Posts
Discussion Starter · #1 ·
I have recently had a thought; feel free to correct me where I go wrong.


Maybe you all can help me with this question that I have. I have been thinking and I am wondering why it is that Digital projectors have the light source shine across the entire image source (LCD, DLP, etc) and not use raster scanning like CRT projectors do. I know that moving a highly focused light beam across an lcd panel is completely different from CRTs using electromagnets to move the electron beam across the tube face; but with laser light shows using high-speed mirrors to move beams of light around with somewhat smooth motion. If raster scanning could be done it seems to me that the resulting image would be Incredibly bright as opposed to tradition projection. As say instead of say 1000 lumens of light shining across an entire lcd panel, if it were focused on say 1/100th of the panel and scanned at a high rate thus making the scanning not visible to the eye the resulting image would be 100 times brighter.
 

·
Registered
Joined
·
271 Posts
Another way to look at it is that any spot on the screen will only be illuminated 1/100th of the time. So averaged over time you would gain nothing, except maybe a terrible flicker.
 

·
Registered
Joined
·
32,172 Posts
Another way to look at this is that the only light source available to do this is a laser, which is an entirely different beast.


There is otherwise no coherent source of light that would output that many lumens onto a small spot.


I mean if you want more lumens today, you can get them with a bigger, more powerful lamp.


So maybe your question should be: Why don't they just make brighter projectors? This as opposed to asking: Why don't they totally and fundamentally redesign digital projectors as they exist today?


mark
 

·
Registered
Joined
·
1,108 Posts
Discussion Starter · #4 ·
I mean this strictly as a theoretical question I really had no interest in practical implementation. My main concept is to move a focused light source across a large (say 15â€) LCD panel thus not requiring a extremely small focused light source. I got my idea from a CRT without its defection yoke—This out this the small point on the center of the screen was just a spec, but moving that beam around the tube yielded increased light output. It the light scanning could be made fast enough we would not sense the flickering. Just think of as children playing with flashlights where you would move them around so fast that you could make solid lines on the wall—same idea.
 

·
Registered
Joined
·
32,172 Posts
And it's years away from your living room -- if it ever is actually sold into that environment.
 

·
Registered
Joined
·
1,494 Posts
Quote:
Originally posted by BizarroTerl
The closest to what you're thinking about is Sony'e GLV - Gated Light Valve. Sony claims it's imminent and has for years.
SONY has only owned the technology for two, possibly three years at this point, so they haven't been claiming imminent introduction for "years".
 

·
Registered
Joined
·
4,720 Posts
Quote:
Originally posted by mikejz84
I mean this strictly as a theoretical question I really had no interest in practical implementation. My main concept is to move a focused light source across a large (say 15?) LCD panel thus not requiring a extremely small focused light source.
Mike,


Viking2000 basically gave the correct answer above - if you focus your lamp so that it is illuminating 1/1000-th of the

screen - it is 1000 times brighter at that spot - but it only illuminates the pixel for 1/1000-th the time.


The reason a CRT can do this has to do with the phosphor which stores some of the electron beams energy, and then

continues to glow after the electron beam has moved on.


However, a pixel in an LCD chip doesn't store energy - it's basically a shutter.


You don't gain energy by focusing - you only redistribute it.
 

·
Registered
Joined
·
1,108 Posts
Discussion Starter · #9 ·
But if you scanned it fash enough, would'nt the eye be unable to detect the scanning?
 

·
Registered
Joined
·
6,520 Posts
As another poster said,


What you're talking about is *exactly* what Sony's GLV is...lasers that "scan" using a digital matrix to reflect off from (or something like that).


Anyway...read about GLV and you'll see that you're onto something. Problems with development seem to be laser cost and durability.
 

·
Registered
Joined
·
4,720 Posts
Quote:
Originally posted by mikejz84
But if you scanned it fash enough, would'nt the eye be unable to detect the scanning?
Yes - your eye won't detect the scanning.


But there's no escaping that you don't get extra energy by scanning.


The brightness is a function of the energy density.


Because of the finite response time of your eye - it is also doing a time-averaging. The scan may be so fast that you

don't notice.


However, when you look at the total amount of energy - and/or the total number of photons - that are reaching

your eye - that isn't changed by focusing them.


If the focus gives you an enhancement factor of "X" but for "1/X" the time - you aren't changing the number of photons

that are reaching your eyes - hence you don't affect the brightness.


Think of it like money [ people always get it right when you put it in monetary terms. ]


If your boss increased your monthly salary by a factor of 12, but only paid you during 1 month of the year, and you

get zip for the other 11 months - are you money ahead?


[ I know it would help if you got the money in January because of the time cost of money - but the analogy breaks

there].
 

·
Registered
Joined
·
5,445 Posts
Laser eith GLV is for some time in late 2005. In mdreport there was the old news from Sony saying two years from introduction.


Interesting Quantum fusion were working on laser crt.
 

·
Registered
Joined
·
1,108 Posts
Discussion Starter · #15 ·
Quote:
Originally posted by David Mendicino
Scanning a digital projector would likely help the black levels greatly, as you could shut of the source as it is scanning.
I am not sure of the responce time on a projector type bulb; but maybe reflecting it off a low rez DLP chip
 

·
Registered
Joined
·
4,720 Posts
Quote:
Originally posted by noah katz
Scanning does not intrinsically have anything to do with the response time of the light source.
Noah,


I believe Mike's response it to David's comment that one could shutoff the sources as the projector scans.


Mike is pointing out that with the lamps presently used in digital projectors - incandescents, and arc lamps - the

lamp can't shutdown fast enough to be of any good in helping black level.


Even if you shut off the current to the lamp momentarily - one still has a hot glowing filament, or blob of plasma

that doesn't stop emitting light.


Hence, it is not analogous to the electron beam of a CRT which can be shutdown quickly.
 

·
Registered
Joined
·
19,586 Posts
To get the black level benefits, you'd still get back to the GLV scenario, where you aren't blasting light at the whole reflective element all the time, so you don't have to worry about stray reflections and the effects of all those mirrors (for instance) constantly flipping back and forth and whipping their bit of stray reflection across from their position to the edge of the screen.


It makes a lot of sense to go that route, with lasers, if it can be made cost effective. There also wouldn't necessarily be a requirement to only have one laser per color either, if scan speed (image decay) was an issue. Have one red scanning the top half or left half and another scanning the other, perhaps. I'm not sure where the price/payoff curve is there, i.e. whether it's cheaper to use one high end laser/mirror system, or to use two cheaper and slower ones.


For DLP type systems, it seems like they'd be working very hard to come up with some surface for the mirrors that has almost zero reflectivity a small fraction of a degree off of it's on position. I'm sure that's hard because it has to be something that can be laid down with chip type technologies. But if you can make the underlying substrate very non-reflective (which they've already made steps on), and get the mirror tops so that they are reflective only within that narrow angle near their target position, it would seem that you could get the black level of DLP down pretty good.
 

·
Premium Member
Joined
·
24,977 Posts
Dean,


"To get the black level benefits, you'd still get back to the GLV scenario, where you aren't blasting light at the whole reflective element all the time,..."


I'm not sure this makes sense; if the light source is bright enough to give a bright time-averaged picture from a very bright point source, the stray light will be commensurately bright as well.


The only way I can see an intrinsic black level advantage is if the lasers' brightness can be modulated a la CRT.


In fact, I'd think all those little fingers would be an inherently worse light scatterer than nice flat DLP mirrors that have only four edges.


Re DLP CR, I don't think there's a problem; the dimple fix will take them over 4000:1.
 

·
Registered
Joined
·
19,586 Posts
Quote:
I'm not sure this makes sense; if the light source is bright enough to give a bright time-averaged picture from a very bright point source, the stray light will be commensurately bright as well.
Even without being able to pulse the laser on and off I think that something like GLV could do better. The reasons is that you don't have to depend on the reflective element to competely turn off the light. Basically, and I'm just talking off the top of my head here and they might not do it this way, you could have two mirrors. There would be one before the reflective element and one after it. The one after the reflective element can do the scanning of the beam after it's been bounced off the reflective element. But the mirror before the reflective element can have a very small, trivial movement that does nothing but move it onto the reflective element or off of the reflective element into a completely absorptive internal 'light well'.


Does that make sense? So, unlike DLP, where turning off the light means trying to bounce it off to the side after it's hit the reflective element, and having to deal with substract be somewhat reflective and also having to sweep the mirrors on and off target, which polutes the image, you can flip the mirror off of the reflective element with a trivial upstream first mirror movement, and not flip it back until the scanning mirror is in the next position where you need to light a pixel. So there would be no pollution of the image by the light sweeping across the screen to get to the off position and no worries about any unwanted reflectivity of the reflective element or it's substrate, because you would have flipped the laser into the light bucket before even moving off the last lighted pixel.
 
1 - 20 of 30 Posts
Status
Not open for further replies.
Top