DV Info Net

DV Info Net (https://www.dvinfo.net/forum/)
-   Alternative Imaging Methods (https://www.dvinfo.net/forum/alternative-imaging-methods/)
-   -   Ground Glass : reflective vs. transmissive - thoughts pls (https://www.dvinfo.net/forum/alternative-imaging-methods/49400-ground-glass-reflective-vs-transmissive-thoughts-pls.html)

Keith Wakeham August 15th, 2005 08:34 PM

Ground Glass : reflective vs. transmissive - thoughts pls
 
I'm just curious about the whole mini35 scene now and I was thinking about how most people are using glass and plastics that are transmissive, light goes through them.

I have a front projector and without starting a front vs. rear projection arguement it seems that the images are brighter, sharper, and more even with a front projector even with a cheap screen that a good rear projector with a really expensive screen.

I was thinking about this when I was focusing stuff outside my window on a wall with a magnifying glass and the picture looked really good.

So have anyone tried to make a mini35 where instead of a ground glass a flat diffused surface is used. Or is the problem of getting the camcorder pointed at the image to much of a problem?

Jim Lafferty August 15th, 2005 08:43 PM

I'm puzzled -- a ground glass is a flat, diffused surface...

edit: oh, unless you're meaning using a mirror to place the incoming image on a flat plane at a 90 degree angle (or thereabouts) with respect to the image's trajectory, then pointing the camera at that (presumably white) surface?

Ben Winter August 15th, 2005 10:14 PM

Good luck finding a material in that situation that won't lose you 209206.44 fstops.

Plus I don't know about you, but I'd prefer not to have a big ugly system of mirrors going from my lens to the camera. And in order to eliminate perspective distortion, the image coming from the lens and the camera lens itself would have to be in the exact same spot.

There's probably other reasons this hasn't been attempted before. But maybe if someone came up with a really good design that somehow countered these problems...

Quyen Le August 16th, 2005 12:19 AM

I tried to use beamsplitter but have no luck finding a reflective screen that can capture image without quality loss and no grain. Plus the image on the GG is much brighter than the image capture using beamsplitter. You can try but it has been done and no luck finding the good enough screen for this purpose. Thanks.

Quyen

Keith Wakeham August 16th, 2005 04:52 AM

Thanks for the replies.

Just to clarify I mean a flat difussed non-transparent surface like a wall or a thick piece of paper and capture from the same side that the image is created on.

From my phyiscs classes a GG works by difussing light. It does this by having a rough surface, and in a GG it should diffuse in 360 degrees. So on paper it seems that the half the light (and likely more) is thrown back at the 35mm lens while only half the light passes through and is available for the camcorder lens. A front projection system can only difusse in a 180 degrees so the light should be brighter based on that.

EDIT:
It will be harder to get a camcorder to capture the image though because of the lens is in the way of placing the camcorder perpendicular with the screen, but I was just curious if anyone tried it.

Wayne Kinney August 16th, 2005 05:01 AM

how about using a 1 way mirror placed at 45 deg. This way the light from the 35mm lens goes through the mirror, hits the white screen, back and hits the mirror, then to the camcorder?

Wayne.

Keith Wakeham August 16th, 2005 05:07 AM

I was thinking about this, Don't know what the transmission amount of a 1 way mirror is. But it could work if its a good transmission level like 70% or more.

If it hasn't been done by now their might be some big limitation other that getting the camera framed properly on the image created but I don't see image brightness as a problem

Bob Hart August 16th, 2005 11:23 AM

1. Reflecting.

It's not impossible. It could be done with a reflecting lens out of an older zoom-capable photocopier. This is quite an old principle, in WW2 times documents were placed face-up on a platen and lit by a very strong light. The image went up on a screen in the similar manner as the more familiar overhead projector which itself has been supplanted by Powerpoint and LCD projectors.

Depending on where the lens is placed with respect the projection screen surface,the image is the same, larger or smaller than the object being projected. The resolution of these lenses is quite good and distortion does not seem to be an issue as image in undergoes the same abberations as image out which self-cancells within limits. Quite a bright projected image is thrown onto a small projection area. However, as this lens is an entire optical system of itself, you cannot place another lens in front of it which defeats the purpose of using it in an Agus style setup. The light loss of a relecting system would be severe by the time the projection image is made lage enough to deal with imperfections in the reflecting screen surface.

Dan Vance August 16th, 2005 02:34 PM

Quote:

Originally Posted by Wayne Kinney
how about using a 1 way mirror placed at 45 deg. This way the light from the 35mm lens goes through the mirror, hits the white screen, back and hits the mirror, then to the camcorder?
Wayne.

What you're describing is a beamsplitter. Since it has to transmit the light from the lens, then reflect the light from the screen, it must be 50% transparent. That means you lose 1 stop from the lens to the screen, then another stop from the screen to the camcorder, plus the loss through the glass of the splitter (a few %), plus the loss from the screen itself. So under absolutely perfect conditions, you lose about 2-1/2 stops. And now your camcorder is oriented 90° from your lens, so you have to deal with that or add prisms or mirrors (more light loss) to reorient the camera.
Note that using a splitter with a different reflectance, say 70%, just shifts the surface losses, but they still add up to 2 stops in any configuration.
Still, if the light loss is acceptable, someone should give it a try.

Bill Porter August 16th, 2005 04:40 PM

What Keith means is, imagine holding an SLR lens at the appropriate registration distance from a white piece of paper that is so thick that nothing can be seen from the back side of the paper.

Standing in front of the paper, you can see a nice, bright image. Keith is suggesting filming that image from the front.

Besides the obvious issues such as having to hold your camera backward and so on, the problem is as Keith says: the SLR lens is in the way.

If you point the camera at an angle, you have parallax and are not filming a image that is perpendicular to the camera and parallel to the planes of your taking lens, the way you do with a current-generation ____35.

Keith Wakeham August 16th, 2005 04:41 PM

It's seeming like a lot of technical limitations and it probably isn't a good way to do it - If it hasn't been done by know their is likely a real reason for it and that answer will proabably be that it is not good. But I really apperciate the feedback, goes to helping understand some of the stuff.

I might try and proved it to myself somehow with an experimental setup, but from the sounds of it it won't work easily.

Ben Winter August 16th, 2005 04:45 PM

The whole GG mechanism works so well and has so little problems to work out that I don't think it's worth it to go a different route.

The current design just rotates or vibrates ground glass and projects the image onto it. Simple, no mirrors and in most cases, very low light loss. Why try anything else?

Keith Wakeham August 16th, 2005 05:45 PM

I was just curious because I though it might act like video projector systems.

Rear projection has hugh light loss and uniformity problems while front projection looks brighter and more uniform on a poor quality wall vs a good rear projection screen.

Thought for an instant that the GG might parallel the same activity, but I guess not.

Bill Porter August 16th, 2005 06:34 PM

Quote:

Originally Posted by Ben Winter
The whole GG mechanism works so well and has so little problems to work out that I don't think it's worth it to go a different route.

The current design just rotates or vibrates ground glass and projects the image onto it. Simple, no mirrors and in most cases, very low light loss. Why try anything else?

You have got to be kidding me. As far as "Why try anything else?", the whole concept wouldn't exist if you applied the same question to the unmodified camcorders. One could say "The original lens your camcorder cam with requires no additional parts, puts no stress on the filter mounting threads, includes auto focus, auto iris, and wasn't developed for a static still camera."

As far as "works so well," it doesn't work tremendously well, just well enough to use in many cases and not well enough to use in all cases.

You also left out not moving the ground glass at all. :)

Dan Vance August 16th, 2005 06:37 PM

Quote:

Originally Posted by Ben Winter
The whole GG mechanism works so well and has so little problems to work out that I don't think it's worth it to go a different route.

The current design just rotates or vibrates ground glass and projects the image onto it. Simple, no mirrors and in most cases, very low light loss. Why try anything else?

Well:
Noise
Power Requirement
Image Softness
Maintenance
Bearing wear (will eventually cause focus issues)
Hot Spot
Grain--yes even with the $10K P+S device you have to be careful of f-stop setting or grain still becomes visible
Inverted Image (unless you add prisms or mirrors)

Sounds like there's tremendous room for improvement to me. By the way, the nebulous "very low light loss" is just wishful thinking. The nature of a transmissive screen dictates a loss of at least 1.5 stops and 2 stops is a more practical minimum. If someone thinks they are getting less than 1.5 stops loss, they are either measuring the loss incorrectly, or putting up with a horrendous hot spot.
(Or they are one of those people who also own a Humvee that gets 75mpg.)


All times are GMT -6. The time now is 10:22 AM.

DV Info Net -- Real Names, Real People, Real Info!
1998-2024 The Digital Video Information Network