I meant that using one prism that equals our two prisms would make things simpler. To me it seems as though the only thing a 4 prism setup accomplishes is to get a larger angle within the glass. Instead of two prisms, why not use just one equal to what the two are accomplishing. That way you may eliminate the extra reflections from two prisms and all those faces back to back.
pinkmouse said:
Eek, yet more internal reflections!
I don't know if it will add more. I tried the diamond stone and it didn't seem to do anything - certainly did nothing bad 🙂
If the projector is the source of the light, then the very first reflection has to be coming from the outer face of the rear prism. If anti-reflection glass was installed there, it has to do some good...
I am going to order some and find out...
Mark
Mark Techer said:
Remember what I said about reflections coming from the transition between two layers?
You will still have the reflection from the back of the prism, and you will have a lesser reflection from the two sides of the glass. The only way to eliminate that is to use a fluid in between the two layers that is exactly the same optical density as both pieces.
pinkmouse said:
Interesting Pinkmouse. I have spent most of the night re-aligning my lens, and only major reflection issue I had was if I attempted too much stretch...
Mark
I am attempting to track down a company that will add antireflective coatings on a custom order basis, but I am not anticipating any luck. I am sure that these small orders are just not worth it for these companies, but I will post what I discover.
Steve Scherrer said:
Yeah that is what I found here. They said that due to the process, it would cost something like $1500 per prism for my lens.
There must be a more cost effective solution out there...
Mark
What about this: http://www.pcconnection.com/ProductDetail?sku=6087979&srccode=cii_9324560&cpncode=09-21555782-2
This seems a little steep, but a 5 pack would give you enough to cover each face of the prisms. Maybe I'll splurge and pick up a pack to see how it fares...
This seems a little steep, but a 5 pack would give you enough to cover each face of the prisms. Maybe I'll splurge and pick up a pack to see how it fares...
I just finished watching "Pirates: The Dead Man's Chest". I am more than happy to report I did NOT see any reflections, even in the darkest passages!
The transfer is brilliant, one of cleanest I've seen. And the sound is AWESOME!
That said, I'm still hoping for an inexpensive anti-reflective coating because there are some seriously offensive scenes in some of my other DVDs! The battle between Darth Vader and Luke in Empire REALLY sucks with these prisms! LOL!
An externally hosted image should be here but it was not working when we last tested it.
The transfer is brilliant, one of cleanest I've seen. And the sound is AWESOME!
That said, I'm still hoping for an inexpensive anti-reflective coating because there are some seriously offensive scenes in some of my other DVDs! The battle between Darth Vader and Luke in Empire REALLY sucks with these prisms! LOL!
artbypaul said:
I can't tell if you're joking about the Empire scenes. That picture you posted of Pirates is pretty stunning. Why would Empire give the nasty reflections, but not Pirates?
Edit: One other thing I wanted to try: as many know, I am trying to go CIH using a 4:3 projector. I believe that I can get the stretch by using these prisms--The stretch required introduces some serious geometry issues, that I believe can be worked out.
However, one other simple solution that I hadn't considered until just now--I own a Panamorph P752. would it be possible to put the two anamorphic lenses in serial? The first to do 33% stretch, and then the other to accomplish the other 33% stretch?
Once again, these prisms are unbelievably mysterious when it comes to SOME scenes showing bad reflections while others that "should" show reflections don't.
I am absolutely positively BAFFLED at the utter randomness of the reflections!!!
When I watch the lightsaber fight with Luke and Darth in "EMPIRE STRIKES BACK" there is a ton of reflections (dual images of bright areas on the left side of the screen mirroring the action to the right).
I am absolutely positively BAFFLED at the utter randomness of the reflections!!!
When I watch the lightsaber fight with Luke and Darth in "EMPIRE STRIKES BACK" there is a ton of reflections (dual images of bright areas on the left side of the screen mirroring the action to the right).
Well, I think I'm getting a handle on the reflections. But I don't have a lot of time to post today, so hopefully this weekend I can put together a more complete post (sorry about that).
But I had to post a great link that I've been playing with to get a better understanding of these prisms. This is a web-browser Java-based optics simulator that will show ray-tracing through a prism and let you change the prism, move it around, change it's shape, etc. It only shows one prism, but it's fun to play with and gives you some idea of where reflections might be coming from. If you play with it, you'll see why it's crucial to mask the ends of the prisms, and you'll get a better feel for internal reflections.
Here is the link: NTNUJAVA Virtual Physics Prism site
Read the instructions on the page. The trick is to click and drag the corners of the prism to change its shape to better match our prisms.
Another cool feature is that the outgoing beam of yellow light will actually change color as you increase the angle of the prism showing those extreme ranges where CA might be a real problem.
Anyway, play with this link...it's pretty fun and interesting. And I'll post more this weekend about what I've determined about the mystery reflections from Cars/James Bond/etc.
But I had to post a great link that I've been playing with to get a better understanding of these prisms. This is a web-browser Java-based optics simulator that will show ray-tracing through a prism and let you change the prism, move it around, change it's shape, etc. It only shows one prism, but it's fun to play with and gives you some idea of where reflections might be coming from. If you play with it, you'll see why it's crucial to mask the ends of the prisms, and you'll get a better feel for internal reflections.
Here is the link: NTNUJAVA Virtual Physics Prism site
Read the instructions on the page. The trick is to click and drag the corners of the prism to change its shape to better match our prisms.
Another cool feature is that the outgoing beam of yellow light will actually change color as you increase the angle of the prism showing those extreme ranges where CA might be a real problem.
Anyway, play with this link...it's pretty fun and interesting. And I'll post more this weekend about what I've determined about the mystery reflections from Cars/James Bond/etc.
After spending a few hours aligning my projector lens and screen, I now have small (but visible) reflections on both sides. But they're not as visible as the single side reflections I had before...
I watched SW2 last night. There is a scene where both Mace and Yoda are discussing the "clone army" with very bright windows behind and to the sides of the charcters. Previously, this sceen would give me grief. It was not too bad last night.
Steve, adding the Panamorph to the mix might produce more problems that it solves. The DVD is based on 33% squeeze, and so are the lenses. If you use the Panamorph first, you then have the 33% offset needed to correct the geometry, so adding the prism then distorts the geometry again...
What are the prisms on the Panamorph made from? I've heard that they are perspex...
Mark
I watched SW2 last night. There is a scene where both Mace and Yoda are discussing the "clone army" with very bright windows behind and to the sides of the charcters. Previously, this sceen would give me grief. It was not too bad last night.
Steve, adding the Panamorph to the mix might produce more problems that it solves. The DVD is based on 33% squeeze, and so are the lenses. If you use the Panamorph first, you then have the 33% offset needed to correct the geometry, so adding the prism then distorts the geometry again...
What are the prisms on the Panamorph made from? I've heard that they are perspex...
Mark
MikeP: That is a phenomenal application with the prisms. I can definitely see where you may get some stray reflections. It looks like you can minimize the amount of reflections based on exactly where the beam enters the prism. Perhaps finding the prisms' "sweet spot" will help minimize internal reflections.
Which leads me to my next question--For those that have prisms that sometimes give reflections in some movies and not others, are the prisms aligned identically, or has the projected image been moved slightly either left or right? Is is possible to take your prisms and move them left or right to eliminate the internal reflections? Alternatively, it appears that you may be able to eliminate internal reflections by not merely blacking out the ends of the prisms, but the faces of the prisms outside of the projected beam. In other words, mask off every part of the prism that isn't utilized by the projected beam.
Which leads me to my next question--For those that have prisms that sometimes give reflections in some movies and not others, are the prisms aligned identically, or has the projected image been moved slightly either left or right? Is is possible to take your prisms and move them left or right to eliminate the internal reflections? Alternatively, it appears that you may be able to eliminate internal reflections by not merely blacking out the ends of the prisms, but the faces of the prisms outside of the projected beam. In other words, mask off every part of the prism that isn't utilized by the projected beam.
I covered half of the prism that sits closest to the projector lens with black tape shortly after I got the prisms, ( I even posted a photo of it here during this LONG thread...Link below). That might help.
Not sure adding anti-reflective sheets will be optimal. We need to cut the reflection from the prism itself. Must find anti-reflective coatings or prisms with anti-reflective coatings already applied to their surfaces.
OK, I've confirmed that the source of the reflection that is visible in the Cars credits and also in the opening to James Bond's Goldeneye (when the white spot light sweeps across the screen just after the MGM logo) is due to internal reflection.
Here is a picture from the ray trace program showing what is happening:
This shows the *second* prism (the one closest to the screen). The incoming yellow beam of light is coming in at a steep angle because it was bent by the first prism (which is not shown in this picture). It is supposed to simulate a bright object appearing on the right edge of the screen.
Notice that in addition to the normal yellow beam on the right edge of the screen (far right side of picture), there is a black reflection line hitting the left edge of the screen.
It's a bit hard to see the details, but this reflection is coming from the main beam reflecting from the final prism surface, then reflecting internally from the next-to-last surface (the surface closer to the projector), then refracting from the final prism surface again.
The ray trace program gives an intensity of 0.1% of the main beam intensity. This doesn't sound like much, but let's put it in perspective...the side reflections (the ones that you get on the side walls if you don't have an enclosure) are only about 4% of the total beam intensity. And the reflection on the left edge of the screen is clearly much dimmer than these side reflections when you are looking for it. That's why it only appears when the left side is normally completely black.
I was able to stand next to the screen where the dim reflection is on the left edge and look back towards the projector. The reflection is clearly visible in the prism, and looks pretty much like the ray-trace picture.
I also rules out other possibilities:
1) By placing a black cloth over the screen along the right side, the reflection on the left side still appears. So this rules out reflection from the screen back towards the lens.
2) By moving a piece of cardboard between the two prism, I can verify that this reflection is not coming from any interaction with the first prism. It's completely from the internal reflection of the second prism.
Unfortunately, this means that coating the prisms with a non-reflective coating isn't going to work. It's not the outside surfaces causing the reflection. It's the internal surfaces of the prism.
By playing with the ray trace program, it appears that a prism with a larger angle will move this reflection more to the left so that it ends up off the edge of the screen. So, a larger prism angle would definitely help. Creating a larger prism out of 2 smaller prisms makes the problem worse because it introduces more internal surfaces to reflect from. But since the final prism is still the same size, it's still going to cause the same internal reflection on the left side of the screen.
I'm going to continue playing with this. Once I start my holiday break, I hope to have some time to write my own ray trace program that actually uses multiple prisms so that we have something better to play with. Now that I see the formulas, it shouldn't be too hard to code. I'll definitely share anything that I come up with.
But for now it looks like the reflection is something we will have to live with
Here is a picture from the ray trace program showing what is happening:
This shows the *second* prism (the one closest to the screen). The incoming yellow beam of light is coming in at a steep angle because it was bent by the first prism (which is not shown in this picture). It is supposed to simulate a bright object appearing on the right edge of the screen.
Notice that in addition to the normal yellow beam on the right edge of the screen (far right side of picture), there is a black reflection line hitting the left edge of the screen.
It's a bit hard to see the details, but this reflection is coming from the main beam reflecting from the final prism surface, then reflecting internally from the next-to-last surface (the surface closer to the projector), then refracting from the final prism surface again.
The ray trace program gives an intensity of 0.1% of the main beam intensity. This doesn't sound like much, but let's put it in perspective...the side reflections (the ones that you get on the side walls if you don't have an enclosure) are only about 4% of the total beam intensity. And the reflection on the left edge of the screen is clearly much dimmer than these side reflections when you are looking for it. That's why it only appears when the left side is normally completely black.
I was able to stand next to the screen where the dim reflection is on the left edge and look back towards the projector. The reflection is clearly visible in the prism, and looks pretty much like the ray-trace picture.
I also rules out other possibilities:
1) By placing a black cloth over the screen along the right side, the reflection on the left side still appears. So this rules out reflection from the screen back towards the lens.
2) By moving a piece of cardboard between the two prism, I can verify that this reflection is not coming from any interaction with the first prism. It's completely from the internal reflection of the second prism.
Unfortunately, this means that coating the prisms with a non-reflective coating isn't going to work. It's not the outside surfaces causing the reflection. It's the internal surfaces of the prism.
By playing with the ray trace program, it appears that a prism with a larger angle will move this reflection more to the left so that it ends up off the edge of the screen. So, a larger prism angle would definitely help. Creating a larger prism out of 2 smaller prisms makes the problem worse because it introduces more internal surfaces to reflect from. But since the final prism is still the same size, it's still going to cause the same internal reflection on the left side of the screen.
I'm going to continue playing with this. Once I start my holiday break, I hope to have some time to write my own ray trace program that actually uses multiple prisms so that we have something better to play with. Now that I see the formulas, it shouldn't be too hard to code. I'll definitely share anything that I come up with.
But for now it looks like the reflection is something we will have to live with
Thanks so much MikeP! You saved the rest of us a lot of headaches in trying to isolate the source of the reflections.
I guess it's no different than people who experience the "rainbow" effect from DLP projectors. Some see it, some don't.
I don't know if I can find a bigger prism than the J25, so I am probably stuck with the reflections.
Fortunately, I've never had anyone mention seeing the reflections on the screen. So, I guess it will bug me a little bit, but as long as other people aren't distracted it's fine.
I guess it's no different than people who experience the "rainbow" effect from DLP projectors. Some see it, some don't.
I don't know if I can find a bigger prism than the J25, so I am probably stuck with the reflections.
Fortunately, I've never had anyone mention seeing the reflections on the screen. So, I guess it will bug me a little bit, but as long as other people aren't distracted it's fine.
I've finally made my way through this entire thread (several hours of reading) and am completely intrigued with this project. This is just too interesting not to try. I know it's a step backwards, but I'm planning on making a few water prisms first just for experimentation before buying some of the trophy wedges.
My only concern is that I have a fairly short throw (about 1.3), so I may have problems with CA and probably perhaps the geometry at the edges, but I still have to give it a try.
The online prism program (thanks MikeP) is fun to use and does a pretty good job of showing where the reflections are. I did want to mention that even if the problem is caused by internal reflections, an AR coating should still help out. If I remember correctly, an AR coating works both ways, so it would reduce reflections for light hitting the front surface from outside, and will also reduce the internal reflections from light that is already inside the prism and hits one of the surfaces from inside the prism.
My only concern is that I have a fairly short throw (about 1.3), so I may have problems with CA and probably perhaps the geometry at the edges, but I still have to give it a try.
The online prism program (thanks MikeP) is fun to use and does a pretty good job of showing where the reflections are. I did want to mention that even if the problem is caused by internal reflections, an AR coating should still help out. If I remember correctly, an AR coating works both ways, so it would reduce reflections for light hitting the front surface from outside, and will also reduce the internal reflections from light that is already inside the prism and hits one of the surfaces from inside the prism.