AVS Forum

AVS Forum (http://www.avsforum.com/forum/)
-   2.35:1 Constant Image Height Chat (http://www.avsforum.com/forum/117-2-35-1-constant-image-height-chat/)
-   -   CIH Naked? (http://www.avsforum.com/forum/117-2-35-1-constant-image-height-chat/1446604-cih-naked.html)

indygreg 12-21-2012 08:25 AM

i am considering replacing my aging infocus 777 with either a sim2 or a titan. i have a CIH setup now as follows:

- vertical compression Panamorph U85 lens.
- 322" throw
- 138" x 58" SMX1 screen
- projector brightness measured at 1552 lumens calibrated

We really like the high light output of the infocus and we really focus on 2.4 viewing. almost never 4:3 and once in a while 16:9 but that is mostly for basketball games. if i were to optimize for something it would be 2.4.

My Problem

I found a Lumix Host for sale with a T2 lens which is 1.75 - 2.48. this would go from 133-181 width which would work for a VC lens but not a HE. for HE i would need the longer throw lens and i seem to remember it impacted picture quality? this also means that i can't zoom between 1.78 and 2.4 since no lens will cover both (their T3 would max out at 122" wide).

Question

since i am largely focused on a great 2.4 experience, i wonder about just doing naked smile.gif If i ran with no lense, 2.4 would be as good as it gets since i could be pixel perfect and no extra glass. the sim2 should have anough brightness for this. of course my 1.78 viewing would be roughly 720P which is what i have today. i hate to throw away all the bightness and pixels though.

i considered the sony pj which has pixels to spare but i doubt it has the lumens to spare. i would think compared to my 777 with VC it would look beautiful but weak.

any thoughts?

greg

John Schuermann 12-21-2012 12:15 PM

Longer throws are actually better for HE lenses (actually, pretty much any anamorphic lens). You have about a 26.6' throw and about a 3 to 1 throw ratio, if I have my calculations right. That means you would have almost no pincushion to speak of.

For a Panamorph HE lens system (either the UH480 or DC1), you would need a corrector added to the lens for optimum focus. The stock UH480 and DC1 come pre-focused for a 14.5 - 17.5 foot throw, so you would need to specify your 26.5' throw distance and the corrector would be installed into the lens for that specific throw. It does add about $1K to the price for this option, FYI. I am not sure how this applies to other brands of anamorphic lens.

indygreg 12-21-2012 09:25 PM

Thanks John. Most Sim2's that i find have the T2 lens (both that i have seen available have that lens at them moment). so i would need to buy the pj, buy a T3 lens then buy an HE lens to replace my VC lens. This is what got me thinking that every time i then watch a 2.4 aspect movie (more than 50% of my viewing), i would be scaling the image up (thus not pixel perfect) then shooting through an extra piece of glass. it got me wondering why go to all this cost and complexity when the PJ with a t2 lens would fill my screen naturally (ie, throw away the top and bottom of the picture). the price i would pay would be some brightness and the ability to watch 16 x 9 at full 1080p. but since the picture size is reduced, the pixel per square foot or the resolution per degree of viewing angle would be the same as my 2.4 AR picture.

i wonder in the end if the picture wouldn't be better (or at least as good) without the extra scaling and glass.

greg

indygreg 12-22-2012 11:09 AM

ok, i am answering my own post here but i just did a test. we just tried running my existing projector with no a-lens and with the scaler bypassed. obviously my pj is older/ower quality than the lumis or titan and my a-lens is also lower quality but for my setup i don't see that the a-lens is buying me anything. running with no lens and bypassing the scaler for a scope blu-ray vs running CIH with the lens and scaler engaged was as good or better. surprisingly i didn't see any noticable difference in brightness and even if i did, i have brightness to spare. picture looked slightly sharper with no lens which i believe was contrast. so this stands to reason for me. my thinking is this:

- Scope picture - the lens lets me stretch 720 vertical lines onto 1080 then reshape the picture using an extra peice of glass. in the end there is still only 720 vertical lines so i traded 1:1 pixel mapping and less glass in order to use more of the dmd and i am not sure the extra light or extra pixels did enough to compensate. Scope is by far the least forgiving of any loss of sharpness given the size of the picture. i think no lens wins.

- 16:9 - without a sled i would be running through the extra glass and scaler then throwing away the sides of the image. with a sled i would have the best possible picture.

so the net is that if i get a really high quality HE lens and a sled i can have an optimal 16:9 and a very close to optimal scope picture. if i get a pj that can fill the screen width with no lens i get an optimal scope picture at the expense of a 720P 16:9 image but both have zero additional CA or pincushion introduced and both are uniform brightness with no loss in contrast.

at the very least this tells me that if you don't get a VERY good A-Lens and a sled, you shouldn't get one at all. you are better off cropping the black bars.

what am i missing?

greg

Aussie Bob II 12-23-2012 09:45 PM

At 322" throw with a 58" high screen, your throw ratio is a whopping 3.0 and your pincushion with anamorphic lens will be 8mm, or 0.5%. This is absolutely trivial.

So using a lens won't affect that part of your presentation. Pincushion is a non-issue with your setup.

The quality of your lens is paramount in this situation. On that you are correct. You just need the right lens.

Using a lens will still be brighter than zooming. You don't fully notice it until the bulb is under stress after a couple of hundred hours. Then it becomes a big issue.

indygreg 12-24-2012 07:03 AM

Thanks for your input bob. My bulb is overdue to be replaced but what we noticed was no significant difference in brightness when switching between the two setups. Maybe side by side you could see it.

I seem to remember that there is a significant difference in the lumens depending on which projector lens I use. Would the longer throw t3 lens allow more light or less light than the medium throw t2? Aren't longer throw lenses often slower than short throw lenses?

G

Aussie Bob II 12-24-2012 05:00 PM

Quote:
I seem to remember that there is a significant difference in the lumens depending on which projector lens I use. Would the longer throw t3 lens allow more light or less light than the medium throw t2? Aren't longer throw lenses often slower than short throw lenses?

Good question.

The theory behind having a range of lenses is that as you enter the range of each lens, from wide angle, to medium (or standard) to narrow angle, the f-stop rating of the lens is reset at each stage.

Due to effect of "Numeric aperture", a "wide angle" lens option that zooms from 25mm to 47mm may have the range of f2 to f3. A "medium angle" lens may zoom from 43mm to 70mm, with an f-stop range again of f2 to f3.

In the overlap area 43mm-47mm you'd choose the "medium range" lens because its aperture for the same focal length is f2, rather than the much dimmer f3 of the "wide angle" lens at the same focal length. The f-stop ratings have been "reset" as you move to each lens in the set of lenses provided.

The f-number is FOCAL LENGTH divided by APERTURE DIAMETER. So an f2 lens at longer focal lengths (the medium and long lens options) will have to be physically bigger to achieve the same light throughput. Bigger lenses mean more expense because it's harder to make big lenses that are flaw-free, the glass is expensive, the mechanics are larger and so on (this is why large aperture - low f-stop - zoom lenses for cameras are so expensive, by the way).

In your case, if the zoom factor that you require is in the "overlap" area between the T2 and T3 lenses, use the T3 lens if you want a brighter image... and if you can afford this very much more expensive lens.

John Schuermann 12-26-2012 05:00 PM

Quote:
Originally Posted by indygreg View Post

ok, i am answering my own post here but i just did a test. we just tried running my existing projector with no a-lens and with the scaler bypassed. obviously my pj is older/ower quality than the lumis or titan and my a-lens is also lower quality but for my setup i don't see that the a-lens is buying me anything. running with no lens and bypassing the scaler for a scope blu-ray vs running CIH with the lens and scaler engaged was as good or better. surprisingly i didn't see any noticable difference in brightness and even if i did, i have brightness to spare. picture looked slightly sharper with no lens which i believe was contrast. so this stands to reason for me. my thinking is this:
- Scope picture - the lens lets me stretch 720 vertical lines onto 1080 then reshape the picture using an extra peice of glass. in the end there is still only 720 vertical lines so i traded 1:1 pixel mapping and less glass in order to use more of the dmd and i am not sure the extra light or extra pixels did enough to compensate. Scope is by far the least forgiving of any loss of sharpness given the size of the picture. i think no lens wins.
- 16:9 - without a sled i would be running through the extra glass and scaler then throwing away the sides of the image. with a sled i would have the best possible picture.
so the net is that if i get a really high quality HE lens and a sled i can have an optimal 16:9 and a very close to optimal scope picture. if i get a pj that can fill the screen width with no lens i get an optimal scope picture at the expense of a 720P 16:9 image but both have zero additional CA or pincushion introduced and both are uniform brightness with no loss in contrast.
at the very least this tells me that if you don't get a VERY good A-Lens and a sled, you shouldn't get one at all. you are better off cropping the black bars.
what am i missing?
greg

I think your basic conclusions are pretty solid. Keep in mind that the U85 is an older lens with acrylic optics. If you wanted to stay with vertical compression, you might try a UV200 / FVX200, which replaces the U85 and has glass optics.

The other thing to make sure of is that you are using the VC lens at 4.3X the screen height or greater, and a throw distance of between 14 - 28 feet.

If you go with an HE lens, there are many excellent options on the market from Panamorph and others.

Here is something to consider with your current setup. Since you are running the InFocus 777 (an excellent projector, BTW), you are already scaling the 1080P native output of the Blu-ray to 720P, and then scaling it again to do the vertical stretch. The double scaling pass could very likely be the culprit when it comes to image sharpness. Going to a 1080P native piece will take the double scaling issue out of the equation.

indygreg 12-26-2012 09:51 PM

Thanks John. I hadn't looked at the fvx200. How does the quality stack up to the iscoiii? (I suspect that is a debated subject)

In looking at this I really wonder why no one does a simple zoom lens sled. Seems like the best no scaling solution would be a lense that simply zooms the image to fill a cih screen. Wouldn't it be easier to build a high quality zoom attachment? In fact I am surprised more pjs dont do what digital proj does with accurate zoom memories for this purpose.

Unfortunately the sim2 doesn't appear to be capable of shooting both 16:9 and scope with the same lens for my setup or I might try that approach.

G

John Schuermann 12-27-2012 02:24 PM

Quote:
Originally Posted by indygreg View Post

Thanks John. I hadn't looked at the fvx200. How does the quality stack up to the iscoiii? (I suspect that is a debated subject)
In looking at this I really wonder why no one does a simple zoom lens sled. Seems like the best no scaling solution would be a lense that simply zooms the image to fill a cih screen. Wouldn't it be easier to build a high quality zoom attachment? In fact I am surprised more pjs dont do what digital proj does with accurate zoom memories for this purpose.
Unfortunately the sim2 doesn't appear to be capable of shooting both 16:9 and scope with the same lens for my setup or I might try that approach.
G

Good question, RE: the Isco III. I will try to answer that as fairly and honestly as I can (you are free to take my answer with a grain of salt if you choose, considering that I am speaking as a Panamorph representative wink.gif

The FVX200, when used within its specified throw range and distance, should actually be the crispest and brightest anamorphic lens on the planet regardless of price point, mostly due to the fact that there are only two elements inside the housing. If you do a search on this forum for Mike Lang (MLang46), he wrote a series of posts a few years back from his standpoint as an optical engineer as to why vertical compression lenses are superior to horizontal expansion lenses in almost every regard. There are a few caveats, though:

  • The FVX200 does not have correction for chromatic aberration, which means there is a slight bit of "color fringing" as you move away from the center and get closer to the top and bottom of the picture (this is true of your U85 as well - you can see it if you bring up a white grid test pattern). Because of the vertical compression design, though, the aberration is only about one pixel wide on a 1080P projector even when you get to the very top and bottom of the image. It is almost impossible to see without a test pattern. We could have corrected for this aberration, but it would have meant a higher price tag and diminished brightness / sharpness to correct an issue almost impossible to see with normal picture content.
  • The FVX200 needs to be used at a very long throw - at least 4.3X the picture height. Closer than that and you run the risk of vignetting or objectionable barrel distortion (barrel distortion is where the image bows outward as you move away from the picture edges, the opposite of pincushion, where the picture bows inward as you move away from the edges). For this reason, the FVX200 should not be used with a curved screen (the same is true of your U85).
  • The FVX200 needs to be used in a fixed configuration when paired with a 2.35:1 / 2.40:1 screen, since if you were to move the lens out of the way, your 16:9 image would overshoot the top and bottom of your screen. This means that you need to scale 16:9 material down to 1440 x 1080 using your internal or external scaler. On the positive side, your vertical resolution remains the same at 1080, with vertical resolution being the most important when it comes to how we perceive picture detail. I believe you are already running this type of configuration with the U85.

So, in direct answer to your question, the FVX200 should actually outperform the Isco III when you take the above under consideration. For that matter, the FVX200 should actually outperform our own UH480 and DC1 in the same regards. If you want to compare the Isco III to the Panamorph UH480 or DC1, that would be a more "apples to apples" comparison since all of those lenses are HE (horizontal expansion) designs.

The problem of course with zoom methods are light overspill top and bottom of the screen, diminished light output, and loss of pixel density.

Aussie Bob II 01-01-2013 11:48 PM

Quote:
The FVX200, when used within its specified throw range and distance, should actually be the crispest and brightest anamorphic lens on the planet regardless of price point, mostly due to the fact that there are only two elements inside the housing.

Oh please...

John Schuermann 01-02-2013 10:59 AM

I challenge you to test it yourself.

Seriously. Shoot it out with any lens you care to.

Aussie Bob II 01-02-2013 04:26 PM

An un-corrected system that exhibits "only" a 1 pixel(!) color smear, and that has astigmatism built-in cannot possibly be as good as a color corrected system with astigmatism control.

Color correction in any system, particularly a cylindrical system (which is what Isco and my own systems are), should be (and is) measured in small fractions of a pixel, the smaller the better. A 1-pixel aberration would get me hung from the nearest lamp post if I tried to pass it off as "crisp", much less "the crispest... on the planet".

Having only two elements is a trivial advantage and has nothing to do with "crispness". Each glass surface if anti-reflection coated reflects around 0.5% of the incident light, on average. The extra 4 surfaces on corrected systems therefore, if coated (which all cylindrical systems are), will introduce around 2% extra loss of brightness, in return for better sharpness and no color aberration.

I'll grant you that a vertical compression system - like the FVX200 - will take advantage of the user having to zoom the image larger (prior to optical compression), thus increasing the numeric aperture (and hence the brightness) of the system and the final image, while constant height systems (horizontal expanders) lose a few per cent compared to the zoom method for the contrary reason, but that is not comparing apples with apples.

For a vertical compression system to work in the same way, and with the same convenience as a horizontal expansion system - Constant Height - you would need to remove the lens for 16x9 presentations to make a true comparison, which you specifically recommend against doing. The reason for this recommendation, I suspect, is that a compression system is inherently Constant Width. In order to make it Constant Height and to use the full pixel array, the user will need to both remove the lens and re-zoom the image smaller, which kind of defeats the purpose, I think, forcing users to tinker around with zooming, which they do not have to do with Constant Height systems.

Each system has its foibles, pros and cons, advantages and disadvantages, but to claim an un-corrected prism system that only uses 3/4 of the horizontal pixels is both "crisper and brighter" than an Isco or any other cylindrical system on the planet - and it's a BIG planet - is not telling the full story of the process with all its "ins" and "outs".

John Schuermann 01-02-2013 04:50 PM

Reading through again, I think you are specifically objecting to this:

" mostly due to the fact that there are only two elements inside the housing."

To be absolutely precise, I should have said "mostly due to the fact of the optical design of the two elements within the housing."

Better?

I still stand by the other claims smile.gif

John Schuermann 01-02-2013 05:25 PM

Quote:
Originally Posted by Aussie Bob II View Post

An un-corrected system that exhibits "only" a 1 pixel(!) color smear, and that has astigmatism built-in cannot possibly be as good as a color corrected system with astigmatism control.
Color correction in any system, particularly a cylindrical system (which is what Isco and my own systems are), should be (and is) measured in small fractions of a pixel, the smaller the better. A 1-pixel aberration would get me hung from the nearest lamp post if I tried to pass it off as "crisp", much less "the crispest... on the planet".
Having only two elements is a trivial advantage and has nothing to do with "crispness". Each glass surface if anti-reflection coated reflects around 0.5% of the incident light, on average. The extra 4 surfaces on corrected systems therefore, if coated (which all cylindrical systems are), will introduce around 2% extra loss of brightness, in return for better sharpness and no color aberration.
I'll grant you that a vertical compression system - like the FVX200 - will take advantage of the user having to zoom the image larger (prior to optical compression), thus increasing the numeric aperture (and hence the brightness) of the system and the final image, while constant height systems (horizontal expanders) lose a few per cent compared to the zoom method for the contrary reason, but that is not comparing apples with apples.
For a vertical compression system to work in the same way, and with the same convenience as a horizontal expansion system - Constant Height - you would need to remove the lens for 16x9 presentations to make a true comparison, which you specifically recommend against doing. The reason for this recommendation, I suspect, is that a compression system is inherently Constant Width. In order to make it Constant Height and to use the full pixel array, the user will need to both remove the lens and re-zoom the image smaller, which kind of defeats the purpose, I think, forcing users to tinker around with zooming, which they do not have to do with Constant Height systems.
Each system has its foibles, pros and cons, advantages and disadvantages, but to claim an un-corrected prism system that only uses 3/4 of the horizontal pixels is both "crisper and brighter" than an Isco or any other cylindrical system on the planet - and it's a BIG planet - is not telling the full story of the process with all its "ins" and "outs".

I agree with some of your points, and will modify one statement:

"The FVX200, when used within its specified throw range and distance, *should* actually be the crispest and brightest anamorphic lens on the planet regardless of price point for viewing 2.35:1 / 2.40:1 material."

I did address the point that you did, regarding fixed installation and the resultant lower resolution you will have for 16:9 material. Of course I was answering the Original Poster's question about his existing setup - he is already working with a VC lens in a fixed configuration, so he is already familiar with how it works. Not sure of where you are going with "only uses 3/4 of the horizontal pixels" unless you are referring to 16:9 material, which I directly addressed by pointing out a resulting resolution of 1440 x 1080. I think I made it clear that the fixed VC solution has its pros and cons, and I genuinely don't see how you can claim otherwise. Why else would I use the term "with a few caveats"? And when I specifically pointed out that my comparison with the Isco III was only valid when you take those caveats under consideration? I mean, you say "not telling the full story of the process with all its "ins' and "outs", but I stand by my answer as doing EXACTLY that. And again, my answer has to be put into the context of the particular setup of the original poster.

As far as crispness, again, I re-iterate my challenge to shoot out the FVX200 within its specified throw ratio and range against any other lens on the market. Have I done this myself? No, not with every single lens, which of course is almost impossible (hence my statement "should be"). However, we have tested internally against *many* other lens systems, including those named here, and we have not found any other lens to be crisper than the FVX200. Of course, it is your right to be skeptical of our internal testing (I certainly would be), which is why I challenged you to do your own tests.

One last thing. I would like to point out once again that Panamorph does make lenses with full chromatic and astigmatism correction - the UH480 and DC1 - that sit at the top of our product line. We have been arguing over the lower end pieces in the Panamorph lineup, which I have acknowledged time and again have some compromises in comparison to our high end lens systems.

Aussie Bob II 01-04-2013 12:13 AM

Quote:
We have been arguing over the lower end pieces in the Panamorph lineup, which I have acknowledged time and again have some compromises in comparison to our high end lens systems.

So it's the "crispest and brightest on the planet" AND a "compromise" conpared to other products at the same time? Illogical. Can't be both.

indygreg 01-06-2013 06:27 AM

For me 1440 x 1080 for 16:9 is no biggie. I am of the opinion that since the 16:9 is smaller than the scope picture in a cih and since I don't watch that much 16:9 material, it is not that big a deal. Certainly not worth the added cost and complexity of a sled for me.

I would love to see a myth busters style shoot out of the various approaches someday. Compare vc with the lens always in place, he with a sled and no lense with zooming in a blind test. I wonder if people would see a noticeable difference and if so how big it would be.

Greg

Gary Lightfoot 01-06-2013 07:01 AM

4 Attachment(s)
I think in a blind test, people will find it very hard to tell the difference between a lens that isn't removed for 16:9 (with scaling down to 16:9) and one that is, all else being equal..

Mind you, a cheap manual sled arrangement using draw runners would work just fine.

Gary

Aussie Bob II 01-07-2013 01:04 AM

Quote:
Mind you, a cheap manual sled arrangement using draw runners would work just fine.

You don't get it.

Using a sled with a vertical compression lens results in constant width, not constant height.

That is why the manufacturer recommends using digital cropping from 16x9 to 4x3 and keeping the lens in place.

Gary Lightfoot 01-07-2013 12:46 PM

4 Attachment(s)
Quote:
Originally Posted by Aussie Bob II View Post

Quote:
Mind you, a cheap manual sled arrangement using draw runners would work just fine.

You don't get it.

Using a sled with a vertical compression lens results in constant width, not constant height.

That is why the manufacturer recommends using digital cropping from 16x9 to 4x3 and keeping the lens in place.

Calm down Sheldon smile.gif

Actually I do (having had both types over the years), I just missed the bit where it was a VC lens. My bad.rolleyes.gif

Gary

Kelvin1965S 01-07-2013 03:10 PM

Quote:
Originally Posted by Gary Lightfoot View Post

Calm down Sheldon smile.gif
Actually I do (having had both types over the years), I just missed the bit where it was a VC lens. My bad.rolleyes.gif
Gary

I don't know, first the zooming thing and now this Gary. wink.gif



I'd missed the VC bit too actually. redface.gif

John Schuermann 01-07-2013 04:51 PM

Quote:
Originally Posted by Aussie Bob II View Post

So it's the "crispest and brightest on the planet" AND a "compromise" conpared to other products at the same time? Illogical. Can't be both.

Of course it can't be both, which is why I said that that statement was only true when you take into account the caveats I listed. To be more specific:

If you take the FVX200 and shoot it out against the UH480, DC1, or Isco III (for example) with a white grid test pattern, you will notice that the FVX200 does the very best job of retaining the sharpness of the individual pixels in the center of the image compared to the other lenses. As you move away from the center and more toward the top and bottom of the image, you start to notice that the colors are starting to spread out due to the lack of correction for chromatic aberration, which I acknowledged in my very first post in this topic. Once you get to the very top and bottom of the image, you will notice the aberration has increased to about the width of a single pixel. As a practical matter, the image appears crisper on the FVX200 - even with a test pattern - until you get about halfway from the center toward the top or bottom, at which point the aberration becomes great enough that the UH480, DC1, or Isco III become noticeably sharper. Since the UH480, DC1, and Isco III are fully corrected for chromatic aberration and can be used in a moveable configuration with a 2.35:1 / 2.40:1 screen, I would say that overall these lens systems are superior.

So, when you factor in the above and add in the slight brightness advantage of the FVX200, hopefully you can see why I still stand by my statement smile.gif

Gary Lightfoot 01-08-2013 12:15 PM

4 Attachment(s)
Quote:
Originally Posted by Kelvin1965S View Post

Quote:
Originally Posted by Gary Lightfoot View Post

Calm down Sheldon smile.gif
Actually I do (having had both types over the years), I just missed the bit where it was a VC lens. My bad.rolleyes.gif
Gary

I don't know, first the zooming thing and now this Gary. wink.gif

I think I'm going to have to move! tongue.gif

Gary


All times are GMT -7. The time now is 05:04 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.