or Connect
AVS › AVS Forum › Display Devices › 2.35:1 Constant Image Height Chat › Zoom vs Lens and Projector Zoom Questions
New Posts  All Forums:Forum Nav:

Zoom vs Lens and Projector Zoom Questions - Page 2

post #31 of 127
Quote:
Originally Posted by CAVX View Post

Yes, zoom and some lens shift. You throw the black bars off the top and bottom of the screen.

CAVX,

I have a question for you regarding scope screen ARs. I think you have recommended a scope screen AR of 2:37 in other threads given that 1.78*1.33 = 2.37 (using an A-lens). Do you mean that, black bar issues aside, regardless of whether a movie has a 2:40 or 2:35 aspect ratio, a 2:37 AR screen would achieve a better geometry vs. a 2:35 or 2:40 screen? I understand that a 2:35 AR movie would have side black bars on a 2:40 screen and that a 2:40AR movie would have black bars top and bottom on a 2:35 AR screen but are you saying that in either case a 2:37 AR screen would achieve better proportions given the standard 1:33x horizontal stretching that a A-lens produces. Apologies if I have misquoted you, I am just trying to make sense of it all. I currently have a 100 16x9 Stewart retractable screen coupled with a JVS RS 25 and I am looking to move to a CIH (with A-lens set up) as I find scope movies to be too small on my screen. Thanks for any feedback.
post #32 of 127
Quote:
Originally Posted by CAVX View Post

Yes, zoom and some lens shift. You throw the black bars off the top and bottom of the screen.

if thats the case, then i dont understand what the big deal is with the AE4000 and its Lens Memory? I plan on using a 2.35 screen.
post #33 of 127
Quote:
Originally Posted by ooms View Post

if thats the case, then i dont understand what the big deal is with the AE4000 and its Lens Memory? I plan on using a 2.35 screen.

The lens memory essentially allows you to switch between ARs without having to manually adjust the lens shift and zoom every time - you select a preset and the projector automatically sets the zoom and lens shift for you. Depending on how often you're changing back and forth between ARs, this may or may not be a big deal for you.
post #34 of 127
I wish it was active with V and H fit...then i would be a happier customer, however, I have 3 clicks of the AR button an i'm there.
post #35 of 127
I have done the zooming and it takes me 30 secs to zoom, lens shift and focus going from 1.78 HDTV to 2.35 Cinemascope.

I am using a Epson 1080UB projector & a Cinema Contour 148" 58x136 1/2 Cinemascope Screen....I had a 119" 58x104 before that...Now I do have black fabric wall on the top and bottom of my screen, so as to not see the black coming from the Projector and a completely light controled dedicated theater room

I may go with a anamorphic lens someday(would need a new PJ or a scaler), but i'm 100% happy at the moment!!

Steve
post #36 of 127
Quote:
Originally Posted by KB1 View Post

CAVX,

I have a question for you regarding scope screen ARs. I think you have recommended a scope screen AR of 2:37 in other threads given that 1.78*1.33 = 2.37 (using an A-lens). Do you mean that, black bar issues aside, regardless of whether a movie has a 2:40 or 2:35 aspect ratio, a 2:37 AR screen would achieve a better geometry vs. a 2:35 or 2:40 screen?

The difference between 2.35 and 2.40 is about 17 vertical pixels on a 1080 projector. Both images are letterboxed in the 1920 x 1080 pixel matrix, so you may still see some of the pixels that make up part of the black bars remaining after the Scaling for CIH is applied.

Quote:


I understand that a 2:35 AR movie would have side black bars on a 2:40 screen and that a 2:40AR movie would have black bars top and bottom on a 2:35 AR screen but are you saying that in either case a 2:37 AR screen would achieve better proportions given the standard 1:33x horizontal stretching that a A-lens produces.

As I mentioned, the Scope films are letterboxed on Blu-ray, so they both use the full width of the pixels, just their height slightly varies (about 17 pixels). Assuming that the image is truely centered (and quite often it is not 100% perfect), you would find that 2.40 films still show a 'sliver' of black bar top and bottom.

Generally people (myself included) apply a slight amount of zoom to rid these slivers and therefore you may notice slight overscaning of the image for a 235 film. There is also the fact that HE A-Lenses introduce pincushion, so your not going to get the image perfectly framed to the exact pixel with a lens onto a flat screen anyway.
post #37 of 127
the black bars that show up when projecting 2.35 on 2.40 are tiny. same thing vice versa. they will not even be noticeable. hardly something to obsess over unless you have doctor has diagnosed you with ocd..
post #38 of 127
Quote:
Originally Posted by ooms View Post

the black bars that show up when projecting 2.35 on 2.40 are tiny. same thing vice versa. they will not even be noticeable.

Actually there is no pillars on a 2.35:1 film at the source level simply because the letterboxing process makes the format CIW over the native AR. The only source material I've noticed small side pillars on is the FOX logo on the Star Wars DVDs after the CIH Scaling. This normally is 16:9, so I am guessing that maybe the AR is actually 16:10 (1.66:1).

Of course when projecting, you can have both side pillars for smaller ARs or slivers of black bars for anything wider than the AR of your Scope screen.
post #39 of 127
Quote:
Originally Posted by CAVX View Post

The difference between 2.35 and 2.40 is about 17 vertical pixels on a 1080 projector. Both images are letterboxed in the 1920 x 1080 pixel matrix, so you may still see some of the pixels that make up part of the black bars remaining after the Scaling for CIH is applied.



As I mentioned, the Scope films are letterboxed on Blu-ray, so they both use the full width of the pixels, just their height slightly varies (about 17 pixels). Assuming that the image is truely centered (and quite often it is not 100% perfect), you would find that 2.40 films still show a 'sliver' of black bar top and bottom.

Generally people (myself included) apply a slight amount of zoom to rid these slivers and therefore you may notice slight overscaning of the image for a 235 film. There is also the fact that HE A-Lenses introduce pincushion, so your not going to get the image perfectly framed to the exact pixel with a lens onto a flat screen anyway.

Thanks a lot for the response. So assuming perfect centering, a 2:40 film and a 2:35 film would have the exact same width on a 2:37 screen (1.33x 1.78 screen width) but the 2:40 BD will have slight bars top and bottom and the 2:35 BD will be slightly cropped top and bottom? I'm looking to potentially upgrade to a 49 high scope screen from my current 16x9 (49x87) screen. In my example, a 2:40 BD would have a width of (87 x 1.33= 115.71) and a height of (87/2.4*1.33 = 48.21) while the 2:35 BD would have a width of (87 x 1.33= 115.71) and a height of (87/2.35*1.33 =49.24). Is that correct? I guess that would mean that, as you mentioned, the 2:40 BD would have slight bars top and bottom but also that the 2:35 BD would be slightly cropped top and bottom? Thx
post #40 of 127
Quote:
Originally Posted by KB1 View Post

So assuming perfect centering, a 2:40 film and a 2:35 film would have the exact same width on a 2:37 screen (1.33x 1.78 screen width) but the 2:40 BD will have slight bars top and bottom and the 2:35 BD will be slightly cropped top and bottom?

Correct as the source is based on CIW once the program exceeds the native AR being 1.78:1.
post #41 of 127
Quote:
Originally Posted by CAVX View Post

Correct as the source is based on CIW once the program exceeds the native AR being 1.78:1.

Thx a lot. I'm assuming the slight cropping top and bottom on 2:35 movies is not that noticeable and cannot be avoided?
post #42 of 127
Quote:
Originally Posted by KB1 View Post

Thx a lot. I'm assuming the slight cropping top and bottom on 2:35 movies is not that noticeable and cannot be avoided?

You can always zoom to trim the image if you want. As for my self, I just want to watch movies, so I would not bother and infact have decided to throw 20 pixels top and bottom off the screen to avoid ever seeing a sliver of black bar. This also means that should a film not be 100% centered (which does happen), I am not going to see that either.
post #43 of 127
Quote:
Originally Posted by CAVX View Post

You can always zoom to trim the image if you want. As for my self, I just want to watch movies, so I would not bother and infact have decided to throw 20 pixels top and bottom off the screen to avoid ever seeing a sliver of black bar. This also means that should a film not be 100% centered (which does happen), I am not going to see that either.

Got it, thx.
post #44 of 127
An earlier post in this thread stated,
"You lose no resolution when zooming. The black bars in material over 16:9 are hard encoded. An anamorphic lens will give you a little extra brightness but will also introduce aberrations of its own and will rob some light as it passes through the extra lenses."

"The zoom on the AE4000 is sufficient for 99% of movies depending on your throw distance, worst case scenario would be very small black bars for something like Ben Hur @ 2.76:1"

Is this correct? Does that mean the source disc will always will have black bars at top and bottom and that there truly aren't many discs in scope?
post #45 of 127
Quote:
Originally Posted by michaelscott73 View Post

An earlier post in this thread stated,
"You lose no resolution when zooming. The black bars in material over 16:9 are hard encoded. An anamorphic lens will give you a little extra brightness but will also introduce aberrations of its own and will rob some light as it passes through the extra lenses."

"The zoom on the AE4000 is sufficient for 99% of movies depending on your throw distance, worst case scenario would be very small black bars for something like Ben Hur @ 2.76:1"

Is this correct? Does that mean the source disc will always will have black bars at top and bottom and that there truly aren't many discs in scope?

You,ve got it exactly right. There are NO truly anamorphic discs, they (2.35 format films) are all letterboxed and all have top and bottom black bars. Exactly why some of us think the A-lens with vertical stretch thingy is a bit of a pointless exercise.
post #46 of 127
BD scope discs are 16:9 but with black bars encoded above and below the image. At the moment there are no anamorphic encoded scope movies.

Zooming will result in 33% larger pixels which, depending on your viewing distance, may make pixels more visible and the image less smooth. A lens with scaling will use all of the displays pixels (removing the black bars) so keeping them smaller and give a smoother image much the same as the 16:9 image. Different lenses offer different quality of image, and not all are visible when you use them. Lumen loss is minimal (around 1% when I measured mine).

Gary
post #47 of 127
Quote:
Originally Posted by taffman View Post

Exactly why some of us think the A-lens with vertical stretch thingy is a bit of a pointless exercise.

Zooming pros:

1:1 pixel mapping.
Ability to precisely trim any AR to fit the screen (often requires the use of masking).

Zooming cons:

Throwing away 25% of the vertical rez.
Increasing pixel size by 33% in both directions.

A-Lens Pros:

True CinemaScope projection.
Use of the full panel.

A-Lens Cons:

Scaling distorts 1:1 Pixel mapping.
post #48 of 127
Quote:
Originally Posted by CAVX View Post

Zooming cons:

Increasing pixel size by 33% in both directions.

Which can result in image distortion if you sit close enough in a CIH set up.

That's the reason why people choose a lens with scaling over zooming.

Gary
post #49 of 127
Quote:


Ability to precisely trim any AR to fit the screen

Has nothing to do with the zoom vs lens debate and everything to do with having an HTPC or a very high end scaler. With either of the two HTPC based aspect ratio controller software packages or the two HTPC based DVD players with full AR control, complete control of the size and shape of any source AR from 1.20:1 to 2.85:1 is easily performed. All four of these solutions allow a minimum of 20 AR presets and support individually adjustable 4 way blanking as well.

Vern
post #50 of 127
Quote:
Increasing pixel size by 33% in both directions.

Quote:
Originally Posted by Gary Lightfoot View Post

Which can result in image distortion if you sit close enough in a CIH set up.

Care to explain how linear optical expansion can result in image distortion.

And why it only happens when applied to both vertical and horizontal when zooming but not when applied horizontally only when using a A-Lens.

Further A-Lens Pros: Increase in On/Off contrast for scope material over zooming

A-Lens Cons: Offset of course by a decrease in ANSI contrast.
post #51 of 127
Quote:
Originally Posted by Ken Tripp View Post

Care to explain how linear optical expansion can result in image distortion.

Sure. When you can see pixel edges and stair stepping rather than smooth transitions. In a direct comparison the difference is quite obvious.

Quote:
Originally Posted by Ken Tripp View Post

And why it only happens when applied to both vertical and horizontal when zooming but not when applied horizontally only when using a A-Lens.

Because you keep the pixel height the same when you're using the full resolution of the display and are therefore using more (smaller) pixels in comparison. The eye is also less sensitive to horizontal resolution.

Quote:
Originally Posted by Ken Tripp View Post

Further A-Lens Pros: Increase in On/Off contrast for scope material over zooming

I assume by that you mean because you tend to use a longer throw from the pj which increases on/off CR and reduces lumens due to the F range of the lens?

Quote:
Originally Posted by Ken Tripp View Post

A-Lens Cons: Offset of course by a decrease in ANSI contrast.

But depending on the lens in use, can be minimal. At least that's what I found when I measured my ISCO and found the ANSI to be pretty much the same as native (around 400:1 with an Optoma H78 in non optimal conditions). Others have also measured spherical lenses with similar results (Coldmachine for example).

Gary
post #52 of 127
Quote:
Originally Posted by Gary Lightfoot View Post

Sure. When you can see pixel edges and stair stepping rather than smooth transitions. In a direct comparison the difference is quite obvious.

Because you keep the pixel height the same when you're using the full resolution of the display and are therefore using more (smaller) pixels in comparison. The eye is also less sensitive to horizontal resolution.

But that is not image distortion as the image has not changed. Unlike the curved horizontal lines due to pincushion caused by a HE anamorphic lens when projecting onto a flat screen or electronic scaling artifacts distortion.

Quote:


I assume by that you mean because you tend to use a longer throw from the pj which increases on/off CR and reduces lumens due to the F range of the lens?

Yes? as the throw for 16x9 material would be the same but in zooming by 1.33 for scope you lose some on/off, but you do gain some brightness. YYMV depending on the PJ, what part of the zoom range you're using etc.

Quote:


But depending on the lens in use, can be minimal. At least that's what I found when I measured my ISCO and found the ANSI to be pretty much the same as native (around 400:1 with an Optoma H78 in non optimal conditions). Others have also measured spherical lenses with similar results (Coldmachine for example).
Gary

Cine4Home measured the loss in ANSI contrast at around 10% for Schneider CineDigitar M and XL lenses. They did also raise the issue of optics and dust because you now have three glass surfaces to keep clean instead of one and nothing kills ANSI as quickly as dirty optics.
post #53 of 127
Quote:
Originally Posted by Gary Lightfoot View Post

Sure. When you can see pixel edges and stair stepping rather than smooth transitions. In a direct comparison the difference is quite obvious.
Gary

Stair stepping is not caused by optical scaling/zooming.......sounds like you are refering to nearest neighbour digital scaling.


Quote:
Originally Posted by Gary Lightfoot View Post

Because you keep the pixel height the same when you're using the full resolution of the display and are therefore using more (smaller) pixels in comparison. The eye is also less sensitive to horizontal resolution.
Gary

Spreading existing detail over more pixels does not give one more detail, in fact there is a softening of the image. As demonstrated by AmirM, scaling a 2k image to 4k.

What is needed for A-Lens uses is true anamorphically encoded BluRays.

If the interpixel gaps/pixel boundries in the horizontal direction are less visible using an A-Lens than via zooming to the same image size, then there is detail that is not being transfered to the screen, this will apply to all the image detail as well.

There appears no scientific data to support that "The eye is also less sensitive to horizontal resolution."

Quote:
Originally Posted by Gary Lightfoot View Post

But depending on the lens in use, can be minimal. At least that's what I found when I measured my ISCO and found the ANSI to be pretty much the same as native (around 400:1 with an Optoma H78 in non optimal conditions). Others have also measured spherical lenses with similar results (Coldmachine for example).
Gary

ANSI is one measurement where the readings are taken in the middle of squares, would be interesting to see readings taken between two adjacent pixel rows.

UMR (W. Jeff Meier) did some readings and found large drops in ANSI using an A-Lens(250:1 to 112:1). No doubt higher quality lenses would have less drop in ANSI.

"I recently aided a test with a Prismasonic H-FE1500R and a JVC DLA-RS35. The projector was mounted at a 2.0 throw ratio. Using lens zoom to fill a 10 ft wide 2.35 screen the ANSI contrast was measured at 250:1. With the anamorphic lens the ANSI contrast fell to 112:1. The light level from the screen only dropped 9 percent using lens zoom versus the lens with the same projector location. That difference in light output would be less if the projector would have been mounted at the minimum throw ratio. The light spill caused by zooming the 2.35 image was not visible on a dark brown wall. The image in this case was clearly better zoomed than not."

http://homecinemaguru.com/?p=698

I think you guys should start a movement to bring True Anamorphically Disks to the market. Perhaps supported by the next BluRay revision, 1920x1080 anamorphic encodes to be optically stretched by an A-Lens or down scaled by the player to fit 1920 x 810 for those without or with 16:9 flat panels.
post #54 of 127
Quote:
Originally Posted by Highjinx View Post

Spreading existing detail over more pixels does not give one more detail, in fact there is a softening of the image. As demonstrated by AmirM, scaling a 2k image to 4k.

Scaling for CIH using 1920 x 1080 projectors and 1.33x A-Lenses is only 33% increase (78% larger than the original letterbox image), not 400% as happens when scaling 2K to 4K which is like spreading that same info out to 4x the original area.

At SMPTE last year, JVC was projecting 4m wide images using both their 4K projector and their HD projectors with an anamorphic lens. With BD as the source, I actually prefered the HD projector + A-Lens images. The 4K projector was astounding when fed 4x HD signals, however, each became just 2m wide.

Quote:
Originally Posted by Highjinx View Post

What is needed for A-Lens uses is true anamorphically encoded BluRays.

Agreed!!!
post #55 of 127
Quote:
Originally Posted by Ken Tripp View Post

But that is not image distortion as the image has not changed. Unlike the curved horizontal lines due to pincushion caused by a HE anamorphic lens when projecting onto a flat screen or electronic scaling artifacts distortion.

It will look distorted due to the enhanced edges and jagged pixel transitions - the image has changed because you have zoomed it larger and if your seating distance is close enough you will the technology more than intended. If the BD was mastered on a plasma it wasn't intended to be viewed from the closer cinematic seating distances that we often have with front projection.

Pincushion is easily masked out whereas enlarged pixels can't be, unless you use something like an IMX lens to smooth the edge transitions, but this still doesn't do anything about the pixel size.

Scaling artefacts with a good scaler aren't visible, unlike enlarged pixels with visible edges.

There are pros and cons for both methods, and you have to consider which visible artefacts there will be with both.

Quote:
Originally Posted by Ken Tripp View Post

Yes? as the throw for 16x9 material would be the same but in zooming by 1.33 for scope you lose some on/off, but you do gain some brightness. YYMV depending on the PJ, what part of the zoom range you're using etc.

Agreed


Quote:
Originally Posted by Ken Tripp View Post

Cine4Home measured the loss in ANSI contrast at around 10% for Schneider CineDigitar M and XL lenses. They did also raise the issue of optics and dust because you now have three glass surfaces to keep clean instead of one and nothing kills ANSI as quickly as dirty optics.

10% is negligible and I would think beyond the perception of the HVS. If you zoom a primary lens so that you're using more surface you will probably lose more ANSI than with a good quality anamorphic lens. Dust is an issue with primary lenses as well as others, and is down to the environment as well as how often the owner cleans their lenses.

Gary
post #56 of 127
Quote:
Originally Posted by Highjinx View Post

Stair stepping is not caused by optical scaling/zooming.......sounds like you are refering to nearest neighbour digital scaling.

Much the same thing since you're enlarging the source pixel. Bicubic or Gaussion give much better results. If you assume NN scaling which results in 4 identical pixels to the original source pixel, what's the difference between that and a 4 x optically enlarged pixel?


Quote:
Originally Posted by Highjinx View Post

Spreading existing detail over more pixels does not give one more detail, in fact there is a softening of the image. As demonstrated by AmirM, scaling a 2k image to 4k.

I never said it gave you more detail, I said you're keeping the pixel size smaller so they remain invisible. AmirM also said that an enlarged image with visible edges is distortion. Using smaller pixels and more display resolution will give you a smoother looking image which may be softer, but is still preferable to visible edges and enlarged pixels. Who wants EE?

2K to 4K is a great deal more than the 33% vertical needed when using a lens so more softening can happen there. Of course if you can use intelligent scaling you can get quite good results.

Which image of these two do you prefer:



Both are from a smaller original and one uses NN which in this case is the same as zooming, and the other uses bicubic scaling IIRC. Image original supplied by AmirM btw. The left one may be more accurate to the original, but has distortion in the form of jaggies. do you prefer the one on the left?

Quote:
Originally Posted by Highjinx View Post

What is needed for A-Lens uses is true anamorphically encoded BluRays.

True. But if you want a CIH set up and sit at closer seating distances then we have to make do with what looks best in a particular set up.

Quote:
Originally Posted by Highjinx View Post

If the interpixel gaps/pixel boundries in the horizontal direction are less visible using an A-Lens than via zooming to the same image size, then there is detail that is not being transfered to the screen, this will apply to all the image detail as well.

They are less tall in the vertical due to the smaller pixels (because you are using more), and are being zoomed in the horizontal only. If zooming on one plane loses detail, then zooming on both does as well. Are you saying that zooming loses more detail?

Quote:
Originally Posted by Highjinx View Post

There appears no scientific data to support that "The eye is also less sensitive to horizontal resolution."

There was no scientific data that supported that Bumblebees could fly, but there they were flying.

If that's the case then maybe someone should do the testing in the same way someone did with the Bumblebees (they finally did that recently and know how they do it now).

Quote:
Originally Posted by Highjinx View Post

ANSI is one measurement where the readings are taken in the middle of squares, would be interesting to see readings taken between two adjacent pixel rows.

good point. I wonder how a zoomed primary lens would fare in that respect too.

Quote:
Originally Posted by Highjinx View Post

UMR (W. Jeff Meier) did some readings and found large drops in ANSI using an A-Lens(250:1 to 112:1). No doubt higher quality lenses would have less drop in ANSI.

"I recently aided a test with a Prismasonic H-FE1500R and a JVC DLA-RS35. The projector was mounted at a 2.0 throw ratio. Using lens zoom to fill a 10 ft wide 2.35 screen the ANSI contrast was measured at 250:1. With the anamorphic lens the ANSI contrast fell to 112:1. The light level from the screen only dropped 9 percent using lens zoom versus the lens with the same projector location. That difference in light output would be less if the projector would have been mounted at the minimum throw ratio. The light spill caused by zooming the 2.35 image was not visible on a dark brown wall. The image in this case was clearly better zoomed than not."

http://homecinemaguru.com/?p=698

I'll have to have a read of that. Some years ago I did own a Prismasonic H1000 lens and although I didn't test the ANSI at the time, I can say that I didn't notice a major drop as that test would suggest (my room at that time was probably a limiting factor). That's considerably more than I found with my ISCO in my room with a DLP though (measured without the room having much of an effect so got around 400:1), and more than even Cine4Home found with the Schneider lens they tested. I think Coldmachine found a negligible drop with the ISCOs he's tested too, but we'll have to ask him.

I've used lens filters and found many of those did make a noticeable difference in ANSI though (none glass ones wee the worst). I tested the prism lens in pass through and stretch mode (for a 16:9 image) and found a drop from 101 lux to 99 lux, which is around 1%, and of course, no drop in on/off contrast.

Quote:
Originally Posted by Highjinx View Post


I think you guys should start a movement to bring True Anamorphically Disks to the market. Perhaps supported by the next BluRay revision, 1920x1080 anamorphic encodes to be optically stretched by an A-Lens or down scaled by the player to fit 1920 x 810 for those without or with 16:9 flat panels.

It would be nice to have, but unlikely to happen since I think people with CIH set ups are probably less than 1% of the projector owning population, and those with lenses an even smaller percentage still.

As for me being one of those 'guys', I don't have a projector set up right now and haven't had one for a while. I just know what looks best under certain conditions going on experience.

you have to consider that many people who do back to back testing of zooming vs lens choose the lens. They don't do it because it produces an inferior image.

Gary
post #57 of 127
Originally Posted by Highjinx
There appears no scientific data to support that "The eye is also less sensitive to horizontal resolution."

There is none. The only place you will read that the eye is more sensitive to vertical resolution is on AV forums, where people are extolling the benefits of A-lenses.
post #58 of 127
Quote:
Originally Posted by Ken Tripp View Post

But that is not image distortion as the image has not changed. Unlike the curved horizontal lines due to pincushion caused by a HE anamorphic lens when projecting onto a flat screen or electronic scaling artifacts distortion.



Cine4Home measured the loss in ANSI contrast at around 10% for Schneider CineDigitar M and XL lenses. They did also raise the issue of optics and dust because you now have three glass surfaces to keep clean instead of one and nothing kills ANSI as quickly as dirty optics.

Coldmachine noted something like 3% or less ANSI and really how hard is it to periodically clean the lens surfaces if you see dust ?

The pincushion is less than 3/8th" over 14' on my set up using the ISCOIII.

It's funny how about every three months these things come up again.

Some of us looked at zooming and had the choice to zoom or to use a high quality A lens.


Art
post #59 of 127
Quote:
Originally Posted by Art Sonneborn View Post

It's funny how about every three months these things come up again.

Quite amusing indeed. I admit I enjoy the topic of CIH and lens vs. zoom, but the back and forth banter is really pointless as that topic has been beaten to death ad nauseum. Although this is the first time that I've seen CAVX attempt to provide a balanced viewpoint on the topic. Perhaps we're making progress after all?

Quote:
Originally Posted by CAVX View Post

Zooming pros:

1:1 pixel mapping.
Ability to precisely trim any AR to fit the screen (often requires the use of masking).

Zooming cons:

Throwing away 25% of the vertical rez.
Increasing pixel size by 33% in both directions.

A-Lens Pros:

True CinemaScope projection.
Use of the full panel.

A-Lens Cons:

Scaling distorts 1:1 Pixel mapping.
post #60 of 127
Quote:
Originally Posted by Gary Lightfoot View Post

It will look distorted due to the enhanced edges and jagged pixel transitions - the image has changed because you have zoomed it larger and if your seating distance is close enough you will the technology more than intended. If the BD was mastered on a plasma it wasn't intended to be viewed from the closer cinematic seating distances that we often have with front projection.

This image is a manipulation of one from a Cine4Home review of the JVC HD950 and shows the pixels and inter pixel gap (aka screendoor).



And here's a coloured and lightened version for anybody thinking I'm trying to hide things in the dark. I actually think this second image highlights the screendoor much more than what I see on my own screen.



On the left is what you'd see zooming and the right has been vertically compressed 33% to represent what an anamorphic lens would do.

It has been sized so that on a XGA (1024 x 768 pixel) display it shows the actual pixel and gap size as if it were projected onto a 150" diagonal scope screen.

At a seating distance of twice screen height which is way closer than most people sit you would be viewing this image from nearly 3 metres (10 feet) away on said XGA monitor.

At that distance I certainly can't see any difference in pixel structure, enhanced edges or jagged pixel transitions. In fact I'm struggling to see any pixel structure at all.

Perhaps I need to see an Optometrist.

And this is for a JVC LCoS panel. The Sony's have even higher fill rates. YMMV (greatly) for inferior technologies (for screendoor only Art) like DLP and LCD.

And once again, zooming an image does not distort it. It gets bigger, it gets smaller, as do the pixels and the gaps between them but the relationships between all of them stay exactly the same.
New Posts  All Forums:Forum Nav:
  Return Home
AVS › AVS Forum › Display Devices › 2.35:1 Constant Image Height Chat › Zoom vs Lens and Projector Zoom Questions