What do you use for your CIH screen: an anamorphic lens or zoom method? - Page 4 - AVS Forum | Home Theater Discussions And Reviews
View Poll Results: How do you display CinemaScope content on your CIH screen?
Anamorphic lens 58 33.53%
Zoom method 97 56.07%
Anamorphic lens, but will start using the zoom method 6 3.47%
Zoom method, but will start using an anamorphic lens 12 6.94%
Voters: 173. You may not vote on this poll

Forum Jump: 
 28Likes
Reply
 
Thread Tools
post #91 of 109 Old 06-08-2019, 06:20 PM
AVS Forum Special Member
 
Gary Lightfoot's Avatar
 
Join Date: Jun 2000
Location: UK
Posts: 6,434
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 1302 Post(s)
Liked: 1053
Quote:
Originally Posted by darksets View Post
If you are of the school of thought that it's just a question of subjective evaluation then like with Killroy we don't have much to talk about since I haven't used an anamorphic lens.
In which case you can't comment or make any conclusions and your assumptions are hardly scientific in nature - part of the definition of the word 'science' is facts learned through experiments and observation none of which have been done by you, so therefore your approach is not scientific.

In science theories have to be tested and proved, as the theorised science may not always be correct, much like in your case here.

Quote:
Originally Posted by darksets View Post
I believe that the subjective experience must be supported by scientific facts. There are people who claim they have seen ghosts and other supernatural phenomena. I dismiss their claims even though I can't dispute their experience. In the video world we have things like sharpness control and edge enhancement which give the appearance of a sharper image although scientific analysis shows that it has the opposite effect. TV manufacturers and stores boost the colors in the settings so that the TVs will look better in the showroom. So arguments of the type such and such dealer does this to improve sales are not going to fly where I'm concerned.
See my comments above regarding the definition of science. If the science says one thing but the testing shows something else, then the underlying science is wrong - that's how a lot of science has evolved over the years and why testing has to be done. The Hadron Colider was built for that very purpose - to prove the scientific theories.

Because video tends to be a visual medium, doing testing and comparisons of real things is a good way to determine image quality IMHO. Bringing ghosts and other nonsense into the discussion to undermine other peoples actual testing (compared to your lack of any) is hardly a valid comparison. It also sounds like you've never met Alan Roser and don't know who he was.

An ISCO lens for example is made from good quality glass optics and is better than many prime lenses found in projectors and can also cost more than the projector - some Sony projectors use a plastic final element in their lens but I wouldn't judge the projector solely based on that fact - I'd have to see the image for myself.

Quote:
Originally Posted by darksets View Post
Now regarding your arguments about pixel density and scaling, let's first clarify some facts. I have a 4k projector and a 2.35:1 screen. Using a Lumagen we have the following scenarios:

- With 4k sources, if the aspect ratio is 2.35:1 or 2.40:1 there is no scaling and we have 1:1 pixel mapping. That's the vast majority of scope movies and the reason I have a scope screen. When the source has an aspect ratio less than 2.35:1 there is scaling.

- With 1080p sources, for 2.35:1 (and possibly 2.40:1 I'm not sure about that) there is doubling which is the most benign form of scaling. Again for aspect ratios less than 2.35:1 there is scaling.

- With lower resolution sources there are various forms of scaling.
I guess the 'benign' scaling you're referring to is nearest neighbour? In which case you're using 4 smaller identical pixels to recreate one on a larger scale which just makes the image coarser so is the worst type of scaling - you may be better off zooming the single pixel as four smaller pixels will have more gaps and less image over the same area (or stick with a 1080 pj for Blu Ray if pixel to pixel is that important to you). All other forms of scaling you are using are making all new pixels, so no real difference between what you're doing and what an A lens user is doing (except they are using more pixels), making your 1:1 pixel mapping argument moot in the most part. If scaling is bad why are you doing it and not just zooming the native image all the time, or do you agree that scaling makes for a better image than native and zooming the non 2.40 ratios?

Other algorithms often look better than nearest neighbour IIRC which will look blocky/coarse in comparison. I don't know if you know Madshi and MadVR but this post shows how much better some scaling algorythms are compared to NN:

https://www.avsforum.com/forum/24-di...l#post46304233

Although it's not discussing A lenses, the scaling element of the discussion is relevant.

Quote:
Originally Posted by darksets View Post
Your main argument is about pixel density and it boils down to whether the higher pixel density compensates for the scaling and other lens effects (geometry, color etc.). We are not in the old days of the terrible screen door effect, modern 4k projectors have extremely high pixel density. You have the advantage of having done comparisons and I can't dispute your preference. It may well be that the extra density does more than compensate for the other artifacts but even then it would be a function of screen size and sitting distance. If you sit far enough you can't see the pixel gaps but you can always see chromatic aberration. I find comfort in the thought that the image I see is the closest possible to the source.
You assume that a lens adds artefacts and that they are visible despite having been told that not all lenses do. I thought you preferred a more scientific approach than just guessing, but don't appear willing to do any form of testing yourself or listen to those who have.

Using more pixels with an A lens and good scaling to render the image does indeed produce better results than using less pixels and zooming them larger. Even with 4K displays, even though the pixels are smaller, the result is the same when comparing an A lens with zooming, just more subtle compared to lower res displays.

Quote:
Originally Posted by darksets View Post
Finally a couple of words on your comments about scaling in post production. Just because they introduce some distortion doesn't mean we should pile on some more on top of it. Every image altering step we put in the path degrades the image further and we want to have as few as possible. I can't control what goes into the blu-rays I buy, in choosing my video equipment my goal is to reproduce what's in the source as faithfully as possible.
This is an interesting watch - part two in particular:

http://yedlin.net/ResDemo/#

Given that you seem happy to be scaling as well and don't appear to have any issues with it, I think it's safe to say that those who use lenses with scaling are having an equally issue free image as a result as well.
ask4me2, Killroy and dschulz like this.

Quote:
Originally Posted by elmalloc
Who says Cameron is "right" and why do we care about him so much - lol!

I trust Gary Lightfoot more than James Cameron.
Gary Lightfoot is offline  
Sponsored Links
Advertisement
 
post #92 of 109 Old 06-09-2019, 01:03 AM
Member
 
ask4me2's Avatar
 
Join Date: Apr 2005
Location: Norway
Posts: 148
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 76 Post(s)
Liked: 48
Quote:
Originally Posted by darksets View Post
I am simply saying that subjective experience must be supported by the underlying science. The image without an anamorphic lens is closer to the source than the image with an anamorphic lens in the path. I have given a more extensive analysis in my answer to Gary Lightfoot. It's quite possible that even I would find the picture with an anamorphic lens more appealing but the knowledge of the underlying processing artifacts would bother me like a thorn in my thumb.
darksets, If you do not like the anamorphic stretch scaling , that is OK, but find it peculiar you like what the extra Lumagen prosessor/scaler is doing in the equation then. Your statement that the " Lumagen by itself is better than an anamorphic lens"... when you have no experience with A-lenses on your own, also sounds a little strange to me...

One reason why i chooses the A-lens path, was to get rid if the " thorn in my thum" feeling knowing about how the chip(s) in the light engine is used when I ~90% of the time is running scope movies on the projector.
Have you done some thinking about how only 66% of the projectors resolution and light output is used, and how the static black bars makes the light engine be heated differently when showing a scope movie where 33% of the pixles on the screen is stuck in one position for hours...? well I did. That is probably not a big problem for the projector, and there is not a big burn in or retention problem doing this, but cannot help for the feeling of doing this to the light engine...

Having used both A-lenses and the zoom method over the years, and because I also like to view the screen watching movies at a distance between 1.5-2x image heights from my e-shift JVC projector and found out that the zoom method looked "more digital" than using the ISCO lenses. So cannot agree with you about any underlying processing artifacts for vertical stretch either. If you want to think about processing when watching a movie, then i hope you know about the processing needed to get the color resolution Chroma that is only 1/4 of the luma resolution from the 4:2:0 source to be a nice looking color image.

You did not mention the make and model of your 4K projector, but if it has a native 4096 x 2160 light engine, a 1.25x stretch or a 0.8 squeeze A-lens will give the projector the full use of its light and resolution potential on scope source material, and using the brilliant Lumagen, i guess there will be rely hard to see any processing artifacts even up close to the screen if you want to do some pixel peeping to ease the thorn in the thumb feeling.

When it comes to how native and 1x1 pixel mapping, we are able to view our BD and 4K UHD BD movies, think Gary Lightfoot have covered a little bit of that earlier.

If we think about the history of movie making and what kind of lenses that is used on the film makers cameras when we typically get the blue horizontal anamorphic
lines, that is even added digitally in movies today. I think we shod not be afraid using rely good A-lenses in the home cinemas too, and good A-lenses will not add any noticeable flares when projecting a picture from a digital projector today.

Screenshot from Terminal


Screenshot from Christine




My point of bringing the use of anamorphic lenses on digital cameras, is that even in the first stage of getting the "original" camera pixels, there is often an optical stretch that need to be digitally done to create the correct AR. The A-lens flare in the movie industry is almost like what the guitar tube amp. distortion in rock music.

Last edited by ask4me2; 06-09-2019 at 08:27 AM.
ask4me2 is offline  
post #93 of 109 Old 06-09-2019, 07:35 PM
Member
 
darksets's Avatar
 
Join Date: Jul 2006
Location: Hoboken, NJ
Posts: 182
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 68
Quote:
Originally Posted by Gary Lightfoot View Post
In which case you can't comment or make any conclusions and your assumptions are hardly scientific in nature - part of the definition of the word 'science' is facts learned through experiments and observation none of which have been done by you, so therefore your approach is not scientific.

In science theories have to be tested and proved, as the theorised science may not always be correct, much like in your case here.
Your understanding of science is severely lacking. I don't have time to expand on it but the central concept is that experiment is supposed to verify the core theories not their application. Observation of projected image is not supposed to cast doubt on the principles of optics.

Quote:
Originally Posted by Gary Lightfoot View Post
I guess the 'benign' scaling you're referring to is nearest neighbour? In which case you're using 4 smaller identical pixels to recreate one on a larger scale which just makes the image coarser so is the worst type of scaling - you may be better off zooming the single pixel as four smaller pixels will have more gaps and less image over the same area (or stick with a 1080 pj for Blu Ray if pixel to pixel is that important to you). All other forms of scaling you are using are making all new pixels, so no real difference between what you're doing and what an A lens user is doing (except they are using more pixels), making your 1:1 pixel mapping argument moot in the most part. If scaling is bad why are you doing it and not just zooming the native image all the time, or do you agree that scaling makes for a better image than native and zooming the non 2.40 ratios?

I get it, you are obsessed with gaps between the pixels and you don't care about creating a pixel that doesn't exist in the source. In your opinion replicating a pixel in the source is worse than creating a new one that wasn't there because of the gaps between the pixels. This is the core of our disagreement.

Quote:
Originally Posted by Gary Lightfoot View Post
You assume that a lens adds artefacts and that they are visible despite having been told that not all lenses do.
Wow, that's great news for the optics industry and all those people who try to make lenses that minimize image distortion. Please tell your secrets to lens manufacturers for vision glasses. Also, since you are so scientifically inclined, please publish your discoveries in Physics journals and advise people to correct the relevant chapters in books, such as Feynman's "Lectures On Physics" Volume 1 Chapter 27 (paragraph 27-6 on aberrations especially).
darksets is offline  
Sponsored Links
Advertisement
 
post #94 of 109 Old 06-10-2019, 07:54 AM
AVS Forum Special Member
 
Gary Lightfoot's Avatar
 
Join Date: Jun 2000
Location: UK
Posts: 6,434
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 1302 Post(s)
Liked: 1053
Quote:
Originally Posted by darksets View Post
Your understanding of science is severely lacking. I don't have time to expand on it but the central concept is that experiment is supposed to verify the core theories not their application. Observation of projected image is not supposed to cast doubt on the principles of optics.
Oh dear.

Well, considering I'm the one that has done some testing to verify which method produces the best image (after all, this is a visual medium), I don't think I'm the one that is lacking anything. You're the one without the knowledge and experience, not me.

In theory, pixel to pixel should look better, in practice (because of zooming), it doesn't. That's science and research for you - you should try it some time. It can prove if theory is wrong.


Quote:
Originally Posted by darksets View Post
I get it, you are obsessed with gaps between the pixels and you don't care about creating a pixel that doesn't exist in the source. In your opinion replicating a pixel in the source is worse than creating a new one that wasn't there because of the gaps between the pixels. This is the core of our disagreement.
I just care about what gives the best image with what I have available to me within my budget. My next step will probably be adding an HTPC back into the set up.

Pointing out the deficiencies in your method isn't an obsession, it's just a statement of fact based on observation and a little common sense. You don't understand the benefits of scaling over just making pixels larger and that's the the point you are missing (even though you use scaling and that's fine when it suits you). You need to do some research - this subject has been covered numerous timed on on this forum and if you had bothered to look at the links I provided and the images they contain you may learn something.

Again, the point is that when you make the pixels 33% larger they become visibly coarser (and as Ask4Me has said, the image also appears more digital). When scaling using more (smaller) pixels it creates a better image that looks closer to the intent, so in this case, keeping as close to the source is actually the worst thing you can do with respect to a pleasing image:

This image is just an example of the differences between nearest neighbour and two other types of scaling. It should (hopefully) give you an idea of why NN is bad, and why other forms of scaling are better (see the link in my earlier post, and please watch the video)



Quote:
Originally Posted by darksets View Post
Wow, that's great news for the optics industry and all those people who try to make lenses that minimize image distortion. Please tell your secrets to lens manufacturers for vision glasses. Also, since you are so scientifically inclined, please publish your discoveries in Physics journals and advise people to correct the relevant chapters in books, such as Feynman's "Lectures On Physics" Volume 1 Chapter 27 (paragraph 27-6 on aberrations especially).
With continuous tone images (video) having a good A lens in the light path is invisible to the image we see. However, if you remove the lens and zoom, you will notice the image looks coarser and more digital and less pleasing. So in this case, no lens results in an image that appears distorted (because it's coarse and blocky) compared to the image with the lens (which appears more natural).


Like Ask4Me I also sit approx 2xSH back from my scope screen, but without the lens it's more like 1.5xSH. Without the lens the image is still watchable (I remove the lens for taller IMAX content), but the image is better with the lens in place so that's why I now continue to use it after having been zooming for 6 months - zooming is OK, but the image from the lens is better.

Rather than argue about something you have no experience or understanding of, why not do what others have done and test and compare?
Attached Thumbnails
Click image for larger version

Name:	Scaling1.jpg
Views:	98
Size:	66.4 KB
ID:	2578286  
ScottAvery, ask4me2 and dimi123 like this.

Quote:
Originally Posted by elmalloc
Who says Cameron is "right" and why do we care about him so much - lol!

I trust Gary Lightfoot more than James Cameron.
Gary Lightfoot is offline  
post #95 of 109 Old 06-10-2019, 09:59 AM
AVS Forum Special Member
 
ScottAvery's Avatar
 
Join Date: Mar 2002
Location: Great Falls, VA
Posts: 1,651
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 632 Post(s)
Liked: 183
Gary, thanks for showing the detailed scaling samples. The doubling comment made my brain melt a little.

I assume there must be an illusory sense of sharpness created by pixel spacing because LCD and DLP reviews are always touting sharpness without acknowledging the significant pixel pitch increase over LCOS solutions which give a more film-like image, especially with an A-lens.
ScottAvery is offline  
post #96 of 109 Old 06-10-2019, 11:58 AM
AVS Forum Special Member
 
Gary Lightfoot's Avatar
 
Join Date: Jun 2000
Location: UK
Posts: 6,434
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 1302 Post(s)
Liked: 1053
Quote:
Originally Posted by ScottAvery View Post
Gary, thanks for showing the detailed scaling samples. The doubling comment made my brain melt a little.

I assume there must be an illusory sense of sharpness created by pixel spacing because LCD and DLP reviews are always touting sharpness without acknowledging the significant pixel pitch increase over LCOS solutions which give a more film-like image, especially with an A-lens.
Hi Scott,

With scaling for an A lens which is just a 33% enlargement, a good scaler will scale up to around double, and then scale back down. That gives better results than just scaling by 33%.

I think LCDs sharpness comments emanate from the greater pixel spacing, even with eshifting (like with the Epsons), and DLP because its usually single chip which doesn't have convergence issues. One thing I found many years ago was keeping image brightness down to cinema levels (12fL or less back in the day of DVD) tended to help with DLP and give that more of an analogue look, but I agree that when I first saw the JVC HD1 back in 2007, it did look very close to the Sony G90 CRT I'd seen just days before. That's changed a little since they started driving the panels digitally and added eshift.

I think how we see the images is a result of illusion anyway, especially with how DLP works, but at the end of the day it's what we see and what we like that counts.

Quote:
Originally Posted by elmalloc
Who says Cameron is "right" and why do we care about him so much - lol!

I trust Gary Lightfoot more than James Cameron.
Gary Lightfoot is offline  
post #97 of 109 Old 06-10-2019, 12:14 PM
AVS Forum Special Member
 
ScottAvery's Avatar
 
Join Date: Mar 2002
Location: Great Falls, VA
Posts: 1,651
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 632 Post(s)
Liked: 183
Quote:
Originally Posted by Gary Lightfoot View Post
Hi Scott,

With scaling for an A lens which is just a 33% enlargement, a good scaler will scale up to around double, and then scale back down. That gives better results than just scaling by 33%.

I think LCDs sharpness comments emanate from the greater pixel spacing, even with eshifting (like with the Epsons), and DLP because its usually single chip which doesn't have convergence issues. One thing I found many years ago was keeping image brightness down to cinema levels (12fL or less back in the day of DVD) tended to help with DLP and give that more of an analogue look, but I agree that when I first saw the JVC HD1 back in 2007, it did look very close to the Sony G90 CRT I'd seen just days before. That's changed a little since they started driving the panels digitally and added eshift.

I think how we see the images is a result of illusion anyway, especially with how DLP works, but at the end of the day it's what we see and what we like that counts.
I'm in your camp on this, but for me there simply was no choice as I wanted CIH with a wider scope screen than I could reach with the natural throw of my room, so a horizontal explansion A-lens lets me go 33% wider. I can't afford a Lumagen but the Oppo 203 at least scales for all of my sources for ratios 2.37 and greater. It would be nice to have custom scaling for all resolutions -- it would make install simpler, for sure -- but the Oppo covers the important ratios for me.
ScottAvery is offline  
post #98 of 109 Old 06-10-2019, 12:26 PM
AVS Forum Special Member
 
Gary Lightfoot's Avatar
 
Join Date: Jun 2000
Location: UK
Posts: 6,434
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 1302 Post(s)
Liked: 1053
Quote:
Originally Posted by ScottAvery View Post
I'm in your camp on this, but for me there simply was no choice as I wanted CIH with a wider scope screen than I could reach with the natural throw of my room, so a horizontal explansion A-lens lets me go 33% wider. I can't afford a Lumagen but the Oppo 203 at least scales for all of my sources for ratios 2.37 and greater. It would be nice to have custom scaling for all resolutions -- it would make install simpler, for sure -- but the Oppo covers the important ratios for me.
I had the choice as my room is quite long compared to the width, but as the lens gave better results over zooming I kept it.

I'm using a Panasonic player which scales BD quite well but I also use the projectors own internal anamorphic modes which works fine with scope and 1.85, but with an HTPC I'll be able to scale all other ratios equally well.

Quote:
Originally Posted by elmalloc
Who says Cameron is "right" and why do we care about him so much - lol!

I trust Gary Lightfoot more than James Cameron.
Gary Lightfoot is offline  
post #99 of 109 Old 06-10-2019, 01:01 PM
AVS Forum Special Member
 
ScottAvery's Avatar
 
Join Date: Mar 2002
Location: Great Falls, VA
Posts: 1,651
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 632 Post(s)
Liked: 183
Quote:
Originally Posted by Gary Lightfoot View Post
I had the choice as my room is quite long compared to the width, but as the lens gave better results over zooming I kept it.

I'm using a Panasonic player which scales BD quite well but I also use the projectors own internal anamorphic modes which works fine with scope and 1.85, but with an HTPC I'll be able to scale all other ratios equally well.
I'm going off topic, I am sure, but am I correct in thinking there is no good way to input to an HTPC to use it to scale external input? Need a solution for Netflix' weird ratios ranging all over between 2.35 and 1.78 and I thought Netflix did not work well out of an HTPC.
ScottAvery is offline  
post #100 of 109 Old 06-10-2019, 01:57 PM
AVS Forum Special Member
 
Gary Lightfoot's Avatar
 
Join Date: Jun 2000
Location: UK
Posts: 6,434
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 1302 Post(s)
Liked: 1053
Quote:
Originally Posted by ScottAvery View Post
I'm going off topic, I am sure, but am I correct in thinking there is no good way to input to an HTPC to use it to scale external input? Need a solution for Netflix' weird ratios ranging all over between 2.35 and 1.78 and I thought Netflix did not work well out of an HTPC.
Currently you're correct, though there may be a MadVR Envy product that will allow that in the not too distant future.

I haven't really looked into Netflix, but I think the problem is that it's restricted to being held within a browser, and most people seem to say that set top boxes or Panasonic players for example do a better job, so we may have to wait a while yet for it to become more HTPC friendly, which is a shame.

Quote:
Originally Posted by elmalloc
Who says Cameron is "right" and why do we care about him so much - lol!

I trust Gary Lightfoot more than James Cameron.
Gary Lightfoot is offline  
post #101 of 109 Old 06-16-2019, 06:52 PM
Member
 
darksets's Avatar
 
Join Date: Jul 2006
Location: Hoboken, NJ
Posts: 182
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 68
Quote:
Originally Posted by Gary Lightfoot View Post
Oh dear.

Well, considering I'm the one that has done some testing to verify which method produces the best image (after all, this is a visual medium), I don't think I'm the one that is lacking anything. You're the one without the knowledge and experience, not me.

In theory, pixel to pixel should look better, in practice (because of zooming), it doesn't. That's science and research for you - you should try it some time. It can prove if theory is wrong.




I just care about what gives the best image with what I have available to me within my budget. My next step will probably be adding an HTPC back into the set up.

Pointing out the deficiencies in your method isn't an obsession, it's just a statement of fact based on observation and a little common sense. You don't understand the benefits of scaling over just making pixels larger and that's the the point you are missing (even though you use scaling and that's fine when it suits you). You need to do some research - this subject has been covered numerous timed on on this forum and if you had bothered to look at the links I provided and the images they contain you may learn something.

Again, the point is that when you make the pixels 33% larger they become visibly coarser (and as Ask4Me has said, the image also appears more digital). When scaling using more (smaller) pixels it creates a better image that looks closer to the intent, so in this case, keeping as close to the source is actually the worst thing you can do with respect to a pleasing image:

This image is just an example of the differences between nearest neighbour and two other types of scaling. It should (hopefully) give you an idea of why NN is bad, and why other forms of scaling are better (see the link in my earlier post, and please watch the video)


This is hilarious! You are citing the first row of pictures as proof of your claims? They exactly prove the distortion produced by these scaling methods. Imagine a picture of a 2-d coordinate system where everything before x<0 is black and the rest is white. These distorting scaling algorithms instead of showing the same divide at a higher resolution would show some weird gray area at the border.

You are trying to scare people with the highly blocky pictures of lower resolution sources. I ask those who have (or had) a 1080p projector showing a 1080p source. Do you look at that image thinking it's horribly blocky? If not, why would it look blocky in the same screen with a 4k projector just copying every pixel 4 times? And if you get scared by the snake oilsmanship at work above, you would lose the clarity of 1-1 pixel mapping with 4k sources.


Anyone who has read my posts carefully should understand my approach of minimum image manipulation and the value of 1-1 pixel mapping. I don't have anything more to add.
darksets is offline  
post #102 of 109 Old 06-18-2019, 08:40 AM
AVS Forum Special Member
 
Gary Lightfoot's Avatar
 
Join Date: Jun 2000
Location: UK
Posts: 6,434
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 1302 Post(s)
Liked: 1053
Quote:
Originally Posted by darksets View Post
This is hilarious! You are citing the first row of pictures as proof of your claims?
They are proof.

The images clearly prove that optically zooming or using a box filter/nearest neighbour scaling (the first image) provides the worst type of image and worst type of scaling, and why using an A lens with scaling produces visibly better results because you're using more pixels with good scaling.

Quote:
Originally Posted by darksets View Post
They exactly prove the distortion produced by these scaling methods. Imagine a picture of a 2-d coordinate system where everything before x<0 is black and the rest is white. These distorting scaling algorithms instead of showing the same divide at a higher resolution would show some weird gray area at the border.
The first image (your own preference for scope movie content) looks more distorted than the other two, yet is accurate to the source. The other two look better yet are not accurate to the source. I think most people would choose the second and third option, or a better still, a scaler such as a Lumagen or MadVR. Many scalers that have the option of nearest neighbour scaling often say it's the worst to use (just ask Madshi who is the guy who created MadVR).

https://forum.doom9.org/showpost.php...ostcount=42595

Quote:
Originally Posted by Madshi
Nearest neighbor was only ever there to demonstrate how bad it is. I never intended anybody to actually use it. Why would you prefer it over other scaling options?
If you google 'nearest neighbour scaling' you'll get a lot of images like the one I posted with similar explanatory comments.

Quote:
Originally Posted by darksets View Post
You are trying to scare people with the highly blocky pictures of lower resolution sources. I ask those who have (or had) a 1080p projector showing a 1080p source. Do you look at that image thinking it's horribly blocky? If not, why would it look blocky in the same screen with a 4k projector just copying every pixel 4 times? And if you get scared by the snake oilsmanship at work above, you would lose the clarity of 1-1 pixel mapping with 4k sources.
I don't think using facts and evidence should scare anybody - it's just a means of education and certainly not snake oil (unless you don't understand it I guess) Plus, those images are clearly enlarged so you can see exactly what's going on. No one is saying that zooming or nearest neighbour scaling will produce unwatchable images (at least from normal seating distances), only that using a good A lens with scaling produces better images. The images prove that beyond any doubt.

Quote:
Originally Posted by darksets View Post
Anyone who has read my posts carefully should understand my approach of minimum image manipulation and the value of 1-1 pixel mapping. I don't have anything more to add.
I think most people who have read your posts can see that you have no experience or understanding of what's being said here (you even admit to have never seen an A lens in action), and even you use scaling and don't zoom everything to retain pixel to pixel 'accuracy', so I'm not really sure what your argument is. If scaling is a bad thing, why do you use it for none scope content?

Quote:
Originally Posted by elmalloc
Who says Cameron is "right" and why do we care about him so much - lol!

I trust Gary Lightfoot more than James Cameron.
Gary Lightfoot is offline  
post #103 of 109 Old 06-18-2019, 01:18 PM
AVS Forum Special Member
 
jeahrens's Avatar
 
Join Date: Dec 2005
Location: Iowa, USA
Posts: 3,765
Mentioned: 85 Post(s)
Tagged: 1 Thread(s)
Quoted: 1831 Post(s)
Liked: 1144
Modern scaling is for the most part excellent. No one watches a 1080P Blu Ray floating in the middle of the screen on a 4K projector. They scale it. Same with a 2K projector and a lower resolution source.

I use full panel scaling (17:9) for the extra light with ARs wider than 1.85:1 on my NX7 and it looks excellent. Same goes for all 3 anamorphic lenses I've seen in person, scaling is excellent and there are zero visible issues.

The only real drawback to a quality anamorphic lens is price. The cost is to high for the benefit provided in my opinion, so I don't own one. But having said that it IS the best scope experience you're going to get. More light for HDR and a smoother picture overall.

jeahrens is online now  
post #104 of 109 Old 06-18-2019, 03:01 PM
AVS Forum Special Member
 
Gary Lightfoot's Avatar
 
Join Date: Jun 2000
Location: UK
Posts: 6,434
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 1302 Post(s)
Liked: 1053
Yes, I think the price to performance can make buying a new A lens seem like poor value, but you can get them much cheaper (though can still be relatively expensive) on the used market. I was lucky to get my ISCO at a good price from this forums classifieds, and they do come up for sale both here and on ebay from time to time.

Quote:
Originally Posted by elmalloc
Who says Cameron is "right" and why do we care about him so much - lol!

I trust Gary Lightfoot more than James Cameron.
Gary Lightfoot is offline  
post #105 of 109 Old 06-18-2019, 04:14 PM
AVS Forum Special Member
 
Killroy's Avatar
 
Join Date: Oct 2008
Location: Ninth Circle of Hell
Posts: 2,333
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 437 Post(s)
Liked: 329
Quote:
Originally Posted by jeahrens View Post
The only real drawback to a quality anamorphic lens is price. The cost is to high for the benefit provided in my opinion, so I don't own one. But having said that it IS the best scope experience you're going to get. More light for HDR and a smoother picture overall.
Define too high (price).

I have kept track of used prices and they are nearing an all time low since lots of people don't think they need them with a 4k projector and they have been sitting unused for months/years.

I would say that ISCO IIIL prices are settling under 2k for gently used but sometimes you can find new-old stock for the same...like mine.
Killroy is online now  
post #106 of 109 Old 06-18-2019, 04:58 PM
Member
 
darksets's Avatar
 
Join Date: Jul 2006
Location: Hoboken, NJ
Posts: 182
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 68
Quote:
Originally Posted by Gary Lightfoot View Post
I think most people who have read your posts can see that you have no experience or understanding of what's being said here (you even admit to have never seen an A lens in action), and even you use scaling and don't zoom everything to retain pixel to pixel 'accuracy', so I'm not really sure what your argument is. If scaling is a bad thing, why do you use it for none scope content?
I use scaling, when necessary, because I don't want to watch a tiny picture in the middle of my screen and I don't want my projector to be zooming in and out all the time. I just want to sit back and change the channel/input and my projector to stay at a fixed zoom setting. That's why I got a Lumagen.

I feel we are going in circles here. Somehow you've made this an argument about scaling methods. My point is that all scaling adds some distortion and I want to keep it to a minimum. Adding an anamorphic lens adds an extra scaling step (among other distortions) when the source resolution and the projector resolution are the same. If we exclude the extra light, your only valid point in favor of the A-lens is the smaller gap between the pixels. I argue that for high resolution projectors that gap is insignificant, so the distortion it adds (scaling being part of it) is a net negative. That's the bottom line and the rest is obfuscation on your part.

Last edited by darksets; 06-18-2019 at 05:01 PM.
darksets is offline  
post #107 of 109 Old 06-19-2019, 07:50 AM
AVS Forum Special Member
 
jeahrens's Avatar
 
Join Date: Dec 2005
Location: Iowa, USA
Posts: 3,765
Mentioned: 85 Post(s)
Tagged: 1 Thread(s)
Quoted: 1831 Post(s)
Liked: 1144
Quote:
Originally Posted by Killroy View Post
Define too high (price).

I have kept track of used prices and they are nearing an all time low since lots of people don't think they need them with a 4k projector and they have been sitting unused for months/years.

I would say that ISCO IIIL prices are settling under 2k for gently used but sometimes you can find new-old stock for the same...like mine.
For me it would have to hit <$2k for something 4K capable to consider it. And with more material falling in between 1.85:1 and 2.35:1 not sure if I would go through the hassle of mounting the sled and moving the lens in/out as zooming works well for me.

Don't take my statements as a negative against anamorphic lenses. I have seen some very nice Panamorphs. They give a gorgeous picture and it is the best scope picture you'll get (short of a native 21:9 projector). It just doesn't hit the cost/benefit threshold for me.
Killroy likes this.

jeahrens is online now  
post #108 of 109 Old 06-19-2019, 08:07 AM
AVS Forum Addicted Member
 
Craig Peer's Avatar
 
Join Date: Aug 2003
Location: Sacramento, CA
Posts: 15,801
Mentioned: 106 Post(s)
Tagged: 0 Thread(s)
Quoted: 6625 Post(s)
Liked: 7700
A good anamorphic lens will last ( with a little care ) - pretty much forever. So while expensive initially, they certainly add picture quality and enjoyment over time. Anyway, define expensive? We live in a world where a really good bicycle can cost $10K. Or even $20K. In fact, I have two road bicycles that cost more than my DCR lens did ( street prices ). And my wife has a tri bike that cost more too.
Craig Peer is online now  
post #109 of 109 Old 06-20-2019, 08:24 AM
Member
 
ask4me2's Avatar
 
Join Date: Apr 2005
Location: Norway
Posts: 148
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 76 Post(s)
Liked: 48
Quote:
Originally Posted by darksets View Post
I use scaling, when necessary, because I don't want to watch a tiny picture in the middle of my screen and I don't want my projector to be zooming in and out all the time. I just want to sit back and change the channel/input and my projector to stay at a fixed zoom setting. That's why I got a Lumagen.

I feel we are going in circles here. Somehow you've made this an argument about scaling methods. My point is that all scaling adds some distortion and I want to keep it to a minimum. Adding an anamorphic lens adds an extra scaling step (among other distortions) when the source resolution and the projector resolution are the same. If we exclude the extra light, your only valid point in favor of the A-lens is the smaller gap between the pixels. I argue that for high resolution projectors that gap is insignificant, so the distortion it adds (scaling being part of it) is a net negative. That's the bottom line and the rest is obfuscation on your part.
In theory i agree with your 1:1 pixel mapping without any scaling way of showing the best possible picture in a setup. I say in theory, because in reality the pixels we get from our source material is not 1:1 true all the way from the camera pixlels (may use a RGGB chip), lossy compression etc. So if the creating of new pixels is something to be afraid of, I think we need to get the RAW format from the movie makers to get it as close to the source as possible. (if the movie is made with the use of anamorpic lenses, the first scaling might be to convert that)

There is some pixel recreating processing from the BD and UHD 4K BD sources until we get a RGB signal the projector use to show the picture. The negative impact you might get from the additional anamorpic prosessing uing an A-lens and getting more pixcles to play with may not be as big as the better picture quality we get with the better use of the resolution and light output the light engine in the projector when setting up a 17:9 or 16:9 chip to be filled with the 2.35:1 scope source.

From my experience there may be wise to actually see these effect in real life, instead of only having a theoretic understanding that 1:1 pixel mapping is the best and the extra processing and an additional lens in the light path is doing more damage than good.

If i remember correctly from reading in other threads here you got the Sony VPL-VW995ES/870ES 4K Laser projector, and in a setup like that you may not need the additional resolution and light output like us that have FullHD with e-shft, but i think the best way to find out is to try or get a demo from someone with the same high end projectors and see if the theory matches the real life experience.

I do think the biggest risk of trying a A-lens is the economic impact that can come out of the experience of that....
ask4me2 is offline  
Sponsored Links
Advertisement
 
Reply 2.35:1 Constant Image Height Chat

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off