How have you been, Roland?
With the incredible use of color in this film, it's a shame those caps don't have much of a range. Where's some of the rich greens, yellows, and purples? I'm not saying caps serve any purpose beyond entertainment, but what is represented there seems a waste of bandwidth.RolandL said:
Here's what's not a shame - the transfer, color-wise. Why are you commenting on caps that bear no relation to what's actually on this disc. Believe me, there are rich greens, yellows, and purples and every other color, a veritable Technicolor feast. Lord I wish there were some way to ban the use of caps everywhere. They cause nothing but confusion, stop people from buying great discs, and bring out the worst on discussion board when they're used as some sort of "evidence" that a transfer is terrible. There's only one thing that's terrible - screen caps. Do away with 'em. Banish 'em. Feed 'em to the dogs.Mark B said:
Their comparisons of Cleopatra and Ben-Hur Blu-ray to DVD really sold me on buying those titles which are from 70mm. I already bought the MM titles as they were only $9.96 each. They do look a lot better than the screen caps.Robert Harris said:
Right. And how do I avoid the bad discs then? Taking some guru's word as gospel and let him/her make my buying decisions? Listen to a random bunch of online commentaries? Buy it anyway and get rid of it when it's not up to my standards? I don't think so. Correctly done stills are paramount to get an impression of the work done for a transfer. There is no replacement for them except watching the disc. But for that, unless it can be rented, a purchase is required (or illegal download of a rip) and I don't want to do that unless I'm confident about the quality I'm getting. For that stills are a very useful, although they can't tell the whole story. Stills are good for sorting out a large part of the garbage. Not more, not less. And of course you need to know what you are seeing and what it means. Not a trivial task by any means. Proper stills in the wrong hands are as useless as incorrect stills in capable hands.haineshisway said:
Michel,
What many do is over time is get to know which people they trust, which people's judgement is similar to their own, and then allow themselves to be influenced by the opinions of those people. It might be an esteemed guru or it might be a humble film lover who contributes regularly to a forum like this.Michel_Hafner said:
Not really. It's not as simple as that. Basically there is the digital image data on the Blu Ray and there is a rendering and display chain that shows you this data, whether you watch it on a PC monitor, a portable Blu Ray player with a built in screen, a TV set or a projector. The image data is always the same, the rendering and display chain can vary a lot, though. When stills are used to judge image characteristics of a transfer put on a Blu Ray then 2 things are required from the technical side to allow this. The stills must be created correctly so the actual data on the disc is in the stills. The stills must be watched on the display without unwanted modifications by the rendering and the display.Ethan Riley said:
Most of the screen caps I see on these forums do not remotely compare to the images I see when I play the discs. Most look as though they've been filtered. They aren't as bright and sharp as the images I see when I watch the BD.Michel_Hafner said:
A very helpul post, it has factual information and common sense. I couldn´t agree more.Michel_Hafner said:Not really. It's not as simple as that. Basically there is the digital image data on the Blu Ray and there is a rendering and display chain that shows you this data, whether you watch it on a PC monitor, a portable Blu Ray player with a built in screen, a TV set or a projector. The image data is always the same, the rendering and display chain can vary a lot, though. When stills are used to judge image characteristics of a
...
...
And the best stills in the world plus the disc itself are useless if you don't know what to make out of what you see. If you have no (proper) reference to compare it to. You might see correctly what there is to see, but you don't know for sure what to make of it. And that affects everybody at times, no matter how experienced or well informed you are, or the opposite of it.
In this case if one wanted to find out why that is one has to go digging and search for the reason(s). To really be sure you need to know how to do stills yourself. You can use test discs with well defined test patterns that will quickly show you if somewhere some sharpening is applied, or some noise processing. If the colour bars look right on the stills or not. If there is a difference between the stills and playing the film itself on the computer you do the stills on. And how it compares to what you see on your TV or projector. All these devices usually have dozens of parameters and options you can activate or deactivate or change the amount of processing that is done. And without calibrating your devices to follow the established standards (colour, contrast, gamma, gray scale, brightness, black level, white level) all bets are off. You can't say if the stills are wrong or your display chain is at fault (or both), when they don't match, without going into the details, unfortunately.Virgoan said:
A very good post.Michel_Hafner said:Not really. It's not as simple as that. Basically there is the digital image data on the Blu Ray and there is a rendering and display chain that shows you this data, whether you watch it on a PC monitor, a portable Blu Ray player with a built in screen, a TV set or a projector. The image data is always the same, the rendering and display chain can vary a lot, though. When stills are used to judge image characteristics of a transfer put on a Blu Ray then 2 things are required from the technical side to allow this. The stills must be created correctly so the actual data on the disc is in the stills. The stills must be watched on the display without unwanted modifications by the rendering and the display.
Correctly done stills are identical to what is on the disc except that
- the compressed data has been decompressed following the proper standards for the compression codec
- the 8 bit YCbCr 4:2:0 data is converted to 8 bit RGB 4:4:4
- no further processing is applied to the data (except converting it to a standard image format such as jpeg etc. which may involve lossy compression and could visibly degrade image fidelity, which has to be avoided)
Given that, the stills are a sound basis for judging the image characteristics of the transfer for the individual frames the stills cover.
When the stills are watched what you will see depends on how you watch them. Which display device? How is it calibrated? What colour gamut does it have? What is the pixel count? What colour depth can it handle? What kind of processing does the display apply on the data it is sent? What kind of processing does the software or hardware apply to the data in the stills before sending it to the display?
So if the goal is to see what is on the disc in the intended way then the rendering and display chain must
- use the HD colour gamut (Rec.709) and suitable gamma (usually 2.2 or somewhat higher)
- use the proper video data range (video versus PC range)
- do no resampling of the stills and use 1:1 pixel mapping instead
- not apply any kind of sharpening, noise filtering or other digital processing
- not modify gama, black level, white level, clip or crush image data
If that is given then it does not matter if you watch it on a computer monitor, on a TV set or project it on a screen. It will look very much the same given the same viewing conditions (viewing distance, room ambient light etc.) as long as the capabilities of the displays are about the same concerning brightness, contrast and absolute black level.
In the real world there will always be differences from small to quite large even with properly done stills. The fault does not lie with the stills, it lies with the different rendering and display chains. And it is relevant or irrelevant depending on what you want to find out using stills. If you want to judge colour fidelity of a transfer the best stills are useless if your display is calibrated to another colour gamut than Rec. 709 (and so is the disc itself played at 24 fps). If you want to judge the amount of digital sharpening on the other hand stills can clearly show you if any was applied which left behind typical artifacts as long as you know your rendering chain does not add additional processing. If you want to know if there is fine detail on the transfer (grain or real world detail) again the stills will tell, if your chain is "transparent¨. If you want to know if there is excessive digital grain removal the stills will tell you if you are familiar with signs of such processing. If you want to know if there is aliasing the stills will tell you. If you are interested in blown out highlights and crushed blacks the stills will tell you. If you want to check for obvious compression issues like blocking the stills will tell you (only for the few frames covered, of course).
And then there are things even the best stills will not tell you. You have to watch the disc to see them. Such things as
- what's going on in all the shots the stills are not covering? (Usually transfers are fairly consistent, though, although the used film elements might not be)
- density fluctuations (a dynamic phenomenon not covered with a static approach)
- compression noise
- unnatural temporal grain behaviour due to grain processing (a dynamic phenomenon not covered with a static approach)
- image stability issues
- overall amount of random dirt/speckles/scratches etc.
- of course all aspects concerning the soundtrack(s)
And the best stills in the world plus the disc itself are useless if you don't know what to make out of what you see. If you have no (proper) reference to compare it to. You might see correctly what there is to see, but you don't know for sure what to make of it. And that affects everybody at times, no matter how experienced or well informed you are, or the opposite of it.
There is a rival site with a huge number of caps and I found them to be pretty good for my needs, they have the advantage of a huge database and of covering most new releases.This is the real catch, isn't it? How many screenshot advocates actually look at those screenshots on the same (presumably) calibrated display that they watch their movies, vs. those that look at the screenshots on some Dynex LCD monitor with a 100:1 contrast ratio that they picked up at Costco for $35?OliverK said:
