Jump to content

Mtf Of Blu-ray/hd-dvds


Recommended Posts

I actually was commenting on bacs post being a good one, I did read the same document as you posted but for the same reasons as below I dis regared it. digital technology has moved along way too fat in 5-6 years

The technology may have improved but the principals are requirements have not changed.

Link to post
Share on other sites
  • Replies 426
  • Created
  • Last Reply

Top Posters In This Topic

  • Moderators
Your analogy to audio is spot on mate.

With HD we effectively have the treble control turned down, the high frequencies are still there, but are hard to hear because the relative level is low in comparison the low frequencies.

Some of use want 20Hz to 20Khz response from our systems, but it seems others are happy with 20Hz to 15Kz -3db. :D

although I agree on hte audio analogy the trouble is in this thread we are not talking replay systems, you are instead contending that the information isnt there in the first place for the system to resolve. which is jsut plain wrong. if the above analogy is an insight into the workings of your inner mind you will never be content. As we have 1920x1080 resoltuion now from various sources, very much here to enjoy. maybe you want 2k or 4k but thats another discussion altogether.

Link to post
Share on other sites
so given you have no actual evidence to support your original claims, jsut some fragments from around the place that you have fabricated together along with yoru own subjective opinion, you then do admit that there is 1920x1080 resolution for hd dvd and blu ray.

Since you think the information I linked is false or fabricated, I’ll turn it around. You provide just ONE link to a camera that can provide 100% MTF at a spatial frequency of 1920 lines per picture width by 1080 lines per picture height. This needs to be substantiated with MTF data.

There is your task for tonight. :D

Link to post
Share on other sites
although I agree on hte audio analogy the trouble is in this thread we are not talking replay systems, you are instead contending that the information isnt there in the first place for the system to resolve. which is jsut plain wrong. if the above analogy is an insight into the workings of your inner mind you will never be content. As we have 1920x1080 resoltuion now from various sources, very much here to enjoy. maybe you want 2k or 4k but thats another discussion altogether.

Sure we have 1920x1080 video, but the response is 3 to 6 db down at the high frequency end of the spectrum, and that is my point.

Even the original source is 3 to 6db down at the top end, so a copy is not going to get any better.

If we had 100% MTF we would have a flat response all the way to the 1920x1080 limit, but we just don’t for none computer generated content.

Link to post
Share on other sites


  • Moderators
Since you think the information I linked is false or fabricated, I’ll turn it around. You provide just ONE link to a camera that can provide 100% MTF at a spatial frequency of 1920 lines per picture width by 1080 lines per picture height. This needs to be substantiated with MTF data.

There is your task for tonight. :D

not necessary at is you who were the proponent of your theory. You were challenged on it and it is your taks to prove which you have failed.

what I have put forward I am yet to see an evidence to dispute

Link to post
Share on other sites
  • Moderators
The technology may have improved but the principals are requirements have not changed.

no thats quite right and as per bacs link the cpabilities of film there are very clear and hard to miss. obviously you dispute what was in that link as well ? it pretty clearly said if I rmemeber that on scanning film even as high as 4K there is information provided. thats 4 times the resolution needed for hd dvd and blu ray. now I wonder how film must be able to achive that given yoru very poorly claims of what it can achieve.

maybe you should actaully read that link and see jsut how much resolution is provided by film. did you infact even read it ?

here it is again

http://www.etconsult.com/papers/Technical%...0Resolution.pdf

Resolution of film

The controversy rages over the ultimate resolution of film. Datasheets, tests and popular belief and actual practice don’t always agree.

Popular belief is that film is a “4K” medium. The current Vision family of film stocks, for example Kodak Vision 200T color negative film - 5274, is specified by Kodak to exceed 4000 lines of resolution (horizontally across the frame), and independent tests (ITU tests) have confirmed that this resolution is captured on the negative.

this was 5-6 years ago. I can only imagine things have improved further in the technology front since then.

on the digital movie cameras & technology and scanning technioques it woudl be hard to imagine they havent have...

Link to post
Share on other sites
not necessary at is you who were the proponent of your theory. You were challenged on it and it is your taks to prove which you have failed.

what I have put forward I am yet to see an evidence to dispute

Sorry that’s not going wash mate. You questioned my data and have nothing to offer in response.

I’ll let forum members make up their own minds based on the information provided.

Link to post
Share on other sites
no thats quite right and as per bacs link the cpabilities of film there are very clear and hard to miss. obviously you dispute what was in that link as well ? it pretty clearly said if I rmemeber that on scanning film even as high as 4K there is information provided. thats 4 times the resolution needed for hd dvd and blu ray. now I wonder how film must be able to achive that given yoru very poorly claims of what it can achieve.

maybe you should actaully read that link and see jsut how much resolution is provided by film. did you infact even read it ?

here it is again

http://www.etconsult.com/papers/Technical%...0Resolution.pdf

this was 7 years ago. I can only imagine things have improved further in the technology front since then.

on the digital movie cameras & technology and scanning technioques it woudl be hard to imagine they havent have...

Sure film has response out to 4k, I have never disputed that. Problem is that its at about 12db down or at about 3-5% MTF. The linked MTF graphs show this.

Link to post
Share on other sites


  • Moderators
Sorry that’s not going wash mate. You questioned my data and have nothing to offer in response.

I’ll let forum members make up there own mind based on the information provided.

what you have offered does not wash. I've been telling that to you all along. and have been asking you for something to back up your claims with some actual evidence rather than yoru fabricated theories. we are yet to see it.

Link to post
Share on other sites

The paper I linked to earlier contained MTF test curves for Kodak 5274 Vision 200T colour negative film using Panavision cameras/lenses.

To summarise, the tests showed the maximum pixel resolution that was attainable at 100% MTF was approximately 350 x 550 (my interpolation from the graph). At a resolution of 1920 x 1080, the MTF dropped to around 15% and 55% respectively (my interpolation as above). BTW, these tests were done with 'normal' lenses - anamorphic lenses would have produced worse results due to optical squeezing.

Granted, improvements in film and lenses probably have occurred since 2001, but 35mm film and lenses are fairly mature technologies and I would suspect the results would still be fairly relevant today. If you accept that, I can't see how you could expect a 35mm film source to be able to be transferred to the digital domain for BluRay/HD-DVD release at anywhere near 100% MTF at a target resolution of 1920x1080.

Does this make sense, or am I missing something here?

Cheers,

BAC

Link to post
Share on other sites
Does this make sense, or am I missing something here?

Cheers,

BAC

BAC, it makes sense. Thanks for going to the trouble of posting.

Link to post
Share on other sites
  • Moderators
Sure film has response out to 4k, I have never disputed that. Problem is that its at about 12db down or at about 3-5% MTF. The linked MTF graphs show this.

thankyou for admitting film does infact have a resolution of 4K which is 4 times the resolution needed for 1920x1080 and the hi-def discs formats. and which as I posted in the links early in this thread is a walk in the park ot achieve the resolution required.

- ok so film has 4K as supported by even your links eg the etconsult one.

- CGI theres no argument.

so is it only digital movie cameras that you miguidingly believe have insufficient resolution ? for HD dvd and blu ray to achieve 1920x1080 ?

Link to post
Share on other sites


thankyou for admitting film does infact have a resolution of 4K which is 4 times the resolution needed for 1920x1080 and the hi-def discs formats. and which as I posted in the links early in this thread is a walk in the park ot achieve the resolution required.

I wasn't 100% sure but I am now. Al, you're mistaking resolution for information. Look back at my chequerboard (sorry for the previous Americanism!) analogy - what you're basically saying is that XYZ is capable of recording pixels that go (in a horizontal line for the sake of ascii):

BBBBWWWWBBBBWWWW (b=black, w=white), where in this case our XYZ camera is recording a 1920x1080 image at 7680x4320, ie 4x hi def in each direction.

In the real world, this won't happen. What you'll get is something like BBBBWWWWBBBB

and so on. That is, it is physically impossible to get a 1:1 mapping between "real" pixels (ie the dots on our chequerboard image) and the pixels in the sensor or the halide crystals on the film stock; there will be some bleeding of information between adjacent sensor pixels simply because that's how physics works. You simply cannot get one piece of 'real' information recorded 100% on a single pixel, or group of pixels, or film crystals, and 0% of that information anywhere else, with no overlap (pure CGI images aside). Hence, regardless of the final resolution of your output format, you won't be able to fully resolve it simply because you don't have discrete information on each pixel. You will have 1920x1080 pixels on your HD disc, but only 960x540 individual "pieces" of information.

Edited by Thudd
Link to post
Share on other sites
really ? and you have some actual evidence to back that up or is this somethign you fabricated as well ?

If you had done MPEG encoding yourself, you would already know about the loss in definition that results, even at high data rates.

Link to post
Share on other sites
  • Moderators
I wasn't 100% sure but I am now. Al, you're mistaking resolution for information. Look back at my chequerboard (sorry for the previous Americanism!) analogy - what you're basically saying is that XYZ is capable of recording pixels that go (in a horizontal line for the sake of ascii):

BBBBWWWWBBBBWWWW (b=black, w=white), where in this case our XYZ camera is recording a 1920x1080 image at 7680x4320, ie 4x hi def in each direction.

In the real world, this won't happen. What you'll get is something like BBBBWWWWBBBB

and so on. That is, it is physically impossible to get a 1:1 mapping between "real" pixels (ie the dots on our chequerboard image) and the pixels in the sensor or the halide crystals on the film stock; there will be some bleeding of information between adjacent sensor pixels simply because that's how physics works. You simply cannot get one piece of 'real' information recorded 100% on a single pixel, or group of pixels, or film crystals, and 0% of that information anywhere else, with no overlap (pure CGI images aside). Hence, regardless of the final resolution of your output format, you won't be able to fully resolve it simply because you don't have discrete information on each pixel. You will have 1920x1080 pixels on your HD disc, but only 960x540 individual "pieces" of information.

ofcourse not thudd, and I have never said it is possible to get 100% interpolation.

the fact of the matter is that this thread isnt about photographing a 1920x1080 image. it is about filming either via digital or film movie camera a real world that does not have pixels. and even if we were filming a 1920x1080 image I'm sure what ever is photographying it would have much higher capture capability eg 4000 line capability of film or the 12mp sensor in a gensis camera and yep the image would be laid out over the sensor or exposed to the film crystals to be captured, ofcourse it goes without saying.

Link to post
Share on other sites
  • Moderators
If you had done MPEG encoding yourself, you would already know about the loss in definition that results, even at high data rates.

so mpeg coding or "loss of information at even high data rates" is the basis behind your theory ?, look forward to reading a bit more about that and how that has anything to do with your hyothesis.

Link to post
Share on other sites


But the principle is the same in the real world; no matter how high the resolution of your screen or the format storing the images, the image that you see won't have the same amount of discrete bits of information as pixels available. That, in a crude sense, is what MTF is - the loss of information along the chain of an optical system. So even with a bazillion pixels available, and the best interpolation and scaling and discomboobilation we'll be losing information between the object being filmed and our eyes.

And then you have to add in the factor of at what distance can our eyes distinguish between two adjacent pixels on our screen, regardless of the information content of those pixels (which is one of Owen's crusades)

Link to post
Share on other sites
  • Moderators
But the principle is the same in the real world; no matter how high the resolution of your screen or the format storing the images, the image that you see won't have the same amount of discrete bits of information as pixels available. That, in a crude sense, is what MTF is - the loss of information along the chain of an optical system. So even with a bazillion pixels available, and the best interpolation and scaling and discomboobilation we'll be losing information between the object being filmed and our eyes.

And then you have to add in the factor of at what distance can our eyes distinguish between two adjacent pixels on our screen, regardless of the information content of those pixels (which is one of Owen's crusades)

I agree thudd, ofcourse thats the case, and I am certaily not suggesting that film or digital cameras start of with a 1920x1080 resolution and thats what they deliver and certainly I am not sayign that cameras be they film or digital have the almost infinite resoltuion capability required to portray the organic real world life forms and surfaces around us. after all look at any what might appear to our eyes as flat surfaces look with a higher resolving device and you will soon notice hills and valleys and crevaces.

but for the purposes of this discussion, the information I've seen so far certainly indicates film & digital cameras are certainly capable of capturing very high resolution images, 4K for film, 2-4K for HD digital. which is quite sufficient for a 1920x1080 resolution image for hd dvd & blu-ray even allowing for losses along the way, given the 2-4 times amount of resolution in the intial image captured.

if you read that etconsult link, thats what its sayign too which makes sense.

Link to post
Share on other sites

Whats more important - the # of pixels on the screen, or the amount of information? If it's the latter, why are there HD screens (again discounting PC use and CGI cartoons etc)? Also, if there is only a maximum of "960x540 individual "pieces" of information", why are viewing distances for 1080 panels scrutinised, when chances are 95% of the time, people watching HD TV (even shows filmed in HD), and/or HD DVD/Blu-ray (actually, I won't guess. In MY case, 95% of viewing is TV/film)?

Link to post
Share on other sites
but for the purposes of this discussion, the information I've seen so far certainly indicates film & digital cameras are certainly capable of capturing very high resolution images, 4K for film, 2-4K for HD digital. which is quite sufficient for a 1920x1080 resolution image for hd dvd & blu-ray even allowing for losses along the way, given the 2-4 times amount of resolution in the intial image captured.

While I feel I need a spare lifetime to gain a full understanding of the material in this thread, I thought what was in dispute was the percentage MTF of 1920x1080 hd dvd & blu-ray material, not the resolution capabilities of the source cameras.

Link to post
Share on other sites
but for the purposes of this discussion
What is that you're actually claiming then? Because I think it's kinda got lost somewhere along the way and I might be trying to point out something here that you already know.

Plain english please, it was a long day at work :wacko:

Link to post
Share on other sites
  • Moderators
While I feel I need a spare lifetime to gain a full understanding of the material in this thread, I thought what was in dispute was the percentage MTF of 1920x1080 hd dvd & blu-ray material, not the resolution capabilities of the source cameras.

well you can go read mlx's pretext in creatign this thread back here

http://www.dtvforum.info/index.php?s=&...st&p=794030

I think you'll find the not only was resolution capability of cameras but also lens and transfer systems both for both film digital were challenged

Link to post
Share on other sites
  • Moderators
What is that you're actually claiming then? Because I think it's kinda got lost somewhere along the way and I might be trying to point out something here that you already know.

Plain english please, it was a long day at work :wacko:

lets see thudd, probably the bit you left off after "for the purposes of this discussion..."

the information I've seen so far certainly indicates film & digital cameras are certainly capable of capturing very high resolution images, 4K for film, 2-4K for HD digital. which is quite sufficient for a 1920x1080 resolution image for hd dvd & blu-ray even allowing for losses along the way, given the 2-4 times amount of resolution in the intial image captured.

and by the way its not me doign the claiming, its owen who was saying in the pio thread that 1080 displays were a waste as hddvd and bluray never have 1920x1080 resolution.

Link to post
Share on other sites
  • Moderators
Whats more important - the # of pixels on the screen, or the amount of information? If it's the latter, why are there HD screens (again discounting PC use and CGI cartoons etc)? Also, if there is only a maximum of "960x540 individual "pieces" of information", why are viewing distances for 1080 panels scrutinised, when chances are 95% of the time, people watching HD TV (even shows filmed in HD), and/or HD DVD/Blu-ray (actually, I won't guess. In MY case, 95% of viewing is TV/film)?

good question davep, I asked owen that as well, since he has always been pushing the visual aquity charts, looks like your still looking for an answer. particualrly if as he says the the amount of information isnt there then why viewing distances scrutinzed which like you have seen time and tie happen on here. :)

Link to post
Share on other sites
well you can go read mlx's pretext in creatign this thread back here

http://www.dtvforum.info/index.php?s=&...st&p=794030

I think you'll find the not only was resolution capability of cameras but also lens and transfer systems both for both film digital

From the above link do you agree or disagree with the following statement ?

Owen is correct in saying that 100% MTF is not attained between adjacent pixels in 1920x1080 Blu-ray or HD-DVDs of real life video. That result follows from Nyquist sampling theory. This result holds true even if the source material is from a very high resolution camera.
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    No registered users viewing this page.





×
×
  • Create New...