Are 35mm films capable of the Rec 2020 colour space? - AVS Forum | Home Theater Discussions And Reviews
Baselworld is only a few weeks away. Getting the latest news is easy, Click Here for info on how to join the Watchuseek.com newsletter list. Follow our team for updates featuring event coverage, new product unveilings, watch industry news & more!


Forum Jump: 
 1Likes
  • 1 Post By Mr.D
 
Thread Tools
post #1 of 27 Old 01-17-2014, 07:48 AM - Thread Starter
Member
 
kristoffer77's Avatar
 
Join Date: Nov 2008
Posts: 134
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 20
If 4K disc comes out and have Rec 2020 colour space support how many movies will be able to take advantage of it? Is the colour space of 35mm on par with Rec 2020 ? Or will rec 2020 have to rely on only new films?
kristoffer77 is offline  
Sponsored Links
Advertisement
 
post #2 of 27 Old 01-22-2014, 08:15 AM
AVS Addicted Member
 
John Mason's Avatar
 
Join Date: Jul 2000
Location: New York, NY
Posts: 10,738
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 49 Post(s)
Liked: 26
Using Google's image-search feature, here's a color gamut plot comparing
Sony's F65 digital cinema camera versus film (top left).

And here's another plot of rec 2020 color gamut (top left). Various search terms might show a direct comparison. Anyway, it's a start. -- John
John Mason is offline  
post #3 of 27 Old 01-22-2014, 11:47 AM
 
bralas's Avatar
 
Join Date: Jan 2014
Location: Lafayette, LA
Posts: 117
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20

Great question,

 

Film is an endowed medium. Modern color film stock (~1985-present) latitude / capture is HDR (~11 stops), film also has a logarithmic response to  light, just like the human eye. Film also has a near infinite color palette. DCI specs are similar to that of 35mm film. Film sets a very high standard to equal.

 

Standard Rec 709 video is limited to ~7 stops or ~200:1 contrast.    

bralas is offline  
post #4 of 27 Old 01-25-2014, 02:20 AM - Thread Starter
Member
 
kristoffer77's Avatar
 
Join Date: Nov 2008
Posts: 134
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 20
Thanks for the answers so far. I got this answer on another forum:

"Well, film color is not a color space, per se. There are boundaries to what can be captured on film and then reproduced by film print, and these boundaries, when mapped to the CIE 1931 XYZ color space chart, establish what might be better termed a film color gamut.

However, to directly answer your question in a nutshell….Yes….because the color coordinates for BT.2020 map out a WIDE color gamut (WCG).

And next if anyone is subsequently wondering….yes….Dolby Vision supports BT.2020 color and….can even be tweaked to support the XYZ color space if needed."
kristoffer77 is offline  
post #5 of 27 Old 01-25-2014, 02:33 AM
 
bralas's Avatar
 
Join Date: Jan 2014
Location: Lafayette, LA
Posts: 117
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20

http://www.creativeplanetnetwork.com/dcp/news/now-you-see-them-now-you-dont-colorimetry-and-electronic-cinema/44388

 

….Kuttner agreed that achieving a "film look" will be the toughest challenge for any electronic cinema system.

 

 "Film has a greater color depth than video and tracks color logarithmically like the human eye. Any electronic projection system will need to support 36-bit color with 12 bits per color plane. 150:1 contrast will probably be the minimum acceptable grayscale, although daylight film stocks can achieve 1,000:1 contrast."

 

- - - - - - - - - -

 

Why I commented 35mm film has potentially near infinite color palette.

bralas is offline  
post #6 of 27 Old 01-25-2014, 04:48 AM
AVS Special Member
 
coolscan's Avatar
 
Join Date: Jan 2006
Posts: 1,814
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 107
Quote:
Originally Posted by bralas View Post

http://www.creativeplanetnetwork.com/dcp/news/now-you-see-them-now-you-dont-colorimetry-and-electronic-cinema/44388

….Kuttner agreed that achieving a "film look" will be the toughest challenge for any electronic cinema system.

 "Film has a greater color depth than video and tracks color logarithmically like the human eye. Any electronic projection system will need to support 36-bit color with 12 bits per color plane. 150:1 contrast will probably be the minimum acceptable grayscale, although daylight film stocks can achieve 1,000:1 contrast."

- - - - - - - - - -

Why I commented 35mm film has potentially near infinite color palette.

rolleyes.gif Please note that the article was written around year 2000, and is completely outdated and the information of film vs. digital is in no way applicable anymore. cool.gif
coolscan is offline  
post #7 of 27 Old 01-25-2014, 05:11 AM
 
bralas's Avatar
 
Join Date: Jan 2014
Location: Lafayette, LA
Posts: 117
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20

Please, expand on exactly what you are saying.

 

A medium that sets the industry benchmark IS paramount in my book. On sale now at Barnes & Nobel:)

bralas is offline  
post #8 of 27 Old 01-26-2014, 07:36 AM
AVS Special Member
 
coolscan's Avatar
 
Join Date: Jan 2006
Posts: 1,814
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 107
Quote:
Originally Posted by bralas View Post

Please, expand on exactly what you are saying.

I don't know what you didn't get? A lot of development has happened in digital cinema since year 2000 which was more like the start of digital cinema.

The article start out with;
"Electronic cinema is a hot topic for 1999,......"
Now it is 2014.
.
"Until recently, the biggest stumbling block to electronic cinema had been the projectors themselves.
It was not technically feasible to project large wide-screen images with brightness, contrast, and color saturation that even came close to 16mm film, let alone 35mm.
But that has all changed int he past six years with the introduction of imaging engines based on transmitted light (liquid-crystal displays), reflected light (Texas Instruments' Digital Light Processing), and hybrid light-shutter(Hughes-JVC's Image Light Amplifier) technology."

This is old-school descriptions.

.
"None of these technologies are ready to replace film immediately, but manufacturers are making impressive demonstrations of film-like projection quality on a daily basis. Texas Instruments has been showcasing standard 4:3, 16:9, and anamorphic 2.35:1 electronic cinema to several studios using three 1280 x 1024-pixel digital micro-mirror devices (DMDs)."

Resolution of 1280 x 1024 was how far they had come at the time.

.
"Hughes-JVC Technology, manufacturer of the super-bright ILA-12K cinema-grade projector, recently announced a joint partnership with QUALCOMM in a new company called CineComm."

JVC is not in cinema projection business anymore.........

.
"Fixed-resolution displays such as DMDs and LCDs have now reached over 1000 vertical pixels and are flirting with 1200,............."

Do I need to say more..............

.
"...............but it also forms the basis of our present-day analog NTSC color television system and the CCIR-601 digital color specification."

CCIR-601 was later renamed Rec.601. We are now on Rec.709 for TV and approaching Rec.2020.
Cinema has been using the wider colorspace of DCI-P3 for many years.

They also talk about Film scanners like The Spirit which is outdated. A lot of development in CCD/CMOS sensors has happened in fifteen years providing higher resolution and better colors from film scans.

Even film has now become a digital media because it is immediately scanned and functions as just another camera sensor like a mix of film and the scanner sensor characteristics.
A film print for distribution has never contained the full color quality of the original positive/negative.

No film has been distributed for many years which has had its distribution copy made optical from the film negative.
Fox just announced that they will stop distributing film copies all together. I you don't have a digital projector in your cinema you won't get the movie.

So you see, information from year 2000 is not applicable anymore.
Quote:
Originally Posted by bralas View Post

A medium that sets the industry benchmark IS paramount in my book.

Why?
Leave it to the benchmark setter of ITU and others, they are paid for the job.

As for thread starters question; Film doesn't have particularly wide color space compared to the modern digital cameras.

Even if Rec.2020 capable displays comes on the market in some years, doesn't mean that all films (or any at all) will contain the full colorspace.

Rec.2020 and the new comparable standard for cinema called ACES are made so wide that they almost contain visible light, but that is done to not have restrictions, not to force everybody to finish movies or TV programs in that colorspace.

Rec.2020 is slightly wider than ProPhoto colorspace.
Rec.7+9 is comparable to sRGB.

This promotion CIE chart from Sony gives an idea of various color spaces. It is not accurate, Sony has cheated a little to make their cameras F65 and F35 colorspace look a little better in comparison.



This one might be more accurate.


Quote:
Originally Posted by bralas View Post

On sale now at Barnes & Nobel:)

Might be fun as an history book, but useles fo anybody that want to have knowledge of what is going on today.
coolscan is offline  
post #9 of 27 Old 01-26-2014, 08:12 AM
 
bralas's Avatar
 
Join Date: Jan 2014
Location: Lafayette, LA
Posts: 117
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
Quote:
Originally Posted by coolscan View Post

Do I need to say more....
 

Histronics, coupled w/large images mean what exactly?

 

The OP was inquiring about the dynamics of 35mm film. I gave succinct empirical data regarding the prowess of the format itself.

 

"Any electronic projection system will need to support 36-bit color with 12 bits per color plane"

 

A statement like this suggest film is quit endowed indeed. In fact the color saturation characteristic is unmatched. Now worries though, most don't

comprehend the quantum difference between log space and what is effectively a "linear" dynamic space. Please don't tell me how Rec 709 gamma is

non linear...please! The medium can only represent 100:1 or at best 200:1 with 8-bit per color.

 

"...CIELAB and CIELUV color spaces target print and video respectively.  L* models contrast approximately 100:1 with peak luminance somewhere around 200 cd/m^2.  The L* value of 8, corresponds to a contrast ratio of 100:1 (linear segment breakpoint).":) 

bralas is offline  
post #10 of 27 Old 01-28-2014, 07:27 PM
AVS Special Member
 
Richard Paul's Avatar
 
Join Date: Sep 2004
Posts: 6,970
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 30
Quote:
Originally Posted by kristoffer77 View Post

If 4K disc comes out and have Rec 2020 colour space support how many movies will be able to take advantage of it? Is the colour space of 35mm on par with Rec 2020 ?
35mm film has a fairly good sized color space as seen in this Sony comparison chart and would benefit from the Rec. 2020 color space.

Quote:
Originally Posted by kristoffer77 View Post

Or will rec 2020 have to rely on only new films?
Almost all digital movies used the DCI P3 color space which is smaller than 35mm film. Hollywood is slowly moving towards the ACES color space, which covers the entire range of visible light, but at the moment it is rare for a movie to use it. Ironically this means that for a period of 10+ years the color space of movies decreased because of the switch from 35mm film to digital projection.
Richard Paul is offline  
post #11 of 27 Old 01-28-2014, 09:16 PM
 
bralas's Avatar
 
Join Date: Jan 2014
Location: Lafayette, LA
Posts: 117
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
Quote:
Originally Posted by Richard Paul View Post


... Ironically this means that for a period of 10+ years the color space of movies decreased because of the switch from 35mm film to digital projection.

+1:) 

 

"...Please note that the article was written around year 2000"

 

"...completely outdated and the information of film vs. digital is in no way applicable anymore"

-2:rolleyes: 

bralas is offline  
post #12 of 27 Old 01-29-2014, 05:05 AM
AVS Special Member
 
coolscan's Avatar
 
Join Date: Jan 2006
Posts: 1,814
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 107
Quote:
Originally Posted by bralas View Post

+1:)  

"...Please note that the article was written around year 2000"

"...completely outdated and the information of film vs. digital is in no way applicable anymore"
-2:rolleyes:  

As a newcomer to this forum you seems to be intent on regularly taking statements out of context and try to provoke an argument where there are nothing to argue about. Your posts seems to display a inability to understand the content of the posts you try to argue about without showing any sign that you have any valuable input.

If you don't understand by yourselves that an article from 2000 that discuss the equipment (scanners and projectors) they had back then, which where unable to extract the full colorspace (and other qualities) of 35mm film, and was unable to digitally project the full and equal quality of 35mm film, then I don't know how someone could explain that to you.

Nobody here are interested in the limitations of equipment made more than fifteen years ago. Even the 35mm film-stock has improved vastly since then.

There are also reasons for why archival movies which has been scanned and distributed digital many years ago are now being rescanned on newer and better equipment, and that is not for the increased resolution alone.
coolscan is offline  
post #13 of 27 Old 08-14-2014, 08:54 AM
Senior Member
 
MLXXX's Avatar
 
Join Date: Jan 2007
Location: Brisbane, Australia
Posts: 352
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 66
Quote:
Originally Posted by bralas View Post

The OP was inquiring about the dynamics of 35mm film.
No, what the OP enquired about was a broader question, about the colour space. Rec 2020 defines a colour space. I think the OP was asking whether modern 35mm film can capture [the negative] the whole of the defined Rec 2020 colour space. And the OP may in addition have been interested in whether a print from a negative would be capable of covering the whole of the Rec 2020 colour space.

If the colour space is limited (e.g. a particular print process is unable to show a particular shade of red at all) it is little consolation that this limitation in the gamut covered by the colour space of the print is maintained over a very wide dynamic range, traceable back to the dynamic range of the negative.

The colour of film prints in the mid to late decades of the 20th century varied considerably as techniques improved. I think it is part of the charm of movies from those past decades that they retain various distinctive looks.

Last edited by MLXXX; 08-14-2014 at 04:46 PM.
MLXXX is offline  
post #14 of 27 Old 01-11-2015, 12:17 AM
AVS Special Member
 
Mr.D's Avatar
 
Join Date: Dec 2001
Location: UK
Posts: 3,319
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 17
Quote:
Originally Posted by coolscan View Post
Quote: Originally Posted by bralas


If you don't understand by yourselves that an article from 2000 that discuss the equipment (scanners and projectors) they had back then, which where unable to extract the full colorspace (and other qualities) of 35mm film, and was unable to digitally project the full and equal quality of 35mm film, then I don't know how someone could explain that to you.
Actually negative has been scanned at 4k to 10 bit log since about 1994. Its exactly the same spec used today. A Cineon Lightning scanner from the mid 90s creates scans that are exacty the same quality as a more recent Northlight.

Quote:
Originally Posted by coolscan View Post

Nobody here are interested in the limitations of equipment made more than fifteen years ago. Even the 35mm film-stock has improved vastly since then.
Like i said film scanning has changed little in 20 years , its gotten faster per frame but the standard is the same. 35mm film stock has improved a bit in terms of grain structure over the last few years but the actual dynamic range has pretty much stayed the same for negative . Makes no difference to the scanner , they were transparent to all intents and purposes back in the mid 90s to the same extent they are today.

Quote:
Originally Posted by coolscan View Post
There are also reasons for why archival movies which has been scanned and distributed digital many years ago are now being rescanned on newer and better equipment, and that is not for the increased resolution alone.
Nope its primarily for the resolution increase , colour and dynamic range are the same as the best scanners from the 90s . They were even 4k but the processing overhead was pretty much impossible to handle for 4k. However Snow White was restored by Cinesite in the 90s at 4k.

digital film janitor
Mr.D is offline  
post #15 of 27 Old 01-21-2015, 11:02 PM
Senior Member
 
MLXXX's Avatar
 
Join Date: Jan 2007
Location: Brisbane, Australia
Posts: 352
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 66
Quote:
Originally Posted by Mr.D View Post
Quote:
Originally Posted by coolscan View Post

If you don't understand by yourselves that an article from 2000 that discuss the equipment (scanners and projectors) they had back then, which where unable to extract the full colorspace (and other qualities) of 35mm film, and was unable to digitally project the full and equal quality of 35mm film, then I don't know how someone could explain that to you.
Actually negative has been scanned at 4k to 10 bit log since about 1994. Its exactly the same spec used today. A Cineon Lightning scanner from the mid 90s creates scans that are exacty the same quality as a more recent Northlight.
I hesitate to comment as this is a field in respect of which I have only a superficial knowledge. However no one else has commented on your last post. I'm not sure whether that is because of agreeing with it, or not agreeing with it but not being inclined to pursue the matter further!

Coolscan refers above to 'colorspace' and you refer to resolution of intensity ("10 bit log"). I note that highly detailed intensity resolution of a colour component is not sufficient if the colour being scanned lies outside the colour gamut of the scanning sensor. For example, certain deep greens and reds may give the same signals to the sensor as not so deep greens, and not so deep reds. Because of that, there is no proper differentiation for colours lying at the extreme edges of a chromaticity diagram.

1. Are you suggesting that scanners of circa 1994 used sensors with a response towards the outer edges of a chromaticity diagram, thus encompassing a very large colour space exceeding that present in the negatives or prints being scanned?

2. Could you provide details of the relevant specifications* for spectral response from a scanner of circa 1994 and compare them with the spectral response specifications for a high performance scanner of today?

* An actual chromaticity diagram showing the extent of the colour space covered by the scanning sensor would be useful.
MLXXX is offline  
post #16 of 27 Old 02-28-2015, 04:31 AM
AVS Special Member
 
Mr.D's Avatar
 
Join Date: Dec 2001
Location: UK
Posts: 3,319
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 17
Quote:
Originally Posted by MLXXX View Post

1. Are you suggesting that scanners of circa 1994 used sensors with a response towards the outer edges of a chromaticity diagram, thus encompassing a very large colour space exceeding that present in the negatives or prints being scanned?
The scanner works in terms of density for the given RGB records of the negative. It densitometers and sets up for the Dmin ( reciprocity failure of the film..unexposed emulsion) ,

Cineon spec is Dmin at 95 code values( 10 bit) , the scanner still scans below that down to 1 to give some tolerance and fall off , in practice you can get density variation down to 40 or so code values.

Ref White sits at 685 code values , what this means is that you can consider everything sat between 95 and 685 as being the image detail that will transition through from the negative to a 1 light print with "nominally exposed" imagery. Everything above this is considered "headroom".

However overexposing negative is a common paradigm in motion picture shooting , the headroom in the whites is actually to allow the negative to be printed back down to allow more range in the blacks to be recorded ( above the toe of the film response curve) whilst allowing the print down to restore good contrast without leading to clipped whites.

Film scans don't target a gamut as such they record the RGB densities on the negative , gamut os about where you map that to , not about how you scan it. All the sensors had to do was record the density variation without visible banding ( hence the 10bit log usuage being nevcessary to represent a 10.5 stop range )

So yes the scans from 20 years ago are essentially as transparent as the ones from today . You only need enough precision in the sensor ...more precision does not bring you better imagery after a point.

2. Could you provide details of the relevant specifications* for spectral response from a scanner of circa 1994 and compare them with the spectral response specifications for a high performance scanner of today?

Quote:
Originally Posted by MLXXX View Post

* An actual chromaticity diagram showing the extent of the colour space covered by the scanning sensor would be useful.
Think of it as a density map , film scans are digital simulations of the original negative in a way. They had to have this functionality because they arose before DIs and had to be transparent through a lab grade.

For example a stop is notionally 98 code values and because of the log nature of the distribution of intensity in the image you can stop up or down by using additive offsets digitally which pretty closely match a lab grading process.

The scanners scans , RED , GREEN , BLUE , densities of each record , chromaticity doesn't come into it you are barking up the wrong tree . Its a film scanner not a video camera.

digital film janitor
Mr.D is offline  
post #17 of 27 Old 02-28-2015, 04:53 AM
AVS Special Member
 
Mr.D's Avatar
 
Join Date: Dec 2001
Location: UK
Posts: 3,319
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 17
http://www.cineon.com/conv_10to8bit.php

google cineon lightning recorder for piles of webchat and references dating back to the late 90s.

generally its still regarded as the best scanner out there , Cinesite in london were still using it alongside Northlights when I left in 2007 and the scans were identical.( having been there for 10 years)

digital film janitor
Mr.D is offline  
post #18 of 27 Old 03-03-2015, 04:37 AM
Senior Member
 
MLXXX's Avatar
 
Join Date: Jan 2007
Location: Brisbane, Australia
Posts: 352
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 66
Quote:
Originally Posted by Mr.D View Post
The scanners scans , RED , GREEN , BLUE , densities of each record , chromaticity doesn't come into it you are barking up the wrong tree . Its a film scanner not a video camera.
Thanks for your detailed replies above. However, it seems you have focussed on the fineness of resolution of the intensity levels captured for RED, GREEN and BLUE. That, with respect, was not the focus of my enquiry. I was more interested in the wavelengths scanned.

You have not stated what wavelength or range of wavelengths was used for what are in everyday conversation loosely referred to as RED, GREEN and BLUE. (I note these are not absolutes. Today's science of human colour vision recognises three types of human cones, S, M, and L, that respond with varying sensitivity to a wide range of wavelengths; with considerable overlap between the spectral response for M and L. See for instance the human cone wavelength response curves on the webpage: http://en.wikipedia.org/wiki/Cone_cell )

From my reading I see that the Cineon scanners used lasers. That suggests that the wavelength spread for the illuminating beams would have been quite narrow for the nominally Red, Green, and Blue scanning "colours".

However I haven't been able to discover what wavelengths were typically used for the nominally red green and blue lasers, and how these would fit onto a modern chromaticity diagram. I presume this information would be known to a number of forum members. Could someone be kind enough to enlighten me as to what frequencies/wavelengths were used for the scanning?


There are more advanced questions that then suggest themselves to me, such as whether prints and negatives had distinct spectral peaks that matched hand in glove with the laser scanning frequencies (which seems to me unlikely), and how different a print might look to the human eye when illuminated with a broad spectrum lamp rather than with narrow spectrum lasers.

I don't want to lead this thread into unnecessarily deep technical waters. Rather, I would be interested in obtaining a broad appreciation as to whether scanning with three narrow spectrum lasers was necessarily going to extract all, or practically all, of the useful colour information present in the negative or print.
MLXXX is offline  
post #19 of 27 Old 03-03-2015, 05:48 AM
AVS Special Member
 
Mr.D's Avatar
 
Join Date: Dec 2001
Location: UK
Posts: 3,319
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 17
Quote:
Originally Posted by MLXXX View Post
Thanks for your detailed replies above. However, it seems you have focussed on the fineness of resolution of the intensity levels captured for RED, GREEN and BLUE. That, with respect, was not the focus of my enquiry. I was more interested in the wavelengths scanned.
The scanner does not use laser. I'm pretty sure the lightning used a xenon lamp for illumination. ( I can't actually see why you would want to use lasers for scanning).

A prism and/or dichroic mirrors or colour filters are used to separate the light into the three: red, green and blue, imaging pick up devices. CCD is the case of the cineon lightning. Older scanners used CRTs.

The Lightning recorder does use lasers to write the image back onto negative. I assume most equivalent modern recorders also do the same.

The negative is scanned in terms of the physical density variation on the film minus the backing colour attributes (Dmin) not in terms of wavelength or gamut. It produced 3 monochrome RGB records. This is about harvesting density variation off of a physical substrate. The actual color this maps to is dependant on the end user ( you cannot view negative meaningfully) ; for example a simulation of known print stock or some other end colorspace target. Notionally negative has no color attributes its 3 monochrome records recorded by a dye seeded from emulsion exposure and development.

It gets its colour by how these monochrome records influence the exposure light path on a specific positive print stock or as I said a digital mapping equivalent to a target colorspace.

Quote:
Originally Posted by MLXXX View Post
You have not stated what wavelength or range of wavelengths was used for what are in everyday conversation loosely referred to as RED, GREEN and BLUE. (I note these are not absolutes. Today's science of human colour vision recognises three types of human cones, S, M, and L, that respond with varying sensitivity to a wide range of wavelengths; with considerable overlap between the spectral response for M and L. See for instance the human cone wavelength response curves on the webpage: http://en.wikipedia.org/wiki/Cone_cell )

From my reading I see that the Cineon scanners used lasers.
No it doesn't and I see no reason why a film scanner would. Same way you wouldn't use a laser as a light source for projecting physical film ...its pointless.

digital film janitor
Mr.D is offline  
post #20 of 27 Old 03-03-2015, 07:34 AM
Senior Member
 
MLXXX's Avatar
 
Join Date: Jan 2007
Location: Brisbane, Australia
Posts: 352
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 66
Quote:
Originally Posted by Mr.D View Post
The scanner does not use laser. I'm pretty sure the lightning used a xenon lamp for illumination. ( I can't actually see why you would want to use lasers for scanning).

A prism and/or dichroic mirrors or colour filters are used to separate the light into the three: red, green and blue, imaging pick up devices. CCD is the case of the cineon lightning. Older scanners used CRTs.

The Lightning recorder does use lasers to write the image back onto negative. I assume most equivalent modern recorders also do the same.
I see from your explanation that it is only the recorder that uses lasers. Thanks for clarifying this.

I am aware of the idea of splitting a right source into R, G and B ranges, using dichroic mirrors and colour filters having owned a Sony SXRD rear-pro TV, which used a light engine based on a single ultra high pressure lamp. [The old SXRD TV did not get particularly favourable reviews for colour accuracy but I personally liked its colour. It produced greens I had never seen before with a CRT, plasma or LCD device. Beyond the gamut of such displays.]

So it seems then that my enquiry may resolve to obtaining the specifications of the colour filtering of the earlier scanners used to create the "Red", "Green" and "Blue" beams, and ascertaining whether there's been any material change/improvement over the years. Also how the ranges of Red, Green and Blue map onto a chromatacity diagram.

(I note that the colour of very old prints apart from being subject to possible deterioration arising from ageing would tend to be rather idiosyncratic to the era and these idiosyncracies might swamp relatively minor variations in the colour capture characteristics of scanning technology.)
MLXXX is offline  
post #21 of 27 Old 03-03-2015, 11:10 AM
AVS Special Member
 
Mr.D's Avatar
 
Join Date: Dec 2001
Location: UK
Posts: 3,319
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 17
Quote:
Originally Posted by MLXXX View Post

(I note that the colour of very old prints apart from being subject to possible deterioration arising from ageing would tend to be rather idiosyncratic to the era and these idiosyncracies might swamp relatively minor variations in the colour capture characteristics of scanning technology.)
They scan the camera NEGATIVE.

Goodnight have fun with your lasers.

digital film janitor
Mr.D is offline  
post #22 of 27 Old 03-03-2015, 12:18 PM
AVS Special Member
 
Joe Bloggs's Avatar
 
Join Date: Apr 2007
Location: UK
Posts: 2,828
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 291 Post(s)
Liked: 90
Quote:
Originally Posted by Mr.D View Post
They scan the camera NEGATIVE.
For creating a Blu-ray release of older films (I'm not taking about scanning for recent films for DI etc.) don't they sometimes (often?) scan a print (or prints) of the film and not always the original camera negative. Also for composited shots of older films (eg. done optically) don't they also often scan prints of the film rather than always going back to the original elements (so they could look worse)? Aren't there quality differences between films releases on Blu-ray by film release date/decade - not just because of the film stock used but because some are said to use an old master (dvd master?) or from a print rather than the original negative - like isn't a a Blu-ray of a film from around 20 years ago likely to look worse (lower effective resolution) than a film of today in general because of those things (not just because of the film negative format used but type/quality of the scan/scanning of print(s) rather than original camera negative/lower quality due to extra generation for any optical composites)?

Last edited by Joe Bloggs; 03-03-2015 at 12:38 PM.
Joe Bloggs is offline  
post #23 of 27 Old 03-03-2015, 03:18 PM
Senior Member
 
MLXXX's Avatar
 
Join Date: Jan 2007
Location: Brisbane, Australia
Posts: 352
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 66
Quote:
Originally Posted by Mr.D View Post
They scan the camera NEGATIVE.

Goodnight have fun with your lasers.
No need for sarcasm. Of course negatives will be scanned if suitable negatives are available. I was referring in that particular post to very old sources where only a print is available. I also note the comments immediately above by Joe Bloggs.

Last edited by MLXXX; 03-03-2015 at 11:29 PM.
MLXXX is offline  
post #24 of 27 Old 03-03-2015, 03:31 PM
AVS Addicted Member
 
scowl's Avatar
 
Join Date: Nov 2003
Location: Portland, OR
Posts: 10,590
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 106 Post(s)
Liked: 37
Also with digital effects, the term "camera negative" has no meaning. This is true for many optical effects as well since they only exist on the print. They don't do fades in the camera any more.

NOW: my post on AVS Forum.
NEXT: someone else's post on AVS Forum.
scowl is offline  
post #25 of 27 Old 03-07-2015, 11:50 PM
AVS Special Member
 
Mr.D's Avatar
 
Join Date: Dec 2001
Location: UK
Posts: 3,319
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 17
Quote:
Originally Posted by scowl View Post
Also with digital effects, the term "camera negative" has no meaning. This is true for many optical effects as well since they only exist on the print. They don't do fades in the camera any more.
Er?? speaking as a VFX supervisor .

The term "camera negative" refers to the exposed negative out of the camera. It most definitely does have "meaning" for VFX or otherwise.

Optical effects ( even simple transitions) were created ultimately on an optical printer which uses multiple elements shot on negative including the camera negative and rephotographs them through a series of multiple passes using mattes ( either hand drawn or created through lab processing key passes).

The end shots are rephotographed back to negative creating an "interpositive which is itself rephotographed back to negative creating an "internegative". They do not work with print other than reference for lab grading and final sign off.

The internegative is then conformed back into the conformed master negative ; which is usually the camera negative "neg cut" together to "conform" to the desired cut of the film. The optical effect works slots in the same as any other bit of negative and then goes through a lab grade, it doesn't come in from another pipe.

digital film janitor

Last edited by Mr.D; 03-08-2015 at 12:17 AM.
Mr.D is offline  
post #26 of 27 Old 03-08-2015, 12:08 AM
AVS Special Member
 
Mr.D's Avatar
 
Join Date: Dec 2001
Location: UK
Posts: 3,319
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 17
Quote:
Originally Posted by Joe Bloggs View Post
For creating a Blu-ray release of older films (I'm not taking about scanning for recent films for DI etc.) don't they sometimes (often?) scan a print (or prints) of the film and not always the original camera negative.
Print is usually avoided it at all possible, what you may get is an interpositive or an internegative scanned but thats not print stock its low contrast intermediate to try and preserve the full latitude of the neg . Print is too high contrast to scan effectively ; you are essentially unable to grade it very well.

Quote:
Originally Posted by Joe Bloggs View Post

Also for composited shots of older films (eg. done optically) don't they also often scan prints of the film rather than always going back to the original elements (so they could look worse)?
No because there should be a conformed negative or an interpositive or an internegative to use before you have to consider scanning print. Prints are last resort. Print stock is not designed to have a long life.

Quote:
Originally Posted by Joe Bloggs View Post

Aren't there quality differences between films releases on Blu-ray by film release date/decade - not just because of the film stock used but because some are said to use an old master (dvd master?) or from a print rather than the original negative - like isn't a a Blu-ray of a film from around 20 years ago likely to look worse (lower effective resolution) than a film of today in general because of those things (not just because of the film negative format used but type/quality of the scan/scanning of print(s) rather than original camera negative/lower quality due to extra generation for any optical composites)?
Scanning technology has literally been about the same quality as today since about 1997. Prior to that point telecine machines couldn't capture quite as much range in the backs or whites.

When DVD arrived most studios remastered their back catalogues because DVD sales were rocketing. It also gave them compatability for 16x9 TVs and digital broadcasting. 1080p was a pipedream initially. The important legacy films were all being scanned to at least 2k 10bit log to preserve and in some cases restore back catalogue material. If you do that you are effectively future proof. 10bit log full latitude scanning of negative is to all intents and purposes transparent and was 20 years ago. Remastering that for video colorpace or DCP deliverables is pretty trivial at that point.

Film negative is a very robust recording material , it captures more dynamic range than is necessary desired on screen. Still at least 3-4 stops beyond what you'll necessary see on the video transfer.

Think why the original Star Trek series looks so good. Think why Dallas went downhill when they started conforming to video rather than negative. They had conformed negatives or lab rolls that they could pull and treat as if it was filmed yesterday.
Joe Bloggs likes this.

digital film janitor
Mr.D is offline  
post #27 of 27 Old 03-10-2015, 02:02 PM
AVS Addicted Member
 
scowl's Avatar
 
Join Date: Nov 2003
Location: Portland, OR
Posts: 10,590
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 106 Post(s)
Liked: 37
Quote:
Originally Posted by Mr.D View Post
Er?? speaking as a VFX supervisor .

The term "camera negative" refers to the exposed negative out of the camera. It most definitely does have "meaning" for VFX or otherwise.
But it doesn't in terms of a video release. Some people think all features can be transferred directly from the negative that was exposed in the camera. As your explanation shows, that's not possible except for a few old features.

NOW: my post on AVS Forum.
NEXT: someone else's post on AVS Forum.
scowl is offline  
Sponsored Links
Advertisement
 
Reply HDTV Software Media Discussion



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