We may have solved the mystery of why film just looks better

Forum covering all aspects of small gauge cinematography! This is the main discussion forum.

Moderator: Andreas Wideroe

User avatar
S8 Booster
Posts: 5857
Joined: Mon May 06, 2002 11:49 pm
Real name: Super Octa Booster
Location: Yeah, it IS the real thing not the Fooleywood Crapitfied Wannabe Copy..

We may have solved the mystery of why film just looks better

Post by S8 Booster »

It just won’t go away, will it? However much you can prove with specifications that digital video is indisputably better than film, there’s a stubborn feeling that there’s more to it than the simple-to-prove facts. We think we've indentified one, subtle, process that helps film to store more visible information than digital.
http://www.redsharknews.com/technology/ ... d-77487181

Image


Shoot....
..tnx for reminding me Michael Lehnert.... or Santo or.... cinematography.com super8 - the forum of Rednex, Wannabees and Pretenders...
JeremyC
Posts: 153
Joined: Fri Jun 01, 2012 2:51 pm
Real name: Jeremy Cavanagh

Re: We may have solved the mystery of why film just looks be

Post by JeremyC »

Its ironic that the author of the article talks about adding noise to pictures produced by solid state sensors i.e. degrading them. The thing the author misses is that the regular arrangement of pixels on a sensors and indeed their shape means adding noise (pseudo random) to try and match that of grain in a film has to over come this structure.
User avatar
S8 Booster
Posts: 5857
Joined: Mon May 06, 2002 11:49 pm
Real name: Super Octa Booster
Location: Yeah, it IS the real thing not the Fooleywood Crapitfied Wannabe Copy..

Re: We may have solved the mystery of why film just looks be

Post by S8 Booster »

well there may be some leeway-ends here and there but in the end it boils down to analog oversampling (ref Roger Evans 15 year or so old definition) down to the Microns (My definition)
..tnx for reminding me Michael Lehnert.... or Santo or.... cinematography.com super8 - the forum of Rednex, Wannabees and Pretenders...
Tommy
Posts: 93
Joined: Sun Oct 23, 2011 5:53 am
Real name: Thomas Dafnides

Re: We may have solved the mystery of why film just looks be

Post by Tommy »

It maybe this.....film is a real image...you can hold the film up to the light and see the picture...whereas digital is a reconstructed image from 1 and 0s...a simulated image not a real one.
User avatar
MovieStuff
Posts: 6135
Joined: Wed May 01, 2002 1:07 am
Real name: Roger Evans
Location: Kerrville, Texas
Contact:

Re: We may have solved the mystery of why film just looks be

Post by MovieStuff »

Indeed, I can't believe this author is just now figuring this out. Film at running speed always has a much higher degree of detail than any single film frame can offer. This is because of the random grain pattern inherent in the image capturing process that spreads the total detail across a large volume of film grain but not all in the same frame and with the film grain never in the same place twice. When the film plays back at speed, these random grain patterns (and the detail they carry) stack up on each other, filling in the "dead spots" between the various grains. It's what I call "accumulative resolution". But digital has a fixed pixel pattern so the degree of detail is no better at running speed than on a single frame because any "dead spots" between the pixels are always dead. Even if you increase the level of resolution on a sensor to reduce or eliminate the dead spots and equal the total accumulative resolution of film at running speed, I think the brain sort of figures out the difference and knows that all the digital information is being delivered at once and not spread out across multiple film frames.

But I also think this is like knowing the subtleties of a fine wine compared to MD2020. The average drunk could care less and neither do most audiences these days.

Roger
JeremyC
Posts: 153
Joined: Fri Jun 01, 2012 2:51 pm
Real name: Jeremy Cavanagh

Re: We may have solved the mystery of why film just looks be

Post by JeremyC »

But I also think this is like knowing the subtleties of a fine wine compared to MD2020. The average drunk could care less and neither do most audiences these days.
You are right if they can only watch things on some of the digital crappy screens here in suburban London or on a large LCD or plasma TV (fixed noise pattern -> compression with artifacts at some stage - > fixed pattern pixel display)
User avatar
MovieStuff
Posts: 6135
Joined: Wed May 01, 2002 1:07 am
Real name: Roger Evans
Location: Kerrville, Texas
Contact:

Re: We may have solved the mystery of why film just looks be

Post by MovieStuff »

Well, modern audiences don't seem to demand much and, frankly, the use of digital over film is hardly the weakest link in the chain when it comes to the problems inherent in most films, today. Audiences tolerate bad scripts, marginal acting abilities, luke warm direction and an over reliance on special effects and keep packing the theaters. They could care less about the look of film vs digital.
PyrodsTechnology
Posts: 64
Joined: Sat Dec 11, 2010 10:45 pm
Real name: Roberto Pirodda
Location: Italy

Re: We may have solved the mystery of why film just looks be

Post by PyrodsTechnology »

Roger, you have described perfectly the "accumulative resolution"of film! It is true that average audience do not care but the point is that who shoots at pro level KNOWS (or should know) , this difference between film and digital
nikonr10
Posts: 421
Joined: Sat Dec 03, 2011 11:41 pm
Real name: Christopher Nigel

Re: We may have solved the mystery of why film just looks be

Post by nikonr10 »

As a famous picture editor once said to me" let then have there digital moments ?
leica have just made a new film camera , so there still have faith in film , and this is not a cheap one , mind you none of then are .

it really is like good food and bad food to me, I shoot film because it lives moves and has alife of it's own in the right hands , and will still been around not like bit rot digital ,
anyway Digital here to stay and that what the mass wants so that what there get and it easy and cheap to do ! And a new better camera will be out and then the one you have is out of date with no spare parts ,

Film has really gone up in price and the kodak moment is not for the poor ? still it/s amazing to think that the camera that we use's are around forty to fifty years old and still going strong ! no one can say that about a digital camera only time will tell .
Tscan
Posts: 548
Joined: Fri Sep 03, 2010 6:44 pm
Real name: Anthony Schilling
Contact:

Re: We may have solved the mystery of why film just looks be

Post by Tscan »

Same could be true with motion blur in which the mechanics and results are very different between film and digital. Blurred motion on a still frame of film makes for smother motion in play. The rolling shutter and fast exposures do no justice to digital in motion.
Reborn member since Sept 2003
User avatar
MovieStuff
Posts: 6135
Joined: Wed May 01, 2002 1:07 am
Real name: Roger Evans
Location: Kerrville, Texas
Contact:

Re: We may have solved the mystery of why film just looks be

Post by MovieStuff »

Only CMOS chips have a rolling shutter. CCDs had them long, long ago but quickly went to a global shutter, which actually handles motion blur the same as film, unless the shutter speed is high. Because the rolling shutter makes things look rubbery, people tend to set the CMOS shutter high to negate the effect, but there's a whole new line of CMOS chips coming out now that have global shutters and CMOS tends to emulate the color and contrast of film better than CCDs. So they're getting closer.
JeremyC
Posts: 153
Joined: Fri Jun 01, 2012 2:51 pm
Real name: Jeremy Cavanagh

Re: We may have solved the mystery of why film just looks be

Post by JeremyC »

Very interesting, so Roger how do CMOS sensors (or the latest CMOS) deal with film images better than CCD sensors?
User avatar
MovieStuff
Posts: 6135
Joined: Wed May 01, 2002 1:07 am
Real name: Roger Evans
Location: Kerrville, Texas
Contact:

Re: We may have solved the mystery of why film just looks be

Post by MovieStuff »

Oh, I have no idea "how" they do it. I just mean that, to my eye, the results from CMOS looks more film like than CCDs.
carllooper
Senior member
Posts: 1206
Joined: Wed Nov 03, 2010 1:00 am
Real name: Carl Looper

Re: We may have solved the mystery of why film just looks be

Post by carllooper »

Yes, an image is not in any single point, in space and/or time, but in the space (and time) between such points. Between the pixels. Between the frames. Between the grain. The random arrangement of the film grain means there is exceedingly more diversity in the potential image. From one grain to another is a change in value and it is this change we see. Not the grain (or the pixels themselves) but the differences between such. We can see this in a binary image.

This image is composed entirely of pixels that are either black or white. No grey pixels here. But subconsciously (if not consciously) we can see the grey tones.

Image

Counter-intuitively we can get an idea of this subconscious image if we blur the dots. Below all I've done is apply an 8X gaussian blur to the above. This is effectively looking into a subconscious aspect of the above - bringing into "focus" (so to speak) what our subconscious mind is otherwise able to appreciate (to see) as occupying the above. We can see grey tones in the above image because they are, in some sense, there. Our eyes roam across the dots, criss-crossing them from different angles and reconstructing something beyond what is otherwise momentarily apparent (as dots). We begin to appreciate instead, a face. With grey tones. Of course all of this happens very fast. But the face that emerges (what we end up seeing) is the image proper. What we see is therefore not an image of a face. It is the face itself which is the image proper. A face image. The dots are not an image. In this sense an image can be understood as much more related to reality than are the dots. Image and reality can be understood, in the limit, as the same thing. The dots are something else. A kind of intermediary. Not yet an image. Not yet a reality. In this sense we can say the image is not an illusion as such, as if the dots were reality. On the contrary, it is the image which is a reality (and reality an image) - not the dots.

Image

In any case what this suggests is that what we see subconsciously, or perhaps super-consciously (!) are not the dots (pixels/grain etc) as such (unless we're specifically concentrating on such of course). It is the image (a face) we see (or feel, or comprehend) and in terms of technical considerations (the domain of dots) this image/reality occupies the space between the dots/pixels/grain. And in the case of video or film, the space (or rather time) between the frames as well - where there is otherwise, on close analytic inspection, nothing there but the apparent absence of any face (of any reality). This is the basis for a realist conception of film (and digital for that matter). It is not in some division between image and reality (leading to an anti-image conception of the real) - but the opposite - of an equivalence between the two. Image and reality become the same thing. A reality that subsists between the dots. Entirely visible. Is defined by it's visibility. Is that which is unable to be invisible (although often hidden). Goddard might argue (or have argued) otherwise: the image as some sort of illusion.

But this is the freaky thing. We can see what is not, on the face of it technically, there. Are we like seers or spirual mediums? Visionaries? Able to see the absent? Yes. In short, precisely so. But such discoveries allow us to understand how we might technically manipulate the intermediaries (the dots, the pixels, the grain, the frames) in order to better bring into "focus" (so to speak) what otherwise strikes us as absent (not really there). Reality we can otherwise theorise is simply not in the moment. It is not locally situated. At any given moment is an intermediary. Dots etc. Reality occupies a different dimension - action at a distance (in time sand space). It refuses to be contained (or erased) in confined spaces. It is always outside the box. It is that which impresses itself from outside, of time and space, into such. Occupying the space/time between the pixels. Between the grain. The image proper. Reality proper. A kind of spooky thing. An apparition. A ghost. A reality. Or a ghostly reality. It insists or insinuates itself. Refuses to be absent. But equally refuses to be entirely present. As such there is no such thing as the perfect crime. Colombo is an angel as Wim Wenders reminds us. He can, and we can, (with the power of angels) see clearly Hitchcock's rope.

But back to a more 'technical' understanding: the spatial randomness of film grain allows the image (reality) more scope (or more freedom) in how it expresses itself (between the grains), than that which digital pixels allow. There is simply more potential. We can reconstruct a far more complex ghost through such. We can feel the image more intimately - in our very being. It is a far more haunting image. Far more powerful.

Carl
Carl Looper
http://artistfilmworkshop.org/
Angus
Senior member
Posts: 3888
Joined: Fri Mar 07, 2003 11:22 am
Contact:

Re: We may have solved the mystery of why film just looks be

Post by Angus »

MovieStuff wrote:Indeed, I can't believe this author is just now figuring this out. Film at running speed always has a much higher degree of detail than any single film frame can offer.
Roger
This reminds me of a rather daft comment made by George Lucas in the early 1990s. At that point, the best video cameras had a resolution of 800-1000 lines which he equated to one 35mm frame (wrong but what the hell). He then made the jump that video had become as good as 35mm film so he'd not need to use film again.

The man made some great movies, but actually hasn't got a clue what made them look so good. As anyone who saw his Star Wars prequels knows. They look *horrible*.
The government says that by 2010 30% of us will be fat....I am merely a trendsetter :)
Post Reply