Film vs Digital - Color Rendition

[mauro_franic]

<p>Les, for your question.</p>

<p>This is the XD11 the winner will receive (along with $100 of film). I am including a new Hoya HMC to protect the baby.</p><div></div>

[mauro_franic]

<p>Andy, (or anyone else),</p>

<p>Please let me know what is the best way to make the RAW files available to you.</p>

[indraneel]

<p><strong>How has the scanner been profiled? </strong>And it does have a fixed spectral response curve?</p>

<blockquote>

<p>For many film users getting accurate colours was a nightmare for years.</p>

</blockquote>

<p>I'm tearing my hair out everyday, try to coax correct colors from slides scanned on my profiled scanner... some batches have rapidly aged, but still...<br>

and, oh... it's a Minolta scanner...</p>

[Dave Luttmann]

<p>Sure Mauro. Love to participate in this. </p>

<p>Regards.</p>

[mauro_franic]

<p>"with the B&W films - did you use any filters and how did you develop them?"</p>

<p>Just a UV(0) on all the lenses.</p>

<p>Developed both on Xtol 1:1 at 75F for 7 minutes.</p>

[mauro_franic]

<p>Dave becomes the first honorary judge.</p>

[mauro_franic]

<p>Dave, after you post your entry I will send you the scans.</p>

[danielleetaylor]

<p><em>First, film (or scanners) do not interpolate color and this translates into higher color resolution which provides a less smoothened (infamously "plasticy") rendition.</em></p>

<p>This is simply not true Mauro. See attached. Note that the 7D has finer resolution of relief texture details, which are color, than 35mm Velvia 50 on an Imacon scanner.</p>

<p>"Plastic" is a complaint that comes up when too much NR is used on high ISO images or when large prints are made from lower resolution files. It is not the result of Bayer interpolation.</p><div></div>

[mauro_franic]

<p>Daniel, would you like to be a judge. Dave and I are the judges so far and two more are needed.</p>

[mauro_franic]

<p>Entries will be posted here:<br>

http://shutterclick.smugmug.com/Photography/Film-and-Digital-Color/16002634_5Eo6Z#1200457847_Makh3</p>

<p> </p><div></div>

[danielleetaylor]

<p><em>Additionally, and most importantly, different films give photographers a characteristic and predictable look, an array of ammunition with different color responses to individual wavelengths.</em></p>

<p>There is some truth to this statement if we're talking about color slide film viewed directly. But there can be a considerable amount of variability once you scan or print any color film, even using traditional techniques. I recall putting a lot of work into making scans match slides on a light box back when I did more scanning. Maybe I just needed a better scanner or better profiles, but it was not automatic. The scanner had more or less difficulty depending not just on film, but on how the film was exposed and what the lighting was like in the exposure. This was even harder with negatives because, well...what's the reference point? This, incidentally, was a big reason why slide film was preferred for publishing.</p>

<p>Still, a photographer can load up a particular film and know that the "look" will be the same within a certain range of variability. A good scanner with good profiles can narrow that range down. That can be useful, but should not be over played as an "array of ammunition" when a photographer can create hundreds, if not thousands of different renditions using RAW converters, PS, and various plugins. Once a photographer has established a look and the digital work flow to achieve it consistency is remarkably high because digital is remarkably consistent to begin with. Skim over the color chart samples at DPReview or Imaging Resource. There's little variation at base settings.</p>

<p>B&W is more interesting because the photographer has so many choices during development and printing. It takes a little bit of work to develop both the eye and experience necessary to do good digital B&W conversions, but they are doable and digital is quite flexible once you get the techniques down. I think it took digital longer to produce satisfactory results in B&W however. The tools needed to evolve and so did the printers. Today B&W work of the highest caliber can be produced with a completely digital work flow.</p>

<p><em>This cannot be reproduced with digital cameras (as of today) since the response of a single sensor to different wavelengths cannot be changed.</em><br>

<em>Consequently, it is impossible to reproduce the look of a particular film (e.g. Velvia) starting with a digital capture unless that particular film is used alongside the digital camera to match the colors later in PS.</em></p>

<p>These are contradictory statements. The response or output of the sensor can be changed which is why it's possible to match images later in PS. PS is changing the response.</p>

<p>For anyone who really loves the look of a particular film and wants to emulate it digitally, shooting both side by side for a couple months under a wide range of conditions and taking the time to match the digital shot to the film shot is the way to go. It's more work than a plugin, but in the end you will have a feel for what needs to be done under different conditions. The plugins tend to fail under certain variations. But if you just want the feel and are not concerned with pixel level accuracy as in a test like this, then the plugins are a simple and effective way to go. Alien Skin Exposure is a lot of fun to work with.</p>

<p>As to the contest...it would be more interesting and fun without the crusade aspect. It also requires the RAW files IMHO, especially since you used AWB rather than daylight balance.</p>

<p>But it doesn't prove anything. Imagine the opposite of this contest, a digital photographer posting a dozen renditions of each image and asking if any films can match them.</p>

[danielleetaylor]

<p><em>Daniel, would you like to be a judge. Dave and I are the judges so far and two more are needed.</em></p>

<p>I appreciate the offer but I probably don't have the time this week. I will be curious as to what different people submit however, so I'll probably drop back in after a couple days. I'm sure the thread will still be going.</p>

[mauro_franic]

<p>Shame. You can follow the entries from an ipad/laptop here:</p>

<p>http://shutterclick.smugmug.com/Photography/Film-and-Digital-Color/16002634_5Eo6Z#1200457847_Makh3</p>

 

[danielleetaylor]

<p><em>Daniel - Why do you keep posting your optimized processed digital result to the none processed straight up scan when we already concluded this - <a rel="nofollow" href="00WErk?start=160"><strong>It all comes down to the print - page 17</strong></a> Conveniently forgot it?</em></p>

<p>Page 17 is highly focused on an area of B&W detail, not color detail, using a level of magnification that degrades the digital image due to interpolation effects. See my comments at Apr 26, 2010; 08:45 a.m. on page 23 http://www.photo.net/film-and-processing-forum/00WErk?start=220.</p>

[mauro_franic]

<p>"<em>This cannot be reproduced with digital cameras (as of today) since the response of a single sensor to different wavelengths cannot be changed."</em></p>

<p><em>This meant that it cannot be reproduced without shooting the film side by side. </em></p>

<p><em>Let me know if there is a shareware file share I can use to upload the RAW files for everyone.<br /></em></p>

[mauro_franic]

<p>Les, I expect you will submit an entry in this contest...</p>

[rodeo_joe1]

<blockquote>

<p>"First, film (or scanners) do not interpolate color and this translates into higher color resolution"</p>

</blockquote>

<p>What nonsense! Where is there any evidence to support this assertion?</p>

<p>The colour in film is captured by 3 and only 3 discrete colours. When film is developed its colour couplers form tiny oily globules of dye in combination with oxidation products from the developer. These tiny blobs, called “dye clouds” are between 2 and 3 microns across, and there is a limit to how many of these dye clouds can be stacked vertically in the thickness of an emulsion layer. It’s this overlay of dye clouds that gives variation in colour density, and we can fairly easily estimate the number of different densities of colour they can represent.<br>

<br>

In addition, each film dye cloud has a fairly uniform density, so shades of colour have to be represented by additions of densities or by the natural dithering of randomly scattered dye clouds. Therefore film colour is definitely a product of an interpolation process.</p>

<p>To put some real numbers on the theory: The thickness of one film emulsion layer will support a stack of around 5 dye clouds, and we can fit between 9 and 16 of them into the same area as one digital pixel (assuming a 12 megapixel full-frame sensor). However, it’s only the vertical stacking of dye clouds that has any bearing on the number of different densities of colour that can be represented. So let’s be generous and say that we can represent 6 levels of cyan, yellow or magenta for each 2 micron by 2 micron square of film. That’s a maximum of 16 x 6 (= 96) levels within the same area as our example pixel above. Now, can anyone see a slight discrepancy there between the 256 levels of each primary that one digital pixel can represent and what we can optimistically expect to get from the same area of film?</p>

<p>Digital pulls even further ahead when RAW capture is used to extract 4096 or more shades of colour from each of its red, green and blue channels.</p>

<p>Sorry, but digital capture is technically capable of far smoother tonal transitions than film could ever hope to achieve, and the very fact that we can use it to imitate the limited range of colours that a particular film reproduces should be ample proof of that.</p>

[Dave Luttmann]

<p>I'll get a chance in a few days. I'll be busy until at least Wednesday.</p>

[Landrum Kelly]

<p>Frankly, I like the subjective element in this exercise--to a point. It really comes down to whether or not film does indeed convey some special quality that digital cannot capture. So, the fact that it is not a particularly rigorous test measured against some objective standard doesn't bother me at all.</p>

<p>On the other hand, I think that I can see the day coming when these kinds of threads will be so totally passé that we will tend to look back at the film v. digital debates as being an inevitable phase that we had to go through. After all, let's face it: digital is so convenient, and the work flow starts with a very direct, very quick scan. </p>

<p>In other words, it is almost up to film users to be able to show why film is worth the extra trouble. I am committed to keeping an open mind, and, when I look at old prints made from film, I have to say that I am looking at something wondrous indeed. But those are tests of PRINTS--the ultimate test for me--and it seems almost impossible to make the case for film on the digital medium of the internet. I HAVE to see the prints in order to decide. That is one of my ultimate dogmas: SHOW ME THE PRINT(S)! I have to see the prints. Since we cannot really do that here, I am not too optimistic that this is going to be at all decisive--but one can hope.</p>

<p>So, for at least one more time, Mauro, let us have another marathon and see if anyone manages to offer something so compelling that we will all set our digital cameras back on our shelves (or on the auction blocks) and say--again, one more time--"I'm going back to film!" </p>

<p>I don't see that happening, but, given the amount of old film equipment sitting around my house, I sort of hope it does.</p>

<p>The brutal fact is that I have not shot film for at least a couple of years, much less made serious prints from film. Nonetheless, Mauro, if only for old time's sake, let us see what you and others can pull out of the hat where film is concerned.</p>

<p>--Lannie</p>

[mauro_franic]

<p>Joe, do you know the size of a pixel from 35mm film scan at 21 megapixels has about 40 microns? Then this is scanned at full 16 RGB after that.<br>

<br /> A pixel of a digital camera is just one color with varying luminosity.</p>

<p>Please help me understand your math. I am not saying your are wrong- I just couldn't follow it.</p>

<p> </p>

[mauro_franic]

<p>Nice Lannie. <br>

Color resolution aside, when it comes to color choice yes, you can try to make a Velvia look from a digital camera, the same way you can try to make a Velvia look from Ektar. Without shooting Velvia it would be almost impossible.</p>

<p>Let's have fun with this exercise. It just came to me this is a great opportunity to test those film emulators people rave about and see how close they come to the real film. Do you use any of them? -> That could be your entry.</p>

[richardsnow]

<p>Well, I don't care about the FvD debate, but I like the game. Here's my go at it.</p>

<p>RS</p><div></div>

[richardsnow]

<p>Edit see post below:</p>

 

[richardsnow]

<p>Ok, after two attempts at uploading to Pnet, here's a link:<br>

<a href="http://www.fairpoint.net/~thesnoman/Pnet/FvD-ColorRendition(screen).jpg">FvD Color Rendition (1400px Wide)</a></p>

[mauro_franic]

<p>Thank you Richard. </p>

<p>Your rendition (so will others that come in) has been posted here.</p>

<p>http://shutterclick.smugmug.com/Photography/Film-and-Digital-Color/16002634_5Eo6Z#1200457847_Makh3</p>

<p> </p>