How good was Kodachrome?

<p>Gentlepersons....</p>

<p>How sharp was Kodachrome 10 and 25 in reality? In 1996 Kodak rated the resolution of Kodachrome 25 to be 100 lp/mm at 1000:1 contrast and 63 lp/mm at 1.6:1. They rated Ektar/Royal Gold at 200 lp/mm and 80 lp/mm respectively. AGFA rated their Ultra 50 at 140 and 50, their Portrait 160 at 150 and 60 and their chrome RSX50 at 125 and 55. Konica rated their Impresa 50 at 130 and 80. Fuji rated their Reala (100) at 125/63, their Velvia at 160 and 80 and their Provia and Sensia 100s at 140 and 60. All these specs would indicate other films beat out the Kodachrome 25. Even Fuji's Provia 1600ASA was rated the same 100 lp/mm at 1000:1.</p>

<p>By 1996 I had some quality sharp lenses like the Contax/CZ 50mm 1.4. I used all these films and can look at them under some good quality microscopes. I found nothing that out resolved Kodachrome although Velvia and Ektar/RG 25 were close. I'm not talking apparent sharpness which can be enhanced by the way the film handles "edges" or tonal change. I'm talking about detail either being there on the film or not. I did not have the same quality of lenses in the 40s and 50s but Kodachrome 10 looked pretty sharp to me too.</p>

<p>Did Kodak down rate Kodachrome to make Ektachrome look as good or better in the advertisements? Who am I supposed to believe anyway, Kodak or my lying eyes?</p>

<p>I'm sure all the film producers tested their competitor's film against their own from time to time. Anybody out there work for Kodak 15-20 years ago and know the truth? Was Kodachrome 25 really the resolution king I think it was?</p>

<p>Tom Burke</p>

<p>P.S. Why would I be suspicious? Because I'm old, I can remember when just prior to the start-up of the 1952 year model production at GM, Cadillac, much to their chagrin, found out that Buick was getting almost 200hp out of their 320 ci old style straight 8 while Cadillac was only getting a little over 150 out of their larger and newer generation 331 ci V8 (but had been advertising it at 160hp). GM corporate step-up marketing could not have that. Part of Buick's secret was their 4bbl carb, new for 1952. GM brass demanded that it was also given to Cadillac with some intercompany costing advantage back to Buick. Even with the 4bbl on both engines, the Buick still out powered the Cad. So....Corporate brass dictated that Cadillac advertise their engine at 190hp and Buick had to state only 170hp in the "official" specs and advertising.</p>

<p>No I dont think it was the resolution king.<br>

You have to remember though, if youre resolving 100 lp/mm, that's still quite a lot of resolution.<br>

Local contrast also comes into play of perceived sharpness, as does the texture of the grain which would leave a somewhat of a diffused imprint of the grain in the dye clouds on colour film.<br>

Which colour negatives typically would have less of, making it more difficult to distinguish detail.<br>

There is also processing, I've found lab processing of C41 was inferior to my home processing of even the tetenal kit.. much thinner range negs.</p>

<p>Im sure in some magazine or publication there would have been resolution charts photographed with these various films.. you'll just have to try and hunt them down with your google fu.<br>

I also dont believe manufacturer claims (Fuji under rating Astia 100f's resolution).. but I dont think such a massive discrepancy would be within their normal realm of exaggerations</p>

<p>there is an implication that Kodachome is being discontinued so ektachome can " look better"<br>

or ektachome has been improved to the point that it is so close that there is no rational to continuing kodachome.<br>

a similar thought was mentioned when panatomic x was discontinued as plus x was "so close: that there was no adavantage to keep making panatomic x.</p>

<p>the truth may be that there was no profit im making more panatomic x ( or in this case kodachrome) This may be closer to the truth.<br>

we do not know what dicussions may have been taking place at kodak HQ.<br>

and the new management are more business people than photo prople.<br>

if kodak thought they could make more money selling microwave popcorn in yellow boxes,m it could happen. ( outrageous)</p>

<p>on the other side, Kodachrome cannot be that big a deal of a small startup liek dynachome could make a clone of kodachome. this implies " how hard can it be"<br>

this is not intended to knock kodachome in any way.<br>

additionally: things have changed and there are no longer reversal papers. all slides have to be scanned and prointed by other methods<br>

third kodak should have made a kodachome that could be used in more cameras<br>

My excellen canon P&S apparently do not accept iso 64 as a valid speed film<br>

if kodak had made iso 100 kodachome. maybe more would use it.<br>

the distribution of any slide film is meager.<br>

nobody sells it except big cities and mail order.<br>

all this has worked to bring about the end of kodachome. and possibly in the near future of all reversal films. ( sad but possible)</p>

<p>I think any big company only looks at the " bottom line"<br>

be thankful we still have what we have and not all forced to use plactic P&S digicams.</p>

<p>as far as GM is concerned,<br>

despite starting out linke an ocean liner<br>

big roar and smooth starts,<br>

Buicks with that dynaflo were<br>

impessive cars. big and very smooth.<br>

some thing that is missing on nerwer tinny cracerbox<br>

we ride around in a 35mpg american crackerbox..<br>

but as I get older I miss my 1963 star chief.<br>

we took apart a olds 98 that had bishaped g metal strips<br>

inside the doors, like the strips they put on the sides of<br>

van/scholbusses. wow what a saftey feasture.<br>

new cars do not have that.</p>

<p>Kodachrome is (was) an high contrast film, known for brilliant reds and yellows. Ektachrome was initially more mellow, with good blues and greens. Ektachrome 64 (mid 60's) brought some parity with Kodachrome's contrast, but still emphasizing blues and greens. High contrast brings the illusion of high resolution, which was questionable in practice.</p>

<p>Neither Kodachrome nor Ektachrome exhibited good resolution when projected or printed, due to many factors. The resolution of projectors is limited by the quality of the optics and transport mechanism. Reversal prints had poor color and contrast, and often poor optical quality. Perhaps more important, the film speeds were low, requiring wide apertures and/or slow shutter speeds. Camera shake was and still is the main limitation to sharpness in hand-held photography.</p>

<p>The published resolution numbers for Kodachrome (and all other Kodak products) are fair representations of the resolution. The problem is that resolution is a very poor measure of sharpness. All of us who worked on Kodachrome knew that the resolution numbers did not portray the product in a favorable light. If you look at the full MTF curves, you find that Kodachrome has a steep curve. At spatial frequencies just below the resolution limit, the response is higher than some other products. At low frequencies, the response is over 100%. This has a lot more to do with what you see than a resolution test.</p>

<p>The other factor is the much debated relief image effect. All Kodachrome users can recognize the product by looking at the emulsion surface under reflected light. This relief image contributes to sharpness when a specular light source is used. Kodachrome slides have better sharpness when projected than when printed or scanned (depending on the scanner). </p>

In the MTF curves of K25 and K64, the "enhancement" at low spatial frequencies up to 20 periods/mm distinguishes the Kodachromes from most other current slide films, resulting in a pronounced plasticity not only on the screen, but also in a good scan. The only similar E-6 film with a "natural" color saturation and MTF exceeding 100 % is Fuji's Provia 100F, which has a finer grain and a higher resolution. At 5400 dpi scanner resolution, Provia 100F gives a marginally better overall impression than K25 Professional. At 2700 dpi, the difference is mainly in the color rendition - even with ICC profiles there is no E-6 sky as in Kodachrome blue.

 

For color offset printing, all the current standard-speed slide films (Kodak and Fuji) will work well. But in the 1960s/1970s, only Kodachrome II, 25, and 64 provided an image quality sufficient for full-page or double-page format from 35 mm slides. Agfa's CT18 was the European standard in medium format with vivid colors and a good archival durability, but not a serious rival as 35 mm film.

 

Kodachrome is demanding to scan correctly. It is a pity that a fitting ICC calibration slide became available for K64 only half a year before Kodak's discontinuance announcement. Meanwhile, I can handle it as well as the E-6 films and enjoy the high image quality from good lenses - as from Cosina and Zeiss.<div></div>

For comparison, Provia 100F:<div></div>

<p>Mr. Andrews...</p>

<p>You are THE expert on Kodachrome, no doubt about it. I was hoping to hear from you on the subject. I have to take what you say at face value. I'll stand corrected. </p>

<p>When I did my sharpness test, I did not yet have a USAF1951 test chart from Edmund Scientific. I took two pictures, about a half hour apart, both in early afternoon sun, down my block and across the highway to the putting green of the local golf course. Both Kodachrome and Ektar were 25 ISO, so I was sure to take the same exposure for both. The in-camera meter did not argue with the same exposure and, as the sky was cloudless, I would not have expected the light to have changed in half an hour in summer early in the afternoon. Both films were the sharpest I shot that day. </p>

<p>When writing my question above, I had it in my mind that I had used my Contax 50mm F:1.4, but I do remember now that I had not yet bought that lens and was using the 50mm F:1.7. F:1.7 lens is not quite as sharp (although damn close) , but probably still out-resolves both films and was not the weakest link in the chain. Of course, I was looking at a positive and a negative, which is sometimes deceiving. I judged the final difference in the ability to see the golf club shaft distinctly with the Kodachrome and maybe a ghost with the Ektar. Unfortunately and of course, the golfers were wearing different clothes so I couldn't catch a plaid pattern or something on a shirt or jacket. Golf club heads in golf bags looked identical enough between the two films that I would have called it a draw, had I not been able to use a golf club shaft at about the same place and angle on the green. The shaft was also at approximately the same position in the exposed frame so I was not looking through a sharper part of the lens in one of the shots. Obviously, the test was not scientific, like using a resolution chart, or even the same item in the exact same spot. </p>

<p>Within a year, Popular Photography (and I don't consider them the most accurate written source of facts rather than opinions) came out with an article comparing the two, showing a massive blowup of the same subject, a Christmas toy with fine detail. They judged the Kodachrome to have the better resolution. The two panels of film, as they printed it, seemed to bear out their conclusion. On the other hand, I know the results I got were so close that with a combination of scanning and printing, I could certainly make either one come out the winner on a print to "prove" a point to someone. Also, if PP did their even best, their equipment that scanned the film and/or the process that put it on a magazine page made out of low-cost paper could easily have altered the results unbeknownst to them. </p>

<p>Furthermore, I have used Kodachrome's 25, 64 and 200 and did not find the resolution the same between them at all. Official Kodachrome specs show 25 and 64 both to have resolution figures of 100/63 and Kodachrome 200 to have 100/50. 25 and 64 should have been the same, with 200 darn close. My eyes see 25 sharper than Velvia and 64 less sharp. The 200, well it was 200. I know the grain was different sized and my eyes are old, but I was looking through fairly high-powered microscopes and I was 10-15 years younger then. That was also back when both films were current stock and on the shelves of the camera stores that are gone now. </p>

<p>Still, Mr. Andrews, I truly yield to your expertise. The above was not to counter or argue with what you said, but to let you know how I had come to my conclusion. I again thank you for your learned input. Photonet is fortunate to have someone with your background available to separate fact from opinion. </p>

<p>Mr. Degroot....</p>

<p>No, sir, I was not meaning to say Kodak is discontinuing Kodachrome to make Ektachrome look better at all. I was meaning to say that I think Kodak was trying to switch people to Ektachrome way back then, as it was more profitable and could compete with other brands locally processed film. That last feature alone would give Kodak a boost in market share. </p>

<p>Off topic, I had a '52 Buick Roadmaster with the aforementioned engine. You're right, it was a little like an ocean liner, and before they had stabilizers, at that. It was a great car. I also had a 1972 Rolls Royce Silver Shadow. The Rolls Royce was almost as smooth driving and had almost as plush a ride as the old Buick. I bought Buicks after the '52, having had a '55 and a '57. They were a lot faster and very Buick-like, but nothing has ever felt like driving my '52. Would I rather drive it now than the relative cracker boxes that I drive today? You bet! Besides, how do you ever enter a cracker box from the curb side and just slide on over to the steering wheel? Of course, that presumes you can twist your body into a shape where you can get into the passenger side of a cracker box from the height of a curb. Yup, "been no good rock and roll since Buddy Holly died," says this old fart. </p>

<p>Other Gentlepersons....</p>

<p>Thank you kindly for your many responses. </p>

<p>Tom Burke </p>

<p>Tom, thanks for your kind words. As with any corporate endevor, I was one of a few dozen staffers and a few hundred production workers who contributed to Kodachrome. Of those who frequent photo.net, Ron Mowrey has more experience in K-14 processing and designing new Kodachrome products. </p>

<p>As for why the official resolution numbers don't match your experience, there is no way I can know for sure. I know the published numbers followed standard procedures. I know the resolutions results for Kodachrome never matched the subjective impressions of the product. My best guess is what I wrote before--that in your pictures, you were looking at spatial frequencies lower than the resolution limit. </p>

<p>One thing I'm sure of--the numbers were not biased to make Ektachrome look better. This is not only unethical, it is counter productive in the long run. </p>

<p>Mr. Andrews....</p>

<p>Modesty, you certainly have. I doubt the other few dozen staffers and few hundred production workers had your level of performance and/or patents granted while at Kodak. It is nice to know that Mr. Mowrey is also a Kodachrome expert. </p>

<p>Of course, a person can never lay their hands on the material when they need it. I cannot find the two old slides of Kodachrome and Ektar that I am referring to. That being said, if memory serves me, I did a comparison of resolved items against later film I had tested with the USAF1951 chart. I also obtained a glass slide from Edmund Scientific with the same USAF1951 chart information sized to test microscopes and scanners. With that, I can look at spacial frequency directly through a microscope on the soda glass slide and then on the film. I have even sandwiched them together to double check my comparisons. I came to the unscientific conclusion that the resolution of the club was around 125 lp/mm. The contrast was a lot closer to the 1:1.6 than 1:1000. That would not be out of line with the spacial frequency of Ektar/RG 25 whose numbers are 200 and 80. It would be impossible at the Kodachrome 100 and 63 level. At the time I made the tests, I was not aware of the official Kodak numbers so there was no bias or prejudice. I could speculate that the silver one can see on the back of a Kodachrome slide may have darkened the shaft to make it more visible with a backlit microscope. </p>

<p>You worked for Kodak at a level where your belief that they did not purposely bias the numbers is certainly more credible than my suspicions. I of course would take you at your word. I agree that it would be unethical and things like that are often counter productive. However, if large corporations acted ethically and properly estimated productive actions vs. counter productive actions, we taxpayers, the stockholders of America, would not have been pumping megabucks into many of America's largest corporations today. I am also reminded of the story about a somewhat recent corporate president who answered a stockholder complaint during a meeting that Kodak was losing its market share by having the reporters/photographers hold up their hand if they had Kodak film in their cameras. It was said the paucity of raised hands ended that attempt of justification. </p>

<p>Kodak no longer publishes resolution or grain RMS specifications for its film. Their print grain index certainly can't be compared against their older films or anybody else's. Would you be comfortable in guessing from your experience and knowledge what the resolution figures might be for the new Ektar 100? "No" would certainly be an acceptable and understandable answer. </p>

<p>I recently bought a used Mamiya 7II with an 80mm lens. As I was on a medical trip when it arrived, I asked a professional photographer friend to run a roll of my Ektar 100 through it to see if it appeared to be working right. In my little town, E 6 is only run once a week, but C 41 is run every day and I did not want my seven-day return period for the camera to run out if there was something wrong. All seemed well. After returning home, I ran some Velvia 50 through it, standing in the same place with the lens pointed at the same objects he had used to test the camera. I was surprised at how much more detail showed up on the Velvia. Writings on distant advertising signs, which didn't even show on the Ektar, could be read, letter by letter, on the Velvia. It's hard sometimes to translate masked negative film results with unmasked positive film, and I would say I metered about 2/3 a stop lower than he did, but even with those differences, we both agreed there was quite a resolution difference. Using a 300 power microscope, the grain (actually dye bits or clumps) on the Ektar appeared smaller than on the Velvia. As an aside, having had a Fuji GW670III, which had a reputation as a reasonably sharp lens, I was surprised at the high level of resolution the Mamiya produced on both films. </p>

<p>Again, I thank you for your many contributions to Photonet and answers to me personally. Upon seeing your offer to Mr. Bayer regarding the Kodachrome project in another thread here on Photonet, I could do nothing less than meet your level myself. Although I once had a pretty good income, I'm certainly not wealthy today, having lived for many many years off the money that I earned when I was younger and working. I don't imagine that you are exactly loaning money to the Rockefellers yourself (has Bill Gates ever hit you up to help him through a cash flow crisis?). But I think it's important that we still pony up a few bucks to assist Mr. Bayer with a project that is certainly dear to our interests. If we both sent him ten times what we've already offered, would it even pay for the film he's already acquired, let alone the processing it will cost him? He has certainly dug deep in his own pocket, and I thank him for acting in our interests. </p>

<p>Tom Burke </p>

http://www.aphog.de/forum/viewtopic.php?f=4&t=9230 (in German)

 

Test conditions: Zeiss ZF Macro Planar 50 mm f/2 af f/4 and f/5,6 ; resolution test chart with 1 : 16 contrast, examination with a microscope.

 

Kodak Ektar 100: 100 - 105 line pairs/mm;

Fujicolor Reala: 115 LP/mm;

Fujicolor C200/Superia 200: 120 LP/mm;

 

Kodak Ektachrome 100G, Fujichrome Provia 100F, Sensia 100, Astia 100F: 130 LP/mm;

Fuji Velvia 100F: 140 LP/mm;

Fuji Provia 400X: 110 LP/mm;

 

Kodachrome: unfortunately not tested - special thanks to Mr. Henning Serger and colleagues for all the test work

<p>I can't guess at resolution numbers because I usually ignored them. The very best measure of sharpness is a signal-to-noise ratio where the signal is the MTF curve and the noise is the Weiner power spectrum (the amount of noise across the range of spatial frequencies). This measure not only correlates with subjective impressions of sharpness, it also correlates with information content of the film. As satisfying is this measure is, the Weiner power spectrum is difficult to measure. Film builders also want a measure of sharpness that is not affected by noise (graininess). For this purpose, a weighted average of the area under the MTF curve works best. There are lots of variations on this idea. The simplest one (and the one that is published) is the System Quality Factor (SQF). To get this, you take the MTF curves for all of the elements in the system (film, lenses, etc.) and multiply them together. Then calculate the area under the resulting curve for the "frequencies of interest". For a target held 14 inches away, the frequencies of interest center around 1 cycle/mm. For 35mm film printed to a 4x6 print, the enlargement is 4X so the critical frequency on the film is 4 cycles/mm. The standard practice is to cover the range from half to double the critical frequency. This is 2 to 8 cycles/mm for a 4x6 print and 4 to 16 cycles/mm for an 8x12 print. With larger prints, the viewing distance increases so the magnification has to be adjusted.</p>

<p>So much for the technical jargon. I don't have a good MTF comparison of Ektar 100 and K-25. The best I can do is a picture comparison of Ektar 100, Ektar 25, and K-64. Pictures are what matter anyway. The Ektar 100 has slighlty finer grain than Ektar 25. The sharpness is close, but not quite as good. The sharpness of K-64 is a notch lower. We know that K-25 sharpness was better than K-64 so it is probably similar to Ektar 100. </p>

<p>My comparison shots are posted at: <a href="http://homepage.mac.com/randrews4/PhotoAlbum27.html">http://homepage.mac.com/randrews4/PhotoAlbum27.html</a></p>

<p>These are 2700 ppi scans. I matched the gray scales, but did no other manipulation. </p>

 

<p>Mr. Anderle....</p>

<p>Always a pleasure to hear from a newer voice. It is especially nice when that voice is as informative and illustrative as yours. Comparing Ektar 100 with Velvia 50, and only by eyeball without using a chart, I had guessed that Ektar 100 would be around 100, which matches the results you pointed out above. </p>

<p>Regarding your answer about another release of Kodachrome, I too had heard that after the previous packaging of Kodachrome, there was still some left on the big roll to be cut. I was wondering if they had cut and packaged all of it for this last release, or perhaps had left some in the vault to be cut and packaged next year, perhaps to make the 75th official. I had hoped there was some left for any reasoning, and thought it might make good marketing to hold back some, with all the hoopla about the 75th. If in fact they didn't, they missed a good publicity opportunity. I thank you for confirming my suspicions with a reference to some science. </p>

<p>I look forward to seeing further comments from you on various topics here on Photonet. </p>

<p>Mr. Andrews....</p>

<p>As usual, you're right on with your comments. For looks, certainly SQFs and MTFs are the most important qualities in general. I was just curious as to an actual figure. Mr. Anderle's research provided that answer. I had seen your referenced shots and had downloaded the door shots onto my main computer, which of course is in the shop today. The little screen on this laptop doesn't do any of it justice. I re-downloaded the images to a thumb drive and am anxious to see them on a good screen Thursday when I get my computer back. As always, it's good to hear from a genuine expert. </p>

<p>A. T. Burke </p>

<blockquote>

<p>Neither Kodachrome nor Ektachrome exhibited good resolution when projected or printed</p>

 

</blockquote>

<p>I have to say I this is a somewhat problematic statement. The reason is that it sounds so negative. Many slide users would state that nothing really beats projected Kodachrome when compared to all other slide films. This is largely due to its color, saturation and high contrast rather than its resolution per se, but your statement seems to imply that somehow it looks inferior to other slides when projected - in my experience this is just not true, if anything the reverse. One can argue that projection is an imperfect medium but with a good Leica Pradovit projector for example then I am quite able to see that in practical terms it does not suffer in any way from a lack of apparent resolution. I do agree that in testing conditions its RMS rating is inferior to Velvia/Astia/Provia, but its apparent sharpness at least for projection exceeds these and one can make a good case that apparent sharpness is what usually "counts".</p>

The testing of the various films and its publication on the German "Aphog" (analog photo group) forum has been done by Mr. Henning Serger and his colleagues - I have just quoted the report and posted the link. My own comparisons are limited to my film scanner's resolution, i. e. 70 - 80 LP/mm with the Minolta Dimage Elite (I) 5400 (if it gets the focus right). A Nikon Scanner with 4000 dpi nominal resolution would as well come close to 70 LP/mm, as tests in computer magazines have shown.

<p>Mr. Anderle....</p>

<p>I have a soda glass slide from Edmund Scientific showing one expression of the USAF1951 resolution test chart. My Minolta Dimage Elite II tests out between 102 and 114 lp/mm. 5400 ppi would equal 106.3. Of course one must try scanning several times because registration is a hit or miss random problem at that resolution. Both my Nikons, the 5000 and 9000, do between 71.8 and 80.6 lp/mm, again after several tries to get the resolution right by random selection. 4000 ppi would equal 78.7 lp/mm. My conclusion is that all those scanners are close to their rated resolution, where all of the flatbeds that I've tried, not only don't get as crisp, but are usually closer to half their advertised resolution. My Nikon IV, rated at 2900 ppi, does between 50.8 and 57.0 lp/mm. Mathematical translation of 2900 ppi would be 57 lp/mm, which I did not get crisply and was barely able to discern, but with that scanner, I ran out of patience trying to get resolution right. I suspect with more patience on my part, it would come up to its rated resolution. </p>

<p>When using a flatbed scanner, the results appear to the eye to be above half their rated resolution, which is better than they test out at with the soda glass slide. I think the software interpretation may do better with random grain and dye clumps in film than it does respectively with a pure test chart. I do not feel that my single expression USAF1951 test card is accurate enough to properly delineate reflective resolution with the flatbed scanners, but it is my impression that the flatbeds do better with reflective material than with film. The last are my impressions, not good science. </p>

<p>A. T. Burke </p>

<p>CORRECTION ********</p>

<p>The sentence in my post above "When using a flatbed scanner, the results appear to the eye to be above half their rated resolution, which is better than they test out at with the soda glass slide." should read: </p>

<p>When using a flatbed scanner IN REFLECTIVE MODE, the results appear to the eye to be above half their rated resolution, which is better than they test out at with the soda glass slide. </p>

<p>A. T. Burke</p>