Kodak Ektar 100 - Is This a Bad Joke from Kodak?

[Les Sarile]

As you said, some pre and/or post work may have to be done and this is to be expected.
However, it seems most reviews won't take these extra steps and simply conclude falsely. Another typical response is their automatic scans with Film B comes out perfectly so why doesn't Film A? Or that their scan software has specific film profiles and still it doesn't come out automatically perfect. Well even profiles are not infallible. In the example below, I used the built-in profile for a specific film Kodak 160VC and selected the various modifiers in Vuescan and the Nikonscan "neutral" setting with my Coolscan 5000. I much prefer the latter as my starting off point.

Link to larger version -> Kodak 160VC

 

[hrst]

Posts #96 (Epson) and #101 (Vuescan) mostly bring back memories from two decades ago; it is unbelievable shit that today's computers and scanners produce results like that. It looks EXACTLY what it could have looked like if consumer film scanners were available in the beginning of 1990's with 16-color CGA/EGA displays! Like "WOW, we can see a shitty PHOTOGRAPH on a COMPUTER SCREEN that normally can only show some text, this is MAGIC!" It was magical in 1980's, now it is just awkward. Especially when the engineers have used EXTRA effort to break their scanners and software instead of just delivering the raw data with minimum processing required.

http://photo.net/classic-cameras-forum/00U9OU shows digital posterization artifact EXACTLY like seen with something like 32- to 64-color palette from 1980's. This is from 2009! This just shouldn't happen, just like nuclear plants just shouldn't blow up. It shouldn't be possible at all, but it happens.

That cyan/turquoise sky is exactly like in some PC games of 1980's/1990's, because the another blue from the 16-color palette was really dark blue, suitable for night sky, and hence they had to use cyan. (This shows that pure cyan is still closer to actual sky color than pure blue.)

Talking about film profiles, scanner profiles, monitor profiles etc. is just bringing more complexity into a system that is actually broken on a very fundamental level and cannot be fixed with those means. A correctly designed system does work without any profiling very well and profiling would just be a "finishing touch", and even then it always involves a risk of going wrong. Computers are not magic. Scanners just record light into numbers, computers store those numbers without changing them and monitors just convert numbers back to light. It would really be this easy but it has been complexified on purpose by idiotic "engineers", mostly software engineers, but HW engineers are not innocent either.

Why buy digital technology in 2012 if the part the user sees has not progressed from 1992 level at all? All it does is to perform 1 000 000 x more computation on the background for the same result but just slower, because computers are only 100 000x faster.

Why use products that resemble space ships in the number of controls (Vuescan), and every single knob turns the result into random horrible disaster?

My suggestion:
1) Get a color enlarger. You use only TWO knobs and both of them work perfectly and do just what they should do!
2) Get RA-4 paper. It is very cheap.
3) Get RA-4 chemicals. They are cheap and can last practically unlimited time even with low usage rates.
4) Enjoy.

I have found it is MORE QUICK to make RA-4 prints than scan and print digitally.

 

[Diapositivo]

My suggestion:
1) Get a color enlarger. You use only TWO knobs and both of them work perfectly and do just what they should do!
2) Get RA-4 paper. It is very cheap.
3) Get RA-4 chemicals. They are cheap and can last practically unlimited time even with low usage rates.
4) Enjoy.

The problem is to show to some other APUGers your pictures. Sooner or later you have to make a scan if you want to talk photography in an internet forum, even if the forum deals with analogue photography. Like it or not, hybrid is our language, our only possibility to talk photography in the four "corners" of the planet.

I don't understand the aversion in this forum for scanning. I understand other "hybrid" techniques might be out of topic, but the mere scanning really is the lingua franca, on the internet, of every analogue photographer. Scanners are what keeps us together as a community! Proper scanning technique do good, not bad, to analogue photographers because it allows them to spread the gospel in a convincing way. Ignorance of proper scanning techniques does bad to analogue photography because it spreads the idea that film photography can only output bad quality results.

Colour-aware people used Amiga since 1985

Anyway the progress from CGA days to today was just enormous. We now use 1280 - 1024 pixels with 32-bit depth per channel and especially a standardised colour management is now very easy implemented with home computers and really allows to exchange colour pictures in a forum like this one with a high degree of reliability, provided those few simple conditions are satisfied.

 

[Les Sarile]

To be sure, the examples I posted that are bad are not always the case in every frame scan. Believe it or not, some do come out reasonable. That of course just complicates matters even more because now they may not know which is the real good or the real bad. However, that is what may trip the casual user. Take the original poster, "I shot a couple rolls of this and I thought the cyan color cast was depressing". They're making judgements on very few samples and unfortunately an obviously negative experience.

Lingua franca indeed . . . how else can we share dialogue from Rome, Finland and me in Atlanta, GA!

 

[DREW WILEY]

It's amazing how many alleged film issues are really scan related or somewhere else in the digital workflow. The short story is, that the smaller the sample, the better the scan you need. This is because a small sample size (typically 35mm film using something less than a drum scan) will not
accurately reproduce nuances in film curve shape one dye layer relative to another. Add this to the
kinds of problems I noted earlier regarding improper exposure in the first place, and you're not going
to get ideal results, to put it mildly. Yeah, someone might be able to dither and paint and who knows
what else for days on end in Fauxtoshop, but it's a helluva lot easier to do things right in the first place. And why bother if it's just for the stupid hokey web anyway?

 

[nworth]

As hinted above, there a couple of surefire tests for whether a film is behaving properly: 1 - Can you make a good analog print (with an enlarger) of the negative? 2 - If you must scan, when using a reliable, calibrated scanner, can you correct any color shift using the controls provided in popular software like Photoshop? If the answer is no to either of these, you really need to find an answer. Since the major manufacturers (Kodak and Fuji) are very fussy about quality control, the problem almost certainly lies in either exposure, processing, or storage.

 

[wogster]

Posts #96 (Epson) and #101 (Vuescan) mostly bring back memories from two decades ago; it is unbelievable shit that today's computers and scanners produce results like that. It looks EXACTLY what it could have looked like if consumer film scanners were available in the beginning of 1990's with 16-color CGA/EGA displays! Like "WOW, we can see a shitty PHOTOGRAPH on a COMPUTER SCREEN that normally can only show some text, this is MAGIC!" It was magical in 1980's, now it is just awkward. Especially when the engineers have used EXTRA effort to break their scanners and software instead of just delivering the raw data with minimum processing required.

http://photo.net/classic-cameras-forum/00U9OU shows digital posterization artifact EXACTLY like seen with something like 32- to 64-color palette from 1980's. This is from 2009! This just shouldn't happen, just like nuclear plants just shouldn't blow up. It shouldn't be possible at all, but it happens.

That cyan/turquoise sky is exactly like in some PC games of 1980's/1990's, because the another blue from the 16-color palette was really dark blue, suitable for night sky, and hence they had to use cyan. (This shows that pure cyan is still closer to actual sky color than pure blue.)

Talking about film profiles, scanner profiles, monitor profiles etc. is just bringing more complexity into a system that is actually broken on a very fundamental level and cannot be fixed with those means. A correctly designed system does work without any profiling very well and profiling would just be a "finishing touch", and even then it always involves a risk of going wrong. Computers are not magic. Scanners just record light into numbers, computers store those numbers without changing them and monitors just convert numbers back to light. It would really be this easy but it has been complexified on purpose by idiotic "engineers", mostly software engineers, but HW engineers are not innocent either.

Why buy digital technology in 2012 if the part the user sees has not progressed from 1992 level at all? All it does is to perform 1 000 000 x more computation on the background for the same result but just slower, because computers are only 100 000x faster.

Why use products that resemble space ships in the number of controls (Vuescan), and every single knob turns the result into random horrible disaster?

My suggestion:
1) Get a color enlarger. You use only TWO knobs and both of them work perfectly and do just what they should do!
2) Get RA-4 paper. It is very cheap.
3) Get RA-4 chemicals. They are cheap and can last practically unlimited time even with low usage rates.
4) Enjoy.

I have found it is MORE QUICK to make RA-4 prints than scan and print digitally.

Scanners have improved quite a bit since 1980, I've had one since 1985 (replaced in 2005), and they have improved quite a bit, even for that one. What we forget is that no two films have the same colour response and the orange mask is not always the same either. Almost all scanner software does some processing of the image, this is often the problem, it makes assumptions about the film image in order to process it.

What I think would make the ultimate film scanner is to remove the assumptions, instead they include a special target, you shoot that target at the beginning of the roll, later on, you put that target image into the scanner, hit a button labelled setup, this then scans the image and sets the scanner up based on the differences between the scanned target and an internal digital copy of the target. That would remove the assumptions. I had an enlarger for many years, the problem is I didn't have a place to set it up, so I finally donated it to a school that teaches photography....

 

[TareqPhoto]

Here are some Ektar shots, some without adjustment and some with adjustments, and one converted to B&W, see which of those you prefer the most? Forget about calibration and accuracy or whatever, just if someone newbie like me don't know about calibration and profiling and just doing simple adjustments.

All taken on one roll, same day, same lighting condition, between shadows and open space, so how the exposure affect the color or that doesn't matter at all.

 

[Diapositivo]

What I think would make the ultimate film scanner is to remove the assumptions

You can "remove the assumptions" with any scanner if you use VueScan or SilverFast. With these two programs you can scan and save a "raw", i.e. a file which contains no "assumptions" but the actual result of the scanning. You can also have the "raw" file to include some basic corrections (dust, black point, white point, and have a profile attached to the raw file).

The nice of these scans is that you save the raw and then you can "develop" the raw with different settings without having to repeat the scan each time. I save my raw scans so that, years from now, I can work again on a picture, maybe with different programs, without having to scan it again.

 

[David A. Goldfarb]

Scanning discussion is simply off topic for APUG outside of whatever minimal discussion is necessary to explain how to post in the APUG galleries. Someone interested in printing optically shouldn't have to wade through all the scanning talk to find the few posts here on printing color neg with RA-4.

Please continue the scanning conversation at:

http://www.dpug.org/forums/f6/scanning-issues-kodak-ektar-100-a-2972/