Slashdot Mirror
Digital Camera Quality Passing Film?
smartbit writes "Luminous Landscape writes in their Preliminary Field Report of the Canon 1Ds 11 Megapixel camera: 'the 1Ds produces the best combination of resolution, colour accuracy and low noise that I've yet seen in a digital camera.
What about a comparison with both 35mm film and medium format? I'm afraid that film has definitively lost the battle. The 1Ds's full-frame 11MP CMOS sensor produces a 32MB file -- as big as a typical scan. But this file is sharper and more noise free than any scan I have ever seen, including drum scans. There simply isn't a contest any longer.'
Kodak's Pro 14n list price is $5000 lower and uses a similar CMOS sensor supplied by Fillfactory "
same website that claimed to be able to print neutral toned B/W from an Epson 1270 with dye inks! -- i'm not impressed with their reporting.. .. digital is great -- ..
Actually, that's not true. Film has a "grain" structure, caused by lumps of silver-halide. The grain is the limiting factor in film resolution.
Film certainly does not provide resolution at the "atomic" level.
The resolution of high-end consumer digital cameras now matches or exceeds that of typical consumer 35mm film.
The biggest advantage that film does have - it will continue to enjoy for some time to come - is dynamic range. You can't even come close with digital. No digital camera - even the most costly professional models - came come anywhere close to the dynamic range of consumer 35mm film and print material - let alone that in an Ansel Adams or Weston print. (And that was the film technology 50 years ago!)
http://clarkvision.com/imagedetail/film.vs.digital .1.html
Check out Roger Clark's analysis for the details.
Conventional photo film has a wider contrast range than any digital technology currently available. Photographers divide this range into 10 levels from total black to total white in measurable steps. Known as the Zone System of photography it is the entire basis Ansel Adams' entire body of work. Digital cameras using either CCD or CMOS chips simply do not have that kind of range. At best the high end cameras might have 7 or 8 zones, resulting in muddy shadows and blown out hightlights. In addition they are slow compared to film, requiring more light to make an exposure. Even though manufacturers might claim that the cameras have an effective ASA/ISO rating of 100 or 400, when compared to film, the digital cameras require a slower shutter speed or wider apeture to make an acceptable exposure. Just like MHz ratings in computers, Mega-Pixel ratings are just a part of the whole when measuring performance.
Comment removed based on user account deletion
> For one, all charge coupled devices (CCD and CMOS)
> with the exception of one camera (The Sigma SD9)
> use a pattern of red, green, and blue sensors, tiled.
I'm not sure what the Sigma uses. But Foveon has developed
a three layer CCD. The products using this CCD are
hardly affordable at the moment. But Canon is rumored
to also work on this. I'd say that those CCDs will be
standard in a few years.
That will take a while. The pixels would have to be able to be smaller than the grains in the finest grained film. I don't know much about colour film, but at least in black and white film, there are several films with grain so fine they can be enlarged to 8x10 or larger with NO grain visible to the naked eye. One way photographers often focus an enlarger is with a "Grain Focuser" which basically magnifies the grain of the image being projected onto the photographic paper. They then focus the enlarger until each grain is sharp. This is much more effective than focusing until, say, a sharp edge in the picture is sharp. Recently, I developed a roll of Fuji Neopan Acros 100. Although I did not dilute the developer when I developed the film, the grain was still so small I had difficulty focusing the enlarger, and this is with 35mm film. Remember there are still cameras around that use 8x10 FILM! An 8 inch by 10 inch ccd with resolution equal to or better than that of good black and white film would cost a fortune to manufacture, and purchase.
/usr/games/fortune
That said, the only time I use film any more is in extreme low light situations and even then I usually don't bother if I can get the equivalent with a long-exposure on my G2.
My Canon G2 Tests
-Matt
Make the switch, it's amazing what you can accomplish with digital - as long as you can think as both a photographer AND a geek.
In this sense Photoshop most certainly does count, and eliminates the "Flexibility in color response and grain" per film. You can adjust the grain to your liking, and get a full range of artistic manipulation with a much greater freedom than traditional paper. I've yet to find an effect or filter I can't reproduce in PhotoShop. It even compensates for some lenses, though I'd still keep those handy (as well as a good polarizer - it's much simpler than photoshopping it).
As for quality of the final print, why go photo printer? I've got one (fairly good quality, 2880x1440 dpi 6 chrome) for proof production, but the cost is beat by going to a good development place with a digital processor. Note: MANY DEVELOPERS NOW USE DIGITAL FOR STANDARD PROCESSING AS WELL. It's just easier, and the results are more consistent.
As for $10000 for a top-shelf camera, pick up a 5-6MP for under $2K unless you have do larger than 20x30 frequently, then wait 6 months and get a 10MP for the same price. Photoshop makes smooth interpolations across the board, really, so that may even be unnecessary.
Any spoon would be too big.
Now, I am tempted not to take this at face value, because there are good reasons why CCDs should essentially never have the dynamic range possible with film. (Essentially: film responds to light non-linearly, such that x photons hitting your camera does not equal the same amount of "brightness" on your image independent of how many previous photons have been registered. CCDs basiclaly are linear in response -- x photons equals x number of counts, modulo factors of gain, etc. -- up to the point where the number of photons registered is a significant fraction (like say 1/2) of the maximum well depth. Note that film is in this way more like your eye: an object that is twice as luminous does not look twice as bright to your eye, and you can simulaneously see things with your eyes that are many orders of magnitude apart in true brightness. To go even more off-topic in this comment: this is basically the reason why the most common stellar magnitude scale is defined logarithmically, where a difference of one magnitude corresponds to a factor of about 2.5 in brightness; it's an historical relic of the fact that when Hipparchos looked out at the stars, he called the brightest ones "1st magnitude" and some of the faintest ones "6th magnitude" ... and the latter turn out to be about 100 times dimmer than the former. Whew.)
Having said that, though, I don't actually have one of these things, and he doesn't really post any objective backup for his statements about dynamic range, so it's hard to prove or disprove them. He probably does know a hell of a lot more about photography than I do, so I'm sort of tempted to believe that they dynamic range issue is ceasing to be a problem, even if only by careful post-processing and choice of exposure. fwiw.
But that ain't a long-term solution. Perpetual admin in the only real way of insuring that my data stays safe, but I consider photos in particular to be important enough to deserve additional safeguards.
So, I have my photos printed (about 40 cents a shot). If they're really good, I send duplicates to my mom, who keeps them in a drawer with the mararoni art I did when I was 3. Pow, I figure I'm at least as safe as film now...
Every year during my review, I just pray the words "slashdot.org" aren't mentioned.
Unfortunately I haven't had a chance myself to carefully compare the results of digital and analog photography for the same subject and light settings. Does anyone know where one can find such comparisons? What do the pro's here think about those two issues: color and contrast ratio?
One: I have yet to see a digital camera beat film in low light level situations. Yeah, the Sony F707 is incredible in low light, but a high quality ISO 1000 film is better. And where long exposure times are okay, ISO 100 film with really long exposures is many times better than the F707.
Although, I think the point of this article is that the new 11MP cameras ARE as good as film.. When the end format is digital. I won't argue that nothing beats film for a print. But if the final destination is digital (or, heck, if it needs to be digital at any point) then these new cameras are better.
Yes, if you're a professional photographer that never digitizes your film, then yes, you have no reason to go digital. But if you're a news reporter, who needs pictures to go digitally to the publisher, then you can finally stop using film. (At least, since I haven't seen the output myself, that's what I'm taking this article to be saying.)
Another non-functioning site was "uncertainty.microsoft.com."
The purpose of that site was not known.
Large DOF has nothing to do with digital vs. analog and everything to do with the fact that the focal length in most consumer digitals is so short that the hyperfocal distance becomes very short. A random shot taken with a DC280 selected from my archive says 6.3mm at f/4, putting the HFD at 1.3 feet and everything from about eight inches to infinity in focus.
I found the excess DOF in my DC280 annoying enough that I sprung for a DSLR, and when I want that much DOF, I can have it.
I'm no photographer but by your numbers 35mm film would have 8.4 million pixels, not 34 million. (24)x(100)x(35)x(100) = 8,400,000
The real comparison is effective resolution and noise, and on these counts it should be clear that the new Canon 1Ds is comparable to 35mm in every meaningful way. At 11x17 print sizes, uncropped shots from the 1Ds will equal in sharpness all but the most carefully handled 35mm negs on top quality gear, and in noise there is no comparison at all. Heavy cropping _might_ still be easier on film, but even that doesn't appear to be the case.
Of course, this is still pretty serious $$$$, but even the D60 is very close to 35 in basically every way. 18-24 months? Guess what - that's when digital will start to rival 645, not 35mm. (of course, that'll be in an MF body, and probably cost $10k++)
Even with a 3 megapixel camera, I frequently have to wait for the CF card to finish storing the data before I can take another picture. Yes, many cameras have a "burst" mode, with lots of internal RAM to hold the images so that they can be written out to the card later, but even then, there is a limit to how much that write-behind caching can do for you. At some point, the RAM fills up, and you have to wait for the flash card to catch up. With a good 35mm SLR autowinder, you can snap several pictures per second until you run out of film, with no waiting.
Free Hans!
> I suspect that it will be a long time before medium or large format photography will have any digital equivalent
:-)
Well, if you have some spare cash to shell out (around 27'000 US$ - and that's just the camera back), there is this baby which looks like a serious contender in the medium format category and delivers whopping 510 MByte TIFFs in 16-Shot mode.
Just in case you're now thinking that the price is a little high: the Peltier element used to cool the imaging sensor and the Firewire port are included in the base price
"There are already a million monkeys on a million typewriters, and Usenet is NOTHING like Shakespeare." - Blair Houghton
b) You won't get anything usable from scanning 35 mm film beyond around 4000 x 3000 anyway, with most film stocks - the grain overwhelms the pixel size.
c) The Canon 1Ds (and Kodak 14n) have 12 bit sensors, which gives a dynamic range of 1:4096.
d) The Kodak DCS 14n is built with a standard Nikon SLR lens mount. The Canon EOS 1Ds is compatible with over 60 of Canon's EF lenses.
OK, a decent SLR is a lot cheaper, but it doesn't have any of the advantages a digital camera gives you, either.
Why would anyone engrave "Elbereth"?
Unfortunately, you're looking at the theoretical maximums for film. The reality of most shots is far different. The contrast range isn't an issue for 99.9% of the population either, since prints only have a contrast range of around 100:1.
The biggest difference right now is color accuracy. Until the Foveon chip reaches 11 Megapixels, we won't have anywhere NEAR the color range of film.
For further reading: here's and excellent summary of the topic.
ummm. I think you got it backwards.
....
How would it be $7.46 per gig, if the DVD-R is 63 cents?
That's 63 cents for 4.7 gigs.
That's (roughly) 13 cents per gig.
Karnal
Funny you mention 5.25" disks. I just started a huge backup of 5.25" disks from as early as 1984 or so. Know how easy it was? Just plug the 5.25" drive into the standard floppy connector on my K7 motherboard and the standard power plug on the power supply. You probably wonder how much data was lost due to bit rot. A _very_ slight percentage. Out of perhaps 500 floppies, only 5-8 would be unreadable. And I'd say that those which are unreadable were because the disks were bad to begin with (infact, I remember having trouble reading the same disks 7+ years ago). With today's technology digital backup is a reality. My entire archive of 5.25" and 3.5" (old 720k) disks will fit on 1-3 CD-Rs easily. And later I will purchase a DVD writer, as everyone will, and toss those CD-Rs onto a backup DVD. The process continues and as long as someone is _caring_ for the material, then it will survive.
Archiving is a constant job. If you let paper photographs sit in a damp, dusty, etc. area then you WILL have problems. Same thing would be leaving 5.25" floppies sit around by magnets, etc. Improper care and treatment is the ONLY reason digital, or otherwise, archiving fails.
Dijkstra Considered Dead