hm

I have a Fuji GSW 690 III with a winder problem. When loading a new roll of film, I wind, shutter, wind, shutter,

then, wind, wind, wind, wind, wind, wind, ... until the whole roll of film has wound through.

I sent the camera off to one of the Fuji "approved" repair companies, Precision Camera in Connecticut. They held

on to it for about a month, then sent it back with a note that the parts needed for the repair were not available.

So what now? Am I stuck with a boat anchor? Are parts really unavailable for this model? Is there a camera "junk

yard" from which parts can be culled from old "wreckers"? Is there somebody else I should have sent the camera

to, instead?

Any advice?

-harry

Unless you're a resolution chart portraitist, then you probably don't really care about the answer to this question.

Your real question is: "how does a big-ass print from whatever digital camera I'm thinking about buying compare to a big-ass print from whatever film system I'm currently using, given my target application and techniques?"

The answer to this question depends on what film system you're currently using, what digital camera you're currently considering, how big you consider "big-ass", and your own subjective tastes, as digital and film do "look different". If you're seriously in search of this answer, then you'll need to try to seek directly for that end result, instead of trying to extrapolate the answer mathematically from an intermediate value determined experimentally (or theoretically).

By the way, as in most venues, the loudest people here are the least reliable, so don't try to glean the response via consensus.

-harry

What should think of camera shake as having a relationship with "angle of view" instead of "focal length". The "1/f" rule of thumb is really just a convenient coincidence for 35mm, if you want to adapt it to other formats (either smaller formats like digicams or larger formats like MF), it's simple enough to just use the "35mm 'equivalent' focal length" and continue to apply a 1/f rule of thumb. All that said, no two pairs of hands are equally shaky, nor is a given pair consistently shaky every day (e.g. consider your recent caffeine intake!), so do consider that one camera may be more or less shaky in your hands due to physical considerations.

A depth of field calculator needs to know your distance, aperture, true focal length (i.e. 7mm), enlargement factor (between sensor/film and print), and required target on-print resolution. These last two elements are typically bundled up into a "circle of confusion" parameter. As you move from one format to another, your enlargement factor changes (assuming your print size remains constant), and so you need to adapt your CoC. These changes are linear with respect to format, so with the Digilux2 you should use a CoC that's 1/4 the CoC you might use on 35mm film, as the sensor is 1/4 the width of a frame of 35mm film.

If you're not using a calculator or chart, but just want to relate the Digilux 2 DOF to your "experience" with 35mm film, apply a 4x factor to the aperture. For example, given comparable subject distance, and a 1/4 true focal length on the Digilux2 to provide comparable subject framing, the Digilux at f/2 will give comparable DOF as a 35mm film camera at f/8.

"How do you archive a digital file. Kodak says put it on film."

Duh. And Sony says you should archive them on CD-Rs, and Maxtor says you should archive them on hard drives.

The Digilux2 sensor is about 1/4 the width of a 35mm frame of film, so to get an estimate of "comparable" depth of field, apply a 4x multiple to the f-number. In other words, the 28-90 (equivalent) zoom lens with f/2-f/2.4 aperture will yield depth of field comparable to 28-90mm on 35mm film with an aperture of f/8 (wide) - f/9.6 (telephoto) .

Note that the sensor has a 4:3 aspect ratio, so you can't directly compare the two without picking one to crop.

<em>"Leica is just buying it from Silicon Film, what risk?"</em>

<p>

The specs for the SiliconFilm sensor and the Leica sensor are very different, though they are both 10MP. The SiliconFilm sensor is very nearly full frame (something like a 1.1x "crop factor"), while the Leica sensor is much smaller, nearly a 1.4x crop factor.

<p>

Any way, Leica was quite clear that they're doing this in partnership with Kodak and Imacon. Presumably, virtually all of the electronics and firmware work is being done by the partners.

Free market dynamics haven't ceased to exist, so there will continue to exist software tools that handle JPG for as long as there is financial gain to be had, i.e. for as long as it seems important to a large enough number of people. Let's not confuse the most popular image format on the planet with an obscure vendor-proprietary RAW format from a super-high-end (price-wise) professional camera from years past.

As far as Mom and Dad saving the baby pictures, it seems clear to me that the answer here is that image archiving becomes a low-cost service (e.g. you keep all your pictures in your AOL account). It's the job of the service provider to ensure that their survival isn't dependent on a CD not going bad, and ultimately their responsibility to migrate those images into new formats as the old ones become obsolete. People will want this service anyway as a mechanism for sharing their pictures with friends and family.

Chip, I'm not sure what you mean by a workaround, but there are strong cryptographic techniques that can reliably authenticate an image as being one that came straight from the camera. A change to a single pixel in the image would be detectable. Obviously, this requires special in-camera support, like the Canon stuff mentioned earlier.

<p>

But if you consider that I can make slides from my Photoshopped digital images, and consider that I can similarly make digital in-camera repros of digital images, then the fact that an image was captured in-camera doesn't necessarily mean that the image is an authentic capture of "the scene", it could just be a capture of the manipulated image. Perhaps an expert could detect this repro, though?

From a technological standpoint, if the goal is to certify that a given image is identical to what the camera originally captured, that technology is possible with digital by having the camera apply a digital "signature" to the image it produces, putting it on an equal level with film in that regard.

<p>

But, of course, whether film or digital, just proving that this image was captured by a sensor or frame of film doesn't prove its authenticity, as you can always "create" an image, and have that doctored image captured (e.g. like dup'ing a slide).

<p>

I'm no lawyer, but ultimately it seems that evidence is as trustworthy as the testimony of the person who vouches for it.

<em>"Digital is: (a) sharp through..."</em>

<p>

Digital point 'n' shoots have very small-format sensors, leading to huge depth of field. They also tend to over-sharpen in their in-camera processing. This isn't an issue with digital itself, the high-end digital SLRs don't suffer from the same limitations.

<p>

<em>More and better films...</em>

<p>

I think that film will continue to be produced for years to come, but I don't see anybody investing significant R&D dollars into it, so I can't see what basis there could possibly be for predicting "more and better", other than wishful thinking.

<p>

<em>Have you taken into consideration...</em>

<p>

5 billion people in the world don't have darkrooms, either, yet somehow they figure out how to get prints from their film cameras.

<em>I beleive that just as mechanical watches and vinyl lps came rebounded, film won't go away</em>

<p>

You can believe anything if you want to believe hard enough, though that don't make it so.

<p>

I don't think film will go away completely any time soon, but you will slowly see companies drop out of the business, the variety will slowly diminish, prices will slowly climb, availability will slowly become more difficult, and processing less convenient.

<p>

I think the worst-case scenario for 20 years from now is that you'll only have a few choices, you'll order and process it via mail-order, and you'll pay about 2 or 3 times what you do today for both film and processing (not considering inflation). Oh yeah, and for all this extra effort and expense you'll be getting inferior results from what digital will offer you at the time, so you'll just be an irrational old codger clinging to the past out of nostalgia.

<p>

If your concern is less along the lines of "will I still be able to use my Leicas", and more along the lines of "will the thousands of dollars I sank into my Leica kit slowly evaporate in value", then you may have a more pressing concern over the next 5 years or so, unless your stuff is in pristine superficial condition and will interest fondlers.

<p>

-harry

<em>20% penetration of digital cameras leaves 80% for film</em>

<p>

There was presumably some year in the mid 80's where 20% of music sales was from CDs. That left 80% for LPs and tapes. So what?

<p>

The important question is "well, what was it the year before that? And the year before that?" Given that trend, what will it be next year, and the year after that?

<em>Yes, people will and do pay $3,000+ for turntables which ARE available.</em>

<p>

Sigh. The point is, given "supply and demand", what happens to the resale value of an item when 95% of the people who use that item transition to something newer? What's going to happen to the value of our M bodies that can only use film, and our M lenses that can only be used with bodies that use film, when 95% of us transition to digital? The argument is "digital sucks, they don't hold value like Leicas do", but, ironically, the value of Leica gear may be dependent on there being a digital future for it.

<p>

<em>... but while driving from your nice house to your office look up from the expressway at the millions of lower middle class houses along the way. How many of them have digital cameras and computers?</em>

<p>

If I'm living in a rich neighborhood, I wish my rich neighbors would stop breaking into my Jeep to steal my loose change.

<p>

Digital cameras need computers like disposable film cameras need darkrooms. The question isn't whether "lower middle class" people are going to go out and buy $500 digital cameras, it's whether their next cell phone (which they do have) is going to come with a camera built-in.

<em>Most Leica users spend a load of money on bodies because they know at least they will last for thirty years plus; and maintian a highish second hand value.</em>

<p>

Does anybody think that their Leica is going to maintain value over the _next_ 30 years?

<p>

A $3000 turntable bought in 1970 was probably still fairly valuable in 1980. But by 1990, after the advent of CDs, how many people were still interested in investing $3000 in a turntable?

<p>

We are on the cusp of a transition here, and it's worth considering whether we're making predictions of the future based on an assumption that it will continue the patterns of the past.

Every individual piece of recording media has a limited lifetime. It's not a question of "if", but "when" it will fail. Don't conclude from this that digital _data_ must thus have a limited lifetime, it can live as long as someone is interested in taking care of it. Part of this process of taking care of it involves keeping redundant copies, periodically verifying integrity, migrating from old media to new media, migrating from old formats to new formats, recovering (from backups) when a piece of media fails.

Consider the difference between the lifespan of a book, as in a bunch of pages bound together, and the lifespan of the contents of those pages, the words that make up that book. The words can be preserved and presented in more current media and formats. A piece of literature written thousands of years ago can still live today, having inhabited a variety of "media" (different generations of printed pages), and "file formats" (having been translated over the years), without any trace of the original media surviving.

A digital camera requires the "shooter" to have a computer to the same extent that a film camera requires the shooter to have a darkroom.

The "drop the film off, pick up the prints an hour later" crowd can do this just as easily with digital. It probably won't be long before the notion of waiting an hour to get photos developed seems like too long, and getting film developed will look a lot like a trip to the ATM (stick the card in one slot, pictures are spit out the other slot).

Hard-drive storage costs about $1/gig, thus redundant hard-drive storage costs about $2/gig. My current 6MP RAW files are about 8MB, so the cost of storing two copies of each image is about 1.6 cents. Assume that a hard drive needs replacement every 5 years, and that the cost of storage doesn't drop over time (a VERY silly assumption), then redundant storage of an image for 30 years costs about 10 cents. How much does film cost as a "storage medium"?

The fact that most people today are saving data to CDs doesn't mean that it's the right approach. Since you asked, my numbers put the cost of saving 600MB for 30 years at $10. There are no giant RAID servers or salt mines involved, just plain, old IDE hard drives hanging off of plain, old PCs, preferably in two different places.

While this data is stored, it's integrity is periodically verified, to ensure it continues to be identical to its original (i.e. we compare current file to a saved checksum taken from the original when it was first created). Since the data is online all the time, this is a trivial, automated operation done by very mundane software (again, no salt mines RAID servers). This is the part of the equation that's lacking from the "burn a CD" approach, it's difficult to do regular verification of integrity on removable media, as every little piece of media requires handling by a human.

The life-span of any single piece of data recording equipment or media is thus irrelevant. If a single byte of data was degraded, it would promptly be detected, and you would revert to the redundant copy. If a hard drive dies, you promptly replace it and re-populate the new one from the redundant copy.

And the most important thing: after 30 years your image is exactly the same as it was 30 years ago. No fading, no scratches.

Dear silly people, please don't confuse the lifetimes of removable storage media formats and devices, or the failure rate of internal storage devices, with the "lifespan" of digital data.

There are simple ways to preserve digital data indefinitely. Burning a CD and forgetting it is not one of these methods.

Can digital images last for 30 years? If you get a 30 year mortgage, will the mortgage company "remember" you owe them money for 30 years, or will it lose that information when 3.5" floppy disks are obsolete? Will the contents of your 401k disappear in 20 years when somebody notices that the DVD that held your information "went bad"?

No, of course not, because people don't use these silly methods for preserving important data. If data cannot be preserved for 30 years, then you have _far_ greater worries than just your photos.

Not without an LCD.

Today, in the real world, the cheapest 35mm full-frame sensor digital camera is around $5000. So your price may be optimistic.

Here's your new digital M from Konica:

<ul>

<a href="http://www.dpreview.com/news/0306/03060301konicakd510z.asp">

http://www.dpreview.com/news/0306/03060301konicakd510z.asp</a>

</ul>

Hmm, well, perhaps not...

Changing the enlargement factor does impact depth of field, it reduces it. But the real point is that if you crop the image (effectively creating a "smaller format"), but want to hold angle of view steady, you need an accompanying change to shorter focal length lens, which increases depth of field to a greater degree than enlargement decreases it.

So taking a shot with a 35mm focal length lens, but then cropping it to match the angle of view of a 50mm lens, only increases the enlargement factor, which decreases depth of field. If we then take a shot with a 50mm lens without changing our distance, and make a print from that frame without cropping, and compare our two prints (the 35mm cropped and 50mm uncropped), we'll see that we have identical perspective and framing, but the 35mm lens crop will display greater depth of field. In fact, the difference will be almost exactly one f-stop's worth of aperture.

This "cropping" is effectively creating a small format smaller than 35mm. The notion of "lens designed for 35mm" doesn't really have meaning in terms of depth of field, it just means that its image circle need be no greater than needed for a 36mm x 24mm frame.

Technically, this has nothing to do with digital, the same phenomenon is true for any format smaller than another format. These days "gobs of depth of field" is associated with digital point 'n' shoots because they have extremely small sensors, and thus are "really small format". The higher end DSLRs are fairly close to 35mm in size, so the effect is not so great, and cameras like the Canon 1Ds are the same size as 35mm, and so have identical depth of field.

If you grab your Leica and stand next to a guy with a 6x9 MF camera, and try to take the same composition of the same subject from the same distance, you'll find that you're using a much shorter focal length lens than he is, a natural result that different format sizes have on angle of view. Your shorter focal length lens will tend to yield far greater depth of field than his much longer lens, as depth of field increases exponentially as focal length drops. Now, to achieve the same on-print resolution, you'll need to enlarge much more than the MF guy will, so your CoC will need to be much smaller. This effect reduces depth of field for the 35mm camera, but only linearly, while the smaller focal length lens increases it exponentially, so the net effect is that the 35mm camera will see greater depth of field at a given aperture.

You'll see the same thing when comparing your Leica (or a 35mm full-frame digital) to a lower-end digital SLR with an APS-sized sensor, or when comparing that DSLR with a digital point 'n' shoot.