Fuji GF670 first impressions

I just received my Fuji GF670 from Dirk Rösler at Japan Exposures. This is a folding medium-format rangefinder camera, an anachronism in many respects, but I regret not getting a G690 when they were still made and since this is a limited edition (apparently quite a popular one at that), I went ahead. I have not yet shot a roll, but here are my first impressions:

• The unfolding mechanism is a bit finnicky. You have to be careful to get the front standard aligned with the film plane. Once deployed it seems fairly stable. Folding it back is also quite tricky.
• The meter indicator LEDs and controls are very reminiscent of the Epson R-D1, not surprising since both are actually made by Cosina.
• The leaf shutter is amazingly quiet. It makes a Leica sound like a clunker in comparison.
• The camera is quite light for MF, it feels lighter than a R-D1 (even though it weighs nearly twice as much) and is not that much larger.
• It does not exude quality like the Fuji-manufactured TX-2 (Hasselblad XPan II).
• The rangefinder patch is bright and clear. The RF base length is very short as in a VC Bessa, and will probably not be as precise as a Leica, XPan or Zeiss Ikon.
• The film loading mechanism is very easy to use, and built as well as other Fuji MF cameras such as the G617.
• You have to remember to reset the lens to infinity focus in order to fold it.
• You get a choice of 6x6 and 6x7, 120 and 220.
• The optional case is a snug fit. I wish it included a belt loop.

In grand old techno-fetishistic tradition, I put up an unboxing gallery.

00:23 - permalink [Photo]

Large sensor compact cameras finally on the horizon

I have stated on the record that my dream camera is a digital Contax T3 with an APS-C size sensor (or larger). Sigma launched the DP1, the first large-sensor compact this year, but it is a flawed camera, very sluggish, with a slow f/4 lens, and its Foveon sensor tops out at ISO 800, making it in practice a less capable low-light camera than my Fuji F31fd.

A few weeks ago, Olympus and Panasonic announced the Micro Four Thirds specification, which would allow for interchangeable-lens compact cameras with a larger sensor than the nasty tiny and noisy ones used on most compacts. Unfortunately it seems the whole misguided Four Thirds effort is destined to flounder, just as APS did compared to 35mm, despite the undeniable convenience. The 18x13.5mm sensor size has almost half the area of an APS-C sensor and all Four Thirds camera made so far have predictably poor low-light performance.

In a much more promising development, Samsung announced today that since it is finding it very hard to dislodge Canon and Nikon from their top position in DSLRs or even make a dent, they are going to create an entire new segment of professional quality compact cameras using the same APS-C sensors as their DSLRs, and due for 2010. Samsung uses the Pentax lens mount for its DSLRs, and has a long established relationship with Schneider Kreuznach. Pentax makes some very nice pancake lenses that combine high optical quality with small size. The only other company is Olympus, but the 25mm f/2 is saddled with the aforementioned Four Thirds sensor with all the limitations that entails.

At the same time, Thom Hogan has echoed rumors of an APS-C size Coolpix compact from Nikon. It looks like the big camera manufacturers can no longer afford to ignore the pent-up demand for this category, as demonstrated by the brisk sales of the DP1 (No. 49 on Amazon's Digital SLR chart).

10:37 - permalink [Photo]

Glimpsed at Chicago O'Hare airport

10:09 - permalink [Photo]

M8, a missed opportunity

Last Saturday, I became the proud owner of a Leica M8. Then, a not-so-proud owner. As of yesterday, an ex-owner...

I returned it and sprung for an Epson R-D1 instead, saving almost 50% in the process. I had already previewed one at MacWorld SF two years ago.

Most people interested in a M8 know by now about its problems with sensitivity to near infrared, which manifests itself as a magenta cast in certain situations. There is a work-around (buy costly Heliopan or B+W IR filters for your lenses, although there are rumors Leica will provide two free filters), but many are legitimately angry at Leica for having rushed the M8 launch despite such a fairly obvious flaw. It's not an ideal situation but I could deal with it, as long as Leica stood behind its product and committed to a free upgrade to the corrected model once a definitive fix becomes available.

The straw that broke this particular camel's back was quality control, however, or the lack thereof. My M8 exhibited almost an entire column of dead pixels (the bottom 3/4 at x=2888). If you must, see this jpeg or the original DNG. This kind of flaw would be unacceptable in a sub-$1000 Canon or Nikon, it is simply outrageous in a camera as expensive as the M8.

The magenta cast is not an edge condition visible in limited conditions, by the way (Leica claims it only affects black synthetics under tungsten light), the photos I took last Sunday indoors in available light are completely unsalvageable, with a strong magenta cast everywhere that cannot be corrected by any amount of custom white balancing. Here is an example: JPEG, DNG.

Last, but not least, noise levels are excessive at ISO 1250, let alone 2500, with smearing in rows where bright highlights are present. Essentially, this camera as it stands today is utterly useless outside broad daylight conditions (I don't have an IR filter, so I can't comment on how effective they are). Of course, pretty much all cameras do reasonably well in daylight, even cheap and nasty point-and-shoots with too many megapixels crammed in a sensor too small. Rangefinders give you a two stop advantage due to the absence of mirror slap, but even with a Noctilux, the M8 has no edge over a Canon DSLR because of the noisy sensor. Then again, it is one of my rules of photographic thumb that Kodak stands for poor quality, and since they make the sensor in the M8, I should have expected the worst.

It's interesting to note how the reviews published so far managed not to mention any of these problems, which are completely obvious, even with the most cursory of inspections. In at least one case (Michael Reichmann of The Luminous Landscape), the reviewer found out about the IR issue, informed Leica about it but neglected to mention it in the review. This confirms me in my belief Phil Askey's reviews at DPReview are the only reliable online reviews of digital cameras.

Leicaphiles seem to be mostly in denial, or minimize the extent of the problem. I am as big a fan as any of Leica's optics and their rangefinder cameras, but the flaws in my M8 were so glaring I can't even begin to fathom the levels of cognitive dissonance required to sustain a positive opinion of this train wreck in the making.

All the reviews I have read so far have been raising hallelujahs and claiming the M8 feels like a real Leica M. It most certainly does not:

• The body feels much thicker than the MP, and is just as thick as the R-D1, in fact, despite not having a flippable LCD like the R-D1.
• The lightweight magnesium body does not have the same level of robustness as the R-D1, let alone a MP, and feels more like a CM. It's not even to the same grade as the original Digilux.
• The lens mount lock does not snap positively and reassuringly as it should, and the release button feels cheap compared to my MP or M6TTL.
• The shutter release is mushy and unpleasant. The shutter sound itself is a loud thunk followed by a noisy motorized re-cocking.
• Setting ISO is buried in a menu and you need even more keystrokes to change it than on a Rebel XT (the R-D1, in comparison, has a genuine knob to set it quickly with direct feedback).
• The rangefinder on mine was slightly misaligned vertically, something one can tolerate in a $300 Bessa, but certainly not in a M (to be fair, rangefinder patch vertical alignment is an endemic problem with the R-D1 as well).
• In another sign of sloppiness and poor quality control, the copy of Capture One LE included in the box was missing the serial number required to activate the program.

After using the R-D1 for a few hours, the superiority of the design over the M8 is readily apparent (with the sole exception of the taller body and short rangefinder base length):

• The R-D1 has perfectly acceptable ISO 800 and 1600, unlike the M8, making suitable for available light shooting.
• The LCD screen pivots and can be turned around to protect it from scratches (or resist the temptation of chimping).
• The viewfinder has an honest to goodness magnification of 1.0x like the original M3, not one that panders to jaded wide-angle junkies (I never shoot wider than 50mm and my MP is a 0.85x mag, so yes, I am biased)
• The power supply is a manageable size and even has a cord, unlike the bloated wall-wart type Leica supplies with the M8.
• The shutter speed dial goes in the traditional direction, not the M6TTL/M7 direction...

One bright light in this fiasco: Doug Thacker at Calumet Photo San Francisco (above) went well above the call of duty to help me with my purchases, all with unfailing good humor (he once sent me an email at 11PM to let me know of the IR sensitivity problem before they started receiving theirs). He even set one M8 aside for me even though I had cancelled my initial pre-order (they are in short supply and are reportedly going for over $6000 on eBay right now, so the opportunity costs are considerable). I think I will switch from B&H to Calumet for the bulk of my photo purchases in the future.

Update (2007-08-25):

I must be a glutton for punishment, as after reading Phil Askey's M8 review, remarkably thorough as usual, I decided to give it another chance and get one for my birthday. The first one I ordered (from Amazon) had a severely misaligned rangefinder - points at infinity would not coincide at all when the lens was at infinity focus. It had a low serial number, suggesting an early model with teething problems. Presumably Amazon does not sell that many, so I returned it and ordered another one from a place with much higher turnover, B&H. That one was a recent vintage (they have an orange sticker on the body cap), but its rangefinder was also misaligned, if not as severely.

In frustration, I went to my local Calumet and finally found one that focuses correctly. Wonder of wonders, it even seems like there are no dead pixels or highlight streaks. Conclusion: Leica's M8 quality control is still spotty, your best bet is to buy locally and test the rangefinder in the store itself.

13:47 - permalink [Photo]

Spare the strap, spoil the camera

There are many ways to carry a camera. Most are supplied with a neck strap (and there is a non-slip shoulder equivalent, the UPstrap). Wearing a camera around the neck gets tiresome really quickly, makes you look like a goofy tourist, and potentially attracts the undesirable attention of thieves and would-be muggers.

I usually carry my camera discreetly inside a shoulder bag. A regular bag, mind you, not one of those obesely over-padded camera bags that are so bulky as to preclude walking around with them. You still need something to secure the camera, prevent it from slipping from your grasp and falling onto the hard pavement.

For pocket cameras, the wrist strap usually supplied will do just fine. You can get a tighter fit by attaching a cord lock (Google comes up with a bewildering variety of them) and reduce the risk of the lanyard slipping off your wrist. For some reason, only Contax had the sense to supply lanyards with a built-in cord lock.

For larger cameras, you need a hand strap. They are very common with camcorders, but unfortunately, very few camera manufacturers think of offering them as an option, or even provide bottom eyelets to make attaching them convenient. You have to hunt for third-party accessories and attach them using the tripod screw mount at the bottom of the camera.

For some time, I have mounted a cheap Sunpak hand strap on my Rebel XT. It does the job, but the plastic tripod mount is flimsy and unscrews all to easily, and the vinyl is not very pleasant to the touch. Another issue is that it precludes the use of an Arca-Swiss type quick-release plate. About a year ago, I wrote to Acratech, the people who make my ballhead and the QR plate on my Rebel XT, to suggest they drill an eyelet in the plate to allow mounting a strap, but never got a reply back.

I recently found out that Markins, a Korean maker of fine photographic ballheads, apparently took a patent on the idea and sells leather hand straps to go with some of their QR plates. Despite the princely price, I immediately ordered a set.

You have to unwind the strap to thread it through the eyelets on the camera and the QR plate, and back through the leather knuckle guard. This is fiendishly difficult to do if you don't know the trick to it: wrap the tip of the strap in packing tape to produce a leader, and cut to a taper with scissors to ease insertion.

This strap works because the Rebel XT has a protruding hand grip. For a camera like the Leica MP, which does not have an ample grip (unless you attach an accessory grip), I use a sturdy strap liberated from my father's old 8mm movie camera.

If you don't have one of these lying around, you can always try one of Gordy Coale's wrist straps, or if they lack snob appeal, Artisan & Artist makes ridiculously fancy (and expensive) ones for Japanese Leica fetishists.

01:03 - permalink [Photo]

Trimming the fat from JPEGs

I use Adobe Photoshop CS2 on my Mac as my primary photo editor. Adobe recently announced that the Intel native port of Photoshop would have to wait for the next release CS3, tentatively scheduled for Spring 2007. This ridiculously long delay is a serious sticking point for Photoshop users, specially those who jumped on the MacBook Pro to finally get an Apple laptop with decent performance, as Photoshop under Rosetta emulation will run at G4 speeds or lower on the new machines.

This nonchalance is not a very smart move on Adobe's part, as it will certainly drive many to explore Apple's Aperture as an alternative, or be more receptive to newcomers like LightZone. I know Aperture and Photoshop are not fully equivalent, but Aperture does take care of a significant proportion of a digital photographer's needs, and combined with Apple's recent $200 price reduction for release 1.1, and their liberal license terms (you can install it on multiple machines as long as you are the only user of those copies, so you only need to buy a single license even if like me you have both a desktop and a laptop).

There is a disaffection for Adobe among artists of late. Their anti-competitive merger with Macromedia is leading to complacency. Adobe's CEO, Bruce Chizen, is also emphasizing corporate customers for the bloatware that is Acrobat as the focus for Adobe, and the demotion of graphics apps shows. Recent releases of Photoshop have been rather ho-hum, and it is starting to accrete the same kind of cruft as Acrobat (to paraphrase Borges, each release of it makes you regret the previous one). Hopefully Thomas Knoll can staunch this worrisome trend.

Adobe is touting its XMP metadata platform. XMP is derived from the obnoxious RDF format, a solution in search of a problem if there ever was one. RDF files are as far from human-readable as a XML-based format can get, and introduce considerable bloat. If Atom people had not taken the RDF cruft out of their syndication format, I would refuse to use it.

I always scan slides and negatives at maximal bit depth and resolution, back up the raw scans to a 1TB external disk array, then apply tonal corrections and spot dust. One bizarre side-effect of XMP is that if I take a 16-bit TIFF straight from the slide scanner, then apply curves and reduce it to 8 bits, somewhere in the XMP metadata that Photoshop "helpfully" embedded in the TIFF the bit depth is not updated and Bridge incorrectly shows the file as being 16-bit. The only way to find out is to open it (Photoshop will show the correct bit depth in the title bar) or look at the file size.

This bug is incredibly annoying, and the only work-around I have found so far is to run ImageMagick's convert utility with the -strip option to remove the offending XMP metadata. I did not pay the princely price for the full version of Photoshop to be required to use open-source software as a stop-gap in my workflow.

Photoshop will embed XMP metadata and other cruft in JPEG files if you use the "Save As..." command. In Photoshop 7, all that extra baggage actually triggered a bug in IE that would break its ability to display images. You have to use the "Save for Web..." command (actually a part of ImageReady) to save files in a usable form. Another example of poor fit-and-finish in Adobe's software: "Save for Web" will not automatically convert images in AdobeRGB or other color profiles to the Web's implied sRGB, so if you forget to do that as a previous step, the colors in the resulting image will be off.

"Save for Web" will also strip EXIF tags that are unnecessary baggage for web graphics (and can actually be a privacy threat). While researching the Fotonotes image annotation scheme, I opened one of my "Save for Web" JPEGs under a hex editor, and I was surprised to see literal strings like "Ducky" and "Adobe" (apparently the ImageReady developers have an obsession with rubber duckies). Photoshop is clearly still embedding some useless metadata in these files, even though it is not supposed to. The overhead corresponds to about 1-2%, which in most cases doesn't require more disk space because files use entire disk blocks, whether they are fully filled or not, but this will lead to increased network bandwidth utilization because packets (which do not have the block size constraints of disks) will have to be bigger than necessary.

I wrote jpegstrip.c, a short C program to strip out Photoshop's unnecessary tags, and other optional JPEG "markers" from JPEG files, like the optional "restart" markers that allow a JPEG decoder to recover if the data was corrupted — it's not really a file format's job to mitigate corruption, more TCP's or the filesystem's. The Independent JPEG Group's jpegtran -copy none actually increased the size of the test file I gave it, so it wasn't going to cut it. jpegstrip is crude and probably breaks in a number of situations (it is the result of a couple of hours' hacking and reading the bare minimum of the JPEG specification required to get it working). The user interface is also pretty crude: it takes an input file over standard input, spits out the stripped JPEG over standard output and diagnostics on standard error (configurable at compile time).

ormag ~/Projects/jpegstrip>gcc -O3 -Wall -o jpegstrip jpegstrip.c
ormag ~/Projects/jpegstrip>./jpegstrip < test.jpg > test_strip.jpg
in=2822 bytes, skipped=35 bytes, out=2787 bytes, saved 1.24%
ormag ~/Projects/jpegstrip>jpegtran -copy none test.jpg > test_jpegtran.jpg
ormag ~/Projects/jpegstrip>jpegtran -restart 1 test.jpg > test_restart.jpg
ormag ~/Projects/jpegstrip>gcc -O3 -Wall -DDEBUG=2 -o jpegstrip jpegstrip.c
ormag ~/Projects/jpegstrip>./jpegstrip < test_restart.jpg > test_restrip.jpg
skipped marker 0xffdd (4 bytes)
skipped restart marker 0xffd0 (2 bytes)
skipped restart marker 0xffd1 (2 bytes)
skipped restart marker 0xffd2 (2 bytes)
skipped restart marker 0xffd3 (2 bytes)
skipped restart marker 0xffd4 (2 bytes)
skipped restart marker 0xffd5 (2 bytes)
skipped restart marker 0xffd6 (2 bytes)
skipped restart marker 0xffd7 (2 bytes)
skipped restart marker 0xffd0 (2 bytes)
in=3168 bytes, skipped=24 bytes, out=3144 bytes, saved 0.76%
ormag ~/Projects/jpegstrip>ls -l *.jpg
-rw-r--r--   1 majid  majid  2822 Apr 22 23:17 test.jpg
-rw-r--r--   1 majid  majid  3131 Apr 22 23:26 test_jpegtran.jpg
-rw-r--r--   1 majid  majid  3168 Apr 22 23:26 test_restart.jpg
-rw-r--r--   1 majid  majid  3144 Apr 22 23:27 test_restrip.jpg
-rw-r--r--   1 majid  majid  2787 Apr 22 23:26 test_strip.jpg

Update (2006-04-24):

Reader "Kam" reports jhead offers JPEG stripping with the -purejpg option, and much much more. Jhead offers an option to strip mostly useless preview thumbnails, but it does not strip out restart markers.

23:36 - permalink [Macintosh, Photo, Soapbox]

Another one bites the dust

After a brief period of 100% digital shooting in 1999–2001, I went back to primarily shooting with film, both black & white and color slides. I process my B&W film at home but my apartment is too small for a darkroom to make prints, not do I have a room dark enough, so I rent time at a shared darkroom. I used to go to the Focus Gallery in Russian Hill, but when I called to book a slot about a month ago, the owner informed me he was shutting down his darkroom rental business and relocating. He did recommend a suitable replacement, which actually has nicer, brand new facilities, albeit in not as nice a neighborhood. Learning new equipment and procedures was still an annoyance

Color is much harder than B&W, and requires toxic chemicals. I shoot slides, which use the E-6 process, not the C-41 process for more common color negative film. For the last five years, I have been going to ChromeWorks, a Mom-and-Pop lab on Bryant Street, San Francisco's closest equivalent to New York's photo district. The only thing they did was E-6 film processing, and they did it exceedingly well, with superlative customer service and quite reasonable rates. When I went there today to hand them a roll for processing, I discovered they closed down two months ago, apparently a mere week after I last went there.

I ended up giving my roll to the NewLab, another pro lab a few blocks away, which is apparently the last E-6 lab in San Francisco (I had used their services before for color negative film, which I almost never use apart from the excellent Fuji Natura 1600).

Needless to say, these developments are not encouraging for a film enthusiast.

Update (2007-12-14):

There is at least one other E-6 lab in San Francisco, Fotodepo (1063 Market @ 7th). They cater mostly to Academy of Arts students and are not a pro lab by any means (I have never seen a more cluttered and untidy lab). In and in any case they are more expensive than the New Lab, if more conveniently located.

20:35 - permalink [Photo, San Francisco]

Shoebox review

For a very long time, the only reason I still used a Windows PC at home (apart from games, of course) was my reliance on IMatch. IMatch is a very powerful image cataloguing database program (a software category also known as Digital Asset Management), The thing that sets IMatch apart from most of its competition is its incredibly powerful category system, which essentially puts the full power of set theory at your fingertips.

Most other asset management programs either pay perfunctory attention to keywords, or require huge amounts of labor to set up, which is part of the cost of doing business for a stock photo agency, but not for an individual. The online photo sharing site Flickr popularized an equivalent system, tagging, which has the advantage of spanning multiple users (you will never be able to get many users to agree on a common classification schema for anything, tags are a reasonable compromise).

Unfortunately, IMatch is not available on the Mac. Canto Cumulus is cross-platform and has recently introduced something similar to IMatch's categories, but it is expensive, and has an obscenely slow image import process (it took more than 30 hours to process 5000 or so photos from my collection on my dual-2GHz PowerMac G5 with 5.5GB of RAM!). Even Aperture is not that slow... I managed to kludge a transfer from IMatch to Cumulus using IMatch's export functions and jury-rigging category import in Cumulus by reverse-engineering one of their data formats.

Cumulus is very clunky compared to IMatch (it does have the edge in some functions like client-server network capabilities for workgroups), and I had resigned myself to using it, until I stumbled upon Shoebox (thanks to Rui Carmo's Tao of Mac). Shoebox (no relation to Kodak's discontinued photo database bearing the same name) offers close to all the power of IMatch, with a much smoother and more usable interface to boot (IMatch is not particularly difficult if you limit yourself to its core functionality, but it does have a sometimes overwhelming array of menus and options).

Andrew Zamler-Carhart, the programmer behind Shoebox, is very responsive to customer feedback, just like Mario Westphal, the author of IMatch — he actually implemented a Cumulus importer just for me, so moving to it was a snap (and much faster than the initial import into Cumulus). That in itself is a good sign that there will always be a place in the software world for the individual programmer, even in the world of "shrinkwrap software", especially since the distribution efficiencies of the Internet have lowered the barrier to entry.

Shoebox is a Mac app through and through, with an attention to detail that shows. It makes excellent use of space, as on larger monitors like mine (click on the screen shot to see it at full resolution) or dual-monitor setups, and image categorization is both streamlined and productive. As an example, Shoebox fully supports using the keyboard to quickly classify images by typing the first few letters of a category name, with auto-completion, without requiring you to shift focus to a specific text box (this non-modal keyboard synergy is quite rare in the Macintosh world). It also has the ability to export categories to Spotlight keywords so your images can be searched by Spotlight. I won't describe the user interface, since Kavasoft has an excellent guided tour.

No application is perfect, and there are a few minor issues or missing features. Shoebox does not know how to deal with XMP, limiting possible synergies with Adobe Photoshop and the many other applications that support XMP like the upcoming Lightroom. It would also benefit from improved RAW support - my Canon Digital Rebel XT CR2 thumbnails are not auto-rotated, for instance, but the blame for that probably lies with Apple. The application icon somehow invariably reminds me of In-n-Out burgers. The earlier versions of Shoebox had some stability problems when I first experimented with them, but the last two have been quite solid.

I haven't started my own list of the top ten "must have" Macintosh applications, but Shoebox certainly makes the cut. If you are a Mac user and photographer, you owe it to yourself to try it and see how it can make your digital photo library emerge from chaos. I used to say IMatch was the best image database bar none, but nowadays I must add the qualification "for Windows", and Shoebox is the new king across all platforms.

17:59 - permalink [Photo]

Opening up Aperture

Apple introduced Aperture, its professional workflow management application at the Photo Plus trade show in New York on 2005-10-19. Initial speculation was that Apple had finally decided to bring its deteriorating relationship with Adobe to a head and release a Photoshop competitor. This was quickly dispelled and its true positioning as a high-end workflow application for professional digital photographers became more apparent. This is not just a fig leaf to appease Adobe — Aperture does indeed lack most of Photoshop (or even Elements') functionality. It is a Bridge-killer of sorts, not that the sluggish, resource-hogging piece of bugware that is Bridge could be really be qualified as "alive". The simplest description of Aperture is that it is iPhoto for professional photographers.

I received my copy today, and will be putting it through its paces in a series of longer articles over the coming weeks, with this article as the central nexus:

1. First impressions
2. Asset management
3. Under the hood: file format internals

00:08 - permalink [Photo]

Resisting camera bloat

I recently upgraded my DSLR from a Canon EOS 10D to a Digital Rebel XT. Thanks to the universal consumer electronics upgrade plan, the final cost ended up quite minimal.

Some may object, how can an entry-level $800 camera be considered an upgrade over an originally $1500 prosumer body with a magnesium shell, glass pentaprism and two control wheels? One word: plastics. More precisely, the weight reduction plastics can offer. I usually carry a professional-grade camera in my gadget bag, and the 10D never made the cut because it is so heavy. A camera that gathers dust at home is not all that useful, so off to eBay it went.

Certainly, the 10D feels better in the hand, its viewfinder is not a claustrophobic little tunnel (although compared to my other cameras like as a Leica MP, the 10D's viewfinder is barely less squinty than the Rebel XT's). The 8 megapixels vs. 6 are immaterial - they amount to only 15% improvement in linear resolution, and megapixels don't matter that much anyway.

Film cameras have the bulk of their body forming an empty cavity to load film into. DSLRs, on the other hand, are densely packed with electronics, making them surprisingly heavy for their size. The 10D weighs 790 grams, compared to 715g for a rugged Nikon F3, 600g for the solid brass MP, and 490g for the Rebel XT. The weight around your shoulders is very perceptible at the end of the day. You are not even getting that much more in build quality, the thin magnesium shell on the 10D is there more for cosmetic effect than any real structural purpose — I have not found the 10D appreciably better constructed than the plastic-shelled D30. It certainly cannot compare with the 1.4mm thick copper-silumin-aluminum alloy walls on the F3.

This brings me to a pet peeve about high-end cameras. It seems Canon and Nikon have decided that for marketing reasons a professional camera has to be a heavy camera. I could easily afford a 1D MkII, but don't feel like carting along a 1.2kg behemoth with all the quiet understated elegance of a Humvee. This camera weighs almost twice as much as a F3 or a MP, both of which are supremely robust professional bodies.

In the era of film, I could understand that an integral motor drive weighs less and is more reliable than an separate one (on the other hand, the film equivalent to the 1D, the EOS 1V, is available without the motor drive to cut down on weight). The bulk of the 1D MkII, and its Nikon equivalents the D2H and D2X, is taken by an oversized portrait grip with slots for heavy batteries.

For digital bodies, however, many of these design choices are unwarranted. The Canon 1D MkII and 1Ds MkII use CMOS sensors that do not require the bulky high-current NiMH battery pack necessary to power the original CCD EOS 1D. Unfortunately Canon have kept the ungainly form instead of adopting the approach, used in their amateur cameras, of providing an optional portrait grip with room for spare batteries for those who absolutely must have them, but not saddle all users with heft and cost they do not want or need. Nikon does no better, their pro cameras all exceed the 1 kilogram mark, as did their film F4 and F5 bodies (the new F6 is under a kilogram without batteries, however). Perhaps that is why the F3 was so enduringly popular compared to the F4. Galen Rowell certainly preferred the F100 over the F4, and the F4 over the F5

There used to be a time when quality and miniaturization went hand in hand. Oskar Barnack invented the lightweight Leica precisely because he was asthmatic and could not lug heavy glass plate view cameras while hiking. Yoshihisa Maitani is justly celebrated for his incredibly light Olympus OM system, accompanied by excellent compact lenses, some of which are still unmatched by Nikon or Canon. Many professional cameras were available in expensive titanium versions to shave a few precious grams. But it now seems that designing a pro camera involves embracing bulk and unnecessary weight, for the simple reason a heavy camera feels more solid and reliable when you handle it in the shop. What next, adding lead ballast? Perhaps lead is not dense enough and depleted uranium will soon be the camera steroid of choice.

I do not see this trend improving over time. I guess my next and probably final digital camera purchase will be a Leica M Digital or the Zeiss Ikon version when they finally arrive on the market. Rangefinder makers still understand ergonomics.

Update (2006-04-08):

It seems pros are doing the same, as in this post from one who downgraded from the Canon EOS 1Ds MkII to a 5D, and apparently it looks like his is not an isolated case. Perhaps Canon will get the message with free-falling sales of big and heavy cameras.

13:40 - permalink [Photo]

Traveling with film

X-rays can fog photographic film. The damage is cumulative, repeated passes will have an effect even on low-sensitivity film that would not suffer too much from a single pass. Most of the films I use are specialized professional emulsions not readily available in ordinary stores (or even in the USA, as with the three rolls of Fuji Natura 1600 I recently ordered from Japan). That is why I travel with film, and on general principle avoid having it passed through X-ray machines by requesting a hand inspection instead.

Grousing about the Transportation Security Administration (TSA) has almost become a national sport. I tend to disagree — TSA has set uniform standards and a measure of courtesy and customer-service orientation compared to the earlier hodgepodge of private contractors. It is much easier to travel with unexposed film in the US today than it used to be post-9/11, pre-TSA. In other countries, screeners will routinely ignore your protests and unceremoniously shove ISO 1600 film into a high-energy scanner, all but ensuring they have fried it.

These common-sense guidelines will ensure speedy processing and avoid aggravation, for yourself as well as for fellow passengers down the line:

• Unpack your film from its cardboard boxes and plastic canisters (or foil wrappers for 120 film), and put it in a transparent zip-lock bag (the ones with the easy-open plastic sliders are best).
• If you are (justifiably) afraid that film outside light-tight canisters may be fogged over time, carry your canisters separately (J&C photo sells inexpensive plastic canisters for 120 film, or you could splurge for the aluminum ones by Acratech). Probably a better option is to use a thick black plastic bag like the ones used to pack photographic paper.
• Pack at least a couple of ISO 1600 or higher rolls of film so you can ask for hand screening. Some TSA personnel will ask you if you have any film higher than ISO 400, having some on hand is simpler than haggling for hand-inspection of low-ISO film. Only once did I have TSA staff separate my higher than ISO 400 rolls for hand screening and pass the rest through the X-ray machine.
• Lead-lined, supposedly X-ray safe film pouches are a waste of money. Not only are they heavy and ungainly, but the X-ray operators will simply increase power when they see an opaque bag that could conceal a weapon or explosives. TSA policy allows you to ask for a hand inspection, just avail yourself of this.
• Pack your film in your carry-on luggage, preferably in an easily accessed compartment

For more authoritative statements, check out the official TSA and ITIP pages.

Digital photographers are not completely off the hook. While ordinary X-ray machines do not affect flash media, the newer high-energy machines currently considered for deployment can alter electronic media. Another interesting aspect is that of cosmic rays, high-energy particles from outer space and which can "flip" bits in a flash card or magnetic media. In a negative, the blip will be imperceptible, but digital files are less tolerant of corrupted bits. IBM has spent quite a bit of R&D effort in quantifying the problem. When you have several billions of bits in RAM or trillions of bits in a hard drive, even an unlikely event like an alpha particle becomes statistically probable, and they have taken special measures against this. Many commonplace materials are also naturally radioactive.

The effect of cosmic rays at ground level is limited, but airplanes flying at 10-25km altitude are much more exposed to them, by a factor of hundred or so — see Dr. Ziegler's article (PDF) on terrestrial cosmic rays. The impact of cosmic rays on flash and microdrive media is poorly understood today and the risk not fully assessed. Due to short-term financial thinking encouraged by Wall Street, few corporations invest today in the kind of R&D that could conclusively answer this question.

13:42 - permalink [Photo]

What to think of pocket digicams

Once you have used a digital SLR (DSLR) with a nice, clean, large, low-noise sensor, the poor image quality of most compact digicams becomes hard to tolerate. This is in contrast with film, where a $70 Olympus Stylus Epic can compete in image quality with thousand-dollar cameras.

Then it hit me: don't consider a pocket digicam as a camera, think of it as a pocket photocopier/scanner instead, like HP's ill-fated CapShare. I use my pocket digicam mostly to record specials in stores, flyers, magazine articles, diagrams on a whiteboard and the like. Japanese otaku teenagers are way ahead of me, as many bookstores in Tokyo now ban cameraphones because the kids would just snap photos of manga comic books and not pay.

A 5 megapixel digicam, pretty mainstream nowadays, with a 4:3 aspect ratio can "scan" a standard US Letter or A4 page at an approximate resolution of 240 spi. This is significantly better than a fax, which scans at 150 spi. Many pocket digicams have lenses that are serviceable in macro mode. The limiting factor is probably setting up the camera, as you can't find portable copy stand like the vintage Leica BOOWU (also shown top left in this outfit photo).

19:04 - permalink [Photo]

MacWorld SF roundup

I work a mere four blocks away from the Moscone Center, where the annual MacWorld SF trade show is held, so naturally I just drift there during my lunch break, possibly extended... Here is a list of strange and wonderful things I saw during the show, and that might have been overlooked by the more mainstream sites:

iLugger

The iLugger is a carrying case for the iMac G5 (it fits both the 17" and 20" models). Most laptops are always connected to the mains and seldom used as real mobile devices, and an iMac G5 will give significantly better performance at 2/3 the price of a PowerBook. Interestingly, the company making it is a blimp manufacturer, clearly a case of someone scratching their own itch.

Epson RD-1

Not a new product, but I got to handle an Epson R-D1, a limited edition Leica M compatible rangefinder digital camera (the only one of its kind) based on a Voigtländer-Cosina Bessa R2 body. I shot a few samples with a 50mm Summicron and Noctilux, and the resulting pictures are remarkable clear and sharp. Noise levels at ISO 800 are significantly better than my Canon EOS 10D, no small feat, and given a rangefinder's 2-3 stop advantage over a SLR, this looks like an ideal available-light camera.

The Bessa R2 has a relatively short rangefinder base length, which reduces its focusing accuracy compared to a Leica. The hardest lens to focus is the Noctilux-M 50mm f/1.0 (yes, you read that right, the fastest production lens in the world), due to its very shallow depth of field at low aperture, as shown in the picture to the left. I took it with a Noctilux (ISO 200, f/1.0, 1/125) at close to its minimum distance of 1 meter, and focusing accuracy seems adequate... Click on the image for the full-size JPEG with EXIF metadata (not including the manually set aperture and focus, of course). For comparison purposes, here is the corresponding JPEG I shot yesterday (ISO 800, Summicron-M 50mm f/2, 1/30, aperture unrecorded, probably f/4).

The gentleman portrayed is an Epson representative who was apparently given the charge of watching over this $3000 camera (apparently his only task). The sight of me pawing over it might explain his expression...

I won't duplicate Luminous Landscape's review, and didn't have that much time to play with the RD-1 in any case. Build quality is good, as good as the 10D at least. It does not have the satisfying heft of my Leica MP, nor its superlative 0.85x viewfinder, but then again what does? Some retro touches like the dials are an affectation, as well as the manually cocked shutter. The shutter cocking lever does not have to advance film, and its short travel feels somewhat odd.

X-Rite Pulse ColorElite

X-Rite, a maker of color calibration hardware, was demonstrating its Pulse ColorElite bundle, resulting from its acquisition of the color management software vendor Monaco Systems. This package allows you to calibrate with precision the color characteristics of a monitor, scanner, digital camera and printer, for consistent, professional-grade color management. It goes much beyond simple and now relatively inexpensive monitor calibration colorimeters, by also using a spectrophotometer (an instrument that measures light across the visible spectrum, wavelength by wavelength), and the price is correspondingly higher. The market-leading product is the GretagMacbeth Eye-One Photo. X-Rite has clearly replicated the Eye-One package, but at a slightly lower price, and with some nice touches that significantly improve usability. The Eye-One spectrophotometer (which is used both for calibrating monitors and prints, a GretagMacbeth patented technology) is reportedly more accurate, however (3nm vs. 20nm). The Pulse bundle retails for $1300, the Eye-One for $1400.

FrogPad

The FrogPad is a small one-handed Bluetooth keyboard designed to be used with PDAs or smartphones, but it can also be used with a Mac or PC as it follows the standard Bluetooth Human Interface Device (HID) profile. You can hold it in one hand and type with the other. I don't know how long it takes to get used to it, but at any rate they are offering $50 off the regular price of $179 if you use the code Apple50. They also has a mockup of a folding version in cloth, for use in wearable computing.

Interwrite Bluetooth tablet

CalComp used to make high-end tablets and digitizers for architects, engineers and artists. The tablet market is pretty much monopolized by Wacom, nowadays, but CalComp is still around (after being bought out by GTCO). They were demonstrating a Bluetooth tablet for use by teachers in a classroom setting (although I am not sure how many cash-strapped school districts can afford the $800 device).

JetPhoto

There was a cluster of small Chinese companies exhibiting. One of the more interesting was Atomix, a company that makes JetPhoto, a digital photo asset management database, similar to Canto Cumulus or Extensis Portfolio. Apparently, their forte is the integration of GPS metadata and the image database, you can do geographical selections on a map to find photos. It also had many export functions with a comprehensive database of cell phones and PDAs to export photos to. Unfortunately, the current version does not support sophisticated hierarchical, set-oriented categories, the one feature in IMatch I find absolutely vital.

The program looked impressively polished for a first version, and is available free to download for now. This is yet another illustration of how the Chinese are rapidly advancing up the value chain, and American firms could be in for a nasty surprise if they maintain the complacent belief high-end jobs are their birthright and only unqualified manufacturing jobs or menial IT tasks are vulnerable to Chinese (or Indian) competition.

Fujitsu ScanSnap

One of the few things I still use my Windows game console PC for is to drive my Canon DR-2080C document scanner. This small machine, the size of a compact fax machine, can scan to PDF 20 pages per minute (and it can scan both sides simultaneously). It is intended for corporate document management, but is also very useful to tame the paper tiger by batch-converting invoices, bills and so onto purely electronic form, in a way that is not practical using a flatbed scanner.

It seems Fujitsu is bringing that functionality to the Mac with the similarly specified ScanSnap fi-5110EOX. The scanner is driven with a bundled version of Adobe Acrobat 6.0. I can well see this becoming popular in small businesses run on Macs, although the Fujitsu reps on the stand implied they were here to gather potential customer feedback to make a stronger case for enhanced Mac support with their management and accelerate the release of Mac drivers for it.

Ovolab Phlink

My office PBX is actually a PC-based CTI unit made by Altigen, and voice mails left to me are automatically forwarded to me as WAV attachments in an email. That has major usability benefits - I can set email rules to drop voice mails when the attachment is too small (usually someone who hanged up on the voice mail prompt), or fast forward and rewind during message replay. This feature is addictive - voice mail still sucks compared to email (disk hogging, not searchable or quickly scannable), but being liberated from excruciatingly slow voice-driven user interfaces, replete with unnecessarily deliberate and verbose prompting, makes it somewhat bearable.

I did not have this kind of functionality at home, however. It is possible VoIP devices will offer it at some point, but that does not seem to be the case in low-end home VoIP for now. I tried experimenting with the open-source Asterisk PBX, but did not have the time to pursue this, and in any case I'd rather not have to install a dedicated Linux machine at home just for this purpose (my home network runs on Solaris/x86, thank you very much).

Fortunately, Ovolab, an Italian company based near Milan, has introduced Phlink, just what I was searching for, and I actually bought one on the spot. It is a small USB telephony attachment that plugs into a phone line and turns your Mac into a sophisticated CTI voice-mail system. It is fully scriptable using AppleScript and supports Caller ID. I have yet to use it extensively (the hardest part, interestingly, is bringing a phone cord close enough to my Mac).

00:37 - permalink [Macintosh, Photo]

Trigonometry for photographers, or not

The photography world learned yesterday the sad but not entirely unexpected news of Henri Cartier-Bresson's demise. Cartier-Bresson was 96 years old, and had prepared his legacy by setting up a retrospective and foundation in Paris. The catalog of the retrospective is one of the finest coffee-table books you can get, by the way. Cartier-Bresson is best known for his theory of the "decisive moment". Although some wags would say the decisive moment was really when he reviewed his contact sheets, Cartier-Bresson clearly perfected a technique of anticipating the event and being ready to capture it on film, helped in this by his Leica rangefinder cameras.

Cartier-Bresson was known for his caustic wit and his often provocative statements. In an interview to Le Monde, he derided the "academic clichés of Weston" (les poncifs académiques de Weston), referring no doubt to Edward Weston's still life studies of peppers. Someone using lightweight equipment like Cartier-Bresson has the luxury of spontaneity large-format photographers like Weston did not. Indeed, Brett Weston, Edward Weston's second son, quipped that "Anything more than 500 yards from the car just isn't photogenic" when working with a 8x10 view camera.

You don't have to carry a behemoth camera to realize the virtues of forward planning. When doing landscape photography, it is helpful to know ahead of time what kind of lens or camera to pack, and the position of the sun. There are many ephemeris tables online to find the latter, but the easiest way to select a lens is to use a map. You could use a protractor to measure angles, but they are relatively small and fiddly to use. As I often shoot with a Fuji G617 panoramic camera and a Hasselblad system, I made a series of translucent templates to help with this - all I need to do is superimpose them on the map (such as a 1:24,000 topographic map produced by a National Geographic map machine).

I wrote a quick program in Python and PostScript to produce templates in PDF format for various film formats and lens focal lengths, ready to print on a laser printer (I used Four Corners Paper IFR Vellum). I hope this will be useful. As an example, here is the template I use with my Fuji G617.

18:44 - permalink [Photo]

Loopy about loupes

I posted a longish comparative review on loupes, and why you get what you pay for.

22:17 - permalink [Photo]

Shutterbabe

Deborah Copaken Kogan

Random House, ISBN: 0375758682 Publisher, Amazon.com

I picked up the hardcover edition of this book from the sale bin at Stacey's Booksellers, as the Leica on the cover just beckoned to me.

This is an autobiography by an American woman, almost a girl, who moved to Paris, fresh out of college, to break into the tightly-knit (and not a little macho) community of photojournalists. Who knows, I might even have crossed paths with her when I studied in Paris. She was certainly not the first female war correspondent, Margaret Bourke-White springs to mind (even though she is not referred to anywhere in the book), but women were still a rarity, specially one as young and inexperienced. She started as a freelancer and eventually ended up working for the Gamma agency, one of the few independent photo agencies left.

For some unknown reason, many of the prestigious photo press agencies are based in Paris, starting with Magnum, founded simultaneously in Paris and New York by Robert Capa (the man who took the only photographs of D-Day), Henri Cartier-Bresson, George Rodger and Chim Seymour. Others like Gamma, Sygma and Sipa followed, but most have been acquired since by large media conglomerates like Bill Gates' Corbis. The move to digital, with the corresponding explosion in equipment costs is one reason - the independent agencies simply couldn't compete with wire services like Reuters or Agence France Presse (AFP), the latter being government-subsidized. Saturation is probably another, and press photographers struggle to make a living in a world with no shortage of wannabes. Just read the Digital Journalist if you are not convinced.

Shutterbabe is not a mere feminist screed, however. Engagingly written, with very candid (sometimes too candid) descriptions of the sexual hijinks and penurious squalor behind her trade, this book is a pleasurable read and features a varied rogues' gallery ranging from the cad (her first partner) to the tragically earnest (her classmate who is executed by Iraqi soldiers while covering Kurdish refugees). It only touches in passing on photographic technique, as the general public was clearly the intended audience, but more surprisingly, does not include that many of her photos either. The main thread reads like a coming of age story, with the young (25 year old at the time) woman moving on from her thrill-seeking ways and discovering true love and marriage in a life marked by death: deaths of friends and colleagues, victims of strife and war in Afghanistan or Russia, but also orphans dying of neglect in Romania.

A photojournalist is always in a rush to get to the next assignments, and she recognizes her involvement with her subjects' culture as superficial, unlike that of her locally based correspondent colleagues or those who would nowadays be called photoethnographers. There is more humanity in a single frame by Karen Nakamura or Dorothea Lange than in all of Deborah Copaken's work. Much like her idol Cartier-Bresson's work, there is a certain glib coldness, perhaps even callousness to her attitude. On her first war coverage, an Afghan who is escorting her (so she can make her ablutions in privacy) has his leg blown off by a landmine, and she hardly elicits any concern for the poor soul. Granted, this is the "Shutterbabe", not the reborn Mom. but it is hard to imagine one's fundamental personality changing that much.

The author is not uncontroversial. She featured in a nasty spat with Jim Nachtwey, one of the most famous photographers alive, and who is obliquely referred to in Shutterbabe's Romanian chapter (where she implies she found out first about the terrible situation in the orphanages, and nobly tipped him so the story could come out). The follow-ups are here and here.

Her observations of the one culture she is immersed in, the French one, seldom go beyond the realm of cliché. Glamorous but feckless and chauvinistic Frenchmen! Sexpot Frenchwomen! Narcissistic French intellectuals!

In the end, she returns to the United States with her husband, and moves into an equally short-lived career in TV production to support her family. A happy enging? One hopes. I for one am curious about how her children will react to the book when they are old enough to read it.

17:21 - permalink [Book Reviews, Photo]

Is the Nikon D70 NEF (RAW) format truly lossless?

Many digital photographers (including myself) prefer shooting in so-called RAW mode. In theory, the camera saves the data exactly as it is read off the sensor, in a proprietary format that can later be processed on a PC or Mac to extract every last drop of performance, dynamic range and detail from the captured image, something the embedded processor on board the camera is hard-pressed to do when it is trying to cook the raw data into a JPEG file in real time.

The debate rages between proponents of JPEG and RAW workflows. What it really reflects is two different approaches to photography, both equally valid.

For people who favor JPEG, the creative moment is when you press the shutter release, and they would rather be out shooting more images than slaving in a darkroom or in front of a computer doing post-processing. This was Henri Cartier-Bresson's philosophy — he was notoriously ignorant of the details of photographic printing, preferring to rely on a trusted master printmaker. This group also includes professionals like wedding photographers or photojournalists for whom the productivity of a streamlined workflow is an economic necessity (even though the overhead of a RAW workflow diminishes with the right software, it is still there).

Advocates of RAW tend to be perfectionists, almost to the point of becoming image control freaks. In the age of film, they would spend long hours in the darkroom getting their prints just right. This is the approach of Ansel Adams, who used every trick in the book (he invented quite a few of them, like the Zone System) to obtain the creative results he wanted. In his later days, he reprinted many of his most famous photographs in ways that made them darker and filled with foreboding. For RAW aficionados, the RAW file is the negative, and the finished output file, which could well be a JPEG file, the equivalent of a print.

Implicit is the assumption that the RAW files are pristine and have not been tampered with, unlike JPEGs that had post-processing such as white balance or Bayer interpolation applied to them, and certainly no lossy compression. This is why the debate can get emotional when a controversy erupts, such as whether a specific camera's RAW format is lossless or not.

The new Nikon D70's predecessor, the D100, had the option of using uncompressed or compressed NEFs. Uncompressed NEFs were about 10MB in size, compressed NEF between 4.5MB and 6MB. In comparison, the Canon 10D lossless CRW format images are around 6MB to 6.5MB in size. In practice, compressed NEFs were not an option as they were simply too slow (the camera would lock up for 20 seconds or so while compressing).

The D70 only offers compressed NEFs as an option, but mercifully they have improved the performance. Ken Rockwell asserts D70 compressed NEFs are lossless, while Thom Hogan claims:

Leaving off Uncompressed NEF is potentially significant--we've been limited in our ability to post process highlight detail, since some of it is destroyed in compression.

To find out which one is correct, I read the C language source code for Dave Coffin's excellent reverse-engineered, open-source RAW converter, dcraw, which supports the D70. The camera has a 12-bit analog to digital converter (ADC) that digitizes the analog signal coming out of the Sony ICX413AQ CCD sensor. In theory a 12-bit sensor should yield up to 2¹² = 4096 possible values, but the RAW conversion reduces these 4096 values into 683 by applying a quantization curve. These 683 values are then encoded using a variable number of bits (1 to 10) with a tree structure similar to the lossless Huffmann or Lempel-Ziv compression schemes used by programs like ZIP.

The decoding curve is embedded in the NEF file (and could thus be changed by a firmware upgrade without having to change NEF converters), I used a D70 NEF file made available by Uwe Steinmuller of Digital Outback Photo.

The quantization discards information by converting 12 bits' worth of data into into log₂(683) = 9.4 bits' worth of resolution. The dynamic range is unchanged. This is a fairly common technique - digital telephony encodes 12 bits' worth of dynamic range in 8 bits using the so-called A-law and mu-law codecs. I modified the program to output the data for the decoding curve (Excel-compatible CSV format), and plotted the curve (PDF) using linear and log-log scales, along with a quadratic regression fit (courtesy of R). The curve resembles a gamma correction curve, linear for values up to 215, then quadratic.

In conclusion, Thom is right - there is some loss of data, mostly in the form of lowered resolution in the highlights.

Does it really matter? You could argue it does not, as most color spaces have gamma correction anyway, but highlights are precisely where digital sensors are weakest, and losing resolution there means less headroom for dynamic range compression in high-contrast scenes. Thom's argument is that RAW mode may not be able to salvage clipped highlights, but truly lossless RAW could allow recovering detail from marginal highlights. I am not sure how practicable this would be as increasing contrast in the highlights will almost certainly yield noise and posterization. But then again, there are also emotional aspects to the lossless vs. lossy debate...

In any case, simply waving the problem away as "curve shaping" as Rockwell does is not a satisfactory answer. His argument that the cNEF compression gain is not all that high, just as with lossless ZIP compression, is risibly fallacious, and his patronizing tone out of place. Lossless compression entails modest compression ratios, but the converse is definitely not true: if I replace the file with a file that is half the size but all zeroes, I have a 2:1 compression ratio, but 100% data loss. Canon does manage to get the close to the same compression level using lossless compression, but Nikon's compressed NEF format has the worst of both world - loss of data, without the high compression ratios of JPEG.

Update (2004-05-12):

Franck Bugnet mentioned this technical article by noted astrophotographer Christian Buil. In addition to the quantization I found, it seems that the D70 runs some kind of low-pass filter or median algorithm on the raw sensor data, at least for long exposures, and this is also done for the (not so) RAW format. Apparently, this was done to hide the higher dark current noise and hot pixels in the Nikon's Sony-sourced CCD sensor compared to the Canon CMOS sensors on the 10D and Digital Rebel/300D, a questionable practice if true. It is not clear if this also applies to normal exposures. The article shows a work-around, but it is too cumbersome for normal usage.

Update (2005-02-15):

Some readers asked whether the loss of data reflected a flaw in dcraw rather than actual loss of data in the NEF itself. I had anticipated that question but never gotten around to publishing the conclusions of my research. Somebody has to vindicate the excellence of Dave Coffin's software, so here goes.

Dcraw reads raw bits sequentially. All bits read are processed, there is no wastage there. It is conceivable, if highly unlikely, that Nikon would keep the low-order bits elsewhere in the file. If that were the case, however, those bits would still take up space somewhere in the file, even with lossless compression.

In the NEF file I used as a test case, dcraw starts processing the raw data sequentially beginning at an offset of 963,776 bytes from the beginning of the file, and reads in 5.15MB of RAW data, i.e. all the way to the end of the 6.07MB NEF file. The 941K before the offset correspond to the EXIF headers and other metadata, the processing curve parameters and the embedded JPEG (which is usually around 700K in size on a D70). There is no room left elsewhere in the file for the missing 2.5 bits by 6 million pixels (roughly 2MB) of missing low-order sensor data. Even if they were compressed using a LZW or equivalent algorithm the way the raw data is, and assuming a typical 50% compression ratio for nontrivial image data, that would still mean something like 1MB of data that is unaccounted for.

Nikon simply could not have tucked the missing data away anywhere else in the file. The only possible conclusion is that dcraw does indeed extract whatever image data is available in the file.

Update (2005-04-17):

In another disturbing development in Nikon's RAW formats saga, it seems they are encrypting white balance information in the D2X and D50 NEF format. This is clearly designed to shut out third-party decoders like Adobe Camera RAW or Phase One Capture One, and a decision that is completely unjustifiable on either technical or quality grounds. Needless to say, these shenanigans on Nikon's part do not inspire respect.

Generally speaking, Nikon's software is usually somewhat crude and inefficient (just for the record, Canon's is far worse). For starters, it does not leverage multi-threading or the AltiVec/SSE3 optimizations in modern CPUs. Nikon Scan displays scanned previews at a glacial pace on my dual 2GHz PowerMac G5, and on a modern multi-tasking operating system, there is no reason for the scanning hardware to pause interminably while the previous frame's data is written to disk.

While Adobe's promotion of the DNG format is partly self-serving, they do know a thing or two about image processing algorithms. Nikon's software development kit (SDK) precludes them from implementing those algorithms instead of Nikon's, and thus disallows Adobe Camera RAW's advanced features like chromatic aberration or vignetting correction. Attempting to lock out alternative image-processing algorithms is more an admission of (justified) insecurity than anything else.

Another important consideration is the long-term accessibility of the RAW image data. Nikon will not support the D70 for ever — Canon has already discontinued support in its SDK for the RAW files produced by the 2001 vintage D30. I have thousands of photos taken with a D30, and the existence of third-party maintained decoders like Adobe Camera RAW, or yet better open-source ones like Dave Coffin's is vital for the long-term viability of those images.

Update (2005-06-23):

The quantization applied to NEF files could conceivably be an artifact of the ADC. Paradoxically, most ADCs digitize a signal by using their opposite circuit, a digital to analog converter (DAC). DACs are much easier to build, so many ADCs combine a precision voltage comparator, a DAC and a counter. The counter increases steadily until the corresponding analog voltage matches the signal to digitize.

The quantization curve on the D70 NEF is simple enough that it could be implemented in hardware, by incrementing by 1 until 215, and then incrementing by the value of a counter afterwards. The resulting non-linear voltage ramp would iterate over at most 683 levels instead of a full 4096 before matching the input signal. The factor of nearly 8 speed-up means faster data capture times, and the D70 was clearly designed for speed. If the D70's ADC (quite possibly one custom-designed for Nikon) is not linear, the quantization of the signal levels would not in itself be lossy as that is indeed the exact data returned by the sensor + ADC combination, but the effect observed by Christian Buil would still mean the D70 NEF format is lossy.

13:02 - permalink [Photo]

The megapixel myth, a pixel too far?

I wrote an article on how questionable digital camera marketing has pushed megapixels to the forefront, to the exclusion of everything else, and how this "megapixel myth" leads to counterproductive design decisions. More than 1200 people read this article in the first few days it was made available, so I promoted it to the "Longer Articles" section.

Click here for the full article

16:36 - permalink [Photo, Soapbox]

First steps in Medium Format

I bought a used Hasselblad 500 C/M last week. I took my first shots last week-end (this is my first medium format camera, and I had to learn how to load it and process 120 format roll film). Today I installed an Epson 3170 scanner capable of scanning medium format negatives (at the highest quality settings of 3200 dpi at 48 bits, this yields 52 megapixel files weighing 300MB each!). The quality is simply amazing, even more than you could expect with the 4x larger negative area than 35mm. Here is a preview scan of one of the shots (the inner marketplace courtyard, Ferry Building, San Francisco), and a 3200 dpi blow-up of the upper left corner of the frame:

Technical details: taken on 2003-10-24, Hasselblad 500 C/M, Carl Zeiss Planar 80mm f/2.8 CF, Fuji Neopan 400 processed in Ilford DD-X, exposure 1/250s at f/4.

02:06 - permalink [Photo, San Francisco]

Bombay Bits

The Economist once called Bombay "The most expensive slum in the world". When I was a child, we used to fly to India every year, stopping for a day on the way there and back at my uncle's place in the Khar district of that city, and I certainly agree with the characterization.

German cinematographer and photographer Lutz Konermann shows you can find beauty in that unlikeliest of places, in his collection called Bombay Bits.

P.S. Yes, I know the city was officially renamed "Mumbai" for political reasons by the extremist BJP party in power there, the way Madras was renamed "Chennai" and Poona became "Pune". Here is an article debunking the controversy.

16:34 - permalink [Photo]

Inkjet printers revisited

In a recent post, I railed at the Inkjet racket. Lest I be perceived as doctrinaire, I do believe there are some good reasons to use inkjet printers, just that price is not one of them.

Here are the cases where I think an inkjet printer is preferable to Fuji Frontier or Noritsu prints on Fuji Crystal Archive paper:

• Speed

  There is no question sometimes speed matters, and the convenience is worth the price, specially if only proofs are required.

• Black & white photography

  Printing black and white photos on color materials usually leads to subtle, but discernible color shifts. Although Ilford makes a paper designed specifically for use in digital enlargers like the Lightjet, very few labs offer true B&W digital photo prints, the only one I know of is Reed Digital, using Kodak Portra BW RC paper. RC papers are not archival in any case.

  Inkjet printers modified to use the PiezographyBW system can yield black and white photos that match or even exceed the quality of gelatin silver prints. If they use carbon pigment inks, they will be at least as archival (the carbon photographic printing technique is quite ancient and is considered one of the highest forms of photographic printing art, along with platinum printing).

  More recently, Hewlett-Packard has introduced the No. 59 photo gray print cartridge for use with its Photosmart 7960 printer. This is a well-supported solution, unlike the finicky Piezography process, but unfortunately, it requires the swellable-polymer HP Premium Plus Photo paper to give decent durability, and is limited to Letter/A4 size. The drying time on swellable polymer paper ranges from a few hours to a day, taking away the immediacy of inkjet prints.

  I compared prints made on Ilford Multigrade IV fiber base (baryta) B&W paper, Agfa Multicontrast Premium (resin-coated) B&W paper, Fuji Crystal Archive on a Lightjet, and the HP 7960 on both the glossy and matte HP Premium Plus Photo papers. The prints were compared under daylight conditions (overcast sky), under an incandescent (tungsten) light bulb, and under a compact fluorescent lamp. The results are summarized in the table below.

  Color casts

  Print paper used	Daylight	Tungsten	Fluorescent
  Ilford MGIV FB	Neutral	Neutral	Neutral
  Agfa RC	Slightly warm	Slightly warm	Neutral
  Fuji Crystal Archive	Slightly blue	Neutral	Slightly purple
  HP Premium Plus Photo	Strong Blue	Slight purple	Strong purple

  The HP prints give better highlight detail than the Fuji, but fall short of the Agfa. The HP solution is not the silver bullet B&W aficionados were waiting for.

  Update (2004-01-22): the color cast seems to be an issue when the printer is new. After a few weeks of use, the ink cartridge "settles down" and seems far more neutral, better than the Lightjet print. It is not clear whether (1) this problem will reoccur with every new No. 59 cartridge, (2) or whether it was a defect in the cartridge I have, (3) or whether it only happens the first few days after a printer is put in service. As HP cartridges include the print head, I suspect it is option 1 or 2. This would increase the cost of the prints further by increasing waste, but the good news is, the grayscale output from the printer is competitive with the darkroom given the superior level of control you get from Photoshop, and this is the first mass-market printer that can really make this claim.

• Matte prints

  Unlike fiber papers, RC papers seldom have true matte finishes. The so-called "lustre" finish is in reality a process in which the surface of the paper is calendared (pressed by textured rollers) to imprint a pebble-grained finish, which is a mere ersatz of real matte photo paper. As this is not a microscopic finish, it diffuses light unevenly and when seen at an angle, the print is washed out by the reflections from the textured surface.

  A printer like the Epson Stylus Photo 2100/2200 using pigmented inks can make durable and almost painterly prints on fine watercolor papers from the likes of Hahnemühle or Crane's. Many fine art photographers favor this printer for that very reason.

23:05 - permalink [Photo]