BlogSplat

I spend a lot of time talking about how digital technology affects audio, particularly in terms of music creation and distribution. Photography is affected in much the same way.

Millions of people enjoy the benefits of digital photography every day, though the advantages are somewhat smaller than for digital audio. Film cameras are cheap. Film is cheap. Processing is affordable. The big advantages are in the ease and speed of “processing” and distribution. Only, there’s a catch.

***

I’m a big fan of film, and I’m glad that I learned how to use it properly – at least from a technical perspective. Like music, there are technical and artistic aspects, and only one can be learned.

Film is awesome. Light sensitive emulsion on a transparent base, which after processing develops a latent image exposed on it through an aperture. Modern films especially feature fairly good sensitivity, high resolution, low noise, and can be made as large as any camera.

Digital cameras use sensors (usually CCDs) which convert light to a voltage, which is sampled on a grid and stored as digital data. The cool thing about this is you can display the image immediately after taking, especially when the camera has a built-in LCD display (as nearly all do). And since the data is digitized, the photographer can get immediate feedback on issues like exposure and framing, giving a chance ot correct problems.

Film takes time to process and print. If you screw up focus or exposure on film, you don’t find out until later, when it is probably too late to fix it. But film offers a level of information (both resolution and dynamic range) far greater than any current digital image sensor. This is mainly because it’s easy to make film frames larger (4″x5″ and 8″x10″ are common large format film sizes, but larger formats exist), but the largest digital sensors are much smaller, and are comparatively very expensive.

Ever since photosensitive emulsions were perfected in the mid-19th Century, the history of photography quality has been a constant retrograde motion in the name of convenience. Early photographs were done on large glass or metal plates hand-coated with a photosensitive emulsion and exposed in large (sometimes room-sized) cameras. With view cameras, film sizes were standardized, typically to 4″x5″ or 8″x10″ frames.

By the beginning of the 20th Century, 120 format film arrived, offering the convenience of multiple exposures on a roll, but did so by reducing negative size to 6×6 cm (or about 2.4″x2.4″) – other sizes exist, such as 6×7 or 6×9 cm, but 6×6 cm is very popular. Shortly after 120 film, the very popular 135 format 35mm film evolved, with an even smaller frame size of 36×24 mm (1.4″x1″).

Granted, film technology has improved considerably in the last century or so, but not enough that an 840 square mm negative (35mm) can produce the same detail that a 51,562 square mm negative (8″x10″) can.

Digital cameras have, for the most part, continued the trend. Digital point-and-shoot cameras and most digital SLRs have sensors smaller than 35mm frames – my Nikon D40 has a 23.7×15.5 mm sensor. High-end DSLRs sometimes offer “full-sized” sensors, which unfortunately means only the same size as a 35mm frame. Larger sensors can be had on professional “medium format” cameras, but they are smaller than even a 120 format film frame, and still cost thousands or tens of thousands of dollars.

To take advantage of the bigger-is-better mentality, digital camera manufacturers keep increasing the megapixel ratings of their cameras. Unfortunately, without increasing the sensor size, megapixel counts cannot be increased without increasing noise, decreasing light sensitivity, or playing interpolation tricks. 6 megapixels seems to be about right for the smaller DSLR sensor size. Past a certain point, megapixels do not matter, and a picture taken on a 6 megapixel camera can look every bit as sharp and good as one taken on a 14 megapixel camera.

Film “resolution” works differently from digital imaging sensors. The grains of photosensitive chemicals in a frame of film are not laid out in a grid, and there are layers and overlap. As a result, film is a very resolution-dense medium. I’ve seen estimates ranging from 50-150 megapixels for resolution equivalents in a frame of 35mm film. For a 6×6 medium format frame, which has more then 4 times the area of a 35mm frame, the resulting resolution would be something like 200-600 megapixels.

This is all theoretical, of course. My medium format film scanner generates a 64-ish megapixel image from a 6×6 cm medium format frame. Which is excessive for publishing to the web, but can be helpful for higher-quality applications like album covers and posters.

Digital image sensors are constantly improving, but until manufacturers begin using significantly larger sensors, or imaging technology sees a quantum leap in sensitivity, film will have the upper hand in resolution and quality.

***

My first camera was a 110 cartridge camera. 110 is fun because you can pretend it is a spy camera. But the negative area is puny, though with the right optics, they can produce detailed images.

My proper introduction to photography was my father’s 35mm Minolta rangefinder, which I took with me on a high school trip to Washington D.C. and accidentally took a great photo of the Lincoln Memorial.

I purchased a used Nikon FN 35mm SLR camera for a photography class, and it served me well for years. The Nikon F series is probably the best 35mm SLR ever made (with all apologies to Leica enthusiasts).

About ten years ago, I bought a used Mamiya C330, a medium format Twin Lens Reflex camera. It’s a whole different experience shooting square pictures on paper-backed roll film, shooting through a top-down viewfinder – and I like it.

The Mamiya is neither a simple nor lightweight camera, and not one that I can carry around with me everywhere. When I heart out about the Holga, I found a great compromise – a medium format rangefinder-style camera that shot 6×6 cm film in a plastic body. It’s light, compact, and it takes very interesting images, mostly because of its plastic lens. I was a devoted Holga photographer for several years. I even bought a fairly expensive film scanner to digitize my pictures.

The main problem with medium format film is the processing and publishing turnaround. The nearby professional photo finishing lab can process in four hours, but I find it takes several hours to scan and clean up a roll of 12 shots. I began a photo-a-day project in 2002, and while I took all the pictures, I got so far behind in the processing that I never got all the pictures up on my web site.

The missus was a digital point-and-shoot digital person when I met her. I began using her camera for anything needing a short turnaround – for posting to the web, mostly. After a while, I became frustrated with the shutter lag and general unresponsiveness, so I decided to get a DSLR.

I bought a Nikon D40. I couldn’t afford a “full-sensor” DSLR, and my research suggested that 6 megapixels was about right. The D40 is a perfect entry-level DSLR. It has all the necessary features and is very reasonably priced. My only issues with the camera have been the lack of a bracketing exposure feature (for HDR photography), and that some lenses do not autofocus.

The best thing about shooting with the D40 is the long-term cost of operation. I paid about $550 for the camera with the very nice general purpose 18-55 mm zoom lens (this kit costs about $425 now). I bought two fast 4 GB SDHC cards (probably spent $25 each, though you can get both for that price now). Each card can hold over 600 images in RAW format, and the included battery gets me over 1000 pictures between charges, so I haven’t needed to purchase a backup. I have purchased two lenses (a 55-200 mm zoom which is my main working lens, and a gorgeous 50mm prime) which cost about $350. So my total cost of $850 total for the camera.

This past weekend, I shot over 300 pictures. Last weekend, I shot over 500. These are unusually high volumes for me, but these two weekends alone have essentially paid for the camera. The cost of this camera, on a per-picture basis, is now well below $1 per picture.

A roll of 120 film costs about $4 to purchase and $8 to develop, for a 12 image roll. So, it costs about $1 per picture to shoot medium format film, not even considering the cost of the camera, so the cost of operating a medium format camera will, over time, approach but never meet or go below $1 per picture.

Granted, when you are paying $1 for a picture, and you only have 12 per roll of film, you get more careful about what you shoot. With digital photography, I might shoot several pictures for a single image to get framing, exposure, and focus right. But even if you only consider a quarter of my images, I’m still way below $1 per picture.

So the big advantage for me is that I use my digital camera more, and I don’t spend as much time scanning and cleaning up images. For posting to Flickr and Facebook it’s more than enough resolution. Shooting film becomes more of a special event, and I do it either to get a high resolution image or because the camera (the Holga) makes interesting pictures.

Today I got an odd-looking piece of mail: a large blue plastic bag closed by a zip tie and sporting a Swedish customs tag. Big and heavy, I wasn’t sure what was in it, mostly because it was something I ordered six months ago and was a month late – the super special super duper edition of the Fireman album, Electric Arguments.

Inside the bag, a large padded envelope. Inside that, a thin paper bag with a bit “F” on it. The top secured by a charcoal-bag sewn paper strip. Inside that, a heavy metal box.

The box is meant to resemble a 1/2″ master tape box. Inside is a plethora of goodies: the album on vinyl, a book, two art prints, the album on CD, a CD of bonus tracks (remixes and such), a DVD of video and 24 bit 96 kHz PCM audio files, and a DVD of multi-track sessions for some of the tracks. Inside the lid of the box was a master tape label, some doodles, and Sir Paul’s autograph.

I was concerned about the cost of the big set, but the packaging is well worth the $80 I spent on it. ($80 was the pre-order cost – the super deluxe package has since gone up to $100.)

There is a design firm now specializing in these premium music packages, called Artist In Residence. They did the Nine Inch Nails Ghosts deluxe package, and they have a new Pixies set which looks cool. Their editions are expensive, but they’re not intended for mass consumption – they definitely target the True Fan demographic. And they show that there is a market for this kind of thing.

The one thing all of these have in common are availability in digital download format. Which should tell the Music Industry something. You can sell digital copies of your music that are easily copied, and still sell physical copies to folks who want it – at a premium if the packaging offers something special.

Fine custom Ukulele luthiers, particularly those in Hawaii, like to brag about their humidity controlled workshops.

Wood is designed to carry water from the ground and air to all parts of the tree. Cutting down the tree does not change that behavior. Meaning that water does horrible things to wood.

This week in Los Angeles weather, freakish thunder storms and June showers have sent the relative humidity through the roof. And I have a half-finished ukulele body for that atmospheric moisture to play tricks on.

Once the heel and neck blocks and linings were installed in the sides, I sanded the top and bottom edges nice and flat, with the quarter inch taper.

The jig has ends that (in theory) match the taper of the body, which is helpful when gluing the blocks, but not so much for clamping when gluing the top to the sides. The kit instructions recommended shimming the neck end by the amount of the taper before clamping – that way the base of the jig and the clamping caul are parallel.

So, imagine my surprise when I tried to dry clamp the back to the body and nothing was fitting. Comparing the body to a piece of plywood showed the body was warped across the body horizontally. The top braces (both running horizontally – a design flaw?) appear to have done their job, preventing warping one way, but doing nothing to prevent it the other.

Ukulele building activities have been on hold for a few days. The humidity appears to have returned to an El Lay nominal 50 percent.

This morning, I clamped the body between two sheets of plywood and applied moderate clamping pressure. This worked with the top when it began to curl after adding the rosette, and on the fretboard after inserting the fret wire. I will leave it clamped overnight, and hopefully tomorrow brings more (relatively) dry weather — and a flat ukulele body.

Oh, the agony of the un-coordinated perfectionist.

This past week, the Stewart-MacDonald ukulele kit arrived, along with the replacement tuners for Ukulele Number 1. The tuners went in easily enough, and they seem to do the trick – the uke keeps its tune much better now, even with brand new strings.

The Stew-Mac kit is going to be an interesting challenge. It comes with pre-bent sides and pre-rough-cut top and back, but requires building the body. The first step is to build a body assembly jig from 3/4″ plywood.

The Grizzly kit went together in about a week, so I didn’t expect to get much done this weekend, but with the missus in Portland for the long weekend, I hoped to get the body built. Didn’t seem too optimistic at the time.

The jig should be easy enough – three pieces of 7″-wide plywood: 11″, 2″, and 2 1/4″ long. Even with my rudimentary tools, how hard could it be? After seeing Star Trek on Saturday, portable workbench set up in the driveway, I planned on spending an hour cutting the pieces.

And had the most frustrating four hours in recent memory. I couldn’t get anything right. I’d measure twice, cut once, and then find that I was 1/32″ off. Consistently. It turns out my FastCap Flatback measuring tape was just a little bit off. The tang is probably not working right.

After figuring that out (and a couple wasted hours), the stupid errors set in. Like deciding that 11″ (the length of the base) + 5″ (the guide width of my circular saw) + 1/8″ (the kerf of the saw) = 17 1/8″. Unfortunately, I don’t have a safe, accurate way to trim exactly one inch off a 7″x12″ piece.

The last cut I made – the last one I could safely make with a circular saw on my stock – was for the 2 1/2″ tail piece. Unfortunately, it needed to be 2 1/4″.

Not wanting to have wasted an entire day, I used a rasp and sandpaper to work the piece down to size. Which took hours that should have been spent assembling the body. So I finally had my pieces.

Sunday, between seeing Terminator: Salvation and Flight Of The Conchords at the Greek Theatre, I assembled the jig. I was happy with how square I got the end pieces, and how flush the ends were with the end of the 7″ x 11″ base. It was frustrating, but at least I had the jig.

Only this morning I realized it was wrong. I decided to measure the space between the end pieces. 9 9/16″. Instead of the 9 1/2″ cited on the plans. And I knew immediately what I had done wrong. The calipers confirmed what I had forgotten – 3/4″ plywood is actually closer to 23/32″ thick. (It’s supposed to be 3/4″ thick, but moisture variations and sanding tend to remove about 1/32″. The sheet I have was actually a little thinner – about 45/64″)

Given that the instructions are supposedly geared towards beginners, it would have been nice for them to mention this. They do actually deal with this by telling you to measure 3/4″ from each of the short ends of the base. That way, you end up with a 9 1/2″ inside measurement. But getting the end pieces accurately lined up to the lines would be difficult.

A better solution, while I was out cutting 7″ wide plywood pieces, would have been to cut a 7″ x 9 1/2″ piece to use as an inside brace. This would get the length exactly right, and would also help square up the end pieces.

When you think about it, a slightly long ukulele body jig is a trivial thing compared to the sacrifices that our armed forces have made, which we remember on this Memorial Day. But it frustrated me to no end nonetheless.

1/16″ might be within tolerances (the pre-bent sides fit, and the top/bottom being cut rough, should work. But I want this to be accurate.

After weighing options, I decided to find some 1/16″ material to shim the jig – mostly because I was certain that in my frustration I would only screw up the cutting again (and possibly injure myself) – if I could find a lumber store open to get more 3/4″ plywood. I found a sheet of 1/16″ basswood at a hobby shop and glued it into the short end of the jig. Not the hardest of woods, but better than balsa (the other hobby store option).

So, after a long weekend and plenty of time, I have the neck and tail blocks glued into the sides and the glue is curing right now. Still lots to do to get to my goal of completing the body this weekend.

Most of my problems so far have been my own fault. But I do have some gripes about the manuals.

The Grizzly kit instructions were a bit spotty, and the Stew-Mac are definitely much better, but they are still a bit rough, and they don’t do as much explaining as I would like. I don’t know how much of this is the kit instructions being written by employees who are not specifically writers, and how much of it is that no one at the company (who didn’t design the kit) has actually built the kit.

For example, the body template is supposed to be cut out and then traced on to the aforementioned jig, and there are two lines marking the waist of the body to be transferred on to the base so that two angle brackets can be installed on the jig to hold the waist in place. But the lines are outside the body shape, so once cut out, you can’t transfer the lines to the jig. Fortunately, I extended the line through the pattern before cutting it out and was able to transfer it to the jig as intended.

Also confusing, at least for me, was the tapering of the body. It is thinner at the neck than it is at the bottom. The neck block is smaller than the tail block, and you have to angle the lining to match, but there is no picture of this. And later, you have to taper the sides down to the lining, but it took a lot of reading ahead to figure this out, and the pictures of this are less than helpful. A little explanation ahead of time would make this a lot more clear to me.

Given my focus on non-realtime music, it is odd that my most recent obsession is with ukuleles.

Neither grade school piano lessons nor a University rock guitar god fantasy turned me into much of a musical performer. I can play a piano, of course – it is a very simple instrument – but just because I can press keys doesn’t mean I’m any good at it. Guitars require much more coordination and technique. I can play a few chords, and a C Major scale, but am jealous of really good guitar players.

The ukulele seems a little bit more manageable – only four strings – and lately, it seems, ever-present. Just search for ukulele on YouTube and you will find over 47,000 videos of people playing David Bowie, The Ramones, The Beatles, Britney Spears, Madonna, Nirvana, Tom Lehrer and Jonathan Coulton songs on the uke.

There is a degree of irony in the instrument choice for many of these, of course. But it takes enough dedication to learn the ukulele to the point where you can post a video of yourself playing it without embarrassment on YouTube to not have some respect.

And there are plenty of professional musicians playing the uke, too. Janet Klein plays songs from the 1920’s and 1930’s on her ukulele. Greg Hawkes of the Cars did a whole album of Beatles songs on the ukulele. Jake Shimabukuro is an amazing ukulele player, whose version of “While My Guitar Gently Weeps” is awe-inspiring. If you haven’t seen the Wellington International Ukulele Orchestra’s version of “Hey Ya,” you need to go see it now. Dent May And His Magnificent Ukulele has a horrible MySpace page but his music is quite lovely.

And of course, my favorite ukulelist is Molly, aka sweetafton23, who plays originals and covers of Jonathan Coulton and Tom Lehrer songs.

My in-laws have a family heirloom Kamaka ukulele from their time living in Hawaii. Ben has been learning the uke – not a big leap from six-string guitar. His wife Kristina has been learning, and my wife wanted to learn as well.

Researching starter ukes, I discovered something that appealed to my music and DIY sensibilities – ukulele kits.

Grizzly Industrial makes power tools, but their president is an avid luthier, so the company offers guitar, violin, mandolin, and ukulele kits. The best part – the uke kit costs about $25 (currently on sale for $20).

We got a nice curly mango tenor ukulele for the missus, and I bought the $25 kit. At the very least, I’d have a bit of fun, I could try to actually learn to play the darn thing, and if I stuck to it, I could justify buying another nice uke.

The Grizzly kit was pretty easy to put together. The body is pre-built, the neck is already carved, the fretboard is pre-slotted, and the bridge is pre-made. All you have to do is glue it together, finish it, and string it up.

The kit has some issues. Some of the instructions don’t make sense. The fingerboard inlay dots do not fit the pre-drilled holes. It’s mostly made out of mahogany plywood, so it’s not a great sounding instrument.

Even so, it was a good first step in instrument building, and of course now I’m planning on uke number 2.

The best part of building an instrument, particularly from a $25 kit, is that I know how its put together and how to adjust and fix it. It’s a great practice platform.

While building uke number 1, a friend told me to check out a book called Clapton’s Guitar, about master guitar builder Wayne Henderson and the guitar that he built for Eric Clapton. It’s a fantastic book and it inspired me to believe I could make a very good ukulele with some practice, care, and careful choice of materials.

So I ordered a kit for uke number 2 from Stewart-MacDonald. They have a kit that costs four times what the Grizzly kit does, but its made from solid mahogany and requires the assembly of the body (they do supply pre-bent sides however) and involves a rosette around the sound hole. It should be a good second step, a little more involved, on the way to the ultimate goal of building a ukulele from scratch.