uneven rates of progress

I just unboxed the new camera and am charging its battery.

 

There is enormous choice these days - enough to make Barry Schwartz shake his head with a bit of disgust.  I spent more than a few hours polling friends who know their way around photography and finally tried a half dozen candidates before settling on the compromise that hopefully will cover my needs.  All of the candidates were compromises.

 

There was a time when I loved photography.  My camera was an ancient  second hand Leica M3 that was nearly perfect as far as I was concerned.  It was completely manual in operation, but the controls were perfectly placed and operated smoothly.

 

The Leica was an extension of my eyes and hands.  I would see something and while my mind worried about composition somehow the shutter would fire at the right time.  

 

I spent time worrying about what film to use and the details of processing. That part was expensive so I didn't shoot as much as I would have liked, but the process of photography was enormously satisfying.  Most of the best images I've ever taken came from that period.

 

I stupidly sold the Leica twenty years ago when I bought into a 35mm Nikon SLR system.  The camera was beautifully made and some of the prime lenses were optically amazing, but it was never an extension of my soul.  I tended not to use it.  

 

At one point I found myself helping out on a photoshoot with a famous fashion and portrait photographer.  Alberto used a 6x7cm Bronica and was so comfortable with it that it seemed to be an extension of his mind.  To check the lighting on the model he pulled out an ancient Polaroid camera, fired a few test shots with the strobes, and made the mental adjustments of how he would use his medium format camera.  There were two bodies - his assistant would load the spare with film, while he shot with the other.  It was amazing to watch him practice his art and to see how many of the negatives were usable.   It was also frightening to watch him burn through several thousand dollars of film and processing in the course of a ten hour shoot.  

 

Digital photography finally changed his world and radically altered the fee structure. In the hands of a pro, a $40k Hasselblad quickly pays for itself.  At the high end user interface and experience are critically important and the pros make their choices on that as well as whatever lenses they have.   The process of photography has changed - art directors and models get immediate feedback from tethered computers and ten hour shoots become five hour shoots.  Amazing cameras.

 

The little digital camera I unboxed earlier today has one of the best user interfaces and experiences of the digital cameras in my price range, but it is jarringly awful compared to the UX of my iMac and iPhone.   I doubt I'll ever be comfortable enough with it that it feels like and extension of myself.  On the other hand it takes good photos and, if I don't could the camera's price,  the cost per image is essentially zero.  The images are in a form that are much more useful to me than film.  Positive progress on several fronts and negative progress in one that is very important to me.

 

Shopping for the camera made me think about the marketing of high technical content items.  Many of the aspects that are really important - the modulation transfer function curves of the optical train,  pixel size and response, dark noise, and so on are rarely communicated.  Pixel count is stressed, although it turns out not to be that important.  On-camera image manipulation is pushed although any image editing program is vastly superior.  The marketing message does not overlap well with what the camera really is - you have to spend some time with one and see if it is going to work for you.  For me this means going to a real camera shop - something that has almost disappeared from the landscape.  I find paying list, but being able to audition what I will have to live with for the next 5 to 10 years a small price to pay.

 

I also found myself reflecting on how far digital photography has come.  I've built several cameras over the years.  Pinhole cameras and a Brownie-like wooden camera that used an old lens someone gave me when I was 12.  It had a shutter and iris from a fleamarket camera and took some surprisingly good photos.  There were also a few cameras for my telescope.

 

In 1984 I made my first digital camera.  I was working at Bell Labs where work had been done on CCD cameras for videophones a decade earlier (resulting in a recent Nobel Prize).  I wanted to build a simple digital camera to capture images on my new Mac 128 and faced some roadblocks.  First it was impossible to get a suitable CCD array and second the port on the Mac was a weird RS-422 standard and programming the beast was awful.  

 

Silicon is photosensitive (duh) and I decided to see how good a DRAM chip would be.  A call to someone in chip packaging at the Western Electric Allentown Works produced a small bag of 16 and 32 kilobit DRAMs nicely mounted in ceramic packages with the top protective cap removed.¹  I put one on a scope and found it to be practical for camera experiments.

 

The trick was to put the DRAM in the focal plane of a simple camera lens mounted on a homemade wooden box.  The first camera had a conventional iris and shutter.  It had 16 kilopixels each with 1 grey level.  I could only read a white or black signal.  That was in November and by then my 128k Mac was now a 512k Mac and I could program the beast.  Writing the i/o for the serial port was the big challenge.  The display was really trivial on the little Mac.  Soon there was a second generation camera with an electronic "shutter".  I would write zeros to all of the memory locations and then read them out later. Generation three happened in mid 1985 with an electronic shutter camera with 3 bits of grey level per pixel.  That was as far as I went.

 

The photo shown is of Meltdown, one of our ferrets, taken with the generation 2 camera with a flash.  The best name ever for a ferret - she was born near the Three Mile Island Nuclear Plant.  It turns out the name matched her personality, but such is the case of silvermitt ferrets.  It turns out to be one of our few surviving photos of her.  Somehow it kept getting backed up and moved from machine to machine over the years, while our negatives, slides and many of our albums were in the cool basement for preservation.  That seemed like a good idea until the 8 year warranty hot water heater failed at year 6 and destroyed our archive.   

 

Sixteen kilopixels to sixteen megapixels and one to twelve bits per pixel and amazing autofocus and other capabilities in around 25 years and, in 2011 dollars my current camera is much less expensive.  

 

But the project deepened my ties to the resident artist at Bell Labs and that took me down several curious paths - an important part of my education.

 

Now all I want is one with a UX that is the equal of my old Leica.  Steve Jobs has legitimized good UX in products that are largely defined by engineering.  Perhaps there is a company that will step up and make a "magical" camera for me.  I rather suspect my next camera will be part of a smartphone. The optical limits are far from tested and there is so much that can be done with increasing computer power.  Smartphones are really just sensor studded computers with a link to the 'net.  The space will expand dramatically over the next decade.  Hopefully the UX will be well thought out.

 

________

¹ This was a home project where I bought everything but the memory chips.  My director thought it was a neat idea and suggested that I get some same chips to play with, although that was probably stretching the rules a bit.   

 

Comments

[Howard Greenstein]

No fair not saying which camera you bought.

[steve]

A Sony NEX-5N ... any camera is a compromise, but this has most of what I want. It feels good in my hands, takes good photos (for its price class), and has what is for me the least objectionable UI of the cameras in its price range.

The UI is a hodgepodge of pieces and it is decidedly "non-analog" in its nature. For a really good UX on a camera you need analog controls. This only has a good analog zoom adjustment. But the other controls are not as bad as other cameras in its class and the image quality is very good for a $600 camera.

Digital cameras UIs, particularly for consumer grade and prosumer cameras, still have a long way to go.