The Pentax 645 dominated aerial photography for years. It was an SLR, relatively small and light for a medium format camera, would cram almost 30 exposures into a 220 roll, offered 2fps and a 1/1000 shutter speed. Then there were the lenses...those glorious Pentax Super Multi-Coating (SMC) optics that couldn't even spell "flare."
The only drawbacks to the P645 system, in my opinion, were the ungainly configuration with the 70mm back (which required an eyepiece extension and destroyed ergonomics), the weight of the 120 prime, and until the latter years, the lack of any zooms.
When I went digital in 2004, I stored all of my P645 lenses (35, 2x45, 2x55, 2x75, 120, 2x150, 200, 80-160) in heated anticipation of a P645 digital, already rumored to be in development. I know that I was not the only one.
Last week (early May, 2007) Pentax was reported to have scrapped its digital 645 development-this, after displaying a mockup of a nearly 4oMP unit on display at PMA 2007!
So...what's next? Bayer Array configurations in the 35mm format are already near their maximum spatial resolutions (it's actually the ability of the format lenses that is limiting in this issue-we'll go through the numbers in a subsequent post). Phase One and Hasselblad have priced themselves out of contention (aerialists need two bodies, minimum, and aren't willing to spend $40K each on units that will be competitively obsolete in 2-3 years). I have actually kicked a P645 out of a helicopter...that memory, and other flirtations with disaster in the field discourage dropping $40K on each body. Driving around certain parts of Phoenix, Sacramento, and Los Angeles in well-marked company vehicles also makes for some concern...
With the 5D and 1DsMII (and their Nikon counterparts) so good, it might be time to fall back on film (gasp!) more regularly for certain applications. I shot with film (Hasselblad 500CM, PlanarPrism, 80mm CF Planar T) only once last year-we had a University customer that wanted their entire campus in a single image (6 square miles) and wanted to be able to read the yard markers on the football field. This had previously been done on a Pentax 67 but the scan wasn't that great, and the yard markers were barely legible. We tried a 5D, and it was awfully close to the 67. The Hassy and a good drum scan embarrassed them both.
As part of my mourning ritual I searched around the 'net for drum scans (I also sold my big scanner when I switched to digital) and was shocked to see that there are service bureaus offering 100-200MB 16bit Tango (a true drum scanner) scans for $35-$60!
Still, a 200MB (~3000spi, 8.5 micron samples) 16bit scan will not provide detail equivalent to that of a 240MB 16bit digital (6.8 micron pixels, Phase One P45+), even with the roughly equivalent spatial sampling. The reasons far exceed any abbreviated explanation, but start with film grain being binary and continue through scanning issues such as Nyquist.
Of course, the film-based workflow is severely disadvantaged.
A moment of silence, please.
Comments