Ultimate digital vs film: 1gp digital vs SR71 reconnaissance cameras

Discussion in '35mm Cameras' started by brian, Dec 4, 2003.

  1. brian

    brian Guest

    I posted this on Max Lyon's stitching forums
    (http://www.tawbaware.com/cgi-bin/forum/cutecast.pl), but thought it
    might be of interest here as well.

    As you may know, Max recently succeeded in stitching 196 images from a
    D60 to form a milestone 1 gigapixel uninterpolated digital image (
    http://www.tawbaware.com/maxlyons/gigapixel.htm ). Naturally, this
    got me thinking about how this might compare with what is arguably the
    highest resolution film "camera" in existence: the SR71
    reconnaissance plane equipped with its ultra-high resolution large
    format cameras.

    I found this declassified CIA document online which talks about
    various aspects of the SR71 reconnaissance plane, including the
    cameras and optics: http://www.blackbirds.net/sr71/successortou2.html
    Evidently several interesting cameras were built for this project,
    including: 1) A Perkin Elmer camera capable of resolving 140 lp/mm on
    6.6" film (2.2 gigapixels), and 2) A Hycon camera with a lens designed
    by James Baker capable of resolving 100 lp/mm on 9.5" film (2.3

    It would certainly be interesting to do a side-by-side shootout
    between one of these cameras and a D60/panotripodhead (talk about
    David vs. Goliath!!). Just based on the numbers, it would appear that
    the recon cameras still have an edge, but I'd bet that the lowly 1gp
    digital image would fare quite well given that the SR71 cameras
    probably just barely reach the stated resolutions.

    Needless to say, ordinary large format equipment can't even begin to
    compare with these examples of cold war excess!

    brian, Dec 4, 2003
    1. Advertisements

  2. brian

    Don Stauffer Guest

    I was still working when the recce community was switching from film to
    electronic focal planes. Things were so secret a Kodak rep I was
    talking to didn't even know his company was involved in electronic recce

    For tactical recon, users were willing to give up resolution in order to
    get immediate results, which could be radioed back even before planes

    I was amazed at resolution even of some of the first sensors, and would
    sure like to see what they are getting now, but of course one of the
    downs of retirement is giving up clearances and need-to-know.
    Don Stauffer, Dec 4, 2003
    1. Advertisements

  3. brian

    jjs Guest

    What year was that? FWIW I was at Upper Heyford in the sixties. As you
    probably know it was a recon base with RF101 and RB66 in house and served
    the U2 as well. We had some fun toys, but nothing like what was mentioned
    in the article Brian pointed to. ;-)
    jjs, Dec 4, 2003
  4. brian

    jjs Guest

    There is some question regarding the way they measured the resolution. Was
    it aerial resolution (I think it was) and was it possibly also factored
    through stereo imaging?
    jjs, Dec 4, 2003
  5. Take those declassified documents with a grain of salt. The military is
    sometimes not-quite-honest in releasing performance envelope stuff.
    The 'declassified' max speed of the SR-71 is about 2100 MPH, but I
    know military radar operators who swear they have tracked them much
    faster. I saw declassified satellite images around 1960 where you could
    read the logo on the side of a truck. I can only imagine what must be
    available now. They admit to 'almost' being able to identify people on
    the ground. NSA has an enormous budget, much more than the CIA. It
    is a good guess they can do much better than they admit, no? :)
    Judson McClendon, Dec 4, 2003
  6. brian

    jjs Guest

    Mind if I Laugh Out Loud? If they (whomever 'they' are) said those were
    satellite images, then it was pure misinformation and served its purpose in
    that regard: you were fooled, and it may have frightened the Soviets into
    further spending themselves to oblivion.
    jjs, Dec 4, 2003
  7. brian

    geo Guest

    But they can't find Osama Bin Laden, Sheik Omar, Saddam Hussein. Maybe
    resolution isn't everything.

    Natural Light Black and White Photography
    geo, Dec 4, 2003
  8. brian

    Gordon Moat Guest

    The first thing that comes to mind is how long it takes to photograph 196
    pictures. Surely a film camera could also take 196 images in the same
    time period, and those could also be stitched together.
    Considering that this is very old technology, imagine what came after
    those. ;-)
    How long would you want to be over a target? Taking one image in a
    fraction of a second, or many images (196), at a much slower rate . . . .
    .. I think I would go for the fastest method.
    One shot at a time, compared to a large roll of film. By the way, Kodak
    still make large rolls of film, especially for aerial photography. Some
    of the old gear can be found as surplus, and could make for an
    interesting camera.
    Here is another oddity to consider: some reconnaissance work has been
    done that involves high speed film and 30 to 60 minute exposures.
    Consider also that battery life of digital gear, and the support
    equipment are other considerations. The working model of reports with
    laptops and satellite phones, plus backup gear, extra batteries, and
    charging equipment, should give some indication of the added weight an
    bulk. With planes, weight is less of an issue, but for a human carrying
    lots of gear, most would welcome extra ammunition, rather than other


    Gordon Moat
    Alliance Graphique Studio
    Gordon Moat, Dec 4, 2003
  9. brian

    brian Guest

    I'd love to know more, although information in this area is probably
    pretty hard to get. In particular, I'd like to know whether the
    resolution is maintained over the entire image plane. Because the
    lenses are relatively fast and used wide-open I assume that the film
    was held extremely flat via vacuum or some other means, and that the
    film itself is capable of very high resolution. If all this is true,
    then the aerial lens resolution *might* translate pretty well into
    actual on-film resolution.

    brian, Dec 4, 2003
  10. brian

    Peter Chant Guest

    Surely this is not a film v digital argument, you could just as well stiched
    196 images from 35 mm film cameras, medium format cameras of 8x10 large
    format cameras.
    Peter Chant, Dec 4, 2003
  11. brian

    jjs Guest

    The cameras I am familiar with had the film moving during exposure in
    relationship to the direction of the airplane, and or used a rotating prism
    to direct the image. These things are HUGE. Not your tripod-mountable kind
    of thing. ;) But you knew that.
    jjs, Dec 4, 2003
  12. brian

    jjs Guest

    Somewhat true. Imagine doing the same survey images with a good, high-rez
    4x5 camera. After you scan in the images, the rest of the process is pure
    digital. How would you combine all those images optically? With good
    jjs, Dec 4, 2003
  13. Not only that, but one can ask if after you've composited a bunch
    of separate photographs whether you still have a photograph. It's
    not a moment in time anymore, it's 196 moments.
    Jan Brittenson, Dec 4, 2003
  14. brian

    jjs Guest

    We can say the same for some of those circuit or panoramic photos - you
    know, the big line-up of people with the same guy on each end.
    jjs, Dec 5, 2003
  15. brian

    Norman Worth Guest

    It may be actual resolution on the film. The claimed 1 foot ground
    resolution works out to a bit more than 140 lines per millimeter at 80000
    feet with an 18 inch lens. If you look at the claimed resolution for Kodak
    Panatomic-X Aerecon film and combine it with the diffraction limited
    resolution of an f/4 lens, you also get something just a little better than
    140 lines per mm.

    Norman Worth, Dec 5, 2003
  16. brian

    Bryan Olson Guest

    It's not so much a 'good' guess as a *pure* guess. It's not
    really plausible that a spy satellite could show the logo on a
    truck in 1960, when the first US spy satellite went into
    operation. That project, 'CORONA', was declassified in 1995,
    and the publicly available images are nowhere close to showing
    that kind of detail.

    Also, the huge-budget agency that does spy satellites is not the
    NSA, but the NRO. See:

    Bryan Olson, Dec 5, 2003
  17. To say nothing of the minor problem that you can't see the _SIDES_ of trucks
    from above the truck...

    That said, I did see some contact prints of U2 test shots sometime in the
    (mid?) 60s, and they were amazing. But no _sides_ of trucks.

    David J. Littleboy
    Tokyo, Japan
    David J. Littleboy, Dec 5, 2003
  18. brian

    jjs Guest

    And those rez test targets - the size of a football field! So we got this
    big idea to cut patterns in farmers' fields. A different one for each
    mission test. Started a whole big thang, we did! (Seriously - mull that
    over for a minute.)
    jjs, Dec 5, 2003
  19. THAT EXPLAINS WHAT CROP CIRCLES ARE!!!!!!!!!!!!!!!!!!!!!!!

    David J. Littleboy
    Tokyo, Japan
    David J. Littleboy, Dec 5, 2003
  20. brian

    jjs Guest

    Yup! It was easy to prototype it England - a friendly constituency,
    easily co-opted with a few pints of bitters. Then it just caught on
    everywhere, even in China and the former Soviet territories. Free help. A
    great game, eh?
    jjs, Dec 5, 2003
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.