Gigapixel images of Charles Babbage's Difference Engine #2

Greg sez, "This project is using a number of computational photography techniques to document Charles Babbage's 'Difference Engine No 2' for the Computer History Museum in Mountain View. There are interactive gigapixel images for the four cardinal views of the device available to view."

Babbage Difference Engine in Gigapixel (Thanks, Greg!)


  1. I tried to photograph this machine (on my iPhone, yeah right) while visiting it earlier this year. Not easy.

    These photos are a good start at documenting the machine. Being static, though, all they show is its complexity and the skill of the fabricators.

    I’m envisioning a few hours of HD video of CAD models with many parts made invisible or see-through to show the operation of each subsystem in detail.

    Sounds like a good project for a federal grant. Oops, a patronage.

  2. That’s pretty darned cool. I find the xRez UI a little counter-intuitive, however. To pan with the mouse, you hold the mouse button down and move the mouse in the direction you want the view to go, rather than the direction you want the object in view to go. (The latter being how almost everything else handles drag panning.)

    1. Thanks Coderjoe :).

      This is one of those things where half the audience likes it one way and the other have likes it the other. There are a few papers on peoples preferences. Generally things like maps and objects are set up for a click and drag type of scenario, but if you look at video games and panoramas (flash or qtvr) or a first person perspective dragging in the direction you want to look is predominate.

      One other argument is that is less fatiguing to click and drag in one continuous motion than to click and drag repeatedly to continue panning.

  3. Something I always wondered: why didn’t he get together with a watchmaker or clockmaker and make the thing a whole lot smaller? Wouldn’t that have solved a lot of the mechanical issues, and enabled the more-complex Analytical Engine to be successfully built?

    1. Making the engine smaller wasn’t really feasible. One of the huge problems that he had was that parts couldn’t be machined to the tolerances required, using technology that was available at the time. If you shrink the mechanism, you also magnify the proportional effect of any mechanical deviation from the design.

      1. Not to mention the fact that if you’re going to such trouble to make the thing, you’d rather it wouldn’t wear out anytime soon.

Comments are closed.