Is it a 3D model for ray tracing purposes or an actual working model?
The group behind the attempt to build Charles Babbage's Analytical Engine has started accepting donated funding for the first stage of the project, a 3D computer simulation. Plan 28 want to start the ancient computer build with a 3D working simulation, computer historian John Graham-Cumming said. For that, the group, started by …
'Simulation' suggests the latter.
The first working model of the Difference Engine No. 2 had more than 8,000 parts. The Analytical Engine is a far more complicated beast, being able to work with "1,000 numbers of 40 decimal digits each". A 'proof of concept' cut-down design by Henry Babbage that could deal with 20 numbers of 25 digits required 10 columns of 15 wheels.
So as far as I can make out, this simulation of the Analytical Engine will have to deal with tens of thousands of moving parts. It doesn't sound too easy, but maybe its modular construction will allow for some shortcuts to be taken. Any ideas?
40 000 moving parts apparently. All kinds of gearing (floating, eccentric, cams the lot)
The Man is open to suggestions for simulation software that could model that system.
My question is: assume perfection or attempt to model friction?. Friction is, I understand the reason the original failed. The torque needed to drive the main shaft sheared the shaft with the materials they had access to.
Plus an engineering lead with his own ideas. As the good Doctor said "You could say it was the first failed government IT project."
Now, drop them a fiver. You know it is the right thing to do.
Why is this so expensive?
1) Why would they go for full physical simulation? If the goal is to prepare for designing a real working model, then simulating the entire machine is an overkill. If the goal is to distribute a virtual model, then a graphical simulation, or one with extremely simplified physics would suffice.
2) As to building the real engine, surely most of the parts can be gotten off-the shelf and the overall design adjusted to fit them.
Graham-Cunning. Plan 28
Came for the Cunning Plan jokes. Left disappointed.
Re: Graham-Cunning. Plan 28
At least, this being The Register, there were some cunning linguistics.
Maybe they can ask Nathan Myhrvold to sponsor one? I've seen his Difference Engine No 2 at work when it was on display at the Computer History Museum in Mountain View, and it was very impressive.
I really hope...
...they get on and use 3D printing for this. Somehow it would seem a very aesthetically pleasing thing to do.
Re: I really hope...
I'd love a 3D printer for Brass, Steel etc.
Re: I really hope...
Me too. They do exist, but are pricey. Selective Laser Sintering.
But you probably wouldn't use a 3D printing technique to produce thousands of examples of the same part.
It was never built as it has rounded corners.
Are there any engineering firms left in the UK? Making the parts for an engine such as Mr Babbage's could provide a fine introduction to machining.
Or maybe James May, with his fascination for Meccano and making things, might help out. Perhaps he could also touch Jeremy Clarkson for a few quid from his recent Top Gear windfall.
Really 2 challengers here
IIRC Babbage was *very* nervous about IP theft and some of the drawings were *deliberately* misleading, although it would be interesting to see if that was the case with the Difference Engine.
The other issue is weather it would have worked with 19th century levels of mechanical engineering precision.
The construction of the DE suggests it *should* but does all that *decision* making logic make the parts alignments *much* tighter?
and of course there is the £5m over 10 years needed...
Kickstarter doesn't always work
A version of the Difference Engine has been built out of Meccano (so much for engineering tolerances),
Perhaps the analytical engine team should consider .....................