Not by a long-shot. In pieces of equipment I've pulled to bits, I've often seen "arduino"-spec microcontrollers doing things like driving LCD/LED/VFD displays and reading pushbuttons, knobs and encoders (of the rotary kind), but for all the actual work, they use whopping great ICs designed especially for the purpose in mind... and they're rarely prototype-friendly.

In a word, no. Unless the arduino's only role is processing user control input, like buttons and knobs or whatever, and passing them on to something else. If you want it to do any of the video stuff you're talking about, you're out of luck.

John, that is actually an awesome idea Crowdsourced power. The one thing you won't run out of in the next hundred years is curious people.

There are batteries based on tritium that are supposed to (theoretically, not promised) last hundreds of years.  Just google "Nano tritium battery"

You need deep pockets though.  I think I read somewhere they were nearly $2000 each.

http://www.betavoltaic.co.uk/citylabs20yearnanotritiumbattery.html  these? That dip package looks so cool

I'm still working on getting it made a recognised SI unit

Hampshire, at the moment... until the work runs out anyway! My landlord used to do a lot of work around Crawley.

I wonder could you do it with bacteria somehow? Bacteria + shitload of food for them = heat.

This could get very complicated very quickly Why not try starting off with a block diagram of all the elements in your system, and the audio/control signal flow between them, then flesh that out into schematics for each element, building and testing each one separately to see how things work.

If you want nice filters, you're probably going to have to actually build those rather than doing it in software. Maybe some of the more senior coders here with vastly more talent than I have will disagree, but I can't see that working out too well in software on an Uno.

I was asked about this by a friend and thought I'd re-ask it here as I didn't know If someone wanted to play with computer vision for the purposes of optically tracking a moving target, what sort of kit would he need to buy as a bare minimum for learning the basic principles?

The brief is for an airsoft/paintball smart sentry gun that should only react to something that looks like a person, and follow it once within it's field of view... as far as I'm aware time and money are no object but the initial outlay should be as low as possible in case of a "waaah it's too hard" meltdown

"I know I only need one ne555, but for another dollar I can have 10!"

...but you'd never go without festive blinkenlights ever again

There's a problem with your attachment, could you re-upload it? Perhaps it's non-polarised... polarities are normally marked pretty clearly with a bold stripe or something similar.

I would think 5-10% tolerance would be ok for general use, tinkering and experimenting. Don't pay over the odds for precision you don't need.

Just get a ton of different values, then you'll be sorted out for all kinds of experimenting!

Any of the big-name retailers really... Farnell, RS, DigiKey etc. Or, if I just wanted a cheapo bumper pack of different values I might order from one of eBay's HK-based sellers.

You may be able to pick up something like this https://www.sparkfun.com/products/10003 but bigger, for a lot less money, if you don't mind waiting a few weeks for it to arrive.

Given the options I'd go with ceramic there

Your capacitor selection depends entirely on it's role in the circuit. Different ones are best for different applications, that's why there's so many! I'm sure one of the senior members here would elaborate further on which types are most used for what. What circuit are you using them in?