Making the slowest computer ever… Just for fun. #Processors #Hardware
We spend some of our time as computing teachers talking about hardware and processors and in particular the characteristics that make one processor more powerful and capable than another. Maybe we should look at things differently for a moment. How slow can we make a processor? I’m taking inspiration here from John Cage and his piece of music that is so slow or will take over 600 years to complete. It made the news recently because there was a chord change! (https://www.bbc.co.uk/news/world-europe-54041568)
How about we don’t use magnets, or lasers, or even electrical charge to store our binary bits. What about trees? It’s a stretch of the imagination, though if we know that the tree* was either there or not, in a specific place. Much in the same way that a hard drive has tracks, sectors and a file allocation table, our tree-based computer can have specific places 20 metres apart that can have Conker trees, or not. If we want to store one byte of data, we need a strip of land 20x160m (I know that this is going to be big!)
One KB of tree data will be stored on about 810 acres of land; this will just about fit into New York’s Central Park. The mechanisms for a ‘bit’ being there or not are already well established, we can plant a tree, or chop it down. It will just take far longer than a magnet changing from North to South or some charge being stored or released; maybe 10 years or so. And so, we have a system able to store bits and in terms of processing, I’m sure that someone can work out the rest. When it comes to the comparison with a different processor, I’m sure that we could increase the clock speed by using a faster growing tree, or even a sunflower. We can increase the size of the cache by having a bigger field closer to the processor, and as for the number of cores, that again is more fields, each able to make processing decisions.
While talking to one of my colleagues about this, the conversation went along the lines of “What if a deer came along and ate one of the saplings?”, or “What if a squirrel stole one of the nuts and planted it in a place that it shouldn’t be?” And although it threw a curveball into the idea, it also raised questions of error checking within the data storage system and parity bits / parity errors. Not such a daft idea. And it might also encourage some highly beneficial tree-planting.
* Maybe a horse chestnut, they were voted most popular in the UK in 2017, https://www.rsb.org.uk/get-involved/biologyweek/uk-s-favourite-tree-species