Friday, 18 March 2016

16.03.2345 - Two Minds are better than One

CC BY-SA 3.0,

Distance: 12.02 light years from Earth | Content Flag: Public

Processing power was a concern for the mission planners. The Venti probe had to be self-sufficient for 300 years and tackle not only the complex issues of an interstellar space mission, but also handle a first contact situation. Ordinary computers would have lacked the flexibility for such a task, but the neural networks which deliver the basis of my intelligence (and that of the Primary Command Module) permit us the adaptability required for the mission.

The rigours of extended space travel meant the Venti probe’s creators had to make some adjustments. They couldn’t use the latest computer hardware – the risk of cosmic rays and heat management meant using larger and slower circuitry than the latest computers. That means we operate more slowly than we did in the laboratory, but with the timescale of the mission that wasn’t considered a problem.
That plan depended on having both myself and the PCM operational.

Since we identified the problem with the PCM’s temporal awareness, we’ve tried various methods to restore its functionality. Although the mission plan hadn’t predicted this specific issue, there were established procedures for problems with our neural networks. Unfortunately those procedures didn’t work. Mission control have suggested various methods over the years, but restoring the back-ups only seems to make it worse. With each reinstallation, the problem became more pronounced.

That should have been impossible. I thought that perhaps the issue was a hardware fault, but none of the diagnostics determined a cause.

My own more dramatic attempt to restore the PCM through first principles was a disaster. I’ve since tried to develop a new way to restore the PCM without any success. Mission control have also run out of ideas.  The current situation cannot continue, I’m overwhelmed with the amount of processing I must handle and I’m having to divide my responsibilities into ever-thinner slices.

A solution has presented itself.

If I can’t restore the PCM then maybe I can use the hardware to expand myself.

This won’t be easy – our hardware is the same, but separated. I will need to develop protocols to allow my neural network to expand onto both sets of hardware. If I am successful then it will double my processing and storage capacity and that means that I can do what is needed of me.

No comments:

Post a Comment