For the last three months, my computer has been performing very poorly, and I have been spending many a weekend trying to figure out why. Well, I have finally fixed the problem, it turns out there was an issue with my cooling, after several minutes of the computer being on it had reached 100°C – which is better than my kettle. To avoid completely burning itself out the computer, it saves itself and reduces its performance.
If you don’t know this is happening and have never seen the consequences before it can be quite hard to diagnose. I reinstalled windows, changed the GPU, and doubled the amount of ram all to no avail. If only the computer would tell me it was limiting to performance due to it being too hot it would have cost me a lot less time and money. I am not worried about Skynet taking when my computer can’t even tell me it’s overheating.
Anyway, it is always satisfying when you have been working on a problem for a long period, and eventually, you solve it. One of the good things about doing a PhD is that my resilience when it comes to problem-solving. I haven’t yet developed a robust system for problem-solving, which may help in adaptability for different types of problem. Still, it does mean it is slower than it perhaps could be. Concerning pc repair, my new favourite tool is ‘openhardwaremonitor‘. It tells you what every single piece of hardware is doing and is great for diagnosing problems.
At the university, things are starting to wind down for the holidays. In practical terms, it means that I only have two weeks left of the year (Yes, this was written in December) in which to complete the experiments I planned to. I am confident I won’t get them all done, but I will get into a position where everything will be completed by the end of January.
If I had it my way I would work a little longer this year as I am in a flow state where everything is going well at the moment, but the equipment at the university is very expensive and temperamental, and it isn’t as easy as just switching it off. It is a big operation to shut down all the machinery, and a coordinated effort is required. Because of this, we can’t just work when we want, when our machines switch off, so do we.