Graduation

Well, five years of Uni sure flew by. Flew by like an Albatross in a storm- fast and undulatory like a bunch of barotropic waves coursing their way through the ocean surface. But it’s done! Not much else to say really- I’ve gotta find something to do post-Uni and have no plans!

I mean obviously there are some plans- I’m going to publish my MSc thesis soon (the cost, well- we’ll see.) I’ve also several photography trips planned- I’ll be putting up as much of my photographic library for stock sale as soon as possible. I’m also working on improving my programming abilities- working on learning Julia at the moment, and intend to further my skills in Python (through combination with Julia, with a focus on improving my skills in parallelism.)

It’s slow going though- Julia is quite a bit different to Python, and the IDE selection available is nothing compared to PyCharm. VSCode is great, but it just can’t compete. Here’s an example of some Julia fun-

The XY Model, run to complete disintegration of all vortices.

The Julia code (a translation of my Python code for it) takes about 300 nanoseconds per sequential update. Pythonically (with Numba/fastmath) this took around 6,000 nanoseconds. If you go to contiguous flattened array instead, you go down to 100 nanoseconds in Julia and around 3,000 in Python- still a marked difference. That in and of itself is certainly a motivation to learn it (and given I hope to be able to code stuff that is fast and utilitarian in the future, being able to call Julia from Python and vice-versa is a great prospect!)

On the agenda next is learning how to use distributed computing. Not that useful for a 256^2 array, but I’m hoping to try it out on much bigger ones and in 3D! It’s really exciting to try a new coding language after using the same one for three years.

And here it is! A few days after this post, I’ve done a quick rehash of my MVis 2022 exam. This time, it’s in 3D (not 2D) with *ahem* distributed computing! That being said, it is only on a 256^3 array. Array sizes get big in 3D, data-wise!!! :_: It’s a simulation of three reacting chemical species (1,2,3.) The code for this is up on the GitHub. In this example, I set eight cores to handle plotting, and two for simulation, giving an iteration speed of 400 milliseconds (600 single-threaded) with plotting being the biggest cost as far as simulation goes. The plots were made with Makie meshscatter.

Other than coding though, there’s photography 🙂

I’ve put up the “Corrour” photo album to the photography section of the site. It has my latest (and tbf, some of my greatest) photography, taken two months ago on a short 2-day trip to Corrour, where I stayed at the Loch Ossian Youth Hostel. I’ll definitely be returning there this summer mind you- absolutely stunning place! Here’s a photograph that I don’t believe I’ve included in the photo set put up on the website-

Staring South across the Highlands, the snowy path up Beinn na Lap up the right.

Anyway, I’m not sure what to do now. We’ll see. For now, I have no plans, hopes, or aspirations- maybe that’ll change in the near future. The one hope I can say I might have is to become a teacher or try the JET Programme out in Japan. Teaching in the U.K isn’t great to do though, so that crushes that hope quite a bit. Take care! 🙂