Monday, 22 June 2020


So it's been 8 months since my last post. Yay, blogging!

There's too much to talk about, so I'll just mention my latest project - following along with in my own way - TypeScript with no library code. My repo is at I hope to provide updates here as I think of things and make progress.

Tuesday, 29 October 2019

I've been feeling pretty crap recently. Nothing really seems to be happening. I make progress in my own projects but it's never enough to make me feel like something has been achieved.

A small highlight has been my GTASC team's continuing good progress. We aren't expecting to win (there are some super devoted people on there, and all three of us work full time) but we are enjoying being strong contenders. We're in the top 20% at least!

One of my projects is the disassembly of the first Super Robot Wars game, and the disassembly tool itself written in C#. It's hard going, but this is the kind of stuff I live for - pointless archaeology of old video games, discovering how things work. This kind of thing has been on my mind since I was translating Wizardry Empire, but I also got inspiration from this complete disassembly of Pokémon Red/Blue.

Neverseed is also slowly coming along. I don't know why I find it so difficult to work on - last time I tried to add a feature, I ended up translating the code into Typescript instead. Hopefully we'll be able to release some kind of demo this year. That sentence seems eerily familiar...

Friday, 6 September 2019

More fun with Japanese

Hi! I'm on holiday right now, on the way to the very North parts of Scotland, so this will be brief. I just wanted to mention this book I bought:

This is more than a little hopeful on my part; while I'm level 18 on WaniKani and therefore know a few hundred kanji and a couple thousand words, I couldn't even completely read the cover!

The main problem is my very loose grasp on Japanese grammar. I know bits and pieces from my university time and what I've looked up online, but after a couple of months it gets muddier and I need a refresher.

The Japanese on the cover is 日本語の短篇小説. Here's how my mind processes that currently:

  • 日本語 is "Japanese language" - nihongo
  • の is the association particle, something like "'s" or a reverse-positioned "of" - no
  • 短 means "short" - tan
  • 篇 ???? I had to look this one up, it means "volume" or "collection" - hen
  • 小説 means "novel" in the sense of "short story" - shousetsu
So, I read it as "collection of Japanese short stories". The first in the book is by the famous Murakami and is indeed very short; only a page and a half! However, the Japanese text remains annoyingly out of reach. I can read many of the kanji words, but the grammar and hiragana-only words are confusing. It may be another year or two before I can read prose, with some dedicated study.

My team is still in the GTASC! We were never really in any danger, but it's getting harder. Especially because I didn't bring my Xbox to Scotland - no room, plus these things don't travel very well. Still, if I can continue to beat large parts of the Bard's Tale Trilogy, I should be fine.

Tuesday, 2 July 2019

Quick pre-holiday update

Today, WaniKani taught me that the kanji for bento is 弁当, which is literally "dialect correct". What a bizarre and beautiful language.

Yesterday, the GTASC began. I've never taken part before but a friend asked me to join his team, so here goes! It's going to be interesting earning gamerscore while in America. Mind you, my whole team is on holiday for parts of July, so hopefully we don't get knocked out too quickly.

I published my Forth on GitHub. It's not done yet - it won't even run on a TS2068, the target machine! It does fine in a Spectrum emulator/assembler I found, but I still haven't figured out how to get it on a tape/cart to run on other emulators. I wonder if I can write the code with conditional compilation so it'll run on both machines.

I wonder if I'll have to write a TS2068 emulator just to debug it.

Thursday, 20 June 2019

Posting in 2019 LUL

So it's been a long time! I've been both very busy and pretty lazy. Here's a run down of my current situation.

DOSJUN is still alive, but in a different form: while I plan to release the full scenario eventually, for now I'm going to concentrate on releasing a smaller and more focused game which will run in the same engine. Hopefully this will help highlight any problems before they become huge issues. Some work has already been done on the content and mechanics for this smaller dungeon.

I'm working on a fairly small platformer that will run in the browser. It's certainly a big change from DOS's limitations! I expect that project might release early next year. It would be nice to finish something :)

I'm very recent news, I've been writing a FORTH implementation designed to run on the Z80 chip. This is just for my own curiosity, though I have access to a real Timex Sinclair 2068 and it would be neat to see it in action. I'm almost at the point where users can type in real programs (like the built in BASIC interpreter mode) but not quite. Once it does that, I'll put it on GitHub and everyone (nobody) can shout at me for writing terrible assembly.

I also just realised it could run on the Gameboy with some modifications, like not having keyboard input. Huh.

I've been spending a fair bit of time on the Xbox this year because I set myself a score challenge. I'm also going to America again in a few weeks so I want to get even more ahead so there's no crunch time later in the year.

I've been semi-regularly streaming all kinds of things on my Twitch channel.

I finally finished my League of Legends placements for this season and got put in Bronze 2. I haven't played any Ranked since.

That's about it. Sorry for being quiet for so long!

Wednesday, 11 April 2018

Tiny 2018 Update

Haha, I've been quiet for almost 2 months!

I'm in America right now, but I've still been ploughing on with DOSJUN. I've added music support for a certain tracker format, wrote new graphics loading code to hopefully avoid some of those horrible "out of memory" errors and done a lot of planning for the setting and denizens.

I realise that there's still plenty of work to be done. However, it is slightly further away from "I want to make a game" than it was. The scope is starting to creep a little, but I'm still fairly confident that I can release a playable version of the game this year. I'm thinking that I'll come back and add more content later, because J and I have come up with some excellent ideas to make the game deeper.

Friday, 26 January 2018


DOSJUN is alive and well. Here's a quick look at the current progress.

You can walk around

You can load and save (in safe spots)

There are scripted events

You can fight random encounters

Technically, there's a lot going on in the background to deal with this short list.

A first person view that uses slices of textures

A scripting language (and compiler)

A dungeon editor (items and monsters too!)

It is my eventual goal to not only release this game when finished, but also annotate the code well enough that people can actually learn from it.

Progress has been slow for a variety of reasons: I play a lot of games (mostly long ones), I don't always want to program after doing it for hours at my day job, I watch people stream gameplay quite a bit. These things have been true for years, but I want to commit to this project a little harder than before.

I will release the game this year. At the very least, I can release a version then add more content later if the story remains unfinished. There will be at least 5 hours of work every week on it. Making the engine easy to extend has been important to me since the beginning; an example post-release task would be removing the hard-coded character creation and using scripting instead.

I'm also thinking about streaming the programming work. This might seem odd, but I recently caught someone using Rust to write a roguelike and enjoyed the experience, so there must be like minds out there. It should be a novelty for many current devs to see C and DOS!