Xerra's Blog

Drunken coder ramblings

Time to start work on a new game —

I’ve been thinking about starting a new game outside of the competition ones that I’ve been doing for the last couple of years finally.

The deadlines that have been imposed from writing competition games has increased my productivity massively during this time and, for the Six competitions I worked on a game, I delivered for Five of them. I even got placed Third and First in two of them which was a massive boost aside from just getting the games themselves completed.

So, now that Creepy Crawlies is finished and there’s not been the need to do any follow up work on it, I’ve been thinking on what to do next. I’ve had a couple of months lay off really, and have done very little coding apart from tweaking to my project file that I’ve tended to improve and add to after each game. It’ll probably never be complete but every game always ends up giving something back to it in a new feature or two, and the development work is much quicker to get going by using it.

Over the last week I’ve come up with three things that I’m considering doing – again with Gamemaker 2 because why not?

Firstly I wanted to do another one of my Vic 20 remasters that I’ve harped on about before. As the missus bought me some more old tapes for display purposes this Xmas, I considered doing Crazy Cavey just so I can say I’ve finally done a platform game. I did a bit of prototyping for this yesterday and, while I’m still keen to throw myself into that, I remembered I’ve not actually finished the first Vic game I was working on almost a year back which I shelved so I could do Damnation Alley. So, if I’m going to get on with a Vic remaster then it should be finishing that off so I’m not going back to my old habit of half-finished games all over the place.

And the second game I came up with is something completely different and not based on any game I’ve played before on computer, although it does have a vague link to the board game Cluedo, if anything. I’ve done a lot of writing ideas and creating a map structure for doing this and have a pretty good idea how it would all come together as a complete game despite not writing a line of code yet. However, it’s an ambitious game compared to the stuff I’ve done up until now and I really feel the urge to do something a bit more conventional first so I’ve written it all up, will be creating a proper map, and borrowing some of the heavy script and database functions I wrote for Bah, Humbug to help me get started on that at a later date.

This nicely brings me onto my third idea which has been incubating since both my nephews have told me how much they’ve been playing Rockman and giving me ideas about what works and what doesn’t. It’s now been a Eighteen months since I put that game to bed and released it and, for all its flaws, it’s still one i come back and play myself every so often. Usually just in the puzzle (easy) mode as I’ve come to realise I pretty much fucked up having bad guys in and making medium and hard levels kind of pointless because it was then almost impossible. In fact, just last night I did a play-through which I considered recording, where I was going to play through every screen and complete the game just to show my nephews that it can be done. Those two are too much into Switch and easier games to probably actually care, but the idea of doing another Rockman and making it much, much better would be quite good fun. I wouldn’t need to actually rush it this time but I also wouldn’t have to remove half of my ideas from the design doc just so I could make it in the Six weeks deadline the original game had.

So here we are and I’ve made a tentative start this evening to get this underway. I had two approaches for this. First I’d load up the original project, rip out all the stuff in the original that I didn’t want to keep and then tidy up the project, import some framework code that I had created or updated since doing this game, and then get to work redoing all the bits that I wanted to change. However, while Rockman was written with the same framework I’m using now, it’s a much older version of it, and there’s been a huge amount of changes to some of the functions and scripts that other parts of the framework relied on which caused no end of compatibility issues. I muddled along fixing stuff, disabling bits, writing little bits of middleware or just modifying function calls and importing replacement files to try and bodge it all to a working state at least but it ended up just an effort in frustration. Even simple stuff like scripts for sound routines had changed since I’d written the original game so I took the second approach instead.

There’s still a lot of functionality my basic framework needs, even after using it for so many games, but I’ve always worked with it as it is at the time I start a new game purely because it’s always straight after finishing a game that a lot of work goes back into the framework itself. Damnation Alley helped get a lot of the retro functions in place which I developed for that game and then put into the framework afterwards. The high score and configuration file i/o stuff from Creepy crawlies has also been copied out and into it now. As Bah, Humbug was a completely different game to the other stuff I’ve done, there’s also a lot of string handling and lists code in it too. Even incomplete projects like RAM have given me code that can go into the framework in case I ever do another card game. So I’ve loaded up my framework and just pulled in some of the objects, graphics and scripts from Rockman that can still be used, and will just rewrite some of this stuff to save a lot of development time. As the sequel is going to have a few new things, such as an editor for starters (I really wish I’d had the time to do that for the first game), I’ll still be doing a shit load of development from scratch.

So a rough list of what I’ll be adding to the sequel to Rockman that I’m tentatively calling “Rockman 2 – Jewel Thieves”

Firstly an Editor. I’m thinking of a bank system so you can load in up to 26 levels (A-Z) from each of the five banks but can only overwrite 2,3,4 and 5 so the original maps are always present. As I’d keep the original map structure for the first bank I’d probably need to have a system like this in place so I can make some new levels where the additional features can be used without breaking them in some way. If I work it out right then the player can also have some kind of system to put screens together and set up their own map from the editor and they can have as few or many screens (up to the limit) as they want for each map as I could have them loadable as text files, or similar.

An update to practically every object in the game, specifically the rocks and player themselves. The way Boulderdash does it works. I’ve written a clone of Boulderdash before and it played really well because I tried to make it just like the original game. With Rockman I switched to pixel movement to simulate sprites instead of character graphics and, as well as being a nightmare to code, also didn’t turn out that well in certain cave layouts where rocks could bug out. The map “Valentine” shows this very well but it’s not the only problem area. Also a lot of players had problems with not being able to move up under a rock but it seemed the right way to go when I was writing the game. So block movement comes in so there’s a speed increase, which will make it a lot more fun, and I’ll balance it to make sure it’s not too tricky. This will also solve the problem of my enemies being too dangerous because I will keep them moving in pixels so it’s a lot easier to avoid them. Or let the dog deal with the buggers – more on that later.

I’ll go back to the artist for some of the graphic stuff that Rockman used and probably get a bit more done. Some of the sprites can stay as they are but I’m sure some new intro screens showing the dog and anything else on the go would be a good idea. Animation and stuff was very lacking in the original game. Hopefully it won’t be too expensive.

Some of the basics can go into the game this time around. High scores, saving configuration data, maybe even a save game system so you haven’t got to crack up to 26 screens in one go. Certainly a map improvement so you can see where you are in the layout. The first game just superimposed an image of the map layout with no clue as to what screen you were on. Rush job that was too. Think the map was the last bit to go in an hour before deadline.

I’m thinking a new char needed to go into the game to add to Roxy and Rocky. So a pet dog is what my nephew suggested. How to implement it is another matter, let alone thinking of a good name, but i’m thinking he could be in a cage on the map and you have to release him. Once free he can run around and deal with the aliens if he runs into them. I’ll need to be working with the same map size that the original game had if I want to retain the original levels so this may need some thought.

Two new objects that have come to mind that can go into the newer levels are the compressor, which turns rocks into diamonds if they fall into it, and the muncher, which just eats anything that hits it full stop. Maybe some more if I come up with them while I’m working on it.

So far I’m at the point where i have all the old elements into the new framework and have patched it up enough so it at least runs. So the next session is going to be developing the editor before anything else as the levels will need to be made using it. I can’t go back to using them from the original game as each map was a seperate game room, which is a hell of a waste of resources doing it that way. I want an editor that will save a map as a text file that can be read both in and out before I start creating anything more than just test designs.


Categorised as: Uncategorised



Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.