Commander X16

Hi all,

I started developing a game for the X16 last September - coming up on a year ago now! I come from same the 1980s 8-bit BASIC programming vintage as Director-in-Chief Murray and probably most of you lot too, but I've never developed a whole game before. The X16 project has inspired me to learn assembly, with the goal of writing this game. More than a whiff of nostalgia about it all too - fond memories of passing POKEs to schoolmates to get infinite lives in Manic Miner and Jet Set Willy.

Loosely, my game is a turn-based strategy/resource management thing with a passing resemblance to the hex-based board game settlers of catan. But the resemblance is only skin deep - this game has a strong story-based adventuring/survival/exploration leaning, all mixed together and served up with a hearty dose of good old fashioned text adventure.

Blimey, I'll tell you what - it's been a steep curve. Simultaneously learning assembly; learning the vera / X16 hardware whilst developing how the game is going to work: combat systems - resource / asset / fatigue management - display; writing the code ; creating the graphics; writing the prose. It turns out taking on a game (even an 8-bit style game) single-handedly is a HUGE undertaking. I doff my hat to @SlithyMatt - you sir are a legend I've no idea how you churn out the code with such amazing regularity!! But the great thing about the X16 is - IT CAN BE DONE. It might take ages, but if I can do it, anyone can!

It's been a long time coming, but I've got to the stage now where I've finalised the gameplay, memory management, and the overarching story of the game. I've also battered my head against my assembly inadequacies sufficiently (with lots of help from proper programmers - again, hat tip to @SlithyMatt , @StephenHorn, @Greg King , @togster510 and others - sorry if I've missed you out!) such that the code for the main game loop is now (pretty much!) in place.

Things I've learned so far:

1. Planning everything out on paper before starting to code ABSOLUTELY VITAL! I was keen to get into the 'interesting stuff' straight away (drawing the graphics, putting things on screen) but in the long run, having the whole game pretty much drawn out in principle on paper first meant I've avoided a number of unpleasantries in the coding thereof. How are you going to address the screen - one layer? two layers? What is going onto each layer? How many bpp will you need for each layer? How many sprites will you need? How many frames of animation? How much vram will all that take? How are you going to encode the various aspects of gameplay? Which memory banks are you going to put them in? Which leads me onto -

2. Plan out your memory management. The X16 only has 40k of low memory + 64 x 8k memory banks to play with (in the base model) plus 128kB of video memory, so you can't just splurge on huge 256 colour graphics all over the place - the limitations of the system require some thought on how much you're going to fit in, and how you're going to fit it in. I've used up three memory banks just storing the hex tiles for the whole game board (64 x 64 hexes in total, although not all visible on screen at once) - made up of 32 different types of hex terrain, each with its own individual replenishing resources, roads, rivers and bridges. 

3. Assembly is HARD but not IMPOSSIBLE. I've lost count the number of times my eyes have glazed over whilst looking at what I've lda'd and what I've sta'd wondering why the hell it doesn't do what I've CLEARLY just told it to do... persevere, and post questions on the software support forum. There are kind people out there who will help you (me included, if it's within my power to do so!!)

4. You will probably end up writing programs that help you write your program. This one was a bit of a surprise for me - but I've done it twice so far already. For example, I wrote a bitmap converter that loads up 16x16 pixel .bmp graphic files I've drawn in photoshop, and it spews them into memory as four sequential 8x8 tiles for storing in the vram tile map (my graphics are 16x16 but they need to be placed on an 8x8 tile grid because of how the hexes work). It took me an afternoon to write but it would have been immensely complicated and taken a LOT longer to hand-convert every terrain and item graphic in a hex-editor! 

5. Things change. Roll with the punches.  If the gameplay has to change slightly to fit within the limitations, so be it. The people playing the finished game aren't bothered about what you thought the game was going to be like. Deciding on the layout of the main screen was a hassle for me - there's more stuff I wanted to show than there was room for, and I tried various ways of cramming it all in, whilst still getting it to look attractive. Eventually I decided I couldn't do it, so I've split the information across two screens that can be toggled between. The main screen now shows a bar that fills up as the amount of stuff you're carrying increases. A handful of grain fills up one backpack 'slot'; an apple fills up 3, and one load of stone fills 16 slots etc. so you have to be careful about choosing what to carry around with you. But you can toggle to the backpack screen that itemises all the resources and equipment you're carrying so you know for example how many apples you've got and how much stone you're carrying etc.



I should add this game is a bit of a nod towards the now legendary Planet X2 -  if you've got half an hour spare, David's 'making of' video https://www.youtube.com/watch?v=NB_VBl7ut9Y is really helpful and has some great tips particularly about memory management - it's aimed specifically at the C64, but a lot of it is pertinent to the X16 too.

I hope this is of some help for others considering embarking on a similar coding journey. At the very least, I'm writing all this down now in the hope that you lovely X16 people will hold me to account and spur me on to actually finish writing this blasted game..! ? 

The hexes await... in the meantime, have a peek at the notes in the photos below, it'll give you a flavour of what kind of things are included in the game. Also a screenshot of how the game currently looks.

Ta ta for now!

Really good report. I hope I'll find the time to eventually do as you did and create my own game.

While we're here, what do you use to create some of your notes? They really look well done.

One thing that have struck me when struggling with my game project is that it’s a huge difference between 1 and 8 MHZ. What everybody with experience from C64 say is that every CPU cycle count. But with the X16 you can afford to structure your code well, have a lot of subroutines that do one thing, avoid global variables etc. Right know I am using between 5 and 14 percent of the available cycles each frame (1/60th of a second). And it is worth noting that updating the graphics takes most of the time, at least for me. It is probably generally true that there is no point to optimize the use of cycles in game logic. There shouldn’t be any problems before you make games with rather advanced graphics like 3D rendering or scaling objects in real time.

On 7/30/2020 at 8:18 PM, Fern+ said:

Really good report. I hope I'll find the time to eventually do as you did and create my own game.



While we're here, what do you use to create some of your notes? They really look well done.

Thanks, I made the notes in Corel draw, which is a graphics package I use in my day job so I'm very familiar with it. Use whatever you're familiar with! Good luck with getting your own game going