VectorStorm Blog

So I see that Geometry Wars 2 has been released on XBox Live Arcade.  I kind of have to acknowledge it as it’s one of the few other glowy vector graphic games out there right now.  And wow, they’ve really nailed the glowy vector thing in a big way.

Granted, in this revision they’re a bit less vector than they’ve appeared to be in the past;  lots of things have been added which are obviously textures and there are plenty of curved lines (which usually aren’t present in old-school vector graphic games), but the overall look is still very nice.

So in reaction to Geometry Wars 2’s beautiful glows, I’ve been revisiting the glow shader used by VectorStorm games.  Here’s a shot of the current map screen in the in-development MMORPG Tycoon 1.1, using my updated glow shader.  For those who’ve played MMORPG Tycoon, you’ll notice a few interface niceties already;  textual level ranges have vanished at this sort of zoom level and in fact, similarly-zoned regions are being drawn as just one larger region, when we’re zoomed this far out.  Also, there’s a cute little compass rose in the northwest corner of the map.  But really, this is showing off the new, brighter, softer, and much more accurate VectorStorm glow shader. I’m pretty pleased with it.

And it doesn’t even cost much more GPU time.  In fact, the optimisations I made to the code probably actually make it cheaper, on most video cards.  (Although I’ll confess that I haven’t actually profiled it)

With bare minutes to spare, I’ve finished Robot Finds Ice Cream!  Go grab it from the sidebar.

I’m vaguely pleased with how it came out, though I do wish I’d had more time to spend with it.  You’ll notice that it feels a little more rushed than the previous GiaW games;  that shows up mostly in the inconsistant art style, its lack of a main menu, and the rather small number of levels (I’d have been happier if there were about fifteen of them).  But hey, what do you want for something that’s free and both designed and implemented in under a week?

And now I’m going to post this, before the clock turns over to midnight.  Hope you all enjoy! 

So it’s time to talk a little about the Game in a Week that I’m ostensibly working on.

This particular GiaW is interesting, in that of all the GiaWs I have done, I got the final game idea for this one the earliest in the week (Tuesday morning), but left the first implementation the latest.  This was primarily due to work and other real-life factors distracting me.  Also, I was introduced to several neat TV shows this week, all of which have conspired to pull me away from the computer.  Also also, I’ve been a little distracted by MMORPG Tycoon bugfixes and development.  Now I know how Petri feels, trying to do his monthly games while still getting work done on Crayon Physics Deluxe!

Anyhow, this is also the GiaW where I’ve used the most words from the initial quote, in setting the game theme.  The initial quote (from J.K. Rowling’s Harry Potter and the Sorcerer’s Stone, although I’m sure that the same sentence appears in the less deprecatingly-titled Philosopher’s Stone), I’ll remind, was this:

It was also very hard to remember where anything was, because it all seemed to move around a lot.

I’ve pulled words from this passage to form my game’s theme, and the game’s theme is this:  “It is very hard to move around, because it remembers a lot.”

So.  In this game, the player takes on the role of a young boy.  This boy has a robot pet/friend/nanny, and is going shopping with that robot.  Unfortunately, the robot is being a little difficult today;  owing to a minor design flaw in its object-avoidance programming, the robot absolutely refuses to cross its own path, for fear of colliding with itself in the past.

Also unfortunately, the robot has the money.  And so the boy must get the robot to each of the vending machines in each level by pulling on the robot’s leash or shoving the robot, but with the robot never agreeing to cross its own path.

This game is strongly grid-based, and there is no time limit;  each level is a maneuverability puzzle which players must solve to reach the next level.  And it’ll be trivially easy for anyone to create new levels for the game.  And I’m hoping to set it up so that the computer can automatically generate new levels itself.  Though that might be a bit of a push, by the end of the day tomorrow.  We’ll see!

Anyhow, I don’t have a title for the game yet;  possible options I’ve entertained are “Robot Pal”, “Robot Goes Shopping”, and my current favourite, “Robot Finds Ice-Cream”.  So for right now, I’m just calling it “Robot”.

New in 1.0.15:

• Autosave support.  The game automatically saves every two minutes.  Just because not doing that was tempting fate far too much. 
• Fixed a silly display bug which caused subscribers who were online at the time of the last save to not be counted in the “Total Subscribers” count.
• Fixed a bug which caused adventurers not to have their XP return all the way to zero after gaining a level, if the player had turned up the “XP Per Level” design value past 10.

And yes, I’m still working on the “Game in a Week”;  sorry for the lack of updates!  I’ll post an update in a few minutes with where that’s going. 

Also, holy cow:  http://www.wired.com/culture/culturereviews/magazine/16-08/st_wte

I’ve gotten very little coding done in the past week;  mostly feeling a little worn out and uninspired.  So I’m going to kick myself into gear by starting another Game in a Week, as a break from development on MMORPG Tycoon.  Not to fret, though;  I’ll be back on MMORPG Tycoon after this is done. 

It’s past midnight at the moment, and I really need my sleep, so I’m going to keep this short.  The book for this Game in a Week is a hardback edition of “Harry Potter and the Sorcerer’s Stone”, by J.K. Rowling (as though I had to tell you that).  This is a US printing of the book, published by Scholastic Press, and the text is copyrighted 1997.  There probably aren’t many copies of the US-titled version of the book here in Australia, but somehow I wound up with one of them, so it’s the one that I’m using.

As always, my topic selection method is to open a book randomly, and take the first sentence which is completely on the leftmost page.  From this sentence, I must choose one or more words to take as the theme for the game I’m to design and implement within the next seven days.  In this case, I’ve opened to page 132, and the first sentence entirely on that page is:

“It was also very hard to remember where anything was, because it all seemed to move around a lot.”

In my mind, that passage kind of conjures images of a cross between Concentration and the Three Card Monte.  But I’m not nearly cruel enough to inflict such a horror on any of you fine people.  At least, not unless it gets to be Sunday and I haven’t come up with anything better. 

My initial thoughts are that good words for starting with would be “remember”, and “move around”.  And if I twist it a little, maybe I could turn “lot” into “lots”, and do something with small stones.  Or the game could take place in an empty lot.  Or there could just be a lot of things to remember or move around.  So many possibilities!

Anyhow, I’ll post more thoughts tomorrow, when I’ve had proper sleep.

So I’ve been a bit busy the last few days.  I’ve once again shrunk subscribers and monsters (or expanded the world, depending on your point of view), and I’ve increased the maximum number of entities allowed in a region.  The new maximum is 300 per region;  I find that 200 monsters and 100 players per region feels about right.. but this will be wired up to options in the “Design” area, so others can tune things differently.

Subscribers are now drawn using the vector sprites I created for “Ill-Timed Petition Damsel”, and the monsters are drawn using the Muncher sprite from “The Muncher’s Labyrinth”.  I know, I’m totally ripping off my own work.  It’s probably just temporary;  I wanted to see how things looked, animated.

So now if you really zoom in on the map screen, munchers actually chase meat (and petition damsels), instead of just having little geometric symbols slide along the ground.  I think it’s a big improvement.

I’ve also fixed a number of extra AI bugs on both the subscriber and the monster side.  Getting monsters and subscribers to actually approach each other to attack is surprisingly non-trivial.  But it’s pretty much working now.

Devs are also now in the world, though they aren’t really doing anything yet;  they’re just little arrows (like the player’s arrow cursor, only they don’t rotate and they’re fainter) that move around in the world.  Eventually, these will scurry around fulfilling the player’s build orders and assisting players in distress.  But that’s not been implemented yet.  Soon!

In other news, I’m pretty sure that next week I’ll be taking a break from development on MMORPG Tycoon, and doing another Game in a Week.  Need to clear my head a bit with a change of pace.  More on that as we get closer.

Wow, has it really been a full week since my last post to the blog?  Time flies..

This has been a week with some real ups and downs, and it’s kept me away from programming rather a lot.  But even so, there’s been some progress on MMORPG Tycoon.  This week, I have gotten monsters and players to fight each other properly.  Players can theoretically gang up on a monster, or a player can be attacked by several monsters at once.  Monsters chase after nearby players, and respawn after a while, when they’re killed.

What’s more, players now try to run away from monsters who are too dangerous for them, and will totally ignore ones that are too low level for them.  This will probably lower the penalties for having users in an incorrectly-zoned region, since most users will attempt to run away from particularly nasty monsters, rather than fight a hopeless fight.

I’ve also done some more map rendering cleanup.  The game map now sits in an infinite ocean, instead of being a big square in the middle of nothing.  I think that makes it look nicer.

There’s also been a lot of code cleanup behind the scenes, in preparation for adding Devs to the game.

Finally, I’ve been rethinking the “placing individual buildings” thing that I’d been talking about before.  The more I think about it, the more I realise that there’s no really good reason for someone to place just one of the buildings; people would just end up placing one of each type of building for every city that they wanted.  So to minimise micromanagement, it’d make more sense for players to still just place towns, but to instead define a few different types of towns for them to place.  For example, we might have three types:  An “outpost” (contains only a small shop and an inn), a “City” (contains a normal shop, a tavern, and an inn), and a “Castle” (contains a shop, a tavern, an inn, and an auction house).  Each of these would cost substantially more than the previous, and you would need to achieve a certain number of subscribers in order to be able to place them.

Oh, and I’ve also fixed that old bug about the “Inspect” window not clearing some lines of information, if you switched from inspecting a player to inspecting a town.  Yay, UI code.

Next step is Devs.  Then the town building, then the subscriber AI to use the new town buildings, and then we’ll see where we’re at.  So there’s still a little ways to go, but progress is being made. 

So I thought I’d post a little bit with what’s going on.

Between work and mild illness, I haven’t actually gotten much at all done on MMORPG Tycoon v1.1.  But here’s my work log from this week:

1. Substantial modifications to the map and subscriber code, in preparation for rendering monsters and devs on the map.
2. Monsters now spawn on the map, but do not yet interact with users.  (In the 1.0 series, monsters were handled using probability calculations, a bit like random encounters in many single-player JRPGs.  In the 1.1 series, monsters now have individual placements on the map, will attack nearby subscribers, and will respawn after being killed, as generally happens in real MMORPGs)
3. Made substantial (CPU-side) performance improvements to the graphic engine.
4. Added a nice effect to the roads when loading into a game, so that they don’t just “pop” into place any more.
5. Small modifications to map rendering, to make it more GPU friendly and less cluttered for complicated maps.
6. Began work modifying the “Build” toolbar to support the new building types (resizing and moving buttons, etc).
7. Dramatically simplified building handling code, to make it easier to add new building types.
8. OS X build now saves preference files in ~/Library/Preferences and save files in ~/Library/Application Support/MMORPG Tycoon, as Apple recommends.  This should stop occasional user permissions problems with preferences not saving correctly in multi-user OS X situations, and will also make it much easier for OS X folks not to lose their saved games when updating versions.

Note that this is all going into the 1.1 build;  there is no new build available for download yet.  It’s just work in progress.

Anyhow, hopefully the progress summary will be a bit longer, next week!