Gogo Quadtree

Hey, got managed to get my quadtree scripts working correctly, the following image was rendered in FF3.0 using the canvas tag to graphically display the information. Admitedly this whole block of functionality was evil nasty feature creep, but now that it is almost complete I have a nice warm feeling inside that you only get form writing something you've never done before :D

The above image took about 1000ms to process, the 1000ms includes loading and processing the whole level. This basically involves a generating 1024x1024 2d array and then populating that array with 1000 rasterised polygonal shapes. Each of these shapes are around 64x64 cells of data. As you can tell quite a lot of processing is done within those 1000ms. It's also worth noting that the map only needs to load once per level/zone/area, meaning its very quick considering some desktop games take minutes to load.
Go me :D 03:16 PM | 0 Comments

/me Goes over the top

Right, delays and going over the top, /me hugs feature creep :D
Basically I've decided to take this experiment in JavaScript programming to a new level, I am currently implementing a Quadtree, the goal for this is to optimise the A* path finding, so that each level can be larger, as there is less data to sift through, meaning bigger maps :D.
So what does this mean, no video yet :D but when the video does land the underpinning technology will be all the more impressive, and hopefully more speedy, or just bigger :D
So far its looking quite possible that the maps can be quite large :D, though there is still the problem of streaming blocks of map data down from the server to the browser. One possible solution to avoid any bandwidth issues due to people streaming map content from my domain is Amazon's S3 service, which once everything is working on my development box I will be looking into :D So with the current feature creep should be around the time of the apocalypse. 02:01 AM | 0 Comments

Ctrl+Z 'Given up :('

Hey, everyone who reads this trash, I've decided to re-re-re-start work on my 'Java Script' game. For those interested, the reason behind this monumental decision on Friday the 11th like all 'Apple-whores™' I fully intend on buying a iPhone 3G, and what better way of motivating myself than to write an application that I can actually touch and use :D. Also I need a reason to justify the £729* it'll cost me.

Now within this new state of fanboyisum I've decided to roll the idea back to something vastly more simplistic. Rather than trying to write an MMO-esk game I've opted to go for something more realistic. My goal is to write a Bomberman clone, then alter/hack the basic game-play mechanics to try and craft something fresh and original. Who knows once a single player version is up and running I may have time to add multi-player, though that's likely to blow up in my face :D
I've managed to get smooth translation working within the 'Engine' so my goal now is to link in my existing path finding stuff and hopefully, within the next few days I'll be able to post a video of some really ugly sprite running around a really simple maze :D

*(That's £99 for the phone and 18 payments of £35, though £729 looks more impressive.) 07:16 AM | 2 Comments