Somnid

Version 0.08

2012-08-25T17:36:00.002-07:00

So I've finally gone through a major refactor of my code. I decided that this release will focus on improvements to make the engine more viable in case I decide to switch gears to another project.

There are several major improvements that should provide a major performance boost.

First, I'm clipping the tile bounds before a screen draw, this decouples the framerate from the map size meaning I can start moving to arbitrary sizes which is good because the game would never have worked if I couldn't do that. This also massively improves performance in general.

Second, I'm doing an offscreen draw and then posting that to the main canvas. This should reduce the amount of draw calls to the GPU as well as reduce refresh artifacts caused by the on screen canvas getting hundreds of tile draws per real frame.

Third, the engine itself has been cleaned out so there is a lot less junk. This meant some features were removed as I didn't want to re-implement everything in one go. So the menu is out but it'll probably return HTML native rather than canvas because it's easier. I also decoupled all the drawing from the engine which is great because now I can swap out the rendering pieces as I need. So if in the future I wanted to use Web GL to squeeze some more performance I can do so without altering the game engine itself.

Also compatibility checks don't work anymore. I'll revisit browser compatibility later as it's a bit too taxing right now (though it's built with compatibility in mind I'm likely missing prefixed functions and CSS). Everything should work in the stable release of Chrome.

Fourth, was utilizing the whole window. I think it made a lot of sense to use the whole window by default and it was necessary to test on different resolution to make sure my new enhancements actually work. It also gets away from the test phase and moves it into a more user-oriented version. It also tolerates window resizing and compensates by changing the buffer size.

Fifth, was to remove debugging interfaces and move them to a new window (press the '~' key ). This was because I didn't want to clutter up my newly full-sized screen with debug crap but also so I don't need that code in the window either. Now it resides in it's own javascript and an external HTML file that can be updated independently of the game engine and interface. The engine just passes a debug info object to the new debug window and that window will display it. It's simple now but I could turn it into a sort of debug console too in the future or a visual debugger for terrain generation or something. All it does right now is show the frame rate.

Performance-wise I get 15 - 30FPS on Chrome at a resolution of about 1600 x 2560 which for now is good but I might have issues once more layers go in. I'm thinking I can make the tiles bigger and that will reduce the draws (64 x 64 tiles would reduce the backbuffer draws by a factor of 4). The other route would be zooming by a factor of 2. Either way something like that will be implemented since at the current size I think things feel a bit small. I can also go full screen which should help too as you won't draw your desktop background and other windows. These features/changes will probably be in the next release and I hope to work a little more on the art.

Anyway this build may not be the most stable due to all the refactoring but it should feel more like an actual game than a test bed and the engine should be better all around.

http://somnid.github.com/Planetite/

Also, there's now a page for current funding. I found a dime in the locker room the other day. Getting money with no money and no begging is hard.

Version 0.07a

2012-08-04T21:03:00.002-07:00

So I finally found myself with some time in between Xenoblade sessions to do some work. However, between now and the last time I touched any code I've kinda changed some feelings. So this update is actually a regression.

Basically changes from 0.07 were reverted. This was because I felt those features were just distracting from the core codebase which needed work. Custom soundtracks was interesting but it was there primarily because I 'm no composer. I think at least for right now it's fine not to have sound and there's somethings vendors need to work out before I try audio again anyway (Chrome, FF and IE have very different audio feature sets). As for ads, they will return but not right now. I'm kinda soured on click ads because I know people won't click them and it's really annoying to see them on websites. Perhaps I need a userbase before I can really do anything here. And I wasn't feeling the settings UI nor is it really necessary at this point.

Instead 0.07a is some stuff from 0.06 that's been refactored and I've also removed all traces jquery (It was for debugging and didn't need to be there). There are no feature enhancements but the code is slightly more organized and I think this will help going forward. There's a lot more work to do on that end but hopefully I can get the core engine and rendering in good shape before I move on. The remaining programmatic slog is partitioning the space so I'm not trying to manipulate the entire map at once which will greatly improve performance and make the game actually feasible. As to if I'll have more time/desire to work, who knows?

http://somnid.github.com/Planetite/

Oh and at least as far as Chrome App Store fees I've collected $0.07 from the ground. Not sure how I'll get to $5 like that...

A little break

2012-04-29T21:06:00.002-07:00

Lately I've had a torrent of interesting app ideas and so I've been taking a little break to do some brain dumping. I've been able to build out a couple javascript libraries that aren't really associated with the project but may come in handy someday. Hopefully this idea purge should let me get back on task. My next couple items are likely going to be terrain generation followed by a new coat of paint.

One thing I looked into on this little excursion was Google Drive integration for cloud saves. It doesn't look like I can use this because I can't make cross-site requests on a website and ironically it won't work for a Chrome App because Google's Oauth callback urls don't accept Chrome:// urls. At least in the Chrome App world Dropbox is still an option though work on the Oauth client is slow and difficult however it seems there might be a couple of libraries now available that I could take a look at.

New Name

2012-03-26T20:19:00.002-07:00

I've been thinking hard for a while about names because I feel to move forward I need to start branding a little bit. So after some thought I've finally come up with a tentative title "Planetite." Essentially I wanted to get across the point of the game in the title since most people aren't going to get much more than that before they stop paying attention. The name is meant to reflect the space setting with resource collection. Other reasons for choosing it: I personally like made-up words that sound like they could be real (hence my username). One word means it's hard to abbreviate so if someone posts about it they have to use the whole title and not obscure it for people who haven't heard of it. In any case I like it a lot more than the old code name "caverns."

Due to the name change I might be renaming and moving stuff not least of which is the game itself. It will now be located at:

http://somnid.github.com/Planetite/

Luckily I'm sure nobody has bookmarked it yet...

Version 0.07 Released!

2012-03-25T14:40:00.003-07:00

Yesterday I released revision 0.07. This version primarily deals with audio. The main areas here were SFX and background music and I took different approaches for each.

SFX

I have no real experience creating my own sound effects so for this iteration I simply grabbed some off the internet that allow me to have a commercial license for free. Credits are given if you click the "License" link so you can see who the effects came from. SFX are implemented in two different ways if you check the AudioPlayer.js file. The baseline is HTML 5 audio. This is pretty simple but has some downsides (notably that I can only play once per object so to here quick succession digs I'd need multiple objects to play). Since this wasn't that important I'm simply resetting the element each time it's called to be played. It cuts off but I doubt anyone notices.

There's also a WebAudio version for use in Chrome. It seems it broke somewhere along the way (probably a bug in my compatibility checking). However this scheme is more powerful and I could for instance add locational sound effects using it as well as filters.

Music

I'm not a musician, I've never played an instrument so I'm clueless as far as composing goes. Instead I added something you probably won't see in any other web-game, custom soundtracks. In the audio panel you can drag-and-drop music and this will create a playlist that loops while you play the game. The downside is I can't keep that music between refreshes so you need to do it each time. I'm thinking about using the File System API to save copies so that you don't need to load each time. Not sure if people care but it's there. I'd rather tap into some other public audio API since I want the app to be as cloud-centric as possible but I've seen nothing so far.

Currently Firefox has issues with soundtracks I'm unable to figure out. If you have ideas let me know!

Ads

Another addition is Google Adsense ads. You may turn them off in the page settings panel as I don't want them getting distractive, in the way or reducing precious battery life if you don't want them to. I know most don't care and won't click them so there's no point forcing them down people's throats after all. If you do see something you like and click it note that this will help me pay for the Chrome Web Store license ($5) and that gives me more options to monetize so support is appreciated.

At this point if you know what you're doing you can also turn this into a Chrome App by downloading the manifest.json file (in root) and packaging it up yourself.

I hope you enjoy this release (though the gameplay is still non-existent). If you have any comments or suggestions please post them. I'd love any feedback I can get.

Also note I'm taking any suggestions on topics you'd like to see here (like "How did you do that?") so post those as well!

Thanks

http://somnid.github.com/Caverns/

Welcome to Caverns (Working title)

2012-03-19T18:53:00.001-07:00

This space will become a blog for my web game under development currently code-named "Caverns."

For this first piece I'd like to make an about page for the project so that people finding it can learn more about it and what I intend to do with it.

Where it started:

Caverns started a little bit after I started playing Terraria and I have taken much influence from it. While I think Terraria is a fantastic game there were a few things I found a bit lacking, but moreover being a web developer looking at the game it seemed to me that there was no reason a game like that couldn't be done totally in browser. I had been looking to start up a new 2D engine using javascript and the idea seemed like a good project to start with.

After a bit of work I thought a bit harder about what I was doing and where I wanted to go with it. Games and the game industry are important to me and I've always wanted to do game development. So with this in mind I also thought about how I could make a game that represents to me some of the ideals that I feel other devs should strive to attain. Also, I started thinking about software projects in general and wanted to try some new models for development and management even if it would just be for myself.

One big such goal I set for myself was to create a game with no budget (and by that I mean I will not donate a single dollar to my own project, it's all done for free). Admittedly, I have a computer (and a rather nice one at that) and I pay for my electricity and internet. However, I argue these are incidental costs, I'd have them anyway and for anyone wanting to follow in my footsteps it's pretty reasonable to expect them to have at least these things (Access to computers, uncensored internet and broadband for all is quite another beast I'm not going to tackle this time). However, I'm not using any software that isn't freely and easily available or at least bundled into Windows. I'm using free services to host my projects and APIs that don't require payment, at least at the starting levels. My hope it to raise an initial $5 via ads to pay for a Chrome Web Store developer license and then I'll be able to further monetize my game.

On this note I'd also like to address my thoughts on new developments like Kickstarter. While I think the idea is great I do not wish to use such a service. For me, it generally feels like panhandling. Certainly I feel it's possible to make something good without having to beg for money. While I can't say I wouldn't accept donations or build in avenues to donate to my project I feel it should be for services and content rendered, non-invasive, and with no expectation of anything on my part or the part of the donator.

Another thing I'd like to use this project for it to demonstrate the web platform. I feel too many budding devs have their hearts set on consoles, Steam or mobile app stores. While not to diminish the success many developers have had on these platforms I feel some people are too platform-centric even though the games they make aren't really utilizing the platform itself. I'd like what I make to be more platform agnostic unless it had a good reason to be on one or the other. Also the web platform is very powerful and can do all the things software does on these others with many other benefits including: platform agnosticism, no need to install, play anywhere, cloud native, social integration and many other free and well supported APIs for ads and analytics. It pains me to think some people still see the web as simple plugin based games.

The last part is agile development. Games have been one of the last software markets to climb on board mainly because they garner hype from secrecy and releasing all at once. However, devs like Relogic and Mojang who have left this behind have found great success taking a small and perhaps buggy game and iterating it into an awesome content powerhouse. I think more indies should adopt this model, bring in funding with a product, not promises and make it better for the fans.

So with this I present the humble beginnings of "Caverns." While it's undoubtedly going to be rough I hope you can enjoy or get inspired by it. Feel free to take any code you find for your own projects as well.

To the app: http://somnid.github.com/Caverns/