The contest seems fairly standard, but there is a catch that might bother some game creators, as Garriott is asking all participating developers to cede their rights to their work to him, going so far as to require them to add “© 1977-2014 Richard Garriott” to their submissions.

Dick move, dude.

Fuck yeah.

cabbagebot:

kr-studios:

I need to have two builds available for my game: the stable release build, and the most recent nightly build. Also, I could add in linux builds, since I’ve got a live USB floating around somewhere. I’ll tag this as a TODO. Problem is, I’ve actually got a lot of TODO comments scattered throughout my code, I probably need to write a utility just to find them all.

I removed the entity system by picking the diff apart for the server’s source file. Here’s the repo, but there’s no visible difference from the stable build.

Anyway, I have work tomorrow, so I guess I need to go to bed :(

Since the game is still in development and not near release I’d suggest not even worrying about something like this.

Keep your master branch in git in a compiling state and any time you want to add a feature make a new branch, build the feature, and merge it with master when it’s all done/compiling. This is essentially what you want except… for you and not other people.

Sidenote, a lot of popular IDEs already have // TODO finder utilities

I never thought about an IDE based util, so I googled mine and found this. Thanks!!!

Whenever I feel down, I just need to remember that I’m not this guy.

I need to have two builds available for my game: the stable release build, and the most recent nightly build. Also, I could add in linux builds, since I’ve got a live USB floating around somewhere. I’ll tag this as a TODO. Problem is, I’ve actually got a lot of TODO comments scattered throughout my code, I probably need to write a utility just to find them all.

I removed the entity system by picking the diff apart for the server’s source file. Here’s the repo, but there’s no visible difference from the stable build.

Anyway, I have work tomorrow, so I guess I need to go to bed :(

A difficult conclusion: The entity system was a bad idea, so now I’m going to go back and undo it, while also trying to preserve the good tweaks (like comments, etc).

I really don’t know what to do next for my game. Should I… should I keep the entity structure? It’d would only need one revert to undo it, so…

The logic itself hasn’t changed, it’s just created a centralized table for all existing objects in the world (the entities), with each element referencing the extra data of that entity (by holding the uid & type of the object). This will theoretically allow the existing logic to handle more than just the player objects.

For some reason I’m having trouble figuring out if that was the right option. A few minor programming gripes is that now I have a few static integers tracking the uid of each object type. 

The next stage would be to adjust the networking and client code, to mesh better with the new system. normally I’d stick with my previous mantra of “make it work”, but doing that would mean reverting to the previous system and just stacking everything else on top. I need to implement this entity system to (hopefully) make things more logical.

Fun fact: sublime text 2 colours the text that gets dumped to a file from a diff. Sweet!

Fun fact: sublime text 2 colours the text that gets dumped to a file from a diff. Sweet!

I just stumbled across the Trillek project again, and I realized how fucking disorganized these guys are. Trillek is based on Notch’s aborted 0x10c, which means that they’ve poured their effort into the computer specs, without really thinking about anything else simply because that’s what Notch started on.

I offered my help as a producer early in the project, because it looked like there was nobody taking that position. Of course, that was during the initial influx of noobs after the media attention, so of course I was passed over. Here’s the advice of someone who doesn’t care:

  1. There’s a reason 0x10c was aborted.
  2. Although there’s a few well done open source games, they’re done mostly by a central team.
  3. Making an MMO is hard.

invaderxir:

thedistanceinsidious:

kr-studios:

thedistanceinsidious:

Apparently I’ve downloaded 60gb of the important 100gb of these projects. I literally don’t even know what the bulk of this is besides a FUCKLOAD of assets.

Personally, I don’t think its the best idea to store assets in…

branches are for different versions, not for different sections. i.e. you wouldn’t use them to separate parts of your project but to separate what you’re working on from what works.

From a single point of origin, yes. However, personally I have an orphan branch for my documentation, and there’s actually a common system where you store the shared libraries, etc. in a separate branch, and include them as a submodule. The tools are there to be used, and it’s up to you how to actually use them. 

Personally, I like the idea of separating shared libraries and assets, as well as tools and such from the master branch.