Dirt! 3D Hack and Slash

Hey folks,

Dirt, the game I’ve referenced a few times on here in the past, has recently advanced to version 0.2. It’s to the point now where we feel comfortable releasing it to a small circle of friends with the understanding that it’s just a very early work in progress.

You can find the most recent version and screenshots here:


Please read the readme file before playing. There are issues we know about and they are mentioned in there.

Looks very pretty. Nice showing of jME features. Few questions:

  1. What are you using to detect hits from the sword to the rat? Collision detection? It didn’t seem very real how she would ‘autoface’ the creature whenever I pressed attack.

  2. Is there a way to display the FPS during the game? Would be nice for testers.

  3. Is NWN a type of keyframe animation or is it Joint based?

    Probs/bugs: The movement seemed akward, and I was getting pauses every few seconds it seemed.

    Can’t wait for version 3

Impressive. Are we fighting on the moon? That’s some bleak terrain :). Lots of fun already, looking forward to some gameplay additions.

We need to start putting some of this stuff on the JGF and apps and demos page. Show what people are doing with jME to the unwashed masses.

@ Cep:

  1. We detect based on distance. The animation is only that, something pretty to look at. Similar to Diablo2 or games in that genre, we’ll be using rules to determine hits rather than intersection of geometry.

  2. Yep, hit the tilde/grave key.

  3. It’s Joint based.

    Yeah, we’re still working on optimizing everything to get rid of the occasional pause. You shouldn’t be getting it every few seconds though… Maybe your fps will tell more. Running it on a P3 450MHz with a GF2 card still gave me a decent 45 FPS.

    .3 should include a new terrain system and a creature edit tool. We are looking to release that in early October. Suggestions on what to add to make it more impressive are always welcome.

Average FPS 55-70. I hit 30FPS twice or so as the camera was panning.

Geforce FX 5600

1.5GHz Athlon

768MB ram

Well done. That is all I can say.

I like your style renanse. This game look promissing so I can’t wait to see it complete.

See ya!

very nice. Looking forward to the terrain editor.


"Anonymous" wrote:
I wasn't able to play it it crashes out loading sound...
Yeah, others I've released to have had the same issue both in DirtV0.1 and 0.2. Eric tells me the cvs version of lwjgl has some fixes that may help... As soon as they release an official new version I'm sure we'll port to that. Maybe that will help.

btw. I noticed on the JGO forums that the lwjgl team was looking to cut a release in the next week or so. (i hope that's true...) http://www.javagaming.org/cgi-bin/JGNetForums/YaBB.cgi?board=LWJGL;action=display;num=1092237566
"DarkProphet" wrote:
very nice. Looking forward to the terrain editor.

It looks like we're going to go a different way with Terrain and not end up using that jME class in the future... (we just don't have the resources needed to generate new terrain content like trees, bushes, rocks, etc.) I'll see about releasing our terrain editor standalone as it is not too shabby.
It looks like we're going to go a different way with Terrain and not end up using that jME class in the future... (we just don't have the resources needed to generate new terrain content like trees, bushes, rocks, etc.) I'll see about releasing our terrain editor standalone as it is not too shabby.

Curious what you mean by this.

In other words, because we are two programmers with limited artistic ability (see our GUI for example!) we’re looking to use more existing content for our world maps, including the land you walk on itself. In fact, we are thinking of going the route of the QuakeWorld apps out there (or see Jake2 – Java Quake2) and requiring you to have the NWN content installed on your drive in order to play, avoiding potential legal issues.

Our aim is to make a great hack and slash with just two programmers (cause that’s all we have) and this seems like a good way to do it.

A requirement like that really limits your user base, including me :frowning:

yeah, :frowning: Unfortunately we don’t have anyway to attract resources for a game without a good game already under our belts. Game #2 will be better. hehe.

Fortunately though, you can pick up a copy of NWN for $15 these days. The nice thing about NWN is that it has a large mod community even now and the game was designed for writing your own adventures which lends itself to an effort like ours.

Have you thought about just making Dirt using the NWN tools then? It sounds like it is a NWN mod more than anything, and the Aurora engine is higher quality than jME for displaying NWN content. I guess it depends on your true goal, do you want to make a game or do you want to do it for the "education".

We aren’t using the actual engine because we want to have complete control over that aspect of the game. Anyhow, we’re only making this a question of where we get our models, which is a game dev issue and not necessarily an engine choice. Getting jME to contain features so it can properly support NWN models also raises the quality of jME (just like adding support for any model type does.)

But overall, yeah, the experience. Why make a Java port of Quake2? Because it’s cool. Dirt won’t be a NWN knock off, but getting Java to show it’s content in a unique and highquality manner will just be fun and a good way to show off. :slight_smile:

How do you edit terrain in real time? (I did not run the program yet but I plan to.)

Thats why i proposed the idea of splitting jME into pieces. So programmers can choose whether to go with the already made hand-crafted piece of perfection from jME :wink: , or roll on their own.

And sometimes, rolling their own is a better option for the developer (not better as in better performing, or better looking) because they can maintain their code easier than maintaining someone else’s. Im sure your familiar with this?


Hey folks,

Thanks for all the comments and suggestions. As Josh mentioned, we’re open to any and all suggestions regarding the development on this project. We’re still a long way from the end.

I noticed that probably the biggest problem people have is Josh’s comments about re-using NWN content, and requiring a copy of NWN to run it because of legal reasons. Basically, Josh and I are not artists and are not proficient in creating 3D models. The reasons are contemplating (the decision is not final yet) going this route are:

  1. We don’t want to spend time creating models and textures when we could be putting that time into the programming.

  2. We want to get a finished product out there, which is playable and fun.

  3. The education.

  4. Generate enough interest that others will possibly want to contribute to round 2.

  5. Chicken shopping cart pineapple of doom

    Erm… ok, maybe not that last one, but you never know what’ll end up in Dirt. Anyways, thanks again and we look forward to any suggestions you may have to better the game.



Hey, you stopped lurking David, there’s hope for you yet. :stuck_out_tongue:

Not yet, but I doubt it would be hard to put one together. The hardest part is the textures though since they get created on the fly from an image and some colors and the .jme format doesn’t really handle dynamic textures.