Previous Blog Next Blog
Prev/Next Blog
by date

MMORPG Contest - 36 Hours Left!

MMORPG Contest - 36 Hours Left!
Name:Tony Richards
Date Posted:Jan 30, 2007
Rating:4.4 out of 5
Public:YES
Comments:YES
RSS Feed:GarageGames Blog feedor Subscribe with .
Profile Page:View profile page for Tony Richards

Blog post
We're making a mad dash to the finish line... not running out of steam, but we're definitely running out of time.

It's tough making the final decisions... looking at the long list of things we'd like to have completed is tough because we've got several weeks worth of things that we'd really like to see done, yet we only have a few hours for the contest.

The coding for the core features that distinguish Fractured Universe from other MMORPG's are still incomplete, but I expected that to be the case because our first goal was to get the basics done, then add the fun stuff. Because of that, I won't be surprised if we don't come off with a win on the contest.... sure, we'll possibly be a little disappointed, but I think nearly all six of the current game developers are in it for the long haul.

The contest has become less important than finishing the game... we'll turn something in, and although it'll be playable, it'll definitely not be complete.

The overall game design / back story / concept art, etc is coming together beautifully. The core code-base is very solid and nearly flawless, with only a couple of bugs that need to be squashed.

Contest Postmortem

Recruiting - Most of the first 10 recruits didn't really work out... although they all had good intentions, most of them simply didn't have the time available to dedicate to a short-term intensive project like this. Even I was hard-pressed to put in the time required, and I usually have plenty of spare time.

Some of the initial recruits were also lacking in skills.... for a longer term project I would have kept them around and trained them a bit, but for something on such a short deadline I simply didn't have the patience.... apologies to the few that I did this;

Being picky paid off for the most part, though... with the four people that stuck it out plus the addition of two more that we picked up in the second round of recruiting, I'm happy to say we have an incredible team. Maybe after the contest I'll be able to recruit another highly skilled coder.

One thing I learned about short-term coding... if something isn't working, drop it immediately. I made the mistake of concentrating too much on Blender / MakeHuman / BVH and I lost over a hundred hours on this sub-project, and although I made significant progress, I didn't finish it in time for the contest. Luckily I abandoned it soon enough for me to work on other things, otherwise I'd probably still be working on it with nothing to show for the contest.

This was a huge mistake as far as the contest goes. Luckily, the time wasn't wasted and for the long-haul game-finishing goal. We'll be able to use this technology for the final game, which will allow us to have a significant number of NPC's / toons / animations with a small team in a short period of time.

Another mistake for the contest but a huge win for the final game was the way in which I did the MMO Kit integration. Instead of taking the MMO Kit and adding / subtracting from it, I started with the stock TGE 1.5 engine, stripped out all of the scripts (yes, I did say all) and used the MMO Kit / TGE starter.fps as a code-by-example resource.

This took significantly longer than using the MMO Kit as a starting point, but the advantage is that I know 100% of the code that's in the game... Dave, Dreamer, etc wrote so much code that I never would've been able to wrap my hands around all of it, so modifying / enhancing it would've put me in a dangerous situation of orphaning code and introducing a bunch of bugs.

The disadvantage is that I didn't have enough time to integrate all of the features included in the MMO Kit... the camera, inventory, quest editing and a few other features simply didn't get put into the game for the contest even though they came stock with MMO Kit.

An advantage, though, was I had TGE 1.5 from the start and AFX integrated within an hour after I downloaded it.

Alpha testing - Although it's way to early to begin beta testing, we started alpha testing a little over two months into game development. This was necessary for the contest and I probably wouldn't have done this in a normal situation because it starts wearing out your primary testers, but in the long run the feedback that we've gotten from the early testing will shape and form the final game into something that's exciting, new and enjoyable to play.

Kicking off Fractured Universe development through this contest has been a great learning experience. We've made a few mistakes along the way, both in making long-term decisions as well as short-term decisions, but nothing disastrous. In short, thanks to this competition, we've got a fantastic start on a game type that many people say cannot be done by indie developers.

I look forward to the day that the nay-sayers are proven wrong, and in a big way because there are at least 5 new MMO RPG/FPS indie games that I expect to be finished within the next 12 months.

Thanks for reading my blogs. Wish us luck!

Recent Blog Posts
List:04/10/08 - Indie 2.0 - Content Packs
03/14/08 - Indie 2.0 - Part 1
01/04/08 - IndieZen Dev Blog, Dec 2007
12/13/07 - Happy Birthday IndieZen!
11/25/07 - IndieZen Dev Blog, Nov 2007
11/17/07 - IMGDC tech talk
10/18/07 - IndieZen Dev Blog, Oct 2007
10/13/07 - Long time no blog

Submit ResourceSubmit your own resources!

Dreamer   (Jan 30, 2007 at 17:17 GMT)   Resource Rating: 5
As always excellent work Tony, and I cannot wait to play this!

Regards,
Dreamer

Jonathon Stevens   (Jan 30, 2007 at 17:42 GMT)
Fantastic read as usual Tony. I'm going to shoot you an email, we should chat again.







Todd Pickens   (Jan 30, 2007 at 19:05 GMT)
Best of Luck!

Mads Laumann   (Jan 30, 2007 at 19:57 GMT)
Nice to read your blog post as always.

Looking forward to try out the game!
Good luck with the contest :)

Stephan (viKKing) Bondier   (Jan 30, 2007 at 20:20 GMT)
This 90 days contest seems to have been too long to put them in a real stress situation...
Shouldn't next one be only 60 days long? ;-)

Good job.
I have not tested it, but I'm sure this project has already all the quality needed to become a very good game.

Jonathon Stevens   (Jan 30, 2007 at 20:41 GMT)
It probably one ups LOTR online *where's my barf bag* It can't dissapoint me more than that pile of crap for an MMO. Took the best IP you could possibly get and screwed it up.







Tom Bentz   (Jan 30, 2007 at 22:31 GMT)
Looking forward to the playable demo!

Paul Usul Fluegel   (Jan 31, 2007 at 00:20 GMT)
Thank you! for writing this blog, has been a thrill reading it, hope you continue..

And I think you did pretty good, really hope to se a litle more of your work on MakeHuman..

Vashner   (Jan 31, 2007 at 00:41 GMT)   Resource Rating: 5
Ah blender education can't be a waste. Keep up the good work.

dienasty   (Jan 31, 2007 at 03:47 GMT)
not a big fan of MMOs but looking forward to seeing what you guys have created!

David Janssens   (Jan 31, 2007 at 09:20 GMT)   Resource Rating: 4
Looking forward to see what you are creating. A fresh wind is always nice in the MMO gamespace.

Could you expand some more about your MakeHuman/Blender project and the issues you've encountered? I was planning to start on such an endeavor myself in a few months and I would like to learn from your experience.

Tony Richards   (Jan 31, 2007 at 14:26 GMT)
@David - The biggest hurdles right now include the vertex group weight assignment automation and the BVH importing. Currently, there's no way (that actually works) to copy weight assignments from one mesh to another based on vertex locations. Also, the BVH import is pretty hackish and with the current version you can't attach the animation to an existing bone heirarchy... instead the process relies on you manually matching the movement using IK, then baking the animation.

So far I can create the mesh, import it into Blender, generate a skin using ambient occlusion and subsurface scattering, bake the skin, reduce the poly count to 3k triangles, apply the baked skin to the new model.

Other than the first step, which is an artistic step that could take anywhere from a few seconds to a few days, the rest of the process takes less than 20 minutes for standard humanoid models. Eventually this process could be 100% automated (again, sans the first step).

The next (missing) steps: Create vertex group weights based on a reference model that already has vertex group weight assignments. Currently it takes a few hours to do this step manually and although the more experience I get in doing it the faster I can do it, it's still way too long of a process.

Once I can get that part automated, no matter the results of my BVH import implementation, I could manually (or hire someone to) create the animations for all of the toons. The drawback would be that all toons must be the same size and basic shape (i.e. humanoid) to share the animations.

I have one toon that I've manually taken through the entire process, including some basic animations.

The final steps are clothing / hair. MMO Kit uses a decent method for this, but the problem is that you duplicate every piece of clothing for every toon, so if I want 100 different toons and 100 different pieces of clothing, I'll have over 10,000 meshes defined and loaded in the game.

Quite likely I'll use the NVN method of doing this. Last year I made a study of the MDL file format and even implemented a native MDL mesh resource for TGEA, minus the animations. I'm seriously headed this direction.

MDL files support mounting of animated objects. You have a root mesh with the animation and sub meshes are mounted on the root mesh much like TGE handles DTS meshes. The main difference is the sub meshes have vertex group assignments that match the root mesh and all animations are applied to the root mesh and the sub meshes.

This will completely solve the last part of the problem, but the engine changes to animate MDL meshes will be quite time consuming / high risk, so I still haven't decided if I'm going to take this path or not.

You must be a member and be logged in to either append comments or rate this resource.