March 26, 2004

Calling All Freaks

I had two separate but similar discussions yesterday about the challenge we (ie, coil) face. Basically it comes down to this: we need artists. There are three of us who can code reasonably well, but only one of us who does art: me. Being such a perfectionist, I've redone at least once almost every asset in the game, and I don't see that trend ending (as I start the third revision of the course). I came up with a good descriptor of our plight: we're Fill-Rate Limited. We have enough programmers to get things done, but not enough artists to keep up with the gameplay.

I'm not sure how to remedy this, other than to hope that when we release something public, people get interested in what we're doing and we can find some talent willing to help. We have big plans for our Phase 4 version, and we'll need all the help we can get to make it.

As I mentioned, I'm starting to redo the course. We want Epic to showcase our mod at E3, and even though the gameplay is there, I doubt they'd show it in its current state. I don't expect to match the level of Links or Tiger, but I'd like to get close to Hot Shots or Swing Away.

Posted by mix at 01:42 PM | Comments (8)

March 17, 2004

Flashback

I updated the sidebar with some links to blogs by current or former GameSpy employees I used to work with. GameSpy was the most unique place to work of all my past employers, and my time there is a pleasant memory. It's interesting to see who's still there and who has moved on. Good times.

Posted by mix at 04:10 PM | Comments (0)

March 16, 2004

Dude, Where's My Ball?

Testing went better this weekend, and we're doing another code/test session this weekend, but we have a problem that's plagued us from the beginning that we haven't quite found the proper fix for. In the Unreal Engine, there are really two types of physics: the default engine physics (PHYS_Walking, PHYS_Flying, PHYS_Falling, etc) and Karma. The default engine physics are great for a typical FPS, and Karma adds enough spice to allow you do some cool (simple) rigid body simulations. However, either are not well suited for a golf simulation. We currently use a combination of physics during ball flight to achieve the effect we want. PHYS_Falling is used for the main ball flight and works well, but you can't continue to use it after the ball slows enough to roll. At that point we switch to PHYS_Karma to allow for more accurate rolling. However, Karma does not have collision prediction like the other physics types, and therefore the ball can become lodged in or pass completely through objects if moving too fast. Since we're dealing with a small object (golf ball), we run into this quite frequently. Our attempts and coding checks to see if we've "fallen through" the ground have proven imperfect and sometimes visibly erratic.

If we could fix that problem, our gameplay would be reaching a very solid point.

Posted by mix at 10:43 AM | Comments (3)

March 10, 2004

Glutton for Punishment

I must be crazy.

I've decided to redo the course. Again. This will mark the third time the course has been redone, and while it sounds like I'm just a perfectionist (which I partly am, but this game is extremely important), it's also becuase changing that way the course is constructed will hopefully solve a few major issues we have. BTW, if anyone knows some tricks to subdivision surfaces, I'm about to embark into new territory.

I find I'm asking myself the same question repeatedly recently, "can I get it all done?" I have no idea, but I'm going to try. Playing the balancing act between career, family, freelance work, and the game is pretty tough, but if it all works out, it'll be a pleasant memory. I've tried recruiting help, but in today's mod environment, getting good help is like trying to get the Pope to come to dinner. It ain't happening.

So the three of us plod along, all of us coding, and me trying to handle all the artwork as well. I guess the benefit is if we win, there's less people to share the winnings with. ;)

Posted by mix at 12:32 PM | Comments (4)

March 09, 2004

Paying Dues

Game Development is rarely easy. If it were, as the old adage goes, everyone would be doing it. As I mentioned in a previous post, we did some testing over the weekend. Well, I should say attempted to do some testing. I scheduled some people to drop by with their machines, and we set up a small LAN. We tried to get into a round, and....  nothing.  It didn't work at all.

We made the mistake of trying to get some features in before the testing session, and neglected to test them properly. We spent the rest of the time trying to fix major issues instead of gathering valuable feedback or just fixing minor problems.

What this means for Duffers is minor in the overall scheme of things, but discouraging nontheless. We're pushing back our date for a closed beta for least a week so we can properly test multiplayer, and try to iron out some gameplay issues.

Baby steps...

Posted by mix at 01:58 PM | Comments (4)

March 01, 2004

The Zen of Game Testing

We're hopefully going to do some hardcore network testing this weekend, in preparation for the closed beta. This is both exciting and terrifying. I'm glad that we're finally at a point where we can play golf and test multiplayer aspects (which I wanted to be doing a year ago, but whatever), but it also means that things could go horribly wrong and I'll have testers sitting around being bored while we fix stuff.

If we did our jobs, it should go fairly well, and we can move ahead with our planned closed beta in two weeks, with our first public version coming out at the beginning of April.

 

Posted by mix at 03:04 PM | Comments (0)