Author Topic: jitspoe's .plan (Read 287149 times)

jitspoe · « **Reply #60 on:** September 14, 2005, 02:17:18 PM »

Fixed the foreign keyboard support, hopefully. That was more work than I bargained for, but it should be better than what most games provide now.

Test it and leave feedback in the build 15 thread, please: http://dpball.com/forums/index.php?topic=1057.0

jitspoe · « **Reply #61 on:** September 19, 2005, 03:00:59 PM »

Spent most of yesterday re-making the texture wals for BSP so that hopefully the lighting will work better. I should have a new BSP pack out soon with these and the updated entity file.

I finally got around to making a fence texture (for cobo), and in doing so, discovered a nice little bug in the rscript code. Somehow I forgot to initialize some values. I hate stuff like that. I spent the better part of the night trying to figure out what was wrong, only to find it was some tiny oversight that took 2 seconds to fix once I figured out what it was.

I also worked on tweaking pong a bit. The ball has a little upward velocity when you kick it and doesn't rely solely on the player velocity, as when you're online, it's much more difficult to jump-kick the ball. Next I need to make it so it can bounce off of multiple surfaces in a single frame so it doesn't wedge itself in corners so much and won't have such slow/choppy movement when rolling on the ground.

Eiii · « **Reply #62 on:** September 19, 2005, 06:48:50 PM »

Cobo · « **Reply #63 on:** September 19, 2005, 08:14:12 PM »

*cobo feels special

(special in a good way

)

Eiii · « **Reply #64 on:** September 19, 2005, 08:39:23 PM »

What, there's a bad special?!

jitspoe · « **Reply #65 on:** September 19, 2005, 09:37:28 PM »

Yes, you and all your classmates that rode on the short bus were a bad special.

Eiii · « **Reply #66 on:** September 19, 2005, 09:58:20 PM »

darn. I was wondering why half of our bus was missing...

jitspoe · « **Reply #67 on:** September 27, 2005, 04:14:06 PM »

Well, the weekend was somewhat productive. I studied R1Q2 some and carefully went through a list of known security holes to make sure paintball2 had no vulnerabilities.

I've been trying to get the pong ball to bounce multiple times per frame if it needs to (since 10hz physics make the ball lag and jerk all over the place). So far my attempts have failed, but they often yield humorous results. I also discovered that the "bounceback" cvar applies to the pong ball, too. Set it to 2.1 and watch the ball act like flubber!

Eiii · « **Reply #68 on:** September 27, 2005, 04:38:35 PM »

You didn't know that?! I TRIED TO TELLYOU!
Also, I think pong is best with bounceback at 0.5 or something.

jitspoe · « **Reply #69 on:** September 28, 2005, 11:29:24 AM »

Finally got the pong ball movement cleaned up the way I wanted to. It's much smoother now.

Bounceback has to be higher than 1.0 to bounce at all, otherwise it just hits a wall and stops. 1.5 is the default.

Eiii · « **Reply #70 on:** September 28, 2005, 04:37:28 PM »

Ooh. I think I meant 1.5 or 1.25 then...

jitspoe · « **Reply #71 on:** September 29, 2005, 11:38:49 PM »

finally fixed the pong ball sticking problem. This is why I'm not actively planning to implement better physics -- you spend weeks trying to resolve some tiny little bug like that.

I also fixed the sticky strafe-jump "bug" where you miss jumps and lose speed. This will require an update on both the server and the client. It's kind of funny. The way q2 physics should have worked was you'd have a brief time after touching the ground in which you couldn't jump. This meant no bunny hopping, double jumping, or pretty much any kind of trick jump. The game would be very different like that. I doubt many people would like it now, though, after being used to strafe-jumping at 60mph.

On the opposite end of the spectrum, I tried disabling the check to see if the jump key was held down. That lets you do all kinds of crazy superjumps just by holding the jump key down. Kinda fun. I'm considering making it a server-side option -- might be good for learning purposes, but I don't know if it would be good if servers started using it all the time...

Eiii · « **Reply #72 on:** September 29, 2005, 11:51:17 PM »

Strafe-jump bug? What?

jitspoe · « **Reply #73 on:** September 30, 2005, 11:39:15 PM »

I came up with a great idea for paintgrens while I was driving home. Picasso was having trouble with the latest release, so I sent him the most current version. In that, the paint grenades can always kill the throwers (but if grenadeffire is 0, they can't kill teammates). I figured this would be the best of both worlds: people wouldn't be able to use cheap throw-at-feet/point-blank tactics (the main complaint about impact grenades), but they also would not be able to intentionally kill teammates. Apparently some people complained to him, though. I wish more people would post complaints on the forum, though. I can't really take "he said she said" complaints that seriously, and I think most of the time, people on the server just whine for the sake of whining and use any kind of change as an excuse for their inability to play well.

Anyway, I'm ranting. On to the idea: Let the players decide. Picasso was saying he wanted people to be able to choose. Since the anti-change whiners prefer the old non-impact grenades, those could be non-self killing (when ffire is disabled). If people tosser impact grenades, though, they run the risk of eliminating themselves.

On another note, as I mentioned in another thread, I implemented the match command finally. I really should have done that as soon as I got the EV1 servers to hopefully standardize the match settings by now... oh well. Kind of funny that people still match using the old public mode settings.

IronFist · « **Reply #74 on:** October 02, 2005, 10:45:48 AM »

Post removed

jitspoe · « **Reply #75 on:** October 13, 2005, 05:05:50 PM »

Let's see, what have I done lately. I implemented the grenade solution I was talking about and added a "nojailffire" worldspawn option. Heh, the conditional that checks if a paintball can kill a player is pretty ridiculous now.

Hmm, also fixed a bug with the skm model code that cause linux to crash when video modes changed. Fixed some bugs in the stdlogging, too. That was all last night, though. Don't remember what I did before that. I just figured I should post an update.

I'm trying to figure out the difference between the way linux (gcc) and win32 (msvc) compiles handle the tracing code. This has bugged me for a long time. It causes a lot of lag/prediction errors on angled surfaces with linux servers.

XtremeBain · « **Reply #76 on:** October 14, 2005, 03:22:47 PM »

I saw you *bump* your year old topic over in one of those dev forums concerning the prediction misses. I hope you can figure it out, it's funny that r1ch hasn't spotted it yet.

jitspoe · « **Reply #77 on:** October 14, 2005, 03:44:10 PM »

I talked to him about it. He said he's aware of it, but hasn't worked on a fix for it yet.

jitspoe · « **Reply #78 on:** October 18, 2005, 02:03:24 AM »

Fixed that silly "sitting on a grenade won't kill you" bug. Also added some checks to reduce wasted cycles on unused entities.

Edit: also tweaked and cleaned up the grenade code a bit. Impact grens should be less likely to fling paint halfway across the map.

Eiii · « **Reply #79 on:** October 18, 2005, 05:05:54 PM »

Did you make tham not last as long?