This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
Messages - Trovaner
Pages: 1 ... 48 49 50 51 52 53 54 [55] 56 57 58 59 60 61 62 ... 64
1081
« on: July 12, 2010, 12:21:35 PM »
Also, while someones havin a look, could someone make the bladewall static (no going up or down) as it'll be a raised outline for the flame grill.
The easiest way of doing this would be to set the mass (listed in the RBCOLLECTION) to 0. If I remember correctly, setting *UNYIELDING to True would also work...
1082
« on: July 11, 2010, 04:38:31 PM »
I don't think I have Dummy's old compiler but I'm definitely able to use both Dummy's newer compiler and FB slightly updated version on my XP computer. Dummy's compiler can be found in the download section of both GTM and RFSHQ while FB's compiler can be found on the beetlebros site (both compilers are based on the latter mentioned compiler in your quote). Also since we're on the topic of GMF compilers, there is also Serge/Megabyte's compiler (before the new compilers that everyone is familiar with) which is based on Dummy's compiler and supports Point to Point constraints (the only compiler besides Serge's new compilers that does) but unfortunately it doesn't support everything that FB's compiler does. Basically its a giant mess that Serge's newer compilers have tried and for the most part fixed.
1083
« on: July 11, 2010, 04:21:19 PM »
Not bad. Its definitely a good tut for beginners.
Pros:
1. I was surprised to see that you got an error but you handled it really well. Whether it was on purpose or not, it helped people understand how easy it is to make a mistake.
2. You explained what you were doing very well. I liked how you explicitly told us what needed to be capitalized due to everything being case sensitive.
3. To tell you the truth I couldn't think of a #3 but I thought you deserved one anyways :P
Cons:
1. The camera was a little wobbly and sometimes you didn't give it enough time for it to regain focus after moving. Using a tripod or placing the camera on top of a sloped or raised surface would help for next time. Otherwise, you can look into screen recorders which allow you to record in and outside of RA2.
2. There could have been slightly more explanation as to why you did things. Something similar to how you explained why you added the SZ but for everything else (why did you edit the Bindings.py, Teams.txt, and rename the bot file).
3. It would help to give a better understanding of what else is possible. Things like how to AI bots using more than just Poker (relating them back to the AI.py files that are in the AI directory), what the other values in the bindings are, how to learn more, etc.
Good job on your tutorial Stag.
1084
« on: July 11, 2010, 03:22:37 PM »
FB's compiler has never worked for me. I read somewhere that it's because I don't have the CD for the game. Is there a copy of the compiler that doesn't need the CD?
To my knowledge, you don't need the CD for it to work (I think thats just one of the CFZ compilers). Although, I may have run into an error (stating that I do need a CD) when I tried using it on a Vista computer. Unfortunately, I've never tried having my CD in the computer while running FB's compiler on Vista nor do I know a way of fixing this.
1085
« on: July 11, 2010, 02:56:23 PM »
Serge's compiler doesn't work very well when there are *GMID_HAVOK_DIS_COLLISION_PAIRS at the end (which is odd because it looks like the compiler should be able to support them). I've always just swapped to Firebeetle's compiler whenever Serge's didn't work (so I've always forgotten to bring this up). It should work, if you delete the unsupported section and change the following:
*NUM_COLLISION_PAIRS 6
}
*GMID_HAVOK_RBCOLLECTION
{
*NODE_NAME RBCollection01
*NUM_DISABLED_PAIRS 6
*SOLVER_TYPE 0
to this: (note that I changed two values)
*NUM_COLLISION_PAIRS 0
}
*GMID_HAVOK_RBCOLLECTION
{
*NODE_NAME RBCollection01
*NUM_DISABLED_PAIRS 0
*SOLVER_TYPE 0
As for the lighting issues, I think the problem is that Dummy's Compiler doesn't support everything that was used. I have a feeling if you used Firebeetle's Upgraded Compiler , it would work. His compiler can be found on the Beetlebro's site.
1086
« on: July 11, 2010, 12:59:51 PM »
POP-sicle's strength "Hard for hammers to hit" made me laugh because there are none in this tournament. The ByeBOT turned out to be a decent contender after its partner threw TAC on top of him. IMO, these were the two best teams in this tourney. To bad this wasn't the final match. It was a battle of good vs evil, a team of popups vs a team of HSers. This was my favorite match, so far... The battle had several unexpected twists which usually made me laugh... On a side note, it looks like G.K. and S_M color coordinated themselves. This one will be a good match... once it gets uploaded... Predictable. The ByeBOT didn't stand a chance (especially after its battle axe got torn off).
1087
« on: July 10, 2010, 08:38:45 PM »
The instructions for adding Plane01 can be found here. I've made a couple different arenas using it but it takes awhile to get used to finding it in the compiled gmf every time you want to test the arena.
1088
« on: July 10, 2010, 08:10:26 PM »
The BBEANS Arena V2 uses a special floor that none of the compilers or decompilers understand. You can either re-enable the floor's collisions (as suggested by Mad) or you can edit the compiled version so that it uses the special floor. I can't remember which hex values needed to be changed to get the Plane01's infinitely flat surface but I remember Click posted something about how to do it.
1089
« on: July 10, 2010, 07:57:31 PM »
It doesn't always give that error message... It depends on what goes wrong in the AI.
1090
« on: July 08, 2010, 05:46:09 PM »
Dos'nt that have something to do with Aing it to avoid the walls so it will fallow them and go around?
Not really... The way I do it, the AI's real controls are overridden by new ones that are given by the Arena.py (the AI is not aware of the obstacles/walls but the arena has already been programmed with all possible routes). Although, I suppose it might be possible to import new tactics for the AI to avoid things but it would not be as specialized and it would be very hard to implement. So when all is said and done, it wouldn't have much of a benefit over the way that I'm doing it.
1091
« on: July 08, 2010, 03:10:06 PM »
It is completely possible to make the AI race around a track (not just in a straight line). I, personally, have gotten the AI to perform complicated tasks in at least 4 unfinished arenas and compared to some of them, it would be easy to make the AI go around a circular path.
1092
« on: July 07, 2010, 10:15:18 AM »
Cool. I like how you got it to rotate (far better than just a picture).
1093
« on: July 06, 2010, 08:57:25 PM »
OK, after some thought, I decided to have a rumble between:
Neglected Waterbug
Insult to Injury V
Shadow Tyrant
Bumble Bee
Which will also be a video and will be posted tomorrow. Also there will be a SFTW rumble between
ErupTion
JAWZ
Tsunami
The Baby Warrior
which will ALSO be a video and posted tomorrow. OK?
Sounds good. I'm looking forward to it.
1094
« on: July 06, 2010, 08:49:34 PM »
Are you trying to balance it for stock or for DSL? Either way, I would be willing to give it a go. I have several efficiency formulas that I use when making bots.
1095
« on: July 06, 2010, 01:57:12 PM »
Huh, I just realized that Shadow Tyrant AND TyranT were in this  . I remade TyranT into Shadow Tyrant so that it was better suited against crawlers (namely Bumble Bee) so its kinda coincidental that TyranT ended up losing to it. By replacing the castors with more ramplates, I lost some of TyranT's wedginess and made it more susceptible to HSers (which is one of the reasons why I'm really surprised ST won against AC). Seism is the not the best but it has repeatedly proven itself as being a superior design. Obviously, there is a rock-paper-scissors effect which prevents it from winning against everything but it is one of the most effective bots around. It has also been the standard/inspiration for several of the bots that can beat it. The BBEANS Arena may not be good for rammers but it is far more balanced (for other designs) than the Combat Arena. The benefit of having the center obstruction is that bots aren't getting thrown out of the arena within the first couple seconds by going too fast up their opponent's wedge.
1096
« on: July 06, 2010, 11:19:37 AM »
In the CONSTRAINTSOLVER section of the GMF, all you have to do is set this:
*IS_LIMITED 0
*FRICTION 0.000000
*ANGLE_LIMITS -1.570796 1.570796
to this:
*IS_LIMITED 1
*FRICTION 0.000000
*ANGLE_LIMITS 1.570000 1.570796
By setting the angle limits so close, you are preventing it from turning. This works because RA2 is only looking at the power values (given in the TXT) when calculating wheel movement. Although it isn't necessary, you may also want to increase torque and decrease speed on your edited 2-mag (so that the treads seem more realistic).
1097
« on: July 01, 2010, 09:26:37 PM »
And that looks like...?
Thats exactly what I was thinking... Depending on how "after" you really mean, it may also be enclosed in the Sarcophagus. The wikipedia pics were taken before the Sarcophagus was made.
1098
« on: July 01, 2010, 05:37:54 PM »
A Lego Arena would be cool but it may have too many vertices for it to be playable. If Serge still plans on making a lego component pack, maybe he'll make an arena to go along with it...
Please expound upon the Chernobyl reactor idea. It sounds like it would be different (to say the least) but I can't seem to draw a picture of how it would look in mind. What kind of signs of impending danger are you thinking of (voice over intercom, siren, some sort of gauge, lights, etc.)? Would kind of effect on the robots would there be?
1099
« on: June 29, 2010, 11:42:02 PM »
Cool popup. It sort of reminds me of Ronin from Inf's RA2 Restocked...
1100
« on: June 29, 2010, 07:57:27 PM »
So are Corner Patrol Zones (CPZs) what you were thinking of in your first post? If so, it wouldn't be too difficult to do and/or teach.
Edit: Okay, I see that your edit answered my question...
Pages: 1 ... 48 49 50 51 52 53 54 [55] 56 57 58 59 60 61 62 ... 64
|
Show Posts