X69RZX
-
Content Count
641 -
Joined
-
Last visited
Content Type
Profiles
Forums
Blogs
Calendar
Posts posted by X69RZX
-
-
I started w/ a AMD K6/2 550mHz w/ 256MB and an onboard video card. I thought I was smokin when I upgraded to a P3 600 w/ 256 that supported AGP, so I could use my super cool GF2MX vid card, lol. I think I almost got 30fps a few times. At least I can play GR now, I have a AMD K7 2400 2GHz, w/ 512MB, and a GF FX 5500, and I can finally get 60fps. I still pwned back in the day tho:devil:, well....some times anyway.
-
Are you hosting? If so, you will need to cinfigure your router to forward the correct port, 2346, to your internal IP address. If you are trying to connect to a remote server, and are having trouble, you may need to turn off your firewall, if you are running one.
-
Updated the d/l and added some more strings to try and prevent clumping and ###### rushing enemy hot spots. Any detailed feedback on scenarios encountered would be great.
-
Clever scripting is the only way to improve upon the AI. They do well in close, but it is up to you to do the rest. Use flags, counters, and queries to create thought processes and to keep them from doing stupid stuff, and to make them behave in certain ways or patterns. It is the only way.
-
Thnks all, @GRT, the dev mode is only in there now sos you can follow them around some, to see how they behave, I will certainly be taking that out. @Jay316 there are three fire teams, 1 enemy each team in recruit, 3 in vet, and 5 per team in elite. I am currently scripting two new strings that will deter them from a) choosing the same path as the pther teams, they may still do it, but it will be by the numbers
and b) not clump in battle, I am adding a flank script that will over-ride their tendancy to clump as they do when engaging an enemy if there is already another team in that position, and they feel like it, of course. 
also, since this is a map with closer spawns, I'm going to make thier movements and what not a bit more agressive, that should help with the catching them all near the spawn issues. I plan to adjust these numbers and settings for each map individually anyway.
*I updated the file at the above link. I have added the script to try to eliminate the teams grouping up in one area too much, as well as decreased the probability that they will end up on the same path, please let me know if they continue to clump. The individual fire teams will clump however, this is due to thier pre-programmed AI, I have them locked to scattered formation, I don't think there is more that I can do there. I can only effect the way the seperate fire teams react to each other and to the player.
-
airwolf, man......
*lol, i googled airwolf, totally wrong, ahhaahaha
-
Thank you. I do have the bots set to a more careful pattern at this point. Let's see what any other peepl say, and if only you play it, I'll set it to w/e you like. Depending on the various variable settings which actually control thier bahaviors, they can act all together differently. I would of course be happy to help you set your own version of the mod to behave as you wish, that info will be available with the full release.
*things I will be adding;
map specific options for the enemies to assault/investigate at random, such as the roof of the embassy, certain interior areas, and any other out of their basic play area, areas (
) that I want them to check out. (the definition of "their basic play area" is available upon request)more complex combat logic inculding strings to prevent the teams from clustering up at one location, but instead to go a different direction and thus create the potential for a flanking type manuver.
and anything else the testers suggest that is workable.
-
As this is a very early test, I am only using the embassy map for this release. The reasone is, I don't want a bunch of unfinished versions floating around, so I figure with just one map, maybe some peeps will test it out, and give me some feedback, so that I can improve upon it, then I will put it on all the maps that are requested(or as many as I can) for a full release. The script is almost pure, which means it can be imported to other maps, and then tweaked for that map. Like I said, once the script is done, I will release it with instructions that even the noobiest of noobs should be able to follow to add the bots to their own or favorite maps.
*just a quick clarification; This is not a game type file. There is really no way to make the AI behave as they do in this mod as a gtf. Instead, there will be a mission for each map. The key is the ease with which the script can be added to a new map, and the random behaviors which create human-like scenarios.
-
I've been working on this for a little while now. The first go around, I scripted this for just one AI player. Since then, I have re-scripted the whole thing from scratch, and added two more teams. Ideally, I would like to have a total of six enemy teams, but the script is rather extensive, at least the programming of it, and so, at this point, I only have three teams. I really need some input on playability before I take the time/trouble to put it on other maps. I would also like to hear which maps people would like to see this on, if any
. Below is a description of what the AI bot will do.3 independant fire teams
1 tango each team in easy - tot 3
3 tangos each team medium - tot 9
5 tangos each team hard - tot 15
the teams will follow infinate random paths around the map untill either you or they are dead
they will not venture to the roof of the embassy, yet
however they will enter any building, and climb to the top of the embassy if they see you, or think you are there.
they may flank you
they may team up and assault
I found the contact points to be very similar to those in real MP games, play the map several times to experience different scenarios. As I said, this is only on one map, for this small test, so take that into account. I am also working on some other logic strings to help the ai be a little more interesting. I am planning to release this, as well as the raw script and instructions on what must be done to use it. I am also willing to divulge any info on exactly what it is they do, if anyone is interested. I'm not doing this for personal glory, it's for love of this game, and my interest in AI and fuzzy logic. There is a dev block that can be enabled by simply opening your command map, not the quick map, and toggled off by simply placing a waypoint, which you can then right click and delete. This will toggle on shadow, run, and names. If so requested, I could modify the dev block to show the enemies on the map as well, so that their behavior can be studied more closely. As it stands, this can be played in coop.
183kb Bot Hunt pre-Beta
-
IIRC, you can create an effect and have the tank shoot at that, but I think you still use the destroy target command.
-
That looks sweet!
-
Hey, lets all just say the stuff over and over, and it will magically apear!
They were developing a GR for the PC. It was said to NOT be a port of GR2 ps, and Xturd(if it was a port, why would it take this long? It wouldn't). Therefore, by deductive reasoning, we can say that it was a completely different game that they were developing. Time passed, it was not complete(this is based on the information gathered from various sources stating the discontent of individuals who beta tested it(hear-say)) and so, most likely needed more developement. More time passed. It is now time to kick out a new title for the xturd2, which will most likely be even more pc-like that it's predecessor. Taking all of that into account, it only makes perfect sense that GR2PC will be ported for the new xturd, and released for PC as GR3. The fact that GR3 was announced, simply puts the nails in the coffin, so to speak. Only time will tell, but that seems to me to be the most logical explaination.
-
D'oh. Where was I?
*I almost PIMP when I saw the last SS of the hostages, w/ the chick diggin in her butt, LOL, thats too funny!!!
-
Man, I would just love to see the idiots be able to backpedal, and or strafe. That in itself would be a marvel.
-
Why would UBI spend a grip of money to develop a game just for us PC users in this day and age? These days, we all should know that the money is in the $50 a pop, no support, no patch, no cares after it's sold console games. GR2PC=GR3PC-->port-->Xbox-->PSw/e. It just makes sense. That way, they get to sell it as a whole new game across three markets, and they basically just have to finish the job they should've already had done. We will get GR2pc when it is done, it will be called GR3pc, and it will be ported to the consoles, and sold as a new game to the eager console people as a brand new GR, as it will, in essence, be. /speculation
Let be written, let it be done.
-
I'll have to check it out at home later, my crappy work puter doesn't even have ppt viewer.
-
Looks like the first block is missing the block preserve operator. The second block you need to change the operator for the compare things query you are using under continue if to is greater than, instead of equal to.
-
I'll check it out again tonight.
-
Here's the deal, I just got off the phone w/ Mrs. Cleo;
They were developing GR2 for PC, then they had to make some revisions, and upgrades to satisfy us picky PC players. The release was looking like it was going to be way, way behind the console versions, and as it is basically a completely different version of the game, the company desided to just keep working on it, and release it as GR3. They will port it out for the consoles, perhaps the Xbox next gen console, and then they can make the whole thing worth while. Same game we've been waiting on, new name. I trust in Mrs. Cleo.
-
Good question. One thing, if you skip the cine, the map is empty.
I also got a CTD when I rushed the NW corner of the map, I think perhaps I killed the driver of one of the trucks that appears, my other team was creeping in the front while I ran rambo style around. -
It's just hard for me to understand why a company would allow their fans to develop such discord towards them, when a simple yes, no, maybe so would really make a huge difference. Of course, I realise the RSE peeps that visit here aren't at the top of the COC, so they can only do what they can, however, it's extremely dissapointing.
-
Sounds great. I can't wait either.
-
I just mean hows it coming?
-
That sounds midly encouraging. Thanks for the find.
Bot Hunt
in GR - General Mod Topics
Posted
Thank you for your input. To clarify;
The bots are indeed wearing camo and are armed with silenced m4 and frags. Because of thier vast limitations, I added the silenced weapon, and boosted the actor stats to create a more challenging game. They actors are set to ghosts since this is supposed to be a MP simulation.
Here is a brief description of thier behavior patterns, to help explain them waiting in the middle of the street;
Thier main thought processes are two, assert, and assert combat.
1) Assert will call them to choose a random direction from the four available directions, north, south, east, or west. If they reach the end of thier path, (the corresponding base) they are called to assert a new direction at random from the other three choices. When they select a plan, they also set a random time, (which is one of the varibles that can be adjusted to control their behavior) when this timer runs out, they will pause, for a random time(another variable adjustment), and then re-assert. They will continue this behavior until you or they are dead, effectively causing them to scour the map. The random travel times, and pause times, create more human-like movement patterns. Because they may move 1/4 of the way across the map, then change directions, only to be repeated infinite times, they are completely random, and so are thier paths. They have a single path point placed at the far poles of the map(bases), the geometry contained within these poles defines their "playable area". The main drawback, and the thing that causes them to sit in the middle of the street is, when they pause to re-assert. they do not have the commands to find cover, or to even cover thier flanks, it is unfortunate, but a shortcoming of the AI. I have tried several different things to try and eliminate this, and will keep working on it. It is a rather tricky task considering they are never in the same spot twice, and you never know where they will stop.
2)Assert combat is looped over all the enemy actors, and will disengage them from any path plan they may have been assigned. It will set them to basically be free to attack or behave however they are programmed to according to the commands of the preset plan, which of course, can be set and or changed to your desire. They are triggered into "combat mode" when they hear or see any member of the player team. There is a timer that is set to a random number(once again, a variable that will control their over-all behavior patterns) and when it expires, it will query if the team still sees or hears the player, if they are not in direct contact, the team will be removed from "combat mode", and called to re-assert, basicallly sending them off in one of four directions(in turn enabling them to return to combat mode at any time).
The actors will toggle back and forth between these two modes indefinately, or untill you or they are dead. By adjusting the plans, of which there are only 8 total(4 path plans, 1 combat plan, 3 pause plans), and the various random times, the teams can be set to exhibit various behaviors. For example, by setting the movement timer to a longer period, they will move much farther before they pause to re-assert, causing them to cover the map much faster, and thus be less cautious. Also, by adjusting the pause times, one can lower the chance that the player will come across the bot running away from or toward them, and increase the probability of the bot catching the player moving instead(except for the fact they like to stand out in the open
). I have also been adding some strings to effect the way the teams interact with each other. For instance, a string providing back-up if another team is in a conflict, and the current team is available, and of course if the odds are right. Also, a string to remove a path plan from assertion if any two teams have already died while on that path plan, effectively removing the possibility of all three teams rushing the player spawn, this is not to say they won't still do it if in combat mode. 
To be honest, I think your best gameplay will be from the recruit setting. That way, there is only one actor on each enemy team, and they seem to function alot better that way. I'm pretty sure I may go ahead and add some more teams, the scripting is just pretty intense, and since there is not much interest in this, it doesn't seem, I may just keep tinkering with the three teams for my own enjoyment, we'll see.