Forgotten Hope Public Forum

Forgotten Hope 2 => Singleplayer and Coop => Topic started by: Zoologic on 02-10-2010, 10:10:51

Title: FH 2.3 SP known errors & bot erronous behavior
Post by: Zoologic on 02-10-2010, 10:10:51
Ah my first free day with FH 2.3! Now I feel that I must pay some contribution by posting this.

Well, known crashes to me:

1. Lebisey crashed when map fully loaded (100% then CTD).
The map has no navmesh.
Fixed by Remick04 1st Fix
The map is now playable.

2. Operation Totalize-64 will eventually crash.
As long as the Canadians don't take the Windmill flag, all will be fine. But then, I play as Canadian and rushed all the way to the windmill (numerous times), and 3 times playing, all end up with this error message:
(http://img.photobucket.com/albums/v242/Johnsy/Battlefield/FH2/th_FH23bug5.jpg) (http://smg.photobucket.com/albums/v242/Johnsy/Battlefield/FH2/?action=view&current=FH23bug5.jpg)

3. Operation Goodwood-64 displays dummy flags capture message.

4. Anctoville 1944-64 displays "ERROR when dealing with (certain object)"
The message will appear periodically, but doesn't affect anything to my observation. The game still plays well and the bots will do their job.
(http://img.photobucket.com/albums/v242/Johnsy/Battlefield/FH2/th_FH23bug1.jpg) (http://smg.photobucket.com/albums/v242/Johnsy/Battlefield/FH2/?action=view&current=FH23bug1.jpg)

5. Port-en-Bessin MG34 gun position in a house adjacent to the Shipyard makes player stuck
I almost given up pressing 'e' key and chose suicide instead, until a clever British bot riflenaded me  :)
(http://img.photobucket.com/albums/v242/Johnsy/th_FH23bug4-1.jpg) (http://smg.photobucket.com/albums/v242/Johnsy/?action=view&current=FH23bug4-1.jpg)

6. Ramelle-Neuville Parker's .30 cal confirmed may cause crash
I was playing as Germans, seeing the bots failed numerous times, i decided to rambo the church tower alone and do whatever i can to clear the path from defending Amis. I man the MG, snipe, and shoot many bots. But one bot with Bazooka slipped my MG sight, I yanked the MG to the right bottom corner trying to aim for him, but suddenly it CTDed without any error message (despite playing in windowed mode).

7. Operation Goodwood now crashes
I applied Remick's first SP patch and ESAI FH2 edition. As Germans, we hold LeMesnil Frementel and LePreiure for long, before it finally crashed. The British are bleeding, because both Cagny flags are in German's hand. Will find out the possible cause.

Bots that do not play properly:

1. Port-en-Bessin bots do not follow the flag cap order
Probably in the StrategicAreas.ai file, the bot British bot will go straight to the uncaptureable Port-en-Bessin from Outskirts. Must follow the push mode order.

Fixed by Remick's SP patch and Void's ESAI FH2 edition

2. Bots flying Hurricanes in Mt. Olympus easily crashed when turning hard
They fly much better in other maps, dunno why.

3. Villers-Bocage hedgerows makes bot stuck on them
The bot driving vehicles seemed not able to find better path to reach their fighting positions. Instead of real tank battle, i only see allied tanks getting into town and get pwn3d by infantries, while the mighty German tanks stuck in hedgerows.
(http://img.photobucket.com/albums/v242/Johnsy/Battlefield/FH2/th_FH23bug3.jpg) (http://smg.photobucket.com/albums/v242/Johnsy/Battlefield/FH2/?action=view&current=FH23bug3.jpg)(http://img.photobucket.com/albums/v242/Johnsy/Battlefield/FH2/th_FH23bug2.jpg) (http://smg.photobucket.com/albums/v242/Johnsy/Battlefield/FH2/?action=view&current=FH23bug2.jpg) Pictures Irrelevant
Bots will still struggle, but managed to get through all of the times, we can now see real tank fightings.

4. Bots in Ramelle-Neuville disregard towed 20 mm FLaK at all.
In small numbers, the Germans pwn3d the Amis by their superior equipment. When given more players, Americans can actually hold the Germans at the bay by using swarms of Zook rockets while the tank drivers are busy finding the right track. Also, the zook rockets seems to kill in pretty large radius too.

5. Bots driving vehicle in Operation Goodwood cannot cross railroad underpass and overpass.
The Germans are pretty much stuck in Goodwood, their counterattack only come with infantries and small armour units. Investigation leads me to their base: they forward and reverse the tanks and halftracks too much, afraid hitting the underpass wall. Probably their collision mesh avoidance played a part in this. Otherwise, we can "waypoint path" this underpasses and overpasses so bots can drive over them easily into the fight.

Fixed by Void's ESAI FH2 edition

6. Bots driving tanks in Anctoville-1944 cannot travel properly.
Bots in Anctoville, where streets and paths are narrow seems to be very clumsy when driving tanks. They just don't get to action too often. They forward and reverse too much, perhaps too afraid of scratching the fences, walls, and others.

Well that's all, will update as soon as I find other problems and its probable cause.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: aserafimov on 02-10-2010, 11:10:25
some feedback form my side, I played st.lo., ramelle, villers

1. Lebisey 16+64 sizes- crash when the map is 100% loaded

2. Ramelle- great town battle.

3.St.Lo.-my new favorite map, just a beauty! :-*

4.Villers- I could not saw any tank/infantry battle, german tanks stay on place in their main base, the brit tanks stay on place around ABC line, for 30 min play only three brit tanks have been crossed their ABC line.
Soft infantry battle.
Definitely I will play my 64-size version of Villers_Bocage, while these problemss are still available.
screens:

(http://i761.photobucket.com/albums/xx254/aserafimov/SP%20problems/th_screen240.jpg) (http://s761.photobucket.com/albums/xx254/aserafimov/SP%20problems/?action=view&current=screen240.jpg) (http://i761.photobucket.com/albums/xx254/aserafimov/SP%20problems/th_screen237.jpg) (http://s761.photobucket.com/albums/xx254/aserafimov/SP%20problems/?action=view&current=screen237.jpg) (http://i761.photobucket.com/albums/xx254/aserafimov/SP%20problems/th_screen234.jpg) (http://s761.photobucket.com/albums/xx254/aserafimov/SP%20problems/?action=view&current=screen234.jpg) (http://i761.photobucket.com/albums/xx254/aserafimov/SP%20problems/th_screen233.jpg) (http://s761.photobucket.com/albums/xx254/aserafimov/SP%20problems/?action=view&current=screen233.jpg) (http://i761.photobucket.com/albums/xx254/aserafimov/SP%20problems/th_screen239.jpg) (http://s761.photobucket.com/albums/xx254/aserafimov/SP%20problems/?action=view&current=screen239.jpg)
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Devilman on 02-10-2010, 12:10:40
Lebisey shows in the SP map list,but contains NO AI files
except for StrategicAreas.ai & Strategies.ai in the SP map size folders
it has no AI or AIPathFinding folders in the server rar file

which is definately the cause of the CTD on join
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Zoologic on 02-10-2010, 12:10:53
Lebisey shows in the SP map list,but contains NO AI files
except for StrategicAreas.ai & Strategies.ai in the SP map size folders
it has no AI or AIPathFinding folders in the server rar file

which is definately the cause of the CTD on join

Just checked too, and found the same thing. The map isn't navmeshed either.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 02-10-2010, 12:10:30
Both sp and coop?

Someone said the Parker's .30cal in ramelle tower also ctds n the p51s fly into each other
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Devilman on 02-10-2010, 13:10:07
Both sp and coop?
 

The (missing) AI & AIPathFinding folders in the server rar file,are used for both SP & CO OP
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 02-10-2010, 16:10:12
So basically, no bot support for Lebisey
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: aserafimov on 02-10-2010, 18:10:58
update:

anctoville 16 size- crash at 14%-16%

(http://i761.photobucket.com/albums/xx254/aserafimov/SP%20problems/th_error.jpg) (http://s761.photobucket.com/albums/xx254/aserafimov/SP%20problems/?action=view&current=error.jpg)

point du hoc 32 size

some bots died, probably due to bad spawn points

(http://i761.photobucket.com/albums/xx254/aserafimov/SP%20problems/th_screen242.jpg) (http://s761.photobucket.com/albums/xx254/aserafimov/SP%20problems/?action=view&current=screen242.jpg)
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Remick04 on 02-10-2010, 18:10:37
Ugh, seems like there are more problems than I anticipated with 2.3... Ramelle, and PHL were the only maps I thought had problems. I tested everything else to death and it all seemed fine.

As you can tell from the Road to 2.3: part 2 update Lebisey was working in SP/coop. I took a couple sceenshots of bots on that map, one of which is in the update. So, somehow between me testing and taking those screenshots, and the files being released it lost its Ai and Aipathfinding folders.

Operation Totalizes' crash I believe is an AIpathfinding issue, which I thought I fixed. Try playing in windows mode and let me know if you get an error.

Port en Bessin's StrategicArea files are not the ones I made that do follow the push order, last minute changes were made to the map and my StrategicArea files got overwritten. Actually there are a couple of maps with a similar situation.

I'm not sure was up with Villers for you aserafimov, unless your SP files for the map are interfering with 2.3's, Because I've never seen that problem before. If anyone else experiences this problem let me know.

And I tested the 30.cal in the bell tower on Ramelle last night and had no crash so I'm hoping it was just a fluke for one person.

I really want to release an unofficial fix to some of the problems now, because a lot of them shouldn't exist
 :-\
 

edit: looks Like Anctoville SP1 file dosen't have it's strategicareas file you can copy it from the Coop 16 folder and it should work.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: VonMudra on 02-10-2010, 18:10:25
Anctoville 64, at least for me, the bots get an error and just stand around doing absolutely nothing :P
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 02-10-2010, 21:10:38
@other SPers
Alryt boys, you heard the man; let's get debugging! I for one hope to get the phl fix in addition.

Should get to play 2.3 finally tonight. Will give a comprehensive feedback then.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: gijas04 on 02-10-2010, 22:10:28
Ramellle at least size x64 was not good for bots. They couldn't capture the first flag and were all bunched together there but nothing happened. The enemy AI just flanked around them. Also, with the terrian of Ramelle being mostly debre and uneven the AI mechenized armour had a hard time moving. Brest is OK but the AI at least on my team doesn't head to the front but stays behind the lines and does me no good.

And to add, I have had the same problem mentioned above on the Villers map. The bots are blocked by the hedge rows which is a path finding issue - kind makes me laugh though as the old nick name for the actual battle was "hedgerow hell" and that's exactly what it is for me and the bots.

Also, when I play coop local against the bots I pick my maps usually five or so and then it seems like after the third map change I get a crash to the desk top everytime? Not sure but I've tested it with random maps and the same thing happens after the third map change? The Cobra map crashed as well when it was loading.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Raziel on 03-10-2010, 11:10:34
Love 2.3!
That's all! The fact that Port en Bessin has been pathmapped makes me very very happy! Was waiting for this for soooo long!! Mareth line wooot Wooot! I still need to find some good quality time to go through everything that has been included!
A BIG Thanks for everyone who worked on this patch! Appreciate it a lot!
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Devilman on 03-10-2010, 12:10:10
just tried mount_olympus 16 size
as soon as you exit the castle towards the battle,your out of the combat area (doesnt seem to effect bots)
deleted all of the combat area code from the bottom of the GPO file,problem fixed
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Slayer on 03-10-2010, 15:10:30
Ramellle at least size x64 was not good for bots. They couldn't capture the first flag and were all bunched together there but nothing happened. The enemy AI just flanked around them. Also, with the terrian of Ramelle being mostly debre and uneven the AI mechenized armour had a hard time moving.
Seconded. And besides: the AT bots on US side shoot 90% of their shots too low so the rocket ends up in the ground 20 ft in front of them.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 03-10-2010, 15:10:35
Ramelle is not meant to be played 50:50. Captain Millers strategy was too good for ze German attackers, and their stupid Tiger TKs alot. I set it to 24:40 bots and the Germans got to the Bridge even though I was American! It plays fine

You do need to work around increasing bots size or changing numbers as the case may be. Download the functional FH Toolkit to set bot count easily and remember, if the bots are set to 64, you need to push the total number ingame to 48 (The very limit of the toggle) and set the ratio to 64 to restore a 50:50, the same is for all numbers you want. 100 bots means, set the total bot count ingame to 48, set the ratio to 100, 96 bots, 96 ration, so on and so forth.... Easy peasy!


Current Errors Detected:
1/ US 105mm howitzer on Cobra uses the Conquest model not the _AI, noticeable by a normal level, rather than raised POV

2/ I think the CTD in Totalize comes from the Wasp. It otherwise plays EXACTLY like it always did, including the buggy Canuck commander telling you to attack across the river after the only a half-ass attempt at the factory, completely forgetting the PAK40, ack-ack and 88 base. I played till we had made about 1 or 2 attempts at that base then I went looking for the new toy. Tried to get a bot to use the flamethrower and he bailed almost instantly. I tried it got a whoosh, then right after the first burst, CTD... might be a coincidence, but it seems like a likely perp

3/Villers... Yes the tanks did NOT cross the ABC line. Wespe did not fire at anything, or did maybe once. I guess its code is that of the marder? The tanks get to the ABC and halt there, as though not sure of themselves, trying then coming back a bit. Halftrrucks don't seem to have this issue

4/ It seems the nebelwerfer uses the old code that makes it fire short. I noted that in Goodwood it fired at infantry which I was impressed at until I realised it was trying to fire at 2 Shermans beyond them, but firing short. Didn't get it to fire in Totalize to compare (My friends PC can only do 64bots before it lags, so not all defenses were manned for me to get the full picture)

5/ Mareth, PeB and Totalize still have dodgy commander AI. Wa-ay better than older maps, but the logic of the attack plan is a bit iffy and only just within the allowable spread of the PUsh system. In mareth bots wait straight for the bunker base with little effort to take the town and then went on for the Tiger base and it all become a scrambled mess after that

6/Parker's .30cal in the church tower on Ramelle does NOT CTD but there is no way for bots to get to it :-( unlike with Sidi Rezegh :-)

7/Quad .50 only works as anti-personelle gun and not AA... It doesn't look up at all

8/ Bots have an issue of sitting on mg positions on vehicles such as the hull gun or turret gun and don't switch positions to drive the otherwise empty tank etc off... might need to rebalance their preference for positions - They also  bail from trucks at random times sometimes getting themselves TKed, might need to play around with this to prevent them camping, but getting them to travel to the flag in it.

Also, some mg positions are not very liked. Can't think of one offhead though. i know wasps flamethrower is one. Tanks have too 'soft' a main position. I saw a good many abondoned tanks. Might need to work on it being closer to that of the static gun, but not quite there

9/ I had a PAK40 on mareth firing HE at my medium tank, and never AP, not sure if that's just a coincidence, or its got its shell priorities wrong

10/Comp. B has the c4 issue again i.e bots drop them defensively causing TKs - Noted this on Ramelle

11/ The planes crashing into each other on Ramelle - are so many supposed to fly by at the same time, or its the bug?

12/ No love for the mobile ack-ack?

13/ Still need to somehow see how we can get AT guns to want to go prone more often. I want them to camp prone, sniping infantry and tanks alike. This I'll make a fan-mod priority

14/ Now that bots have a lowered hip-fire 3rd person animation, we need to see if we can get them to fire down the sights at range only using hip fire at close range i.e when you get a bot firing at you at point blannk range rather than using knife

15/ I think mobile mortar still needs raise perspective. On maps like Brest and St. Lo, breaking through to cap the final base would only have been possible IF we had concealed close-support arty in the form of deployed mortars, which we could not get as the open terrain made them vulnerable to attack from great range and concealing them nullifies their use. We should try to get the 2 POVs (normal for players, raised for bots) implemented also, so we can get mobile arillery to use something between the assault gun for mobility and thw howitzer with raised POV for range

16/ Tanks get a tad too distracted by aircrafts. I think we should stop them using their hull mg to fire at those altogether. They lose focus

17/ We have an increasing number of double-cap flags i.e flags that are capped by one side, but still show the other side's flag and this causes bots to camp it. Noted on Cobra's mill flag, no longer the farm (Or so it seems), and Ramelle Neuville's first flag if they get too much resistance


However,


1/ Love the speed of firing i.e reasonable but generally enough to suppress a human, even with rifles

2/ Bots like mortars, mg42 more, which is a welcome improvement

3/ Range of mgs is EXCELLENT. on St Lo and from distant tanks, this is beautiful

4/ infantry don't spam greandes or rifle grenades but use them well

5/ The new maps play incredibly well. Ramelle is tricky for tanks, but that keeps them at par with infantry and shouls be left exactly so. Never did get to cross the bridge though, zook wielding bots and BAR men are too many. Want to see if kit limiting can somehow be applied to, or at least emulated by bot preference.

6/ I'm so grateful for bots being able to use almost every, if not every single building humans can - Even saw bots trying to climb a ladder into a barn of St. Lo from which I was firng. My men spawned on me and jumped down :-)

7/ Range of arty is functional. I'm not sure extending it further would make it a rapefest, but will need to play more to see. For now, as seen on St. Lo and Mareth, it works WELL

8/ For the first time, a new Ai Patch did not introduce new unseemly problems like is the SNAFU of former AI in past patches. New weapons all work well, the new system did not bring new bugs, and it goes without saying that there must have been round the clock work to make such things happen  

9/ 88 fires HE as far as AP and with its inaccuracy makes for a really fun objective each time me and a few bots need to charge one to put it out of action without air or armor support :-)

10/ Smoke use seems to have been reduced. So its done once and left. Its so nice to see it in use but not dominate bot priority

11/ Spawnpoint mgs are excellent, and it seems so you DID move the cap area for Sidi underground. Great work there. For the first time ever, Supercharge was not spammy with radio commands, the Germans put up a tough defense each way, even counter-attacking a good many times, the new howitzers position makes it more useful like in CQ and the bunker next to the 88s is a terrific defense to attack thanks to both mgs being priority

12/ Fixing commander Issues really solves a good many problems. I didn't see camping tanks in El Alamein, Supercharge or any of those, and the battle flows for all maps!

11/ Remick, you did an excellent work moving Drawde's mod into the release. i feared you having to cut it apart might mean key aspects would be lost, but its clear you actually improved on the final version.. ALOT. Great work man. I think even the ignorant conquest players are impressed :-D

12/ We have only Falaise and Lebisey without AI... Falaise works fine with  Aserafimovs's current version although I wish the bots would cross the river at more points, and Lebisey, we've heard already has AI, just not in yet. It was a sheer joy to scroll though the maps, knowing every SINGLE one has Ai now. No small feat! I plugged my own 64-size Bardia in. That also has limited AI. Great to see PDH 32 in too... Still have craxy ideas about bots driving to the hills and assaulting the hill defenses etc on PDH 64, but let me make that a practical and complete thought ;-)


Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Zoologic on 03-10-2010, 17:10:37
Operation Totalizes' crash I believe is an AIpathfinding issue, which I thought I fixed. Try playing in windows mode and let me know if you get an error.

Still working on this Remick04, I love the bot's massive improvement, you guys are awesome. They are no longer brainless & clumsy, they now really know how to use the tool quite well.

Quote
Port en Bessin's StrategicArea files are not the ones I made that do follow the push order, last minute changes were made to the map and my StrategicArea files got overwritten. Actually there are a couple of maps with a similar situation.

Yeah, i found several maps with this issue as well. The commander AI send their troops to the wrong flags. But the issues are minor as i notice that only half of the squad of the team were sent to the wrong flag, and the rest were sent to the correct ones. In some maps, the commander AI will split the team into three different attack orders.

The real problem with Port-en-Bessin is because we have to attack the Germans bases at the far right and left corner before actually going to the port itself. The game really stops here.

Quote
I'm not sure was up with Villers for you aserafimov, unless your SP files for the map are interfering with 2.3's, Because I've never seen that problem before. If anyone else experiences this problem let me know.

I get the problem as well. Have never installed aserafimov's file for Villers-Bocage, just the Falaise Pocket. The problem is only with vehicles, the soldier bots can find their way out easily.

Quote
And I tested the 30.cal in the bell tower on Ramelle last night and had no crash so I'm hoping it was just a fluke for one person.

Yes, me too. the .30 cal in the bell tower of Ramelle-Neuville is perfectly fine for me. The map runs smoothly for me, except a few bot behavior issues.


2/ I think the CTD in Totalize comes from the Wasp. It otherwise plays EXACTLY like it always did, including the buggy Canuck commander telling you to attack across the river after the only a half-ass attempt at the factory, completely forgetting the PAK40, ack-ack and 88 base. I played till we had made about 1 or 2 attempts at that base then I went looking for the new toy. Tried to get a bot to use the flamethrower and he bailed almost instantly. I tried it got a whoosh, then right after the first burst, CTD... might be a coincidence, but it seems like a likely perp

The Wespe doesn't really crash the game. In Villers-Bocage at least, it doesn't CTD the map. I'll try to remove them from the GamePlayObjects.con file to see if it really affects anything.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Remick04 on 03-10-2010, 19:10:57
I tried Villers last night and noted the same problems you’re all reporting. It seems a bunch of new Hedgerows were added to the map that weren't there when I navmeshed it, so bots are getting stuck on them. I’m debating whether I want to re-navmesh it or try and edit the mesh I have now. Not sure which would take less time.

I've seen bots using the Waspe on Totalize even firing the flame thrower when assaulting a position. But it is rare, it's not one of their favorite vehicles.

I know the Terrain on Ramelle isn't exactly bot friendly... made worse by the fact that there were changes made to the terrain after I had navmeshed it. I did my best to correct and fix things, but its hard working on adding bot support to a map that is still under development because changes are still being made. And ultimately I'd rather have the vehicles in there even if bots aren't great with them, then have an infantry only map. I'm working on fixing the fly over problem, I may have a fix for you guys to help me test.

Keep the feedback coming, I'll do what I can to address as much of it as I can. Hopefully Drawde can clean some of the AI stuff up for me, and I'll focus on fixing map specific issues.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: gijas04 on 03-10-2010, 21:10:28


You do need to work around increasing bots size or changing numbers as the case may be. Download the functional FH Toolkit to set bot count easily and remember, if the bots are set to 64, you need to push the total number ingame to 48 (The very limit of the toggle) and set the ratio to 64 to restore a 50:50, the same is for all numbers you want. 100 bots means, set the total bot count ingame to 48, set the ratio to 100, 96 bots, 96 ration, so on and so forth.... Easy peasy!


The Toolkit in FH2 2.3 doesn't work for me and I get a error. Where can I download a functional FH2 Toolkit from? I think I understand the bot ratio tips you described but if I have say, 48 bots and set the ratio to 50/50 then I check the number of bots in-game and they are correct which is 24/24 on each side plus one for me. I edited the aidefault file and increased the numbers to 40 for number of bots as suggested. I did this in older versions of FH2 as well. I like that you can also increase the Bots skill level there. I set mine to 0.90 because I think 1.0 would be to hard for me.


Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 03-10-2010, 22:10:21
No, the one I'm refering to is in the DL Sticky of the Singleplayer section of the forum. Goto http://fhpubforum.warumdarum.de/index.php?topic=6539.0 and find the DL there

I could've given you the direct link, but I want SPers to get used to the DL sticky for maps and fixes as they come along

Just replace the toolbox you have with the one at the link. The only other DLs you need in addition are Falaise Pocket, Bardia and Aserafimiv's Lebisey infantry and maybe, his Villers, Bocage. remeber to backup whatever you are replacing in case you need to restore it or get an error ;-)
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: gijas04 on 04-10-2010, 00:10:59

Just replace the toolbox you have with the one at the link. The only other DLs you need in addition are Falaise Pocket, Bardia and Aserafimiv's Lebisey infantry and maybe, his Villers, Bocage.


I Downloaded the FH2 Toolbox file from the link but I still get this error when clicking on the FH2 Toolbox.exe in the FH2 Toolbox folder: Run-time error '339: Component 'MSWINSCK.OCX' or one of its dependencies not correctly registered: a file is missing or invalid?

Also, there are two final versions of Lebisey and Villers, Bocage maps on the second page of the link you gave me. Do I want to download the push mode versions on the first page or these final versions?
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 04-10-2010, 01:10:44
Well, for Villers, you might actually want to look at the thread of the same name. Not sure if cF updated the DL links, but that one would be the latest

For Lebisey, tbh, I'd say wait for Remick to release, unless you don't mind a 16 only size map. Not sure which is which for that one though - Read comments around each and see


Sorry I couldn't be more help here. Had my PC on fritz for a while so I'm kinda outta date
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: aserafimov on 04-10-2010, 04:10:20
@gijas04

final link of Lebisey 16 size is:
http://www.4shared.com/file/lLPS8pj6/lebisey_push_mode.html

Villers Bocage

link of Villers 64 size:
http://www.4shared.com/file/8vFkURp2/villers_bocage_64size.html

link of Villers, alternative map layout:
http://www.4shared.com/file/tkMx1oOO/villers_bocage_push_mode.html
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Ghost on 04-10-2010, 05:10:43
Hi,

Yes, me too. the .30 cal in the bell tower of Ramelle-Neuville is perfectly fine for me. The map runs smoothly for me, except a few bot behavior issues.

Hm, i noticed the problem with that mg too. when i try to use it, i get a CTD. I noticed another problem with an mg, on Port en Bessin. The mg in the church tower, if you use it, you get stuck and cant leav it. But i think this is just a map bug.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: gijas04 on 04-10-2010, 06:10:48
Hi,

I noticed another problem with an mg, on Port en Bessin. The mg in the church tower, if you use it, you get stuck and cant leav it. But i think this is just a map bug.

I second that bug. I was on the same MG near the church tower and was stuck there.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Zoologic on 04-10-2010, 07:10:09
It seems every .30 cal and MG34 MG emplacement in Port-en-Bessin makes you stuck.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Devilman on 04-10-2010, 08:10:21
It seems every .30 cal and MG34 MG emplacement in Port-en-Bessin makes you stuck.

There are several stationary weapons,that have 2 versions..example
mg34_bipod & mg34_bipod_ai

the mg34_bipod_ai is coded to NOT allow exit,this is done to keep bots always on the weapon (they still can be killed)
Quote
ObjectTemplate.dontAllowExit 1

ideally these should mainly be used on the non human side

if you want these weapons exitable,on certain maps,open the GPO and change mg34_bipod_ai to mg34_bipod
or if you want them exitable from all maps,remove this line from all stationary weapons ending in _ai
Quote
ObjectTemplate.dontAllowExit 1
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 04-10-2010, 10:10:42
that's wierd, I never knew there wasc _ai mgs too. I thought the values for not getting off them were simply set so bots 'liked them too much' but not in such a way that humans got stuck?!

Maybe Remick or Drawde might have to clarify this one. Because that will suck if you cannot use certain mgs because they are bots-only.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Devilman on 04-10-2010, 10:10:38

Maybe Remick or Drawde might have to clarify this one.

There is no need for anyone to clarify "this one"
I know what i am talking about

A human can always enter a vehicle or weapon which is occupied by a bot,regardless wether it does,or doesnt allow exiting

(http://i135.photobucket.com/albums/q137/DEVILMAN_09/FH2WEAPONS.jpg)
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 04-10-2010, 11:10:52
Sorry, Devilman. Didn't intend to pull ranks. I just wanted to understadn 'why' they made it so is all. I can appreciate that humans can use the vehicle, but not being able to exit afterwards - I didn't know they made it so. I thought it was just bot values that made them stay on, as opposed to the gun making anyone on it get stuck

Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Devilman on 04-10-2010, 11:10:15
I just wanted to understadn 'why' they made it so is all. I thought it was just bot values that made them stay on, as opposed to the gun making anyone on it get stuck

Short story,bots are dumb,they are preprogrammed with many priorites
The best way to guarantee,they do a certain thing,is to guarentee they cant do anything else
There are some mods,and some maps where the exit positions are accidently blocked,which will also stop users exiting

Under normal circumstances a bot will remain in a vehicle/stationary weapon,until something with higher priorites comes within their range,map strategies and AI behaviours,can also play a large role in these situations
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 04-10-2010, 12:10:20
Here's my problem with that approach. Fist off, I find it hard to believe that the code affects players also because I would have gotten stuck on more mgs in the past, considering that bots camp ALL mgs.

Whats more, if we were to be dismissive about these 'dumb bots', we'd have settled for 2.0-style Ai, and we havent.

I saw a thread about mgs that players got stuck on at the Bug reporting section, and I'm wondering if this is only a bot issue, or if it has a solution, besides stopping them from camping altogether - which we can all agree is a bad idea
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Devilman on 04-10-2010, 13:10:30
Here's my problem with that approach. Fist off, I find it hard to believe that the code affects players also because I would have gotten stuck on more mgs in the past, considering that bots camp ALL mgs.

Whats more, if we were to be dismissive about these 'dumb bots', we'd have settled for 2.0-style Ai, and we havent.

I saw a thread about mgs that players got stuck on at the Bug reporting section, and I'm wondering if this is only a bot issue, or if it has a solution, besides stopping them from camping altogether - which we can all agree is a bad idea

Yep,your right and i'm wrong
I know FUCKALL
Now if you will excuse,i,m going to check out some modding videos,by some guy called Devilman

http://www.youtube.com/user/DevilmanWhiteCats?feature=mhsn
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 04-10-2010, 14:10:44
Erm, dude. Now YOU'RE pulling rank. I'm not judging your knowlege. But fact is if we conclude that bots cant be made better, we'd have AI only at the level of AIX i.e nice, simple, arcade. We are trying to make a gamemode that still works with logistics, if not tactics.

I can confidently say FH2 AI is now better than AIX and we didnt get there dismissively 'accepting' the facts but trying to diagnose situations conclusively (Ask Remick and Drawde. We always get multiple voices on any argument irrispective of who it is). That way we know if we can push harder to fix it or need a workaround.

Sorry if I came across dismissive.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Remick04 on 04-10-2010, 18:10:01
The AI MGs were created by Winterhilf prior to Drawde's or my involvement as his solution to getting bots to stay on MG positions. Drawde's AI files do render them useless, so they could probably be taken out of the mod files. I used them on a couple of maps before Drawde's mini mod was applied, but replaced them with standard MGs once the AI tweaks were in. Port en Bessin wasn't one of those maps though, but like I said PeB was updated by the mapper prior to release. So, the GPO and Strategicarea files are not mine.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 05-10-2010, 20:10:55
Cool, then its about changing the guns then. Kewl.

Updated possible project list in the 'singleplayer fan project' thread. Check out the thread and lets do all we can to fix stuff for a possible Drawde 2.3 compatible mod/fix - if, Drawde, you'd oblige us.

This thread is still the best for bug detection. zM, fret not. Im not torpedoing this. Just funneling results from here to proactive work.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Devilman on 06-10-2010, 01:10:13
Fist off, I find it hard to believe that the code affects players also because I would have gotten stuck on more mgs in the past, considering that bots camp ALL mgs.

First off,I am still stunned that you are questioning my information

Not every MG is non exitable
as I posted earlier there are 2 types,exitable and non exitable
some maps have exitable,some maps have non exitable
and some maps even contain both

 
from port_en_bessin SP 64

Quote
rem [ObjectSpawnerTemplate: cq64_wn57_mg42]
ObjectTemplate.create ObjectSpawner cq64_wn57_mg42
ObjectTemplate.activeSafe ObjectSpawner cq64_wn57_mg42
ObjectTemplate.modifiedByUser "ramon"
ObjectTemplate.isNotSaveable 1
ObjectTemplate.hasMobilePhysics 0
ObjectTemplate.setObjectTemplate 1 mg42_bipod
ObjectTemplate.minSpawnDelay 999
ObjectTemplate.maxSpawnDelay 999

rem [ObjectSpawnerTemplate: cq64_wn56_kwk5]
ObjectTemplate.create ObjectSpawner cq64_wn56_kwk5
ObjectTemplate.activeSafe ObjectSpawner cq64_wn56_kwk5
ObjectTemplate.modifiedByUser "ramon"
ObjectTemplate.isNotSaveable 1
ObjectTemplate.hasMobilePhysics 0
ObjectTemplate.setObjectTemplate 1 kwk_5cm_fr
ObjectTemplate.minSpawnDelay 999
ObjectTemplate.maxSpawnDelay 999

rem [ObjectSpawnerTemplate: cq64_wn56_mg34]
ObjectTemplate.create ObjectSpawner cq64_wn56_mg34
ObjectTemplate.activeSafe ObjectSpawner cq64_wn56_mg34
ObjectTemplate.modifiedByUser "ramon"
ObjectTemplate.isNotSaveable 1
ObjectTemplate.hasMobilePhysics 0
ObjectTemplate.setObjectTemplate 1 mg34_bipod_ai
ObjectTemplate.minSpawnDelay 999
ObjectTemplate.maxSpawnDelay 999

rem [ObjectSpawnerTemplate: cq64_wn56_mg42]
ObjectTemplate.create ObjectSpawner cq64_wn56_mg42
ObjectTemplate.activeSafe ObjectSpawner cq64_wn56_mg42
ObjectTemplate.modifiedByUser "ramon"
ObjectTemplate.isNotSaveable 1
ObjectTemplate.hasMobilePhysics 0
ObjectTemplate.setObjectTemplate 1 mg42_bipod_ai
ObjectTemplate.minSpawnDelay 999
ObjectTemplate.maxSpawnDelay 999


Also as I posted before
Quote
ObjectTemplate.dontAllowExit 1
dontAllowExit and 1
1 meaning YES
So,Do not allow exit..YES
That works for humans and bots
it is written in black and white
which part dont you understand  ???

Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 06-10-2010, 01:10:37
We're good here - thanks.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 06-10-2010, 01:10:17
Actually, Devilman, could you help us at the 'fan project' thread with code to get Bardia 64 sp3 work in coop too, maybe someone could make a new updated DL for it then cF could post it on the link section?
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Devilman on 06-10-2010, 02:10:39
WOW,look what I found
Another post of mine clearly explaining about non exitable MG's

(http://i135.photobucket.com/albums/q137/DEVILMAN_09/exitablemg.jpg)

http://fhpubforum.warumdarum.de/index.php?topic=10628.0

Post #10
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Zoologic on 06-10-2010, 08:10:58
Updated another bot erroneous behavior... check the first post.

Operation Goodwood bots cannot drive through the railway underpass and overpass.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: gijas04 on 07-10-2010, 03:10:52
Updated another bot erroneous behavior... check the first post.

Operation Goodwood bots cannot drive through the railway underpass and overpass.

I noticed this to on Goodwood. Especially with half tracks.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 07-10-2010, 15:10:39
Bots now love both lafette and bipod mg42, but never use .30cal tripod - not tried other deployable mgs. I recommend making them not bail from it like with the static guns or they may not fire it, bail and have it lost to the player

Also, please look at vehicle, esp. Tank 'heat' for each seat. Bots tend to know how long to stay riding tanks, but no other vehicle's passenger seats. They bail instantly.

They also dont switch between turret or hull mg to driver seat when the tank driver leaves which was fixed in the minimod

Also, they abondon healthy tanks for no reason. Solution also in minimod heat value for primary tank seat.

mobile 20mm on ramelle never used

All reverb sounds i.e 88, pz2, 20mm, mg34 etc are back. They make firing sound between actual shots.

allied Tanks on brest get to the 3rd or 4th flag and jurt sit there unless you drive beyond it yourself and give it back to them. Must be navmesh?

Pak 40 fires only HE at tanks.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Mud Buddha on 07-10-2010, 19:10:32
Don't play a lot SP FH2, but since the good servers are always full now 2.3 is out (which is a good thing) I started getting a little more into SP.

First thing I noticed that on Olympus allied bots don't use planes. As a german I did a couple of runs on the enemy airfield in a 109 but there was no one to shoot at, just three planes in a row. Only after I switched teams and grabbed a Hurricane myself did they start spawning on the airfield and taking the occasional plane.

Also twice in a row I chased a german pilot attacking the NZ airfield  into the no-fly/swim/walk/drive zone at the right of the airfield and both times the pilot didn't make any effort of trying to fly out of that area. He did try to avoid me but was basically flying himself to death by not flying into the "normal" part of the map. I did die before I got him, because you know how it goes - you just have to try and get that kill, but apparently the bot pilot didn't. Does the out of bounds area affect bots at all?
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 07-10-2010, 19:10:34
Yes it does.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Zoologic on 08-10-2010, 19:10:57
Updated:

Finally i crashed in Ramelle-Neuville when using Parker's MG.

They sometimes fire just fine, but it definitely crashes when we do something with it.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Ghost on 08-10-2010, 23:10:53
St. Lo crashed. I played the map a few times without any problems, and now i got the first crash on it. 
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Zoologic on 09-10-2010, 15:10:17
Update:

I think i found the Totalize bug that cause CTD. It confirmed Remick04's suspicion, it is indeed AIPathFinding issue as shown by the screenshot i posted.

See first post.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: zxx43 on 09-10-2010, 15:10:20
Yeah,the St. Lo has some problem.It crashed when I played it for the first time.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Zoologic on 09-10-2010, 15:10:27
Did you guys played St. Lo in SP or Coop mode? The 64 or 32 size map?

Cause i haven't experience any crash yet with St. Lo.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: zxx43 on 09-10-2010, 15:10:37
SP,32 size.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Ghost on 09-10-2010, 16:10:58
Did you guys played St. Lo in SP or Coop mode? The 64 or 32 size map?

Cause i haven't experience any crash yet with St. Lo.

SP,64 size.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 10-10-2010, 08:10:12
I think i get why vehicles in goodwood cant cross the rail @ either under or overpass

It has more to with ai than navmesh, although that plays a factor.

Bots used to only reverse slightly to try a path again, now they reverse all the way, making whatever initial attempt completely nullified, so unless they make the pass in 1 take, they will try n try again forever.

That said, didnt goodwood have multiple paths across the rail itself? I recall winterhilf adding those along with removing the barnhouse spawn.

Also, i recall drawde making vehicles push against dodgy paths rather than reverse. dont know if @ some point they gave up and tried again, but i recall that system worked wonders in goodwood especially - could we try that and see? Drawde, Remick?
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Zoologic on 10-10-2010, 09:10:13
I tried to force the bots to went through the obstacles by waypointing them.

Somehow, they managed to cross them, although it wasn't a smooth one, but at least they went through it. Try this in Bardia (to ford the river) and Goodwood (the railway crossings and underpass). I used to saw the Tiger II attacking the British frontline before getting knocked out by the Firefly.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Remick04 on 10-10-2010, 22:10:32
2.3 isn't using the most up to date navmesh by Winterhilf. I didn't realize this until it was too late to make the change. I have tomarrow off from work so I'll try and compile some of the fixes I've made to a few of the problems in this thread and post it for you guys. I'll include Winterhilfs updated navmesh for Goodwood.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 10-10-2010, 22:10:04
Dont forget devilmans discovery of the lafette ctd in phl and the german spawn order for that map also.

Germans leaving the first 3 flags for yanks and no flag at the nebelwerfer is 2.2, not 2.25.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Remick04 on 10-10-2010, 23:10:14
I'm probably not going to include Devilman's PHL GPO file in my small fixes, for a couple reasons. For one, it’s Devilman's fix, so I'm not going to unintentionally take credit for it by including it among mine. And two, because I'm hoping it's a temporary fix until we can find out what’s wrong with the lafette. They are too important to the defensive positions of the map to remove entirely, and I don't want to have to release more files adding them back in once a fix is found.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 11-10-2010, 10:10:20
Anyone else note the nebelwerfer firing short?

I know it did in Drawde's 1.0 then a fix was found at the same time by winterhilf and Drawde.

If its the same thing, then the fix should be out there. Till then, the nebelwerfer will be nothing but effects, no risk
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Zoologic on 11-10-2010, 15:10:34
Nebelwerfer did shot furiously, i got knocked out numerous time when trying to kill it up close.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Remick04 on 11-10-2010, 22:10:23
Ok, here are a couple of fixups to a few of the problems reported in this thread

http://www.filefront.com/17377677/FH2.3_SPfixups_01.rar/ (http://www.filefront.com/17377677/FH2.3_SPfixups_01.rar/)

Fixes included
-Proper AI files for Lebisey
-Fixed navmesh for Villers Bocage
-Updated Gameplayobjects for Villers, Olympus, and Port en Bessin
-Fixed StategicAreas file for Port en bessin
-Winterhilfs fixed Goodwood navmesh and updated GPO files

I didn't have time to write up an elaborate installation instruction. But I'm hoping you guys know enough about what you're doing. Just be sure to place the files for each map into that maps server.zip, and keep backups of the original server.zips incase something goes wrong.

Keep posting any future issues, I probably won't be able to post any future updated files but I will keep working on any reported issues. Just these particular fixes are files you should have.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 11-10-2010, 22:10:06
Gr8 work, remick! Dont worry, we got Drawde and now devilman. We covered ;-)

Once i can run this i will create entire map files to Dl for the casual SPers
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Devilman on 12-10-2010, 04:10:04
Gr8 work, remick! Dont worry, we got Drawde and now devilman. We covered ;-)

You always had me,its just that you, never listened to me
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Rumia on 12-10-2010, 08:10:52
gonna test that out
thx remick
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: cannonfodder on 12-10-2010, 11:10:33
*crawls out from under his rock*

Sorry I haven't updated the DL thread yet, I'll fix it this weekend.

I upgraded to Win7 and bought Civ 5 and a new CPU. Between playing with my new toys and work, I've barely had time to scratch my ass.


Yesterday I finally went through the BF2/FH2 patch-fest so I could get stuck into 2.3. I've only tried a round of PeB and Brest so far, no dramas I noticed...except for one.

One that nobody has thus far mentioned (unless I missed it), and from this I assume that either: my copy of FH2 is broken, or you lot enjoy waiting half a minute to respawn... ???

That fucking mandown timer is still broken......*sigh*...I guess it's time to "fix" it... ::)


If it annoys anyone else, this is the "fix":
You want no spawntime in you singleplayer game?

# Solution 1:
Start your game, open console (~) and type:
Code: [Select]
sv.spawntime 0
You have to do this everytime your game starts

# Solution 2:
Open gamelogicinit.con and add
Code: [Select]
sv.spawntime 0 to the end of the file

You may also want to change 0 to 1, as 0 gives (really) instant spawn, which may be irritating...




Oh, and by the way kids, Legion's back... ;)

 - http://www.battlefieldsingleplayer.com/forum/index.php?s=&showtopic=14578&view=findpost&p=166534
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Barryman on 13-10-2010, 00:10:22
Tested two of your fixes for now Remick04.

Lebisey:
Startsup now, but as soon as it is started, the map shows up, but the half of it is not on screen and I cant select any point on the map to start from.

Villers-Bocage:
Works perfectly now, The AI doesnt struggle anymore with the Hedgerows.

Three things I noticed:
- I spawned inside a truck one time I dont know if this is intended
- I had had a strange problem with the crossroads flag. The capture bar was full while I was Allies, but the Axis flag remained as Icon on the map and flag and our tanks stayed at the point as if it was not captured yet (Dont know if this is rare, because it only happend once for now)
- Allied tanks having a struggle getting pass the point shown on the screenshot below.

(http://img96.imageshack.us/img96/6591/strugglec.jpg) (http://img96.imageshack.us/i/strugglec.jpg/)
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: aserafimov on 13-10-2010, 04:10:07
Lebisey 64 size

into: levels/lebisey/server.zip/gamemode/sp3/64/ the AI folder and the GPO file are from map "mount olympus"
because of this you could not select any spawn point in SP regime.

How to fix this problem?......Just copy and paste AI folder and GPO file from:
levels/lebisey/server.zip/gamemode/gpm_coop/64 into sp3/64
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Devilman on 13-10-2010, 09:10:27
Lebisey 64 size

into: levels/lebisey/server.zip/gamemode/sp3/64/ the AI folder and the GPO file are from map "mount olympus"
because of this you could not select any spawn point in SP regime.

How to fix this problem?......Just copy and paste AI folder and GPO file from:
levels/lebisey/server.zip/gamemode/gpm_coop/64 into sp3/64

you may also want to delete the AI/Strategies.ai file from SP/CO-OP 64 as both these are from MT Olympus
Note...not sure what difference having these 2 wrong files make,but better to play it safe
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: cannonfodder on 13-10-2010, 10:10:35
Anyone else notice AT bots going to sleep if there's nothing to fire their bazooka at?


@Remick: What happened to Goodwood-16?
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Barryman on 13-10-2010, 11:10:47
Thank you guys, the map is working!
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 13-10-2010, 17:10:42
Yer, the cap issue has to do with the issu of 'Double-cap' which plagues FH2 Singleplayer, sadly

Its in alot of maps. Usually happens when the enemy doesn't give the flag away without a fight right in the cap area - But it sometimes occurs for no reason - Enemy continue to spawn there, the flag is shown as friendly, and your side will remain camped there indefinitely...

The trick is to somehow get them to attack else where or get them defeated by the enemy

Either:
a/ Become the enemy and destroy them. Getting the flag back to mid point, changes it to a blank flag and nullifies the issue for that bout. Capping it all the way, might just restore things to a new instance of the issue

b/ Become a SL or commander and send them else where by command interface or commando radio
c/ Destroy them all and allow the enemy to overrun the place

I know it sucks...


I think we should really find a way of fixing this once and for all. Winterhilf did, but he never told anyone (I don't think) before he left. He did fix all maps which he'd created for the official patch ie. Fall of Tobruk's first flag etc

New challenge!!



Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Tubesteak on 13-10-2010, 21:10:14
Any chance of getting one of the Python guru's to revamp an anti bot camp script?  It was in BF2 Editor forum's ages ago, and also the XWW2 mod has a version of it incorporated.

Basically it just kills off any bot that sits idle for a pre-determined amount of time.

Also has anyone tried this AI strategy written by Void, to try and spice up the bots into attacking/defending more vigorously? 

http://www.battlefieldsingleplayer.com/forum/index.php?showtopic=15109

I've run it on some normal BF2 SP maps and it seems to work quite effectively.  I'm going to tinker with FH2 and see if there is a difference.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 13-10-2010, 21:10:21
I'm actually quite interested in the die-off script as it solves alot of idling and allows Point Du Hoc to work, cliffs and all.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Remick04 on 14-10-2010, 01:10:25
@Remick: What happened to Goodwood-16?

Goodwood-16 fell to laziness and exhaustion. I realized when putting together the files that I had forgotten it, and almost added the files last minute, but it would require editing the info's .desc file and that would require slightly more complicated installation instructions than simply "add to the server.zip". And I just wanted to get you guys the files I had as soon as possible so I skipped it. But Winterhilf's original files for Goodwood-16 should still work. Or one of the other talented guys on these forums can set it up for everyone. ;D

And I apologize for the Lebisey files mix-up. I was aware that some of Mount Olympus' files had gotten mixed up with Lebisey, and let the devs know, to which they said they fixed it. But something must have changed between then and release because Lebisey lost its AI files and Mount Olympus is still mix up in its Gameplay objects. I have a half dozen versions of Fh2 on my computer and I'm loosing track of who has what files and which files are in which version. I'm very appreciative of those of you who are picking up my slack because I have been slacking and I apologize for that.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 14-10-2010, 11:10:15
Hey, that's why we're here - Your stopgap :-)


With the sheer amount of great work done on AI, bug-fixing and map-support, this is barely a scratch


Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: cannonfodder on 14-10-2010, 11:10:00
...Also has anyone tried this AI strategy written by Void, to try and spice up the bots into attacking/defending more vigorously? 

http://www.battlefieldsingleplayer.com/forum/index.php?showtopic=15109

I've run it on some normal BF2 SP maps and it seems to work quite effectively.  I'm going to tinker with FH2 and see if there is a difference.
Nope. I've been meaning to DL it and try it ever since it first appeared, but just haven't gotten around to it. From all reports though, there's a definite improvements in the AI when using it.

May be a little premature, but it might be worth considering incorporating this into FH... :-\


Goodwood-16 fell to laziness and exhaustion...
Luckily for you, Brest is doing a great job of satisfying my CQB itch, so I'll forgive your laziness...just this once... ;)  ;D

I'll add the files on the weekend. If there's no drama's or CTD's, I'll upload the updated files and post a link in the DL thread.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 14-10-2010, 15:10:41
Hey cF, you think you could upload with the Lebisey fix (i.e to get rid of the Olympus stuff) and Goodwood 16?

If it wont be too much trouble. Would be nice to have the patches all tidy, esp. for new SPers
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Darth Nisis on 14-10-2010, 20:10:04
It seems  the GPO for Giarabub SP 64 has not been updated in 2.3, because the cannone da 47/32 M35 is not there, so i put the Conquest 64 GPO file in place of the SP GPO, and the italian gun works fine with the ai.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 14-10-2010, 20:10:16
Cool... Glad it at least has code

Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: cannonfodder on 18-10-2010, 12:10:16
DL thread updated...

I added Remick's fixes, the Lafette fix and GW-16. I'll add anything I've missed later.


@aserafimov: I took the links to your maps down because I wasn't sure if they work with 2.3 (I don't have time to check them atm). If they do let me know and I'll put them back up.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 23-10-2010, 09:10:59
... it seems they added a no go zone for New Zealand on Mount Olympus and the bots keep running out into it. Would that kill the bots? There's a lot of "_______ is no more" messages in the upper left and the tickets run awfully fast.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 26-10-2010, 03:10:14
Lebisey suffers from the spammy commander once the Brits cap all the frontline flags

Also, I note that the Germans in PHL don't spawn at the front at all, so until the 88, there is practically no resistance. No mg, use whatsoever. I wonder if their being defensive at the first line bases would aid this

Also, ALOT of maps have the double-cap issue where the defender lose the flag, but it stil shows as theirs in the map, they can spawn there and the attackers wont move on till its capped - Which it never will until counter-attacked and reclaimed

I really think we need to make this a priority to fix once and for all. I think its now the biggest issue now that Devilman has repaired the PHL crash

Remick04, you might want to look at his mod patch and see the fix to the lafettes.

Curious, am I even talking to anyone on this forum section at all. Its seems like its gone quiet in the last few weeks... Drawde? CannonFodder, Devilman, anyone?
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: cannonfodder on 26-10-2010, 10:10:41
...I really think we need to make this a priority to fix once and for all. I think its now the biggest issue now that Devilman has repaired the PHL crash...
I'd like to know why the double-capping problem is getting worse with each release.

Fall of Tobruk suffers from it at the lower right flag (the one closet to the Axis main) but, happily, the Brits have finally got their tanks back.


Have you forgotten about the "deaf" bots? They do function alot better in 2.3 now that they can get on/off vehicles when they want to, but it's still not good enough...

E.g.: Start of the round on Lebisey, I'm sitting in a tank surrounded by bots. A couple of minutes later, I drive away by myself cursing the stupid bastards because none of them felt like getting in... ::)


The solution to the other major drama (stoopid CO) has already been pointed out. I haven't had time to try it yet, so I can't confirm it, but hopefully I'll have time this weekend. The solution? ESAI. Why? To quote it's creator:
Quote
ESAI is Strategic AI code. It cannot alleviate the basic limitations of BF2 bots, some of which were enumerated by SP4Me. It works at the AI Commander level, controlling which Flags are attacked, when they are attacked, and with what level of resources.

The Vanilla BF2 Commander only has one basic strategy to work with 99% of the time - just a basic attack strategy. ESAI gives the AI commander more options. At the ESAI core there are sets of strategies for when a team is behind on flags, when the number of flags are even, and for when a team is ahead on flags.
- http://www.battlefieldsingleplayer.com/forum/index.php?showtopic=15109&st=0



EDIT: I didn't realize the double-cap happened on neutral flags as well (St. Lo-32: 1st flag south of US main)... :P

Also, Anctoville-16 crashes while loading (at 14%).
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Zoologic on 26-10-2010, 17:10:44
Double cap happens almost everywhere.

Op Cobra, Fall of Tobruk, Pointe du Hoc (all 64), St. Lo 32, has it.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 26-10-2010, 19:10:32
First off, weird behaviour by the King Tiger. Not sure if its in other tanks. I know they do it a bit, but this one took the cake: Tiger II moves up on Goodwood, open area behind the base close to the main, then reverses all the way back to the underpass - I mean like 7secs of reversing. I watch it do this repeatedly for like 5minutes then decide I'd had enough.

In Goodwood in general, German tanks tend to have real issue crossing the underpass to get to the action. Not sure if this is the version of Winterhilf where the tanks move over the dikes and across the rail at multiple points other than the overpass, but it really kills tank combat in this map.

Also, Goodwood has weird strategy that sometimes tells the Allies to attack the flag next to the German main without Cagney being capped.

About the double cap, here are some I know offhead:

1/ Operation cobra, the North-Western German position i.e the house over the river or whatever - It USED to be the farm before the unofficial version, but its changed

2/ Fall of Tobruk, the first Brit base. This was fixed in 2.26, but it might be because it was renavmeshed or something, but now its back

3/Ramelle, First American base.

4/ Luttich, the farm HAD it in 2.26, but I haven't playe it enough in 2.3 to know if its fixed

5/PDH, St. Lo? Where?

To be honest, there are fewer maps that this is in than I thought, but its a problem with a solution we are yet to know and it plagues the maps its in - I know Winterhilf figured it out when he solved in Fall, Tobruk and El Alamein but never told anyone before he left.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: winterhilf on 26-10-2010, 19:10:38
Re: Python flag capping bug.

Tried a few theories (seemed to work only for FoT), but still don't know for sure what the cause is.

Theories ranged from 2 or more CPs with the same or partly the same name & whether or not bots had vanilla kit ai templates.
But seems likely a piece of python code is the cause, python coders are a bit thin on the ground.

Try copying the gpm_cq over from the vanilla python folder & see if it's still happening. (back up 1st!)

Think also remming out "objecttemplate.forsoldieronly" helps, this doesn't occur in vanilla & could be causing python havoc for the bots.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 26-10-2010, 20:10:28
Holy shiat! Winterhif!!!!!!

Thanks man. Would want to ask if you are back in the community. But that might just jinx it.
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Remick04 on 27-10-2010, 02:10:32
Sorry guys I've been really busy with non-FH related work but I still check these forums daily.

I would love to figure out what causes the double cap issue. It's frustrating as hell, and really breaks the flow of gameplay. But I've been unsuccessful in finding the problem, and I agree it seems to have gotten worse and more widespread. I unfortunately don't have much free time for testing and bug tracking. So I'm relying on you guys for the time being. Hopefully Winterhilf, Devilman, Drawde or one of the many talented people on these forums will find the problem... and hopefully I will get some free time in the near future so I can be some help :P
 
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 27-10-2010, 10:10:06
Also noted, Dodgy AI commander on Mareth line - doesn't follow push-order - Also on The St. Lo breakout... but not as bad

Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 28-10-2010, 00:10:51
EDIT (27/10/10): Anyone figured out the problem with all the p51s coming in at once on Ramelle, Neuville? I watched it happen a couple of times, 2 planes at a time come in from each direction, one, two, three, then its like someone says 'altogether now' and WHAM!! Epic lag for like 10 secs. Also noted, after a few times of this cycle, the explosion and plane flyover go out of sync... Actually one set of planes goes missing -

You still get the planes returning after their bombing run, but their moving to bomb is missing, just the explosion, with no plane to go with it...

So these are the big priorities:
/ Planes of Ramelle
/ CTD on Totalize
/ Double-cap... you hear Winterhilf - zM, cF. I think this is up your alley. You guys seem to know what you are about editting map stuff.

I wonder where Devilman, Drawde and Aserafimov are. Haven't heard from any of those chaps almost as long as Lobo. We could really use a mapper and AI person for the other issues, like Goodwood

Also noted, bots wont use deployable weapons well. First, there is ofcourse the unraised perspective. Remick... Ple-ease! And more important, there is the fact that bots don't STAY on deployables. They treat them like mobile vehicles. I can't position an mg to give cover fire if the bot gets on, loses interest and bails before the bad guys show up. I think they should camp these. Its easy to kill them later or more likely, they get killed, but when it was camped in Drawde's first mod version, it did wonders in PHL - Mortars especially.


EDIT (28/10/10)
Noted that bots throw Frag grenades at the Crusader in Mersa Matru  :o
And I deploy an mg, they get on, they bail, mg is lost. I  position a mobile gun, they get on, they bail, defense is vulnerable. I did deploy a vikers mg on Lebisey and had them get on facing visible enemy troops. The bot pointed at everyone but didn't fire a single shot till he was shot himself.

Also, tanks should use HE on AT guns, or both HE and AP with HE being the main thing for mgs if they have it.




Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 03-11-2010, 07:11:50
Olympus 64 also crashes - Anyone else confirm that? Very early in the game
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 22-11-2010, 18:11:57
Ok, a few more

The British Crusader AA and the American quad-50 for some reason don't see themselves as AA guns but infantry cannons. Perhaps if they shared code with the flakpanzer this would change, as that one is deadly both ways


Also, the wespe uses code from either the marder of something similar. Perhaps switching its HE and AP round position (Requires a _ai version) should fix that

Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: djinn on 24-11-2010, 08:11:19
oh right. Was not ready in time for 2.3. You need to DL Remick_04's fixes for some maps, listed here:
http://www.mediafire.com/file/djfvypp7pzutmhw/2.3SP_fixed_files.rar

for next time, note that all these are listed here: http://fhpubforum.warumdarum.de/index.php?topic=6539.0

Then once that's in, apply Void's ESAI map strategy

I did note something though,
After the Tommies capture all the bases on Lebisey, I kept getting commander spam to defend the chateau and the town. I didn't edit out the Olympus issue from Remick's files first, but I thought Void's own would replace them. Did I leave something out, or is this an issue with the current Void Lebisey strategy?
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: cannonfodder on 24-11-2010, 10:11:29
Don't recall any CO spam on Lebisey, but I'll give it a go and see if I get any.

I can't remember what the Olympus file was, but if it was Strategies.ai or StrategicAreas.ai, Void's files should've overwritten it... :-\
Title: Re: FH 2.3 SP known errors & bot erronous behavior
Post by: Void on 25-11-2010, 23:11:47
Quote
After the Tommies capture all the bases on Lebisey, I kept getting commander spam to defend the chateau and the town. I didn't edit out the Olympus issue from Remick's files first, but I thought Void's own would replace them. Did I leave something out, or is this an issue with the current Void Lebisey strategy?

Let me check something, I'll be right back...

Right then. That behavior is intentional Djinn. After all cappable flags are taken by the Allies in Lebisey, the SAI is coded to order a defense on "3rd Company Command Post" and "Lebisey Headquarters". So if it is spam, it is spam by design.

 If you are only getting a few orders to defend those flags that is one thing, but if you are truly getting spammed (esp. if the CO keeps switching between the flag it wants you to defend) then I have some adjustments to make. Post more details about it over in the ESAI thread, and I will see if I need to make any changes.

It's not a bug, it's a feature...