Hawk666

Yes, that would be very great. But if the development resources are not available, a dynamic engine like that of falcon 3 would also be good: It did not have a real time engine, but it generated missions which are dynamic and not just randomized. For example it checked the availability of ammo&logistics, the condition of the infrastructure and so on. Another example (for armor combat) is M1 Tank Platoon 1 from Microprose. It had a dynamic mission generator but was not so sophisticated as the engine of Falcon3, for example it did not consider logistics etc. On the other hand the terrain was created on the fly which forced you to choose different tatics for each new game.

I 100% agree. A truely dynamic campaign would be a unique feature of ARMA 2 and in every review that would be highlighted. Sometimes I don't understand why many people only have a look at the graphics. I still play from time to time "Red Storm Rising" (nuclear submarine simulation), which was developed in 1988. Surely not due to its graphics but because of its true dynamic campaign which gurantess endless replayibility.

The question is what are real issues and topics. Probably everybody has a different understanding which topics are important and not. A user who wants to develop complex content/missions is probably interested in an improvement of the script language (introduce OO etc.) ...a user who just plays 100% probably not. As I already mentioned in another thread: I would appreciate if the Devs would give a little feedback what they intend to do and what not (due to resource constraints ...) The suggestion list is growing and growing but wouldn't it be more effective if the Devs just say they have % resources to build community wishes and then just do a poll to decide what key 'whishes' should be built etc. ?

Sure, you're right, AI development is very tough, but there are improvements in AI research and in implementing them in projects. With the introduction of multi-core CPUs there is now the possibility to give the AI algorithms more CPU time. I think when BIS achieves it to make the AI more dynamic and enhance the corresponding script commands and introduce RPG elements...that would be a great step in the right direction. Nobody expects wonders

I totally disagree...how do you want to create a realistic environment with a non-capable AI? Have a look at other games and you'll realize that the AI is continuously improved and the should be done with the Arma-engine as well.

Personal, I have to say, that I like the 'development environment' of Arma, apart of the known and already discussed gaps in the other threads. Ok, there is no real IDE, but thanks to some very talented members of the community we have a FSM editor and more than one script editor which helps a lot! Arma2 should just provide an object-oriented script language to enable the community to build real reusable functionality, a debugger (one with a very basic functionality would serve as well) and full access to the AI and I'm happy . If the developers (BIS) think that it is feasible, I would appreciate it if they would discuss some design issues (script language etc.) with the community in advance. Probably both side would benefit.

Hi, I like in general the 'open' engine of Arma and OFP and have tried to start to create an highly dynamic cqb mission (incl. enhancing the AI due to the existing AI of ArmA is just not capable enough) but it's difficult to realize, so here my suggestions: - Ok, the script language/FSM is quite capable but since the language does not really support reuse of code (only isolated functions, no OO etc.) it's very hard not to loose the overview, especially if you want to reuse script/code of the community. That the language does not provide a 'real' debugger does not help either. To be honest, I've hoped, that you have more time to develop Arma2/Game2, so they you may consider to drop your property script language and switch to a SDK by supporting a standard language (python etc.) with a clean class library. I know the compatibility issues to former mods etc. but what's the benefit of a script language that gets more and more commands and results in 'spaghetti code' ; probably you have more and more work with the maintenance of your 'own' code too? - You've introduced FSM in ArmA which is an improvement to OFP. But if you want to (or have to) micromanage the AI, you just have not all the needed (sensor) informations available (LOS as an example). So, please, even if you try to vastly improve the AI this time (thumps up :-)) , give us the needed support to micromanage the AI by the script language.

Hi, sorry if this is a noobe question, but at least I've tried to search for the solution ;-): Is it possible to loop over all existing groups (regardless of the side) ? I've the problem, that some external scripts creates groups on the fly and I want to include these groups in my scripts. Of course I could directly access the global variables which store the groups but that's ugly coding: When I add new scripts I've to adapt my coding etc. Thanks in advance guys.

I totally agree. This aspect is one of the main missing parts in the engine so far. AI soldiers should have some EQ (emotional intelligence), which can be accessed by mission creators.

Hi, I hope, that this is not a silly question ;-): I want to play Coop from time to time, but don't want to join a clan (time issue). I've tried out the internal ArmA MP list, but I never know, who plays 'serious' and so on. Can you recommend me a server etc.? Thanks for feedback guys!

Hi, does anybody know, what enableAttack is good for? In the wiki there is just the following comment available: "Set if leader can issue attack commands.". Thx.

I don't know the exact behavior, but I'm currently working on a prototype of a rescue mission, where the west side has to rescue some civilians. In the beginning I want to simulate the neutrality between 'West' and the Resistance/EAST side. So I've put the following in the init script: //init Script //set EAST friendly to WEST & vice versa in the beginning east setfriend [west,0.99]; west setfriend [east,0.99]; and it semms to work

Did you call the command in the following way? EAST setFriend[WEST,0.99] I've used the command in the initializer of an unit of the east side and that unit doesn't fire on western soldiers any more.

Brute-force AI algorithms work 'fine' for chess-like games, in which the CPU has plenty of time to solve problems. This is definitely not true for ArmA, neither is the environment the same. If NNs could be part of a possible solution has to be evaluated. BTW: There are several non-commercial NN toolkits available (open source etc.) You don't have to buy one.

Reasonable behavior does not always mean to win a mission etc. but to retreat etc. if the losses are too high/the opponent is too strong. That's right, group behaviour should be the focus, not the individual behaviour (that should handled by the ArmA engine anyway). Perhaps it can slightly be modified by forcing the FSM to a certain state (doFSM) in some situations.

Very well written post. Total randomization won't work. I was more thinking of a predefined kit, from which the script chooses randomly from and adapts it (positions etc.) The challenge is to write a generator which fits some of these components to a believable mission. I think the generator needs a kind of a simplified scoring algorithm (like which is used in chess engines to determine the 'score' for a given player) to construct balanced scenarios. BTW I do not think that every generated mission has to be 100% balanced...is that real life? surely not. You are right that the (script)engine should only deal with the local area around the player and abstract all other events. We cannot build a Falcon4 dynamic engine :-). And: we need a reusable function library which encapsulates complex algorithms (AI,planing algorithms).

With 'need-driven' AI driven I mean, that the AI acts in a reasonable way. For example when a soldier is injured, the commander has to organize a rescue-helicopter and so on. If an AI-commander 'recognizes' that the enemy is too strong he has either to call for support or has to retreat. I'm an experiecend programmer but not with the Arma-Script lanugage (I'm currently learning the basics). Does you or anybody know, if compiled scripts/functions perform well enough to use complex algorithms (like AI-methods: genetics/neuronal nets)?

Nice ideas, but with such a huge scope it needs huge efforts too. I'm planning to create a dynamic mission but which is limited to a specific area to reduce complexity. The mission has a civilian-rescue topic. One thing is important too: The AI should be need-driven to make it more realistic.

Hi, will there be a 'real' SDK for Game2? The best thing, from my point of view, would be a .Net API because that would mean you can create scripts etc. with your favorite language. The script language of ArmA is quite powerful but it lacks the benefits of an object-oriented language. The main pain point to create missions in ArmA (and its predecessors) is, that you need knowledge about dozens of files/technologies (config, script, functions, FSM etc.) and it's not transparent for users, which don't have time to get used to all that things. Another pain point of property technologies is, that you can't use existing open source libraries and their proven algorithms (like AI methods (neuronal nets,generics etc.)) A clean API with a .NET/Javadocs common approach would be very nice Any chance to see it in Game2?

can please someone confirms, that setSkill does not currently work? Thx.

Set all visual settings in video options to low. This will increase the speed a lot while editing. You can always switch back to higher settings once you have the mission ready for testing or want to play some Arma missions. Thanks. I've set the resolution to <1024x and now the leg is gone. Hope, that this issue will be addressed in a later patch anyway...

Hi, currently (1.02,German version) there is a disturbing lag (terrain streaming engine does not seem to buffer the map) if you test a mission inside the editor, go back to the editor and test it again and so on. You've the same problem, if you like to write a script, switch to ArmA and so on etc. Are there any recommendations how you can minimize this lag? Thanks.

<!--QuoteBegin-- -Dawnrazor+Nov. 20 2006,10:00--><table border="0" align="center" width="95%" cellpadding="0" cellspacing="0"><tr><td>Quote ( -Dawnrazor @ Nov. 20 2006,10:00)</td></tr><tr><td id="QUOTE">PC Action ArmA 75% -15% because of extreme bad AI I can just hope, that the developers read this thread and improve the AI (at least to an acceptable level) in the first patch. I know that AI programming is very difficult but this game needs a good one, else it will never reach its potential... I'm a little disappointed because in the previews it was always stated, that the AI will be sophisticated...

Hi, I would like to ask the dev team of ArmA if they plan to release a first patch before Christmas. Nobody has the expectations, that the 1.00 version of ArmA is free of bugs/problems. What I've read from the threads+review is, that the AI is one of the main topics, which should be addressed. Do you think, that the situation awareness of the AI and the unrealistic 'super AI soldier' which hits an enemy with the first shot at a great distance etc. can be enhanced in the first patch? Thanks for feedback!

Does the AI now search for cover if he's under fire or what parts are better than in OFP? Thanks...