AI going to be player centric?

[Big Dawg KS 6]

No way. There's already enough new technology coming in ArmA 3, adding xaitment middleware would throw it over the top. Maybe for ArmA 4. Let's let our community get familiar with the PhysX implementation & new animation system first.

[NeMeSiS 11]

No way. There's already enough new technology coming in ArmA 3, adding xaitment middleware would throw it over the top. Maybe for ArmA 4. Let's let our community get familiar with the PhysX implementation & new animation system first.

Of all the reasons that could be given to not put it in ArmA3, 'letting the community get familiar with other changes first for a couple of years' is pretty much on the bottom of the list.

"Wow! These boxes fall over when i drive into them! Good thing the AI is still roughly the same, i couldnt handle that much change all at once!" :p

[Big Dawg KS 6]

Of all the reasons that could be given to not put it in ArmA3, 'letting the community get familiar with other changes first for a couple of years' is pretty much on the bottom of the list.

"Wow! These boxes fall over when i drive into them! Good thing the AI is still roughly the same, i couldnt handle that much change all at once!" :p

Fine, another reason: Rushing to cram as many features/updates in as possible is never a good idea. ;) BIS needs to take things slow and really perfect what they already have before moving on.

Let's not act spoiled and start asking for BIS to turn ArmA 3 into VBS2 overnight.

[NeMeSiS 11]

Fine, another reason: Rushing to cram as many features/updates in as possible is never a good idea. ;)

And neither is rehashing the same stuff over and over again with some slight modifications. ;)

Obviously BI is the only one who can judge wether they have too much on their plate or not, but if this doesnt make it in and PhysX does i am going to be somewhat dissapointed by that decision.

[Big Dawg KS 6]

And neither is rehashing the same stuff over and over again with some slight modifications. ;)

It's not rehashing, it's iterative development, and it works a hell of a lot better than rushing into everything breadth first. First you perfect one thing (physX & animations), then you build on it. Trying to modify something you've already built off of is only going to cause more trouble.

Obviously BI is the only one who can judge wether they have too much on their plate or not, but if this doesnt make it in and PhysX does i am going to be somewhat dissapointed by that decision.

I bet your disappointment won't last very long. Besides if BIS threw this into ArmA 3 what would we have to look forward for in ArmA 4? ;)

BIS needs to really focus on what new features they've already anounced, especially since screwing them up can have very serious consequences.

[NeMeSiS 11]

It's not rehashing, it's iterative development, and it works a hell of a lot better than rushing into everything breadth first. First you perfect one thing (physX & animations), then you build on it. Trying to modify something you've already built off of is only going to cause more trouble.

Come on, you know what i ment. ;)

I bet your disappointment won't last very long. Besides if BIS threw this into ArmA 3 what would we have to look forward for in ArmA 4? ;)

BIS needs to really focus on what new features they've already anounced, especially since screwing them up can have very serious consequences.

My point was that they chose physics over AI, which i think is a terrible decision.

[Big Dawg KS 6]

Come on, you know what i ment. ;)

My point was that they chose physics over AI, which i think is a terrible decision.

Necessary IMO, since AI has a certain level of dependence on physics. Physics first, then AI that can cope with it. Of course animations and AI are also very closely related, so hopefully with the new animation improvements AI will become a much higher priority/will see improvement.

In otherwords, I like to think of it as working from the ground up. Physics is a very core element that's pretty close to the bottom (foundation), where AI is a lot higher up at the top. If you want to build a new expansion on the top floor of a building but you have a weak foundation, which do you address first? ;)

Edited June 18, 2011 by Big Dawg KS

[maturin 12]

No way. There's already enough new technology coming in ArmA 3, adding xaitment middleware would throw it over the top. Maybe for ArmA 4. Let's let our community get familiar with the PhysX implementation & new animation system first.

Why do we need to get used to Physics and animations? You just play the game and watch.

My point was that they chose physics over AI, which i think is a terrible decision.

I don't like those priorities either, but the physics were a disaster, while the AI is not. BIS can do AI themselves, and AI as used in this game is rather rare. Physics, not so much.

[Big Dawg KS 6]

Why do we need to get used to Physics and animations? You just play the game and watch.

Mods? :rolleyes:

[froggyluv 2136]

If BI says they've implemented a middleware AI pathfinding software as well as new animations for the next release -I'd wait for the modders to catch up :p

[NeMeSiS 11]

Necessary IMO, since AI has a certain level of dependence on physics. Physics first, then AI that can cope with it. Of course animations and AI are also very closely related, so hopefully with the new animation improvements AI will become a much higher priority/will see improvement.

Animations and AI, yeah, on a deeper level they are related. However the high-level stuff (FSM's, basically everything behaviour related)? Barely, if at all.

Also, i fail to see what physics have to do with anything AI related, wether the AI can use cover properly or not does not depend on gravity.

Changes related the flying AI may be needed, but since physX is added this is would be necessary anyway.

The only proper reasons i can think of are BI not having enough manpower/money to implement everything, and VBS2 needing more unique selling points. Seriously, the community not being able to cope with changes is a rather ridiculous argument. Unless we all decide we dont want to learn new things anymore, but in that case we might as well stick with ArmA2 forever.

Edited June 18, 2011 by NeMeSiS

[Big Dawg KS 6]

Also, i fail to see what physics have to do with anything AI related

Well like you said, steering in vehicles mostly (which is totally dependent on physics), but also perhaps pathfinding through physics enabled objects. I never said there was a high amount of dependence, but physics is still at a much lower leven than AI and like I said it's better to start at the bottom. Either way, I think AI is going to have to wait for some other things before it can get totally revamped/outsourced (but I could be wrong, since BIS is always full of surprises).

Edited June 18, 2011 by Big Dawg KS

[NeMeSiS 11]

To be honest, everything you mentioned would need changes to the AI, wether it is the current system or xaitment. And while adding xaitment would obviously add extra workload, i still fail to see why that would be increased because of physX and the new animation system, as the 'old' AI would also require the changes you mentioned.

[froggyluv 2136]

This new game '7554' supposedly to feature xaiment AI and looks pretty interesting.

A Vietnamese developer and featuring play as a freedom fighting Vietnamese against Colonialists.

0Lt89MiuvKc

tcEU4Ihs69Y

[CarlGustaffa 4]

Pretty sweet actually. Wouldn't mind another Vietnam based game, but fighting in jungle - the success for me depends on how the AI simulates vision, and particularly block thereof and the chance of loosing what it's tracking. Arma AI, when set bad enough, gets bad enough but for the wrong reasons - their vision is very problematic at times.

Edit: About middleware. If BIS decided to use xaitment, do you think we would get access to the same editing tools? For a fraction of the price? Would modding survive? I don't know, just thought I'd air the issue.

Edited June 18, 2011 by CarlGustaffa

[roberthammer 582]

Pretty sweet actually. Wouldn't mind another Vietnam based game, but fighting in jungle - the success for me depends on how the AI simulates vision, and particularly block thereof and the chance of loosing what it's tracking. Arma AI, when set bad enough, gets bad enough but for the wrong reasons - their vision is very problematic at times.

Edit: About middleware. If BIS decided to use xaitment, do you think we would get access to the same editing tools? For a fraction of the price? Would modding survive? I don't know, just thought I'd air the issue.

xaiment AI will be only for VBS2 - not for ArmA 3

[froggyluv 2136]

Your probably right that it won't make it into Arma3 but I'm wondering how the AI will be upgraded to navigate in urban environments especially as all houses are supposedly enterable.

Also apparently VBS2 has new window damage modelling and level of destruction and although not knowing Greek Islands, I'm assuming they have glass windows (unlike OA).

[Make Love Not War 10]

No way. There's already enough new technology coming in ArmA 3, adding xaitment middleware would throw it over the top. Maybe for ArmA 4. Let's let our community get familiar with the PhysX implementation & new animation system first.

Pffft... we can handle it. Give it to us NAUW!

But, seriously, considering all the other new features already slated for A3, it's not realistic to expect BIS to be able to implement xaitment as well. So, yes, we agree. Too bad though. A4, maybe?

[CameronMcDonald 146]

Arrowhead-esque standalone expansion much?