AI Artillery - not WAD?

[Paper Tiger]

Okay, nobody wants to reply about changes to the AI artillery system. I can understand if you guys want to keep this under your hat. However, I was trying some experiments with the AI artillery earlier today and it seems that no matter which setting you choose, 'Destroy', 'Damage' or 'Suppress', it's one artillery module per zone. It doesn't appear to affect either the intensity or the duration of the strike.

I got the idea from reading the manual that the 'Suppress' setting would hit one zone first with a few shots and then hit another zone and so on until it ran out of artillery. That WOULD be cool! So I set this up but it still hammered away at the first selected location until it was all out of ammo. I counted the shell holes and there were about 48 packed into the one target zone, same as a Destroy strike.

Further, it says in the manual that the only requirement is that a valid spotter have LoS to the target zone but thais has never worked as far as I can remember. I know that nobody has LoS to ANY target zones in the scenario's I'm working on and the artillery always starts after 20-25 seconds regardless.

Also, it says that any artillery assets not used adainst the support targets are available to the AI to use during the regular course of the battle. Hmm, not quite sure what that means but I interpret that as, if it's not used in the first strike, it'll be available later. It says that further use of artillery is up to the TacAI's assessment of the tactical situation etc etc, but I've NEVER seen it use artillery after the first strikes.

Now, one enormous caveat to all this is that I'm almost always giving the AI mortars to use. I once gave the US AI a battery of Paladins and they fired FOREVER. Well, nearly 20 mins at least when they were on 'Destroy'. They did fire for a shorter length of time (but not much)when I selected 'Damage' but they didn't then shift their fire to another target.

It seems that, at the moment, AI artillery fires off everything at the start and there's nothing for later, even if it arrives as a reinforcement. Are any of these issues on the list of items to be fixed?

[Paper Tiger]

[bump]

I'll give this one more try and then I'll post it in the Bugs thread.

[Cpl Steiner]

I'm pretty sure there is no game code to make the AI choose targets for artillery mid-game, like it used to in CMx1. This is a major oversight IMHO. When coupled with the AI's inability to acquire Javelins it makes AI control of US forces pretty lame right now.

[SlowMotion]

If the AI was to use artillery according to a plan, this would require new tools to Scenario Editor. Scenario designers would need new kinds of triggers that would decide when to use arty to some target.

[SgtMuhammed]

The AI though should call for fire on its own if it has access to the required assets, FO's and such. So far I have never seen it do this.

[thelmia]

I am getting 1/2 as much artillery fire out of the AI as I used to. I had a scenario that tested AI fire, and I ran it in 1.05. It was obvious that the volume of fire was drastically reduced.

Instead of firing everything at the start of the scenario, the second half didn't fire for about 15 minutes, when the AI troops were at the objective.... Pretty funny watching the fratricide, but not what I had in mind.

[Paper Tiger]

Everything that is in my first post is suggested by my reading of the manual. therefore, I am a little surprised that none of these features appear to work as designed. Of course, if it's already noted and is going to be corrected sometime soon then that's fine by me. It'll be a pretty good system if it works like the manual says it should. For example, if the 'valid spotter' thing worked properly, I could do some clever things with this system but it clearly isn't. I always make sure that NOBODY has LoS to the target zones in my experiments.

Thelmia:

Wow, what can I say. I've never seen AI artillery fire later in the game and I've tried a few experiments of my own too. I sometimes wonder if some of the problems we see in the game are not so much bugs as problems with our own individual kits. We've both got the same game, patched to 1.05 and I've never seen that happen. My experiment was done on Saturday morning and the artillery fire was intense man, no change as far as I can see.

[Paper Tiger]

Well, I'm bringing this one back up instead of starting another thread because I'd really like to hear what Battlefront's position is on the state of AI artillery planning.

Yeah, I can live without it coming in later but if those 'Destroy', 'Damage' and 'Suppress' orders start working as the manual says, it will radically alter the way some of the scenarios I've been working on function. We are actually allowed to choose these on the interface but I can't see any functional difference between the three settings. They are all fire until the ammo is exhausted.

If changes are a LONG way off in the future then that's fine, I won't have to change anything so that they're not made redundant by significant changes that appear in a monoth or two.

[Rokossovski]

It seems to me that revising the AI to make semi-intelligent use of artillery assets should be a high priority.

Now that I have made my pronouncement on the subject, I assume the developers will scurry to carry out my will.

[Paper Tiger]

Rokossovski

-Now that I have made my pronouncement on the subject, I assume the developers will scurry to carry out my will.-

I wish ! I get the feeling that this is only an issue for those of us who design scenarios rather than exclusively play them. I know Battlefront have a position on this as I read a thread Steve posted a LONG time ago saying that there were definitely plans to improve it. That's fine but I'd like to know if there are plans to fix what's obviously not WAD. From reading the manual, it's obvious that it's not.

I think the AI artillery is a very important aspect of the game as, the way it stands just now, it unbalances it in favour of the human player. Firing off everything in the first minutes isn't exactly satisfying and to get a result, the scenario designer has to place the target zones close to or on top of the human player's set up zones. Except in a very few situations, I would imagine most people would complain solidly to a designer if he did that but that's pretty much all the AI artillery is capable of doing at the moment.

Also, if the AI is given a module larger that a mortar, it fires forever and in some cases, it is still firing when it's own troops are approaching the zone. Okay, maybe that's bad scenario design on my part but I'd like to be able to do it without reducing the amount of ammo the module has.

[missinginreality]

Yep I tried some experiments with AI Arty and just wasn't happening. Seemed pretty simple to me to bring it in as a reinforcement mid game but nothing would ever happen Be good if this aspect could be fixed up as it says in the manual.

[Paper Tiger]

Let's give it another try. Probably a waste of time but this question matters to me.

[Meach]

I remember in the dim and distant past Steve said something about AI using artillery as it was in CM1. i.e calling it down mid-battle.

As usual I could be wrong but I distinctly remember seeing a post with something to that effect. Now whether this post was saying it could be CM:WW2 or CM:SF is another matter.

[Paper Tiger]

Yup, I remember reading that post somewhere too. I know they're super busy getting 1.06 finished but this thread's been around since the beginning of the month and I get no replies from Steve. I really hope to hear something as this is a very important part of the game.

[Meach]

I have faith that it will be fixed or modified. What I do now is call down a barrage right at the start of the battle if I have artillery and if the AI has artillery it does the same.

Not perfect but it deals a prepatory bombardment and then you can get on with the fight.

[Cpl Steiner]

There were many things "unfinished" in the initial release, and AI use of artillery is clearly one of them. As the manual clearly states, the AI should use it without scripting during a game. The game wasn't finished when it was released, and it still isn't finished now some 8 months later. To my mind, BFC should concentrate on putting in all the missing bits of code as a priority rather than adding in things like dynamic lighting and enhanced LOS. These things are welcome additions but could have waited until fundamental areas of missing code were addressed.

[SlowMotion]

Don't you think that well working LOS code is one of those fundamental areas? I thought that was the biggest reason behind introducing ELOS now - to really make that part of the game work well.

[Cpl Steiner]

Originally posted by SlowMotion:

Don't you think that well working LOS code is one of those fundamental areas? I thought that was the biggest reason behind introducing ELOS now - to really make that part of the game work well.

Correctly working LOS certainly is important but it is an "enhancement". They called it enhanced LOS for a reason. Correctly working artillery is a requirement IMHO, not an enhancement.

[Meach]

I would rather have ELOS than AI artillery calls.

I can work round not getting blasted by artillery but not so much no hull down for my tanks.

Everyone has different priorities.

[Paper Tiger]

Yup, can't argue with that. What would I rather have, dynamic lighting at night or AI artillery WAD? Dynamic lighting!

What would I rather have, ELoS or AI artillery WAD? ELoS!

What would I rather have, improved TACAI or etc...

Still, I'm hoping that some importance is attached to this aspect of the game. If it were working as the manual suggests it should, I could be doing some pretty cool things with it in my scenarios.

[Cpl Steiner]

My fear is that with each successive patch BFC will be less willing to do "yet one more" patch just for a few people who moan about this, that or the other. At some point they are bound to say the product is effectively complete, whether it's finished or not, so they can divert resources to the modules and the WWII title. By then it will be too late to get things like working AI artillery into the game. Even if it works in the WWII module, I have serious doubts that BFC would waste time retrofitting it into CM:SF when sales will have bottomed out to next to nothing.

[SlowMotion]

My understanding is that new features can be still added when modules become available. People who don't buy some module would still benefit from the new features, they just wouldn't get any new units.

[Elmar Bijlsma]

Yes, BFC have already stated that, unless unreasonably laborious, fixes and engine improvements in one game or module will go in the previous iterations. Or sumfink.

[missinginreality]

I'd second [or third or whoever many] that a mid-game working AI artillery is a pretty important factor. there's no reason why AI would just call in arty at the start and then leave a handy resource like that sitting doing nothing when there's ample reason to use it. seems like a pretty darn important thing missing to me.

[Paper Tiger]

Yeah, it's a very important part of the game if you play the game against the AI which most of us do. The current system, 'fire everything after 30 secs and then that's it folks' isn't really very good and it doesn't provide a challenge for the human player. It's never going to be as effective as artillery in the hands of a human opponent but a good scenario design might make a player think twice about his game plan if the AI could use it more effectively.

Since the designers obviously intend it to do what it says in the manual, it would be nice to hear when they are going to get around to fixing it. If it were WAD, it would be a very good system.