[WIP][MOD][OXCE] From the Apocalypse 0.29.0

[robin]

I'm rather curious robin. What are you going to use most of these vehicles for? I am assuming they are civilian vehicles? I'm interested in just how they are going to be integrated into your mod's ground missions.

In Apocalypse there are "illegal flyer" events.
I'm picking up and expanding the idea. All these civilian (you got it right) vehicles will be used in illegal flyer missions. I'd rather not spoil anything else (also because I'm still thinking about it ).

[Zharkov]

Well it has been very hard but it's paying off I guess.

I must say, these look like something I'd like to shoot down and plunder, but I might have played piratez too much.

[robin]

The thing I hate about this is doing routes.

[Zharkov]

Say, Robin, may I use your Spitter resources for a little addon for piratez?

[Hobbes]

The thing I hate about this is doing routes.

Good that Map View has been updated to make it a lot easier.

One tip: it is useless to add a lot of nodes to try to force the AI to move as you want when patrolling or create a lot of alternative patrolling routes. Keep the patrol routes simple as possible.

[Solarius Scorch]

Good that Map View has been updated to make it a lot easier.

One tip: it is useless to add a lot of nodes to try to force the AI to move as you want when patrolling or create a lot of alternative patrolling routes. Keep the patrol routes simple as possible.

I personally think there is an added value in more complicated node maps (as long as they're done sensibly, and not just about adding more nodes), because I think the AI tends to fare better thanks to more useful attack vectors being created. For example, I like to place high-priority node near windows, because it makes the aliens look outside periodically and therefore have better detection.

It's all a bit fuzzy/susceptible to attribution errors and therefore I won't be defending this position to the death, but I think it's at least worth considering.

[Dioxine]

Or placing nodes to the side of a door instead of in the front of it, or otherwise in cover, makes the AI a bit less likely to get easily killed.

[Hobbes]

I personally think there is an added value in more complicated node maps (as long as they're done sensibly, and not just about adding more nodes), because I think the AI tends to fare better thanks to more useful attack vectors being created.

The AI doesn't use the nodes to determine attack vectors. I think there might be a misunderstanding how the AI/nodes works.

At the beginning of a mission all the aliens are set for Patrol mode and spawned in the nodes. While on Patrol mode they will use the nodes to move around and look for XCom units.

When an alien unit spots an XCom one, then the AI will switch the mode from Patrol to Attack/Ambush/Retreat, depending on its aggression, etc., and it will stop using the nodes for movement. Instead it will move the alien to where there is cover, etc., and completely ignore the nodes.

If the alien stops seeing XCom units for X turns (X = intelligence stat) then it will revert back to Patrol mode and use again the nodes.

Now, you can place nodes in locations where you think it may be more advantageous for the aliens to reaction fire your soldiers (to the side of the door instead of in front of it, etc.) but this can be a rather useless effort for a number of reasons:
* You can't control the direction that the alien will be facing. If you place a node like Dioxine said, then it will be likely that the alien will be facing the wall next to the door rather than the door. Then XCom unit enters, turns around and shoots alien on its side.
* You can't control how much an alien moves during its turn, thus the alien might not have TUs to reach a certain node during its turn or have enough TUs for go through several nodes during its turn.

So, when you are designing nodes you need to take into account all of these factors. And it is really better and easier to keep the patrol routes simple. And the best example comes from TFTD - if you compare its maps with UFO's you'll see that its nodes are much more simplified.

[Arthanor]

But wouldn't it make sense for aliens to take a look through the window when in patrol mode instead of just walking through a room?

I think that's what Solarius was meaning. When the alien doesn't know of any XCom soldiers, having a node next to a window will make it walk to the window and hopefully take a peak, which might allow it to see an XCom soldier and enter "fighting" mode from its previous patrol mode.

The side/front of door one is more difficult, but there still should be an advantage. If the alien is right in front of the door, it opens and, at worst, you get mutual surprise, at best the alien is not facing you. Either way, you are free to engage with either an autoshot or a melee swing. If it's next to the door, melee might still work, but shooting will be hampered by the wall, forcing you to take a step into the other room and potentially triggering reactions.

[Dioxine]

Yeah, pretty much that. Not that much, but sometimes nasty. I was never suggesting that these nodes will allow me to control the AI, but might, maybe in 10% of cases, cause more 'bastardly' behaviour on the part of the AI.

[Solarius Scorch]

Hobbes, I do know how nodes work. I think you've explained it to me several times already.

Okay, what I meant was that AI units walk between nodes when not aware of any enemy units. Which means they are very often in Patrol mode when they are discovered, which means the nodes determine pretty strongly where aliens will be found.

If an important, high-priority node is strategically located, it is a boon to the AI, because it gives an average alien a better position to spot the enemy as well as start shooting them. But it means there should be more nodes per map block, because you often have several such places, also need to make connecting nodes etc.

Also, more nodes mean less predictable alien spawning and movements.

[Hobbes]

But wouldn't it make sense for aliens to take a look through the window when in patrol mode instead of just walking through a room?

I think that's what Solarius was meaning. When the alien doesn't know of any XCom soldiers, having a node next to a window will make it walk to the window and hopefully take a peak, which might allow it to see an XCom soldier and enter "fighting" mode from its previous patrol mode.

Yeah, this makes sense: you have a room with 4 windows, you set nodes where the alien can spot XCom units upon reaching that node (and they don't have to be right by the window, in fact it can better to place them 1 or 2 tiles away from the window since you'll make it harder to be spotted).

The side/front of door one is more difficult, but there still should be an advantage. If the alien is right in front of the door, it opens and, at worst, you get mutual surprise, at best the alien is not facing you. Either way, you are free to engage with either an autoshot or a melee swing. If it's next to the door, melee might still work, but shooting will be hampered by the wall, forcing you to take a step into the other room and potentially triggering reactions.

Even better: place the node a couple of tiles away from the UFO door. This forces the player to move its units to try a melee attack, and it makes it harder for shots coming from the side.

Yeah, pretty much that. Not that much, but sometimes nasty. I was never suggesting that these nodes will allow me to control the AI, but might, maybe in 10% of cases, cause more 'bastardly' behaviour on the part of the AI.

The other required part for 'bastardly' behavior is map design. Specially designing maps that prevent the player of leveling down several walls to target an alien. But that's another issue

I have also been guilty of trying to manipulate the AI: place one node there and another one there and another there, so that the alien will make this really complex movement and surprise the player... until I realized that all of this was happening only in my head, not the actual game

[Dioxine]

The other required part for 'bastardly' behavior is map design. Specially designing maps that prevent the player of leveling down several walls to target an alien. But that's another issue

I consider this to be low. Indestructible walls are uncool if they're unrealistically so. I'm rather trying to reduce the explosives spam in my mod

[Hobbes]

Hobbes, I do know how nodes work. I think you've explained it to me several times already.

Okay, what I meant was that AI units walk between nodes when not aware of any enemy units. Which means they are very often in Patrol mode when they are discovered, which means the nodes determine pretty strongly where aliens will be found.

If an important, high-priority node is strategically located, it is a boon to the AI, because it gives an average alien a better position to spot the enemy as well as start shooting them. But it means there should be more nodes per map block, because you often have several such places, also need to make connecting nodes etc.

Also, more nodes mean less predictable alien spawning and movements.

When you mentioned 'attack vectors' before it just made me worry.

Well placed nodes on strategic locations are important but if you want to ensure that there will be an alien on that location to spot XCom units then the best way is not connect those nodes with the rest to ensure that the alien doesn't wander off to some other less important node, like some TFTD maps.

I consider this to be low. Indestructible walls are uncool if they're unrealistically so. I'm rather trying to reduce the explosives spam in my mod

Discussions about 'realism' are almost irrelevant IMHO regarding games. To me it is completely unrealistic to expect that a laser or plasma beam would completely wreck a wall rather than simply carving a small hole through it. Even a grenade or a HE shell from a handheld cannon busting a wall made of masonry, metal or stone is highly 'unrealistic': you'd need a rocket launcher like a Carl Gustav or and AT-4 or an artillery/tank round to do the trick, or some well placed high explosive.

But I'm playing a game for fun so I don't really think much about if it's realistic or not. And maps that can be leveled down with explosives or long range sniping are too much of an advantage for the human player, since the AI can't do either mass explosions or scout/snipe. So I really prefer to go the TFTD route and increase the armor values of terrain - it's not just Lobstermen & Co. that make TFTD more challenging - it's also that its map design limits both explosives and sniping.

[Solarius Scorch]

When you mentioned 'attack vectors' before it just made me worry.

Well placed nodes on strategic locations are important but if you want to ensure that there will be an alien on that location to spot XCom units then the best way is not connect those nodes with the rest to ensure that the alien doesn't wander off to some other less important node, like some TFTD maps.

Yes, certainly. But I think it would be too extreme in most cases, unless we specifically talk about a guarding post of sorts. Playing with probabilities should suffice, probably.

Discussions about 'realism' are almost irrelevant IMHO regarding games. To me it is completely unrealistic to expect that a laser or plasma beam would completely wreck a wall rather than simply carving a small hole through it. Even a grenade or a HE shell from a handheld cannon busting a wall made of masonry, metal or stone is highly 'unrealistic': you'd need a rocket launcher like a Carl Gustav or and AT-4 or an artillery/tank round to do the trick, or some well placed high explosive.

But I'm playing a game for fun so I don't really think much about if it's realistic or not. And maps that can be leveled down with explosives or long range sniping are too much of an advantage for the human player, since the AI can't do either mass explosions or scout/snipe. So I really prefer to go the TFTD route and increase the armor values of terrain - it's not just Lobstermen & Co. that make TFTD more challenging - it's also that its map design limits both explosives and sniping.

While the word "realism" was maybe a bit unfortunate, I too find certain aspects of your decisions... well, jarring. Particularly walls that are perfectly ordinary, yet put up almost as much resistance as Alien Alloys. It's simply misleading and inconsistent to have a type of brick wall with armour, say, 50 and another type of brick wall with armour 100. (These numbers are completely made up.) Even if you learn which is which, it hurts the immersion.

I do appreciate your objective of making maps that are more challenging, or different to play on, but I also think these superwalls (and particularly superceilings!) have more downsides than upsides. A map that is sufficiently claustrophobic will limit sniping (and explosives too, to a degree) much less viable in many places. Your maps are mostly quite complex and I don't think they need such drastic measures to do their job.

Of course one could go all the way and edit all walls in the game to the same high-durability standard, which in my opinion would be better than the schizophrenic present. But I'm not sure it would make the game more fun to play, and it certainly would be controversial to do so. Or we could edit all weapons to make them less effective against terrain, but I think it's only possible in OpenXCom Extended (and is also controversial).