Iron War - Official Thread

[General_Zod]

If anyone is interested in taking Iron War for a spin today slap me in lobby. I prefer a multi if possible 2v2.

[Hepps]

@frostion Well done getting this turned out so quickly to make use of the new code.

Looking forward to testing it all out.

[redrum]

@Frostion I played about a round to test it out and things seem to work well. One question I had is that I realized AA guns are essentially the fodder unit of the map instead of infantry, is that intended? It feels a little strange to mass AA guns for fodder.

[Frostion]

@redrum the AA is the cheapest and weakest unit. It is not the best cost/effective defence unit, that is the Infantry. And it is certainly also not the best attack unit. Actually it is probably the worst unit. I think that the AI choses the unit as first casualty as it has the weakest stats, 1/10 attack and 1/10 defence.

In the battle calculator the AA and the Tank-Destroyer are worthwhile units to have, in small numbers. Here are example results, that should shown this. Both attacker and defender has precisely 155 to attack with and 140 PUs to defend with:

• 10 inf + 5 art attack 14 inf = 60% defender win

• 10 inf + 5 art attack 10 Tank-Dest + 1 Inf = 38% defender win

• 10 inf + 5 art attack 20 AA = 11% defender win

(So Inf is best defender vs inf + art army)

• 7 Inf + 2 art + 1 Mech-Inf + 1 L-Tank + 1 M-Tank + 1 Fighter attack 14 inf = 53% defender win

(So mixed attack army is better than an all inf + art attack army (60% defender win))

• 7 Inf + 2 art + 1 Mech-Inf + 1 L-Tank + 1 M-Tank + 1 Fighter attack 1 inf + 10 Tank-Dest = 68% defender win

(So Tank-Destroyers are better than Inf vs mixed army)

• 7 Inf + 2 art + 1 Mech-Inf + 1 L-Tank + 1 M-Tank + 1 Fighter attack 10 inf + 2 Tank-Dest + 2 AA = 73% defender win

(So inf heavy defence with a bit of AA and Tank-Dest seems to be the optimal defence)

[redrum]

@frostion Fair points but consider some of these alternatives which focus on using AA guns as fodder and other units to provide attack/defense power:

1. Optimal 140 PU defense vs (10 inf + 5 art) attackers -> (7 inf + 10 AA) = 68% defender win
   -- Essentially uses AA as fodder and inf to add defense power

2. Optimal 155 PU attack vs (14 inf) defenders -> (4 inf + 4 art + 10 AA) = 42% defender win (only 154 PU)
   -- Essentially uses AA as fodder and inf/art combo for attack power, there might be even more optimal ways to get attack power but figure I'd keep the example simple

3. Optimal 140 PU defense vs (7 Inf + 2 art + 1 Mech-Inf + 1 L-Tank + 1 M-Tank + 1 Fighter) attackers -> (7 inf + 10 AA) = 79% defender win

Point being is that for almost every attack and defense, its optimal to have a good portion of your army as AA guns to use as fodder. This also doesn't even take into account that if you add a few planes to the opposing attacking/defending forces then AA guns are just crazy OP.

PS. For Russia turn 1, building almost all AA guns is probably close to an optimal purchase.

[Frostion]

@redrum I can try to run some calculations with the AA costing 8 PUs, instead of 7?

It would not make sense if AA had a 10 PU cost like infantry, since the AA is just 1A/1D(+1AA) and infantry is 2A/3D(+attack supportable).

[redrum]

@frostion Might be better to consider making AA more expensive but stronger so that they aren't so 'hitpoint' efficient otherwise they become 'fodder' instead of the infantry units. Something like 3 AA attack/defense and cost maybe 12 PUs. More aligned with Tank Destroyers. Or change them to be infrastructure units that don't have hitpoints.

[Black_Elk]

Looking good. Just downloaded and played a few rounds of a German solo under the new set up. One thing I noticed immediately is that Germany is under a lot more pressure to take the southern USSR right away. That whole swath around the Black Sea and Caucasus becomes really critical to war machine. Historically it makes sense, though it does push a somewhat more one dimensional play pattern on G right out the gate. If you mess around with a navy or try to go on escapades to the periphery (like Africa/England or wherever beforehand) you'll almost certainly be running out of fuel in the third round. After the starting surplus is spent things start grinding to a halt pretty quickly going that way. First 2 turns you can only move like 15 fuel units to their max range per round until G passes the 30 fuel threshold (which pretty much requires cracking Stalingrad). I think that definitely favors the infantry push dynamic, since if you move your fleet beyond the Baltic/North Sea area, or try to expand the Luftwaffe and bounce around with it to make light trades too soon, then the tanks/mech will likely run out of gas before they can drive far enough to take the Russian oil fields.Similarly trying to use the Kriegsmarine for much of anything too soon puts you in a real bind. I think it cost me 20 fuel just to clear the Baltic of Russian ships after which point the starting reserves are more or less spent. I think its entertaining trying to juggle the choices, but the learning curve might be a little steep. Anyhow here was my first run at it. Going south with the mobile ground and trying to maintain a surplus I was able to keep the fuel up pretty high vs the fastAI... Soviets are still grinding it out in the Pacific though.

0_1521537494931_Elk vs FastAI Allies New Fuel Germany 15.tsvg

If people find that the fuel costs are too rigid, I think the easiest approach would just be to increase the starting reserves slightly for whichever power might need it (rather than introducing a bunch of new oil fields points across the map or anything too drastic.) Basically that would just push everything out another couple rounds, giving players a little more time to maneuver their heavy equipment before they run dry. Or try the resource exchange/development thing. Seems pretty fun though right now at least for messing around vs the machine.

[General_Zod]

@black_elk @Frostion

Based on @Black_Elk's assessment. You might wanna eventually consider 2 modes for human players. One "historical mode" where fuel is, as is and realistic. And "ahistorical mode" where fuel is more abundant for axis. It can be done with a custom game property in game options before the game. Or if you prefer on game start with a "user action".

Another suggestion is "synthetic fuel" which the axis can lean on, to keep the war machine moving. Perhaps just let the axis purchase it out right or build refineries to produce it.

[Frostion]

"historical mode" and "ahistorical mode" is something that I would like to avoid. I would rather try to create conditions for nations like Germany, that allowed them to win with what fuel and strategic options they got.

The fuel-needs of course pressures specific players towards capturing specific territories nearby, but hopefully there can still be alternative actions to do, like using more inf/art, spending more PUs on minor nations, fortifying a turn instead of advancing/attacking, aiming for high PU-territories instead of fuel-territories etc. It would be optimal if optimal fuel income was not required to win.

Hopefully a fuel pressured nation is also fighting nations with their own fuel needs, or at least all the major nations are in some fuel need.

In regards to how nations can gain fuel, besides losing battles and fuel units to gain a fuel surplus, a player could also think about selecting units like Mech-Inf as casualties instead of Inf, or self propelled artillery instead of artillery as the units have the same combat stats and the surviving units are probably at the front line where 1 could be most likely.

Actually this process of balancing fuel is probably more complex than I first thought. Maybe it would be a good idea to make a nation list and then focus on one nation at a time. Maybe and logically starting from the top with Germany. Maybe each nation should have their "historical correct" actions/options tested as well as alternatives. And maybe when all nations have had a fuel tuning, another from the top down fine tuning would be needed, as all enemies have been altered a bit.

I would like to play a bit as Germany vs AI and also investigate black elks approach, but it will be a few days before I can play a few more games.

[Black_Elk]

I dig the concept of drilling or perhaps creating synthetic fuel (maybe that latter as a technology advance?) as a way to introduce more fuel over the course of the game.

The challenge of having most of the fuel front-loaded as part of the starting reserve is fun (since it forces you to make some hard choices), but somewhat unforgiving if don't realize what's going on. Basically if you do everything you want to do and move everything you want to move during the first 2 turns, then you'll likely run out of fuel halfway through turn 3. Whereas if you are more conservative with fuel in the opener then its easier to strike a balance to keep you running after turn 3. The trick of it is that the consequences don't come till later, so if you aren't paying close attention to what's happening and tracking the fuel consumption in the economy tab, you will probably burn through all the oil in the opener and screw yourself for the midgame.

I played a Japanese solo vs the fastAI this time to see how their fuel holds up, and the thirst is probably more pronounced with them than it is with Germany hehe. Again more historical that Japan should be immediately tripping about oil (since it was part of their reason for starting the pacific war in the first place) but they are definitely hard pressed to navigate the naval movement to fuel balance. If you move all your warships on the first turn, the way people usually do to maximize the attack advantage, that costs basically the entire fuel surplus. Just trying to shuck the 7 starting transports every turn will burn like half the fuel that Japan collects per round, to say nothing of the warships, air and mobile ground.

I think the option to develop an oil field for a cost in PU's could be a cool way to ease the pressure. A unit that could be bought/destroyed would be cool. But in its simplest formulation you could just make it another purchasing decision. Like purchasing a limited amount of fuel for a larger amount of PUs at the end of the turn, as a way to overcome a chronic shortage? Maybe if they hit a synthetic tech advance, then players can buy a little more fuel for the same cost or whatever. But basically a simple mechanism to give the player some breathing room might be cool.

In this game I left about third of the IJN fleet and most of the mobile ground in place on the first turn. Just moving the stuff at Truk out of harms way, and a Carrier+Battleship to cover the necessary transport movements to take some oil from the French Colonies and the Dutch. The Gulf of Thailand, Japan sea zone, East and South China Sea zones, are the obvious safe spots for a fleet that isn't moving much, so I figured just try to park the bulk of the IJN somewhere around there and save as much fuel as possible for the ground war in Asia. Then I built factories everywhere I could (probably more than necessary) and dropped as many infantry units as I could afford to hold onto the oil while working on the Soviets and Chinese. The 4 oil at Truk seemed especially critical so put like 100 PUs into holding it. I think the distribution of oilfields in the Pacific is pretty nice to model the historical war. It puts a big focus on China worth 3 fuel, and the Dutch and French colonies worth 5. Once you got those then its easier to move around. I think it makes it somewhat less likely that a player just guns for India or Irkutsk or Australia full force out the gate, since it costs a shit ton of fuel to diverge from the more historical play pattern. I guess balancing the game for the AI will probably need a sizable fuel bonus for the machine (such that fuel wouldn't really effect the computer opponent) otherwise I think they'll burn through it too quickly and get stuck with a bunch of TUV that can't move.

Anyhow, here is that Japanese solo game in late 44...

0_1521613055401_Elk vs FastAI Allies New Fuel Japan 10.tsvg

[zlefin]

how is the ai handling the new fuel system? I haven't gotten to play in awhile, but i'm curious, as it seems like something the ai isn' tsetup to handle well. is the ai given some bonuses to handle it? especially with moving around that's something tricky for the ai to evaluate whether it's worth it or not.

[General_Zod]

@Frostion

Here's another suggestion for fuel. Create areas that have fuel/oil stockpiles. These are areas that will give a one time immediate fuel boost if captured.

Similar to national objectives but only one time. Further more they should not give any bonus to the original owners of the territory (for simplicity). The original owners will want to defend the stock piles regardless, so they don't help the enemies war effort.

Eg. Hawaii had huge fuel reserves stockpiled there, although it does not have actual fuel production.

Since the xml doesn't support placing actual resources on map, for capture. It would need to be done via a special unit and triggers, so when the unit is captured it gives the one time fuel stockpiles then it disappears. Alternative method would be triggers/conditions for the territory resources property.

A step further than this would be to not only make those fuel stockpiles capturable but destroyable. This would require more depth though, since it would require the stockpiles to be just as useful to the original owners, as to the capturers, if it's to be logical.

[Black_Elk]

Right now the AI doesn't handle the fuel thing very well. I see most of the weirdness happen on non-com, like transports with floaters, fleets splitting up recklessly, newly purchased tanks or aircraft staying put etc. Even as a human player the starting fuel is probably a little low, which is why it would be cool to see some gradual influx. I like zods idea of stockpiles. It reminds me of an idea floated a while back to have gold reserves, but using fuel instead.

I think for the AI to have a chance to remain competative it should just be swimming in oil. Especially if its a naval power like Japan, we should assume that the machine will be moving those ships all over the place.

I played a few more solo games, one as the USA and the other as Russia to see how things felt on the Allied side. Will post when I get home in a little bit.

In general I'd say that the fuel=movement thing pushes a lot of the other usual stuff into the background. It drives purchasing and informs the conquest objectives more than cash or steel I think, since you can't do much if you can't move the heavy hitters. The game is probably more realistic, though somewhat less dynamic. I don't know, my rough sense of it is that the fuel probably needs to be doubled across the board, at least for the AI big dogs.

Here is the first save, as the US I just waited until I had an opening then made a B line for Tokyo...
0_1521665733124_Elk vs FastAI Allies New Fuel USA 8.tsvg

In the next game when I played as Russia you can see how the AI struggles with purchasing under the new fuel system. US AI for example has like 500 TUV in fighter aircraft stuck in North America unable to move them out across the pond. At various points I saw AI Japan crash fighters into the sea because they couldn't move their carrier on non com. A bunch of loaded transports in the middle of the sea. Stuff like that. Meanwhile with a little conservation and a focused campaign the Soviets have built up a substantial reserve (larger than the rest of the Allies combined). Might be cool if they could sell it to their buddies or something hehe.

0_1521670544947_Elk vs FastAI Allies New Fuel Russia 17.tsvg

Anyhow, I think the issue gets a bit more pronounced the further into the endgame you go, which is why it would be cool if there was some sort of incremental increase in the total available fuel, so you could unlock more reserves as time goes on.

[Frostion]

Iron War 0.2.21 to 0.2.22 (requires the latest prerelease) 0_1521675408191_IronWar0.2.22.zip
• The Anti-Air unit now costs 8 PUs, not 7.
• 1 starting Fuel-Barrel placed in Norway.

The AA gun is now set to cost 8 PUs, not 7. According to the battle calculator, the optimal defense unit composition / cost effective defense against an Infantry + artillery army is around 76% Infantry and 24% AA guns. If is like 16% or 32% AA, the odds will be more in favor of attacker. (Anyone is welcome to recalculate. I used an infantry defense army of 500 PUs where 10, 15 or 20 AA guns could be inserted and be part of the 500 PUs by removing Infantry)

I think 8 PUs is reasonable, as the AA is weaker in attack because it cannot be supported. I would like the unit to be in play, and not be an almost never used unit. As I see it now, players would now maybe build AAs to guard against bombers, air-transports moving infantry and also build them for mixed armies.

Germany:
I have done a few games with Germany again. I could see that their options, if going on a fuel grab mission, should be expanded, so I added a barrel in the Norway (http://www.norskolje.museum.no/en/vallo-refinery-closed-down/). Yet an incentive to attack Norway.

For me, the Germans need fuel, but it is not crucial for winning. There seems to be a general trend when I play Germany: Round 1-2 has fuel enough, round 3-4 lacks fuel, 5+ again has enough fuel as most units have arrived at the eastern front + without Germany necessarily sitting on Ukraine, Stalingrad and Caucasus. My play testing is probably affected a lot by my normal choices. I build a factory in East Prussia. I give 20-30 PUs away every turn to Iran, Iraq and Finland. And I keep my navy in the North Sea after destroying the USSR Baltic fleet. This probably keeps my fuel needs down. I would see it as a good thing if other players felt the urge to go historical and try to take the oil fields in Caucasus, but this is not something I go all in to do. Overall I can live with the German fuel need right now.

Japan:
I will also try to play with Japan again next time, when I got time.

Synthetic fuel / fuel purchase / capturable one-time fuel depos:
I would rather not implement new fuel sources or new purchasable stuff to the game, and this one would also complicate the fuel aspect of the game further. Right now the income concept is pretty simple, and I would like to keep it simple. It might take a few games to learn how to handle and conserve the resources, but this will hopefully just keep layers playing.

AI and fuel needs:
I know that the AI is not conserving fuel in any way. It does not think in any way about its fuel needs, it just moves and runs dry quickly. I will have to consider what way to go about this, and how to maybe give AI some extra fuel.

[CrazyG]

The AA-Gun vs Infantry thing isn't necessarily about big battles. Often you are capturing territories with a handful of defenders and lose a unit. It doesn't matter how bad the combat stats are, the cheapest unit always has a niche as fodder for this type of battle. Also as fodder for the defender in these small battles, this it makes it risky for the opponent to use aircraft to trade territory.

[General_Zod]

@frostion I don't see the Norway barrel.

[Black_Elk]

Sounds good will check it out when I get home later.

I think the main thing that happens under the fuel=movement scheme is that its a lot harder to globe trot. For example I think you're less likely to see Germany in South America or Japan in Africa etc, because it just burns up too much fuel to move a large invasion fleet that distance. Similarly its probably harder to transit a ton of fighters or bombers to prop up a buddy on the other side of the world, because of the fuel requirements. Stuff like that grounds the game as a more historically realistic scenario, though I do worry that we lose some of that world domination charm if the fuel runs out too quickly.

I think there should be enough fuel in the starting reserve to move the starting naval units their max distance every round at least until 1942. So basically running dry closer to round 5-6 rather than round 3. That way the big dogs on the water have enough gas in the tank to position their fleets for the battle of the Atlantic, and for island hopping in the Pacific.

Still just my first impression though. I'd like to get in a few more games beating up on the machine and report back. Catch you in a few

[Black_Elk]

ps. on the subject of Germany and Norway etc. One thing I've noticed is that the German AI has a really tough time dealing with Denmark and the Danish straits. The fact that the Danish straits start the game neutral and closed, means that the German AI invariably sends the bulk of their fleet north to Murmansk/Archangel rather than defending in the North Sea or trying to retreat into the Baltic Sea the following round.

I know you've said before that you didn't want to cater the design of the map to the AI's deficiencies overmuch, but I think this one is particularly rough on the game balance. The German AI is doing stuff on the first turn that falls way outside the scope of what any reasonable human player might do under the same circumstances. It got me thinking again about the timeline for early 1940.

The invasion of Denmark and Norway occurred on April 9th, 1940.

The Battle of France and the invasion of the low countries began on May 10th, 1940.

This would make it pretty easy to give starting control of Norway and Denmark to Germany from the outset, and say that the game begins immediately after the Scandinavian invasions, but just prior to the invasion of France. (The situation should be familiar from the A&A Europe 1940 game which does the same.) I think it would be a lot easier to design a sensible opener for the German AI under those conditions. Right now they just don't seem to be very interested in Denmark or defending the North Sea fleet, which should be like priority number one (even more than beating back the Russians) since it makes it easy for the Allies to set up shop in striking distance of their core territories.

Just to see how the AI would respond, I edited Denmark and Norway to German control and watched to see what the Fast AI would do. Here they made a pretty sensible decision to consolidate their fleet in the Baltic, which feels like it would set up a somewhat more historical play pattern from G heading into the second round.

0_1521694068534_AI test New Fuel Denmark and Norway edit to Germany.tsvg

[Frostion]

@General_Zod
In the XML posted above, with the ingame version name 0.2.22, there should be a Fuel-Barrel in Norway. The same version has AA gun costing 8 PUs, not 7. Have you downloaded the zip, extracted the XML and placed it in the game dir of the Iron War zip?