• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
You can't code creativity into an AI. Not yet, anyway, and not into a game AI. But the idea is that the Theater level AI only ever formulates very simple goals - such as, "broad offensive against country X," and then the rest is handled by lower echelons. You can make it more complicated by having it prioritize certain strategies already at the Theater level (say, emphasize taking province Z instead of Y, and have Army Group commands take note of that), but the point is to get it down to a problem of a Corps choosing to attack 1 province out of 3. If you don't start the real decision-making at that level, anything on top of that will collapse like a house of cards :p

The problem is that if say I have at whatever level the decision made to surround and isolate the units in Province A the last thing I want to have happen is before the overall forces are assembled the I Corps commander to attack the weak link province I am planning on ending the encirclement on and starting a chain of reactions by the enemy which likely blows the plan out of the water.

Information needs to pass in both directions up and down the link. However, speaking only for myself I don't start the planning process by worrying about which province I Corps will attack but by setting the overall objective. Only once I have set the objective "Isolate and destroy all German Army Units west of the line running north of province eksupsolon" do I start worrying about what I Corps, I Armoured Corps, etc will do to implement the plan. And only once I have decided that I Corps will attack in that direction and I Armoured Corps will attack in that other direction do I worry about exactly which provinces will I Corps attack in what order. Finally at the end I worry about which division will do what.

I may modify start dates, change the orders and so forth of the various divisions depending on organization, type, terrain, leadership, enemy forces etc in a "refinement phase" of the planning process.

Again this is without considering airpower, naval lift, or naval support since all three add a lot of complication.

For me the process is fully top down. You are correct that for the plan to accually work what province(s) is(are) attacked by what unit(s) is the make or break decision for the plan. But without an objective to the action it is just attacking randomly. It is not much different then playing a poor chess player who basically moves pieces at random. That can, infact, be maddening but is unlikely to result in consistant victories.

But I agree creativity won't exist as in reality I am not sure that the term AI is fully applicable. AO (Artificial Opponent) is likely far more accurate as I am dubious about "intellegence" in the extreme.

I guess we will have to see what is done and how it works in reality.
 
Please let the new combat command system do away with the "Lone Ranger" divisions.
You know the ones, you blitz into Poland and a single division retreats into the heart of a German province. It is immediately terminated by flanking German divisions. It is totally not just unhistoric but suicidal and is an emersion killer.
No unit would advance into enemy territory after being defeated, or when your country is collapsing.
The commander would either dissolve his unit defend to the death or retreat in a direction ordered by his C in C. But never mount a one man band invasion of Germany.
I know what we have is a front ai at the moment so if a German player leaves a province open it is attacked even if suicidally by a lone division which leaves the area it is guardng open.
Very easy for a human player to exploit.
Actions should only take place if there is a rational.
Which brings us to the importances of Theatre, Army, Corps and division ai commanders having a plan.
 
The fact that human's beat chess playing computer all the time... well enough said.

The computer can fight battles but it has a hard problem sticking to a plan, it is forever evaluating your moves and countering them.
You retreat it attacks. You fall back it follows. You slam the door and its army is encircled. Did the AI do anything wrong? No.
Would a human player have seen the bigger picture, maybe? So getting the AI to do the right thing is a lot harder than you think.
Now add in to the equation that there is no Comtern, Allies or even Axis for the AI.
Each country is fighting by itself and you can understand just one of the problems the AI programmers’ face.
Now lets throw in research, logistics, supply, diplomacy leadership and another 10 000 odd statistics and the movement of just one division by you can cause the AI to change literally thousands of variables.
Yes it theoretically should be able to fight better but fighting is only a minuscule part of the game.
My hats go off to the game designers, their task is thankless and despite having two highly successful games under their belt, no one will be totally happy with what HO3 eventually is.
Modders will go to work, forums will spring up and thread after thread will complain and later a much awaited patch/expansion will fix it… that is until there is talk of HO4 :rofl:

If the AI was that good then I would have been happy lol. Instead, the AI tended to be absolutely idiotic. They had no battle plan like you claim they do. For example, instead of the allies having a giant invasion as in Normandy, the AI would invade with about 5 divisions, which I easily destroyed each time. The Allies lost about 50 divisions doing that while I concentrated in Russia. The grand strategy part of the game the AI hardly does worse than the human. The problem is on the stragic and tactical level the AI is plain dumb.
 
Please let the new combat command system do away with the "Lone Ranger" divisions.
You know the ones, you blitz into Poland and a single division retreats into the heart of a German province. It is immediately terminated by flanking German divisions. It is totally not just unhistoric but suicidal and is an emersion killer.
No unit would advance into enemy territory after being defeated, or when your country is collapsing.
The commander would either dissolve his unit defend to the death or retreat in a direction ordered by his C in C. But never mount a one man band invasion of Germany.
I know what we have is a front ai at the moment so if a German player leaves a province open it is attacked even if suicidally by a lone division which leaves the area it is guardng open.
Very easy for a human player to exploit.
Actions should only take place if there is a rational.
Which brings us to the importances of Theatre, Army, Corps and division ai commanders having a plan.

Very rarely do any unit irl "defend to the death" simply since it is pointless. Soldiers don't fight well (at all mostly) when faced with the option to live or die.
 
Rarely but that happens - or better call it 'fight to the last bullet'. Polish campaign for example:
- Westerplatte (Danzig) 180 held for 7 days http://en.wikipedia.org/wiki/Westerplatte
- Wizna: 4 days ~700 troops on 10km agains german divisions (with 1 tank div) http://en.wikipedia.org/wiki/Battle_of_Wizna
- Hel
http://en.wikipedia.org/wiki/Battle_of_Hel
thats just a few examples in one 5 week campaigne - there were more on other fronts and should be included.
Many sorrounded units will try to hold on as long as its possible because they hope for a counter that will help em, belive that it can help the rest of forces to bind as much enemy troops as possible or because of its military tradition and honour/dignity.
 
If the AI was that good then I would have been happy lol. Instead, the AI tended to be absolutely idiotic. They had no battle plan like you claim they do. For example, instead of the allies having a giant invasion as in Normandy, the AI would invade with about 5 divisions, which I easily destroyed each time. The Allies lost about 50 divisions doing that while I concentrated in Russia. The grand strategy part of the game the AI hardly does worse than the human. The problem is on the stragic and tactical level the AI is plain dumb.

That thread has come back to bite me twice. I was using it to illustrate how hard it would be to create AI capable of doing the above.
As you have quite rightly pointed out the AI is dumb and will land small forces. Have 250 divisions along an entire front including six or sever mech and armoured divisions yet attack with a stack of 20 infantry divisions?

Very rarely do any unit irl "defend to the death" simply since it is pointless. Soldiers don't fight well (at all mostly) when faced with the option to live or die.
The Japanese did, without question.
Several SS units did obviously for different reasons.
So a Japanese commander ordering his troops to hold position (must be an option to ignore all casualties even if it is nation specific) should be a last stand. And that is historical. or as you put it irl.:)
 
landing with 5-6 divs isnt wrong - its how it goes IRL but the problem is that there are no troops to follow. Landing is performed by only part of army (size depends on the fleet assigned for that task - Normandy IIRC was made by 3 para and ~6 divs including some independent brigades baons) - they are suppose to secure the beaches. Then comes the time for the rest of the Army. Thats the problem with the AI - it cant plan farther.
 
landing with 5-6 divs isnt wrong - its how it goes IRL but the problem is that there are no troops to follow. Landing is performed by only part of army (size depends on the fleet assigned for that task - Normandy IIRC was made by 3 para and ~6 divs including some independent brigades baons) - they are suppose to secure the beaches. Then comes the time for the rest of the Army. Thats the problem with the AI - it cant plan farther.

It can with the top-down approach. If it deems that it has sufficient strategic interest in staging an invasion of France, then it needs to assign enough forces to have a good chance of completing the objective within a reasonable time. FE by creating an Army, Army Group or Theatre especially for the task, or redeploying existing ones.

For instance, if it assigns an Army to the task, lands three Divs in a province, creates a bridgehead, then I would expect it to follow up by moving the rest of the Army into the bridgehead. Or with an Army Group it might land in more than one province at the same time, with each Army assigned a province to carry out the landing, then get ashore and consolidate their position. To me this is all relatively easy to code.

A much better AI comes from a few very simple rules:
1. Create a proper OOB
2. Assign objectives to high-level commands which require long-term (weeks/months) actions (capture the whole of the region which includes Paris, rather than capture the single province of Bayeux on the coast)
3. Have the AI reconsider/change those objectives only occasionally. Maybe only once per week to reconsider, and only change in extreme circumstances. So the Army Group's objective to capture Paris would only be changed if they suffered significant losses or there was something like an invasion of the USA.
4. Prevent the AI from considering removing Divs from a Corps, and Corps from an Army every hour/day. Again, once a unit is assigned in the command structure it's position should only be reconsidered occasionally, and only changed for very good reasons. Personally, I would only reconsider Corps organisation if the Corps has lost some of it's Divs, and only reconsider Army organisation, such as removing a whole Corps and assigning it to another another Army, for strong strategic reasons. Again maybe only once per week.
5.Try to keep all of the Divs in a Corps in adjoining or the same province(s). Try to keep Corps in the same Army in contact with each other, in adjoining provinces. So once the leading Div from a Corps has landed, this has the effect of pulling the rest over.

The problem in HOI2 was there was no plan, and it often seems like units deployed to the UK ready for an invasion of France, were just as likely to be actually landed in Southern Africa as in France by the AI.

What I would hope to see in HOI3 is the AI to FE assign 20 Divs in an Army Group for an invasion of France, and two months after the landing still see all of those 20 Divs fighting in France in the same Corps/Army structure, rather than finding half of them scattered across the world in different Corps and under different commanders.
 
a bit late but....
Divisions now fit into a multilevel command structure. From Theatre, Army Group, Army, Corps through to Division. Each level has its own commander that gives its own bonus according to his skill level. At division level you get a bonus to combat while a corps commander increases the chances of reserves joining combat.
This is just great. Keep up the good Paradox :)
 
build times

Why are the build times so long for a brigade? I think the build times need some serious structure. Based on #7 diary photo your going to have a division with 2 inf brigades with 2 amor car brigades and its going to take 3 years to complete for battle ready? 1080/365 =2.95...years. that is an unacceptable algorithm for 1 division.

even if you did 4 inf brigade to make a division, that is 95 days * 4 = 380 days? that is nuts. people are just going to be building a bunch of 1 brigade divisions to have a enough units to cover the massive front line land mass.

again there is no way to build all the historical divisions of the major nations with the current build times.

USSR had 250 rifle divisions. In HOI2 it doesn't comes close to that many before I invade it on 22 July 1941.

I like the idea of custom divisions that are created based on brigade units but the build time for all brigades (4) that make up a division unit need to be cut by 1/4th.
 
Why are the build times so long for a brigade? I think the build times need some serious structure. Based on #7 diary photo your going to have a division with 2 inf brigades with 2 amor car brigades and its going to take 3 years to complete for battle ready? 1080/365 =2.95...years. that is an unacceptable algorithm for 1 division.

Actually, that was 4 unit serial build, so one division is build in (1080 / 4 = 270 days).

And, those values are just placeholders. If you look at superheavy tanks, you notice that they don't use any fuel and you can build them in 60 days.
 
AWESOME!!! You folks at PI are easily keeping the title of BEST Strategic WWII era game ever.

The concept of Army group-Army-Corps-Division structure must be a definite keeper. I always was sad to have to promote good divisional commanders to a higher rank just to have a realistic feel of Corps or Army leadership. I always felt the single division- in a group of 6 divs- with individual leaders was not optimum. Easy to control stacks of 4-5 divs with divisional, corps, and theater commanders takes HOI to the top level of play and realistic looks.

I dont know if I can make it until 3rd quarter 2009.
 
"Divisions now fit into a multilevel command structure. From Theatre, Army Group, Army, Corps through to Division. Each level has its own commander that gives its own bonus according to his skill level. At division level you get a bonus to combat while a corps commander increases the chances of reserves joining combat. "

I'm starting to wonder if Johan et al have played the old SSI "Second Front: Germany turns east" game (1990).

This command aspect - especially the fact that a senior commander increases the chances of reserves being committed - is almost an exact copy in principle.

Some of the logistics also seem to mirror aspects of it.

I probably should point out that I thought that game was one of the most entertaining simulations of Barbarossa I've ever played, so if Paradox chooses to parallel some aspects of it I'm all in favour of it!

It's entirely likely, mind you, that Paradox is reaching these conclusions (about logistics, chain of command etc) entirely as a result of their own studies and developments from HoI2, and it makes sense that what they arrive at shares similarities with another successful simulation of them.
 
It's entirely likely, mind you, that Paradox is reaching these conclusions (about logistics, chain of command etc) entirely as a result of their own studies and developments from HoI2, and it makes sense that what they arrive at shares similarities with another successful simulation of them.

Not sure how they decide what to use. I hope they cherry-pick some of our good ideas. I dont mind complexity in strategic games. As long as it is complexity that is fairly historical and adds to the gaming experience. Good game principles and concepts are interchangeable in my book too.
 
To end this weeks development diary, we’re going with a few screenshot of South America, where so many of the major battles of WW2 was.

South American and Central America are some of the zones that I never really cared a bit about in HoI2

Judging by how dynamic this game is shaping up to be.. I'm think anything is possible anywhere on the map...

this is good :D