God that would be nice.
They are lacking Italian and Soviet focus trees that are a bit more detailed, if they want ahistorical nation play for those…
Using AI to command troops is never going to be as good as a human doing it, but it might suffice as it is now, for holding more irrelevant fronts or activating a large amount of units at once.
Its still the best 4x ever made :)
I agree it certainly could be better, but it is really the first wargame, where you can grab a bunch of units and draw arrows on the map like general do in real life and the computer will do what you tell it. Now they are bunch of caveats and unless you have about 3 to 1 advantage in strength you need to micro. Honestly, my problem with game is more Naval warfare than land, and the amount of micro needed to train your air units.
But as you point out, it is almost certainly the case that core AI code has been patched and special cased to death to reach the level of competence that it has today. It is almost certainly incomprehensible to B level programmer.
At this point the way to make the game better is to start fresh, with a new code base and new design philosophy, it would be great if they could bring back some of the original team.
Realistically, I doubt either of us is going to get our wishes. I don’t anticipate an HOI V any time soon, you aren’t going getting better a battleline and Paradox is going to adopt the EU IV model and keep cranking out increasingly niche DLC.
This feels very true.
I find naval warfare to work quite okay, it exposes how vulnerable ships were to airplanes and that was no joke anyway…
As for the battle line stuff, it is so FRICKING good to just assign that to the AI at times, and then I go in and micro the heavy hitters myself…it REALLY works if you ask me.
I feel like what they’ve accomplished with it was probably pretty damned hard. It still has its issues and frequently leads to frustration on my part, but I figure most of the fixes at this point are probably of the whack-a-mole variety: address one issue and two new problems surface to take its place. There’s so much to juggle, in real-time at that, I can only imagine how problematic it is.
Don’t get me wrong it is a great game and easily worth may times what I paid for it. Hell even at $.50/hour I got the bargain of the century. The problem is it doesn’t have a random maps and so the replay value is only a couple thousand hours :-).
I’ve seen maps done but you have a point Strollen, GIEF random maps.:)
Playing as the Soviets in the Ultra Mod can be tough. I haven’t played Soviets much, and when I did, I tended to hide behind the major river lines. The mod makes it pretty tough on the soviets in the early going, and I might have overdone it with bonuses to Germany and penalties to me. I figured I’d fight defensively everywhere I could, and I’ve held the Germans to far less than historical gains in 1941, but my army is pretty much a shadow of it’s former self. Even avoiding large encirclements, the grinding battle put me in a giant equipment hole. I’ve ruthlessly combined divisions but I am still working at a deficit.
I think this is the first time I’ve played the USSR in HOI and felt some of that desperation.
I have been playing this lately, particularly in eastern Europe. I’ve been going for the Bulgaria and Romania achievements and some are real tough. Sometimes I will spend hours on an attempt, learn I failed due to some stupid mistake or technicality, and then think of moving on to another game before the urge to try again keeps finding its way back! It’s really satisfying though, to take on a new country and gradually learn the best min/max opening moves and the intricacies of the national focus tree.
Here’s my latest one, ‘Death or Dishonor or Cake’:
I ended up having to fight both the comintern and the axis for my slices of cake. (I love the smiling general on the right).
Nice. I get why they abstracted away supply as much as they did as it was a real mess for a large part of HOI3’s life, but I’ve felt like it’s abstracted a little too far. This looks pretty interesting, I’m looking forward to seeing more detail. Also looking forward to them shaking up the division meta.
Especially since he mentions the emphasis on large divisions, which is caused by some issues with HoI4’s very basic combat model. I’d be super happy if they can address the root causes here.
Also the railways and requirement for trains and trucks to move supplies over land is basically what people have been asking about for a long time.
If they can actually improve the battle plan AI on top of these, it will be one heck of an update. On the other hand, if the AI doesn’t understand stuff like railroads it’s going to suck a lot of the fun out of it.
This is likely the sticking point.
That said, giving air something to attack or the like would be cool.
https://gfycat.com/neatglisteningfishingcat
Yep it’s always the sticking point for these strategy games. The Civ series (they have a lot of company in this) has been example of adding things that really ends up subtracting from the experience because the AI is completely incapable of handling/understanding the mechanics.
At a surface level, the railroad stuff seems to be something an AI could be taught to handle. Could and will are two different things, of course. Basically, my thoughts are that it’s a great idea, just waiting to see how the implementation works out. :)
It may actually be something that can help the AI, assuming connecting nodes now becomes one of its primary drivers. That should be a pretty easy thing for an AI to do. The bigger challenge would then become when should the AI make a decision that prioritizes something at the expense of logistics.
That’s how it seemed to me as well, but massive caveat that I’m no AI developer so I really have to hedge. :) I’m cautiously optimistic this is something the AI could do well. I’m hoping!
For real. The battle-line system also seems like something the AI could be decent at, but the devil is in the details. One user (bitmode) decompiled the code and looked at the AI decision-making system, laying out the logic step-by-step and this was his conclusion:
To me the unit controller almost feels like it has been written by someone who wanted to subtly sabotage the game. On a high level view everything looks in order, even if not particularly inspiring. But when it comes to the details, there are just so many things going wrong or step by step diluting the player’s input into almost nothing that it is really hard to ignore.
This – once again – is on my list to make it through the tutorial on this weekend.
I’ve stopped paying attention to the HOI IV forums, and so missed this gem. I got a good laugh out of it.