Purge
The reasons for developing a second sequel to Hearts of Iron are probably at least partially economic. But if you were expecting HOI 3 to be a rushed-out moneymaker for a niche studio in a somewhat less-than-stellar global economy, you are going to be pleasantly surprised. This new Hearts of Iron is an almost completely different beast – the preview copy I've been playing through feels more like Hearts of Iron 4 than Hearts of Iron 2.5.
The basics are still the same. Taking control of any nation on God's green earth at one of several stages of the unfolding of the Second World War, players conduct diplomacy, run their economies, engineer social policy, direct research, and of course orchestrate mass carnage with ships, planes, tanks, artillery, etc. through a highly detailed combat model. A great deal about the specifics of each of those systems has changed dramatically, however.
The biggest change, the one that applies broadly to the most areas of the game, is the integration of the AI in various forms. When starting a game, you not only pick your country but also decide which aspects of its historical experience will be under your direct control. If you want to focus on the research and economic side of things and leave the military to someone else, you can elect to hand it over to the AI. After you've given the AI objectives (and satisfied the AI's demands for new units, delivered via the production interface), you should more or less be able to sit back and watch your delicately tuned units with their brand-new sidearms and supply theories paint the map your color one province at a time.
The AI can be told to handle production, diplomacy, your internal government (laws, etc.), research and the actual movement of the military on the map. It's conceivable (and maybe a good option for newbies) that the AI runs everything and the player sits back and provides guidance on certain specific aspects of the overall picture, leaving the heavy-lifting and stack-shuffling to the computer. A premonition of humanity's future as robotic slaves, or just innovative programming? You decide.
The AI's nice in theory, and a few parts of it I was able to experience were nice in practice as well. I let my commanders handle the steamrolling of Ethiopia as Italy in early 1936 and, as one would predict, they did a pretty good job. A more severe test of the AI's capacity came when I played as the Soviet Union during the opening stages of Barbarossa. According to the little scrap of paper I received with the preview build, “At this moment in time the Naval and Strategic warfare AI's are not implemented.” I can't say for sure what this means. When I turned my HQ units over to the AI they did a terrible job at holding back the Germans, but then again I didn't do that much better. However, the fact that something as important-sounding as the “strategic AI” isn't yet fleshed out raises an issue, probably the only real issue I have with what is otherwise turning out to be a very cool game.
What Paradox has done here is actually a very risky thing. The level of detail in this game – never low in the series – is now almost beyond the capacity of the average human being to comfortably manage. Especially for larger nations, incorporating the AI isn't so much a decision as a requirement for success. If you want to concentrate on the combat, you better not be worrying about your resource stocks, because when you're bogged down in unit widths and frontage penalties and keeping your complicated hierarchy straight, you're not going to have time for anything else. I didn't have time to do all the testing of the AI I wanted to, so I'm not confident that it's fully up to the task. In the limited areas where I employed it, it did an OK job, aside from the occasional boner that would result in encirclement or something equally disastrous. It's not good form to use cliches like this, but time really will tell whether or not this gamble will have paid off. If you're reading this preview sweetly ignorant of all previous HOI games, bear this in mind.
Progress
The proper division of systems in HOI 3 is between “combat” and “everything else.” The game is supposed to be about fighting a war, not running supply convoys or delving into research minutiae. Thus, any improvement in the non-combat systems has to be examined in this light. The economic model, for instance, of resource collection and factories, hasn't really changed and is still extremely simplistic. But this isn't a game about factories, so the worthiness of the play-experience isn't really in jeopardy.
Technology has been basically reworked from the ground up and more closely resembles the system in the original Hearts of Iron, where numerous specific upgrades added up into large improvements over time. Tech teams are gone, replaced by an allocation of scarce “leadership” points which can either be given over to research, diplomacy, intelligence, or to replenishing the ranks of officers. There are still a bunch of technological categories – infantry, armor, fighter, bomber, naval, theory, etc. – but in each of them there are now a multitude of small advancements which are researched quickly, grant small bonuses (and some drawbacks, even) , and are highly inter-related. There are now external values which store your nation's expertise in a certain area of research. The more you concentrate on, say, armor, the more your values relating to armor experience grow, and the easier your next bit of research in armor will be. Your tanks will be built faster, too. This new wrinkle allows players to really customize their armies, especially if they pick the 1936 scenario and thus have a minimum of three years before the first shots are fired. Going back to the example of the Soviet Union, you start out with specialties in Human Wave doctrine and militia theory. However, you can totally abandon those sectors of research and attempt to confront the Germans with high-quality tanks and maybe even a decent navy.
Diplomacy is totally different, as well. There's a nifty new triangle on the diplomacy screen, where every single nation is plotted by their relationships to the three corners of the triangle – Comintern, Axis, and Allies. Most nations are clustered in the center between the Axis and Allies, and a few stragglers stick out more towards the commies. As a member of one of the factions, you can try and move nations more towards your corner of the triangle month by month, by diverting some of your leadership points (mentioned earlier under technology) towards the training of diplomats. As a visual aid it's a little stroke of genius. You can form alliances, pacts, declare war, etc. all from this single screen.
Intelligence is (surprise!) new, too. Instead of rolling the dice every few days on a coup attempt, you now allocate spies from that same leadership pool to work in Country X and either support rebels, support your party in the country, conduct sabotage, etc. It's a more organic and slightly less frustrating system than the old one, though it still takes some doing and a lot of patience to really upset things.
Rust
I had a few crashes to desktop during my time with the beta, but for the most part my technical problems were more amusing than game-breaking. One time after building radars in Kiev, I was able to see the force compositions of armies hanging out in the Pyrenees. Aside from the rather glaring omission of some AI functionality, the game is basically done, and what's there is very solid, all things considered.
Census
So what is this game? It's gigantic, packed with detail, and positively overflowing with potential. It could be the theoretical end of WW II gaming; if Hearts of Iron 3 lives up to its pedigree and to the ambitiousness of its design it might make every subsequent WW II war game seem like a weak imitation by comparison. It all rests on how the question of the AI is resolved. Everything else is in place and more or less competent, at times even brilliant. On the other hand, if the computer is incompetent at running things and players have to rely on their own fragile brains to process everything at once, I think the whole thing will fall to pieces. Here's hoping, regardless!
