The fourth and final part of the Heroes V, Deconstructed series is up. Corribus tells us what he though of combat (one of the hot topics of the beta test, because of the size of the battlefield) and the artificial intelligence (something that could very well remain a hot topic forever), and concludes the series by giving his final appreciation of the game. Did he love it or hate it? Read on to find out.

For his next column, Corribus needs your help. He would like to start a strategy debate where each participant will support a specific Heroes of Might and Magic V town. Send him a private message if you'd like to learn more and possibly join the little project.
  1. Battles

When I entered my first battle, I have to say I was very happy to see that the isometric view from H4, which I absolutely loathed, was gone, replaced with a side-on grid similar to what I remember from H2 and H3. Thank god, and may the isometric H4 battlefield rot in hell for eternity. Finally, I can once again see what creature can go where, and how they can get there. H4 had some nice tactical battle innovations, like being able to shield back-row units from enemy shooters with your own front row units, but to me they were all sadly destroyed by that stupid, nonintuitive isometric view.

Ok enough about that godawful, heinous isometric view.

Where was I? Oh, yes. Another thing I noticed right away is that while the side-on H2 and H3 battlefields used a hex grid, in H5 we get squares, which seems a little clumsy. I’m not sure why they opted for squares, as hexagons are much more natural for tactical war-games, given the whole corner thing. It’s not an ideal solution to my mind, but at least it is simple and for the most part intuitive. But not always: for instance it’s not always logical to me what square a creature will occupy when attacking another creature from a certain direction. Still, the battles work very similarly to those in H2 and H3, with some differences, not all of them good.

The first is the initiative bar, which is a nice idea, as it allows me to easily plan for future moves. Yet, while H5 does seem to operate on a “rounds, turns” sort of structure (i.e., battle is separated into rounds, in which each units gets a turn, ordered by unit speed), the separation of one round from the next is not really marked very well. Thus it is hard for me to determine when one round begins and one round ends, and sometimes it seems like some units get multiple attacks per round. So I’m not sure if there are even rounds at all. Which makes it sort of hard for me to plan out my actions in advance, because I’m not sure when an enemy unit’s retaliation is reset. Also the exact effect of haste and slow spells, as well as morale bonuses, isn’t quite clear. Maybe I missed that explanation somewhere. Speaking of morale bonuses, Nival decided to go sort of H4-style on this one, so that morale bonuses no longer give you a second immediate action; they give you another one sooner than you would normally have. The end result is that, sort of like H4, combat in H5 is sort of an unstructured mush that I like less than the rigorous pen-and-paper RPG-like “round, turn” structure of H2 and H3 combat. At least the initiative bar gives you some idea about how that mush will proceed, though, a nice feature we lacked in H4.

Another change is that, as opposed to H2 and H3, in which your hero could act (i.e., cast a spell) once at any time one of your units has a turn during a given round, in H5 your hero actually has a turn just like every other unit, and must cast his/her spell at that time. This seemingly minor change actually alters the impact of Heroes on battles quite significantly, and increases tactical depth by forcing you to more carefully consider your spell selection. It’s taken me some getting used to. Also, of course I knew that Heroes would not be on the battlefield as they were in H4, but now heroes have an option beyond casting a spell; they can also attack enemy units. This is nice for might heroes, as it gives them something to do, but it also sort of renders some of the weak spells like magic arrow a little useless, as your hero’s attack often does as much damage as the spell. Still, let me emphasize that I do not miss having Heroes on the battlefield AT ALL. At least that major balance issue no longer seems to be with us in H5.

Finally, I have noticed that siege battles seem to have undergone a subtle, but significant, change in difficulty. Assaulting a fully built castle seems to be much harder than in earlier Heroes games, particularly for armies lacking fliers or shooters. Without artillery skill, my catapult is next to useless, and some of the “moat” effects are very harmful. I still need some more exposure to siege warfare, but that’s my initial impression, anyway.

Given the creature skills mentioned earlier, I find the battles to be rife with tactical options. Because of the adventure AI’s weaknesses (see below), the number of good battles with enemy heroes that I’ve had has been small, except for a few instances against stationary Heroes (such as against the final Sylvan hero in the penultimate Inferno campaign map). Also, I miss the random stacks of mixed creature types from H4, as fighting random stacks composed of single creature types is repetitive and boring, and these battles are tactically simple. I haven’t talked about the magic system much, mostly because I haven’t been exposed much to it yet, but the spells so far seem to be rather limited, so I don’t expect them to really enhance the tactical experience. I guess my ultimate conclusion here from what little I’ve played is that, quirks and nitpicks aside, the battle system in H5 has the potential for depth, but some other gameplay features sort of short-circuit that potential by limiting the situations where good battles arise.

  1. Adventure AI, IQ 75.

Heroes 4, despite its poor implementation of some radical changes to the classic HoMM formula, might nevertheless have turned out to be a game for the ages if it hadn’t been for the disastrous adventure AI. For lack of better words, it was as dumb as a doornail. Dumber, if that’s possible. I remember countless maps where, having spent weeks building up my army, I would march into enemy territory to find an AI-controlled faction that hadn’t even bothered to flag its own mines or recruit creatures from its undeveloped towns. Other times, a stray AI hero would march into my own territory while I was away plundering the countryside, and blissfully ignore my undefended castles and creature dwellings. I’ve heard it said that the Equilibris mod makes the AI a little more aggressive, and others claim that a crafty mapmaker, by using complicated scripts, can make a fun H4 battle-map. Whether or not that’s true, the AI’s strange, noncompetitive behavior really ruined the single-player H4 experience for me. After the thrill of building an uber-hero wore off, what else was there? A strategy game with no opponent isn’t much of a strategy game. The H2 and H3 AI were much more functional, even if they followed predictable patterns, and a stiff challenge could be had in these earlier HoMM games if the mapmaker merely loaded the AI up with a significant amount of starting resources. Thus it was clear to me early on that, if H5 was to be a success, much work had to be put into writing an artificial intelligence capable of making life difficult for the player.

I will admit to feeling a little trepidation when I saw my first enemy hero in H5 march across the screen, and I will also admit to feeling a little bit of a sinking feeling in my stomach when that hero marched right past two undefended mines and right back into the blackness. A decidedly H4-like maneuver, that was. I continue to be a little saddened by how timid and confused the H5 adventure AI seems to be. I do not feel it is quite as bad as the H4 AI, which never seemed to have a clue at all, but it still seems very incapable of prioritizing its actions on any given turn. I have seen numerous examples of it failing to recognize undefended castles and mines, and it only bothers to attack me if the hero in question is trapped or if it has an overwhelming advantage in numbers. Even then, I have to be less than a day’s march from the hero for it to bother, which leads me to believe that the AI does not “think” more than a day in advance. Its priority seems to be to secure the area immediately surrounding its base-town (which I guess is more than I can say for the H4 AI) and then patrol that area. It doesn’t seem to do a lot of exploring, and when it does, it doesn’t move with much of a purpose. It doesn’t do a good job of amassing an army either, and I have seen countless high level enemy heroes with armies that would be barely suited for castle defense. So… I don’t know. For me H5 is still very enjoyable, but eventually the “new game” thrill will wear off, and I will be left with a strategy game whose quality and longevity will ultimately rest on this AI, which to me seems to have sub-standard intelligence. I am worried. Very worried.

  1. Tilt.

For my last “impression”, I will just spend a moment on my global perceptions of the game. I might have come of sounding a bit negative at times, and I do think there are weak spots in the game and places where concern is justified. I also miss some of the features of H4 (there are some I don’t miss also). For instance, I wish I could send a “creature only” army onto the battlefield. And the caravans, of course, but I understand they have returned in the expansion.

Nevertheless, overall I love what little of H5 I have experienced. I am happy that it seems to be more than just a graphically enhanced version of H3, although I would have been fine with that to be honest. The skill system is different, at least in the available options, and the battles seem to have a lot of tactical depth offered by the diverse creature skills. Some other subtle changes like the hero’s turn during battles greatly impact the way the battles play out. Maybe my overall positive feelings about this game despite some acknowledged problems derive from its newness. Obviously, my impressions may change over time as the initial thrill of discovery wears off. I fully understand that some of these downsides may in the end limit the game’s long-lasting appeal. Then again, maybe future patches will fix these problems.

One thing however which I have omitted completely from this discussion but which is possibly more important than everything else is the map editor. I have yet to open it but I have heard whispers on the wind that it is anything but intuitive. If H4 showed us anything, it is that no matter how poor the AI is, or how poorly implemented some gameplay features are, the game will find die-hard fans and, more importantly, longevity, if (and only if) there is a dedicated body of mapmakers out there to churn out new adventures. H4 may have had its problems, but its editor was reasonably accessible (though not as simple as those which accompanied earlier HoMM games) and some creative mapmakers were able to create a lot of gems, giving H4 a long life, certainly longer than you would reasonably expect from such crippled software. If it is true that the map editor for H5 is as opaque as I’ve heard, then H5’s future is uncertain, no matter what its other merits are. One of most endearing features of H2 and H3 was that their editors were extraordinarily simple, making it easy for just about anyone to turn out a reasonably complex map or campaign. New H3 maps are STILL being generated. I had really hoped that the creators of H5 would have understood how important the editor was to its success. It’s not too late.

I will reserve judgment of where I think H5 fits among its predecessors in terms of overall quality. As there is still another expansion on the way that appears to be quite ambitious, I think it is premature for anyone to make that final call. However, from what I’ve seen so far, the designers did what was necessary for them to do: they have created an attractive, captivating game which has breathed new life into the franchise. The gameplay is not perfect, but it is solid. There are lots of things to see and do. It’s pretty and modern. Though maligned by some people, it is nevertheless evident that now, a year past its initial release, HoMM5’s various gameplay elements are still hotly discussed and the game still is being played by many people. That’s no small feat, and that is also good news, as it means that Ubisoft should have a continued interest in creating new software that bears the HoMM brand name, which is, at the most fundamental level, what I really wanted most from Heroes of Might and Magic V.