Shockingly (to me at least) I've made thirty videos matching up one faction against another (31, actually, because I redid Castle vs Rampart, my first one, after I settled on a better format, materials and methods for the experiment.) Equally shocking, I've only got fifteen more to go. I've done two-thirds of the experiment.
I'm getting kind of tired of it too, though. It's only my commitment to see it all the way through that will in fact see me all the way through. It's a little frustrating sometimes to spend so much of my free time with a nearly twenty-five year old computer game running experiments rather than just chillin' and playin'. Once I'm done, I'm done and I'm going to do something totally different. In any case, here's the current status:
Castle, as I expected, performed admirably, and didn't lose a single match (closest it got was it's 6-4 win over Stronghold.) Rampart and Tower also performed quite well, only losing to Castle each, and to Cove in Tower's case. Dungeon has surprisingly, performed very poorly, only beating Necropolis, Stronghold and Conflux. The Neutral towns have been all over the map; sometimes Stronghold or Fortress will win a lot of battles within a match, and other times get completely blown out. Cove has (so far) performed quite well; nearly at the level of Rampart or Tower, perhaps, although we'll see as it gets more matches under its belt. Not only do I like the concept of the Cove, but it's nice to see that a late, unofficial add is balanced fairly well. It seems that there's a tendency to go overboard and make it weaker to not be seen as making it overpowered.
Talking about it has recharged my energy about the project, but I still might take a couple of days where I actually play the game normally rather than do something like this. We'll see. I am determined to finish this, especially noting that I've done two-thirds of the work already. The next matches will go down fairly fast. I'm already at the point, on Inferno if you note above, where I only have to do five sets to finish the match. Necropolis will only need four, Stronghold three (no reference to Sky High intended), Fortress two, Conflux one, and by the time we get to Cove, I literally don't have to do it; they'll already have fought everyone. In fact, I do the (now shortened) Inferno run, and I've done another third of everything that's left!
Because I have four factions who've now played everyone and are out of the action, it's maybe appropriate to review their record:
Castle: 9-1 No losses to anyone.
Rampart: 7-1-1 Only a loss to Castle, and a tie with Tower.
Tower: 6-2-1 A loss to Castle and Cove and a tie with Rampart
Dungeon: 3-7 The first really bad performance and overall loser. I'm really surprised and disappointed. I had thought Dungeon was one of the stronger players, or rather, one of the easier factions to play with relatively strong troops and a forgiving playstyle. Even though I admit that I've used them pretty infrequently, especially as a starting main.
Note that none of this means that any of the factions aren't viable. Although Dungeon performed poorly, the nature of this test is that it flattens any other variable, and many factions' success depends on strategic variables that are not represented at all in the experiment, which is just raw 1x1 unit power. For instance, Dungeon is a rather magic-heavy faction, with special magical bonuses in its castle, and the nature of the test completely eliminated magic from consideration, except in the cases where the faction's might hero came with a spell book and a modest spell (Tower and Necropolis.) It also has extraordinarily fragile 6th level troops, who the computer AI seems to want to throw into harm's way as quickly as possible because of their speed. They're not alone in this (Fortress has the same problem) but I'm just pointing out that there are certain parameters of the experiment which will not paint factions in the best light, because their best light is not in straight up combat troop strength, but is rather in a strategic assessment of resources. Necropolis is largely considered OP, for instance, but it is unlikely to win a single match, or maybe even a single combat. Their strength is in the Necromancy skill, not in creature growth in town. To use one obvious example.
Anyway, the experiment has been fun, but I'm ready to finish it up and move on to some other kind of HOMM3 enjoyment. I'll also switch to my alternative soundtrack at that point.
No comments:
Post a Comment