Heroes of Might and Magic

[Deceptus]

And a Wizard castle does better, with more shooter damage, boars for equally fast scouts and equal fighting against anything other than fast shooters, and one hired warlock for a few gargoyles to block shooters.  Then the wizard gets better with rocs over griffins and mages over hydras.  The warlock only catches up in power with minotaur kings and dragons, which is past the effective timeframe to speedrun most standard maps.  Sure, the warlock does fine in Archibald's campaign, but you're not getting choices there between warlock and wizard.

Many of the maps for the Roland campaign, where you generally choose between Wizard, Knight and Sorceress, actually prefer the Sorceress town due to guaranteed Navigation (very good in Mission 2, for instance) and generally fast unit speed. How do you feel about Sorceress for speedrunning? Even without the HoMM3-style Wait command, I generally like Sprite powerstacks.

Yeah, it has sounded a bit antagonistic, but reconciliation accepted.

I will work on that - I let my enthusiasm for Heroes get the better of me, and I certainly wanted to come off as inviting rather than confrontational. Sorry again!

[T-hawk]

I actually never played the campaigns much, I think twice each at most. They were already fairly well solved by the time I got into the game a couple years after its release. And they were so driven by the dwarf/ogre alliance rewards rather than properly building up your economy. Every map was like, restart until the nearest monster stacks are all dwarves/ogres. Maybe I'd embrace such singleminded exploitation more nowadays than I did back then.

I don't like the Sorceress because it seems it's never possible to keep the sprites from dying. You have no other flyer that will attract the AI's attention until phoenixes. At best the AI is attacking the druids instead, but that's not good either. Of course I see where it's useful if you need Navigation, and also sometimes the sorceress can be the best situated to reach a killer tier-six troop in week one, sometimes maps give out enough mercury and gold to manage that. But without those situations I'd always pick Wizard or Knight instead.

I hated that Wait command in HOMM 3. Felt like it made every battle quadratically more complicated for little real strategic payoff.

[Deceptus]

I hated that Wait command in HOMM 3.  Felt like it made every battle quadratically more complicated for little real strategic payoff.

The order of how things progress after the Wait command is applied is a little bit confusing, I agree. It does also lend itself to heavy micromanagement styles, such as hit'n'running with fast, no-retal units and using Wait to dance around stacks.

Have you considered doing your own playthroughs of well-liked community maps from either game? HoMM3 has plenty of good, challenging maps like Unleashing the Bloodthirsty - though they're also heavy on large-scale combat management, which I'm getting the vibe you're not a big fan of. I did read a very nice report from a fast playthrough of The Empire of the World II, available over here.

Oh, and BTW, do you have any source for the HoMM2 speedrun times? I'd like to do some research here and there, and I can't find any well-documented information through simple Google-Fu. I wouldn't be surprised if the community hosting such competitions were a bit secluded, or perhaps not English-speaking.

[T-hawk]

Yeah, I specifically remember having a single Devil troop that was able to wait, hit without retaliation, and run, to kill non-shooters of any arbitrary amount under a certain speed.  Micromanagement is the whole point of the tactical battles, but that wasn't strategy, just an hourlong grind.  Large-scale combat management is great in HOMM 2 but became bigger than I wanted to do in 3.  The complexity increases quadratically: 14 stacks in a HOMM 3 battle is twice as many possible interactions as 10 stacks in HOMM 2, and the Wait command increases their possible orderings quadratically as well.  Also I continually screwed up the speed bonuses for favored-terrain and hero specialties, which became more frustrating than fun.  HOMM 3 was a kitchen-sink design that threw in every fan suggestion without thinking about whether each rule was pulling its weight.

I don't have any desire to play HOMM 3 now.  I'd love a game about halfway between 2 and 3 in complexity.  I haven't looked into any of the later HOMM games; they all seem to have significant negative opinions written about them.  I played a few community maps of HOMM 2 back in the day but just as a filthy casual and not optimized speedrunning.  I don't know where to find any audience other than the few people here; the few active forums for HOMM now all write about only 3 and not 2.

Finding references for the campaign speedruns: Problem is that it wasn't on the web back then.  It happened in two places: the "Statesman's Quill" email list run by astralwizard.com, and Usenet at comp.sys.ibm.pc.games.strategic .  The email list wasn't archived anywhere; bits and pieces were posted on astralwizard.com and still accessible through archive.org, but nothing comprehensive that I can find.  Usenet archives exist but searching them is a big pain.

I did find this, that isn't exhaustive on analyzing the times, but seems to be basically everything from astralwizard.com pasted together: http://the-spoiler.com/STRATEGY/New.worl...ic2.1.html

[TheMeInTeam]

Some of the HOMM 3 maps were nasty, I remember botching pestilence lake really badly before improving a bit. Now that I've rusted away skill after doing the video series it might be worth breaking HOMM 3 out again to do some of the maps.

I didn't play HOMM 2 nearly as much. The dwarf/ogre thing did seem pretty broken in its campaign, though spells in HOMM 3 get pretty stupid in the expansion campaigns. I had Gem summoning > 100 stacks of elementals with blind + resurrection in shadow of death, this beats anything the AI does as long as you get a single turn to cast the first elementals.

HOMM 5's setup is pretty decent. After that I've heard some pretty rough stuff.

[RFS-81]

It's been ages since I played any HOMM, and I never did speedrunning or multiplayer (except for very casual hot-seat games with my brother - only TBS game he'd ever touch), so not much for me to contribute here. They're also fun for casual play. Strangely, I like the graphics in 2 better than 3, especially the city screens

About other HOMM games: 4 did shake things up a lot. Creatures can now move on the strategic map without heroes and heroes move on the battlefield. It's like a mix of Heroes and Age of Wonders. You can't move units over long distances anymore by passing them on to another hero at the end of a hero's move. However, there's a building that lets you send caravans between cities that move units faster. The game also ramps up the complexity a lot, so I doubt you'd like it. Skills have now several sub-skills, and a hero's class changes depending on what skills he learns - though that may have been only cosmetic, I don't remember. One thing I found neat was that there were fewer unit tiers, but each faction has two unit types at every tier, and each castle can produce only one of them. But there you have a blowup of the number of possible interactions again...

HOMM5 was basically 3 with 3D graphics, I think. I don't know anything about the ones after.

[TheMeInTeam]

HOMM 5 initiative system was pretty different from round-based setup in HOMM 3. It has a lot of interesting ideas in its own right. Demon gating, using hero ability to spam actions on champions, mark of necromancer chains, etc. I actually like the way it handles magic schools better than the way HOMM 3 does, and its magic is less imbalanced w/o mods.

Still, it does carry a lot of the HOMM 3 feel.

[Jowy]

HoMM3 could use a dynamic turn order list in the battle UI. Kinda like FFX if anyone played that.

I never played HoMM2 so I don't know if this run is any good, but someone ran it at ESA this year:



The main thing I remember from HoMM5 is that it barely ran on my PC at the time when it came out.

[TheMeInTeam]

When I read speed run here I was thinking in terms of day/week/month optimization, not IRL time owing to the kinds of games usually discussed on this forum. I expect the strategies and skillset to do well in each to be pretty disparate! Both are interesting to me.

[Modo]

Day 4 "This is the last thing I would do at my castle in the game."

Yeap, this is T-Hawk playthrough alright. Always enjoyed this ability to move back and forth from present to future to arrange the steps in an optimized fashion.
Not that it's uncommon for strategy games in general but given the streamlined HOMM2 gameplay how much of this did you see before Day 1? I'm assuming you've had some contingency plans and implicit options set for different outcomes as well.