Adventures In Heroic

[playforfun]

This is a really interesting thread. When I first read Winterfate's post I was agreeing with him 100% that heroic adventures were pretty impossible. Then, we have a lot of replies here from people who share their experience with being able to handle the heroic level with relative ease. Curiosity forces me to write this post. For a more meaningful discussion I'd like to appeal to everyone who reply here to share your playing habits and experience, most of which I consider crucial to our discussion of the real challenges from heroic difficulty:

Reload habit: How often do you reload? If you always/often reload, you can greatly improve your chances of winning with zero or minimal casualties. You can restart the game at the beginning if you are not satisfied with the random faction you’ve got or the random starting tier-1 stacks guarding your mills. You will also have extra scouting ability e.g. you go to one side of the map, and then reload, and go to the other side for a more rewarding trip.

The version you’re playing: Let’s concentrate our experience on the game with the latest patch (patch 1.4) where some of the more serious game exploits (phantom, raise dead, etc) have been fixed.

The map(s) you’re playing: obviously, in 1v1v1 maps, you get better chances at heroic level than a 1v1 map.

How long have you been playing HoMM series games: from the beginning since H1 or is H5 your first game?

If you don’t reload at all, you’ll probably lose the game after you lose an important battle on a 1v1 map. If you are playing patch 1.4, you have less bugs to exploit. If you’ve been playing HoMM for a long time, you know you have known almost all “tricks” to give yourself the best chance at beating the AI.

Here comes two most important questions:
1) How often do you win at heroic difficulty in a 1v1 map (pls include the map name) in this situation? 1 out of 2, 5, 10 or more?
2) How often do you think you SHOULD win in that situation?

[winterfate]

playforfun wrote:

This is a really interesting thread. When I first read Winterfate's post I was agreeing with him 100% that heroic adventures were pretty impossible.

Hello playforfun! . I still stand by that (it is beatable, but barely, and it's way too dependent on external factors like the map, the map size, mine guards, etc.).

Then, we have a lot of replies here from people who share their experience with being able to handle the heroic level with relative ease. Curiosity forces me to write this post. For a more meaningful discussion I'd like to appeal to everyone who reply here to share your playing habits and experience, most of which I consider crucial to our discussion of the real challenges from heroic difficulty:

Reload habit: How often do you reload? If you always/often reload, you can greatly improve your chances of winning with zero or minimal casualties. You can restart the game at the beginning if you are not satisfied with the random faction you’ve got or the random starting tier-1 stacks guarding your mills. You will also have extra scouting ability e.g. you go to one side of the map, and then reload, and go to the other side for a more rewarding trip.

The version you’re playing: Let’s concentrate our experience on the game with the latest patch (patch 1.4) where some of the more serious game exploits (phantom, raise dead, etc) have been fixed.

The map(s) you’re playing: obviously, in 1v1v1 maps, you get better chances at heroic level than a 1v1 map.

How long have you been playing HoMM series games: from the beginning since H1 or is H5 your first game?

If you don’t reload at all, you’ll probably lose the game after you lose an important battle on a 1v1 map. If you are playing patch 1.4, you have less bugs to exploit. If you’ve been playing HoMM for a long time, you know you have known almost all “tricks” to give yourself the best chance at beating the AI.

Here comes two most important questions:
1) How often do you win at heroic difficulty in a 1v1 map (pls include the map name) in this situation? 1 out of 2, 5, 10 or more?
2) How often do you think you SHOULD win in that situation?

Many, many questions...ok, let's see:

Reloading: After the first three weeks, I'm either dead or above water. Before that, I'll admit that I reload (playing perfect is impossible, and you have to because of the way the computer cheats on Heroic).

I'm still playing 1.3 myself (must...get...expansion...pack).

My two Heroic wins have been in Land of the Outcasts, disabling two of the computer opponents at startup (thus 1v1).

I've been playing since Heroes 2.

How often do I win ? Ugh...one out of ten...maybe.

Excuse me one moment...*winterfate looks up into the stars and spits out his Galactic Gargle, aiming at no particular direction*.
OK, thanks for waiting. (Too bad that bonus rank lasts one post ...thought it was neat.)

But, back on topic.

Gaidal Cain wrote:

Eh? The H3 AI sucked at picking skills. We're talking H4-strategic AI bad. The H5 AI seems a bit better, though that might be due to there being fewer totally worthless skills and abilities.

Alamar wrote

I'm pretty sure that both the H3 and H5 AIs [more or less] pick skills totally at random with only barely detectable logic behind their choices. I agree that the H5 AI looks better on the surface because there are fewer truly STINKER skills.

Just hiring high level AI-developed heroes is enough to prove the point ....

[L19 Necros with only basic light magic, Wizard with NO magic schools except Sorcery && Enlightenment, Warlocks without Destruction, etc.

*shudders at the thought Impossible could've been even more Impossible* And to think that all this time I thought the computer picked it's skills wisely at the higher difficulty levels in Heroes 3 (it's hard to tell when he's attacking your capitol with 5 times the army strength ).

[Jolly Joker]

H 3 picking was even worse because, believe it or not, the AI had the crappy skills actually preferred while the good skills - especially the magic skills were last on the list. You'd see the Might heroes of the AI with Scouting, War Machines, Ballistics, Pathfinding and so on and the magic heroes with First Aid, Sorcery, Mysticism, Eagle Eye and so on. Which was one of the reasons why the AI is such a cake walk: ever seen the AI actually cast a decent mass spell on any normal mp map?

My playing habits are the following:
IN H 3 I always keep only the autosave (that means, the last turn only). I played exclusively on impossible.
In H 5 I have autosave option NOT activated and keep mostly ONE save only. I tend to save before and after large battles - mostly, anyway. Of course I had my only gane crash in a big battle I had not saved before. . With H 5 I'm mixing difficulty levels. I like to play hard on regular MP maps, but heroic on the single player maps.
I DO reload, but only when I want to try something out - outside of the normal game - if you want to, let's say what kind of Elementals would I get here if I summoned them, or can I beat a certain type of creature with a new tactic I'd like to try and so on.
In the beginning I did reload quite often. Now I don't. I know you can trust the AI to check what you are doing, and if you give it an opening it makes use of it.

[azzy]

I have been playing heroes since HoMM 2 (but also tried HoMM 1 and Kings Bounty)

Now I play version 2.0 (expansion)
The first try against AI on Heroic was on v.1.0 with Necro. I dont remember the map, but it was large allied map. I wanted to have AI as an ally so I could see how he develops and study better its style of playing - it was a relatively easy win, because I had 3k skellies at the end and the allied AI took its time with the other team while I was preparing to rock (the last battle though was pretty tough and I wouldn't have won it with the current status of Raise dead spell).

Since than I have been playing almost entirely on Heroic (and sometimes Hard when I play allied with my girlfriend against AI) The only challanges are 1vs1 on small maps (like dragon's cape, war of the pauperers, land of outcasts, etc.) The first 2 times 1vs1 the AI beat me, but I was persistant. The first time I won was with Sylvan on the pauperers map against dungeon. Than I tried with Castle on land of outcasts and did it again... Since than my wins have been on the rate of 5:1 for me (i've played 30+ games but here I exclude my tries with deamons which had disasterous results I got my deamon ass beaten up every time, I am lame with deamons I guess...)

I play with all other factions, incuding dwarves, and my tactics go like this:
1. the first two weeks I rely almost entirely on magic to creep (doesnt count for Heaven because of the archers... train peasants and upgrade archers => 100+ crosbowmen in week 2!!!), while trying to minimize the casualties. I use quick save before most battles so if something goes wrong I can reload. In the first week focus on MageGuild and 2 units to build, see what magics you have => develop your hero (take exp from chests always before you reach lvl 20)

2. weeks from 3 till the AI arrives - develop hero, creep, collect useful artifacts, FLAG ALL MINES AS SOON AS POSSIBLE (by the middle of week 3 you should have all of them, but that is the deadline). Nasty creeps are druids, magi and priests, but by the end of week 2/mid 3, you should have enough spells and army to overcome them with minimal casualties. Focus on two more strong units to build as you will not have the resources for all (i.e. sylvan: 1st week - sprites and hunters; 2nd week druids and treants + later on dragons or unicorns, but they are just bonus the other are the core of your army)

3. when the AI comes (usually week 5) it will have an army in general 2 times stronger than you. You will have advantage in skills and magic. Use your other heroes (I usually have 4) to distract the AI and lure it away from your castle untill the next week. Go back with main hero collect creatures (this makes 2 weekly polulations advantage for you) even thaugh your army is still much weaker than the AI you have huge advantage in skills and magic and of course tactics on the battle field. By that time you should win the battle (although pretty devastated and with minimal army left), but at least you have to fght from now on against low lvl AI opponents with no artifacts, and you become a monster.

The above is not very specific, but it is just a frame I use to defeat AI and it is not that difficult at all. Just needs a little practice, preparation and patience :p

the easiest faction to play on heroic for me is Wizzard, than Necro, Dwarf, Dungeon, Castle, Sylvan, Inferno

ps.
last night I had a nice game on dragon's cape, when the AI rushed in the cape in week 4 and took the utopia, that's when i seised my chance and pursued him with my 4 titans 33 magi 300+ gremlins and 19 ginn. It had 2 black dragons left, 20+ hidras, 20+ matriarchs, 30+ minotaur, 17 riders, 25 furys, 70+ assasins (and tons of artifacts incl. immune to implosion, -2 luck for opponent, dragon boots, 2att 2def ring etc. and implosion and armageddon) It was a tough battle but i came out with 100+ gremlins ans 15-20 magi, which was cool given the artifacts I recieved. Than I rushed in and summoned my next population from town and dealth with the AI fairly easy.
My point is that you also have to carefully watch what the AI does and utilize best your opportunities. If i missed the hero after he took the utopia and he went back to his castle for another population, i would have lost the game. So good luck!!!

[Mytical]

Yeah in a campain I found out how silly the AI can really be (in heroic even). I had two heroes, and had to split my army ammong the two. So I went out and creeped almost the entire map without seeing the AI. Well both of my heroes had Town Portal (and summon creature but that is not material). So as soon as I seen the enemy was getting close I town portaled back to my castle and combined forces. Now he still was much bigger then me. So I sent my hero without any forces running (if you loose him the campain is over). The AI obligingly went for that hero while my other one stayed in the town (and the protection it afforded). I would lead the AI by the nose with my 1 creature hero and then just Town Portal back to town and start all over. The AI just could not seem to grasp what I was doing and chased this guy for 2-3 weeks! By that time I had an army that was actually more sizable then his. So on the next pass my hero that was in town waited till he was close and popped out and grabbed him. It wasn't even a contest by then. I still had 2/3 of my creatures when the battle was over. Now had he attacked the town instead of chasing my hero with one creature (who he could never catch regardless because that hero had boots of the open road, expert logistics, and pathfinding) he would have defeated me easily (even with the towers).

Was it underhanded of me to use this tactics, absolutely. Should it have worked, never in a million years. Is it a cheat? Hey think of it the way you want, it worked, and it shouldn't have. Just goes to show you, the AI really needs improved.

[DaemianLucifer]

I dont find satisfaction in exploiting the game mechanics so I can win,and having an AI thats a chalenge only because it cheats,so I almost never play the highest difficulty,in any game(plus the thief thaught me never to try it again).As for hero games,I only play story driven maps now,because great stories remove the feeling of the game beeing too easy.If I want a chalenge against the AI,I play chess(or gal civ now).As for reloading,I do that in campaigns when I power game,but in single player maps I use only autosaves(even in HV I use autosaves)because of the crashes.

[Jolly Joker]

I don't understand the problem here. The campaigns are scripted. The AI is scripted and has a certain task, in this case a very good opposing hero has the task: hunt down Raelag (or whoever it was) and kill him for good. Easy, simple and in terms of the story understandable.
That has nothing to do with exploiting game mechanics. In fact it's an often used ploy in every kind of story that someone wanted is acting as a decoy to occupy the opposition, while the rest of whatever rebel or whatever else force is busily laying the foundations for winning the war.
Traditionally the henchmen hunting the wanted guys are not the brightest examples of their race, so there is nothing wrong with those things in a campaign.

[aulfgar]

So its ok for the AI to be so predictable? The ability to accurately predict what your opponent will do EVERY SINGLE TIME seems rather silly.

Don't try to defend the AI it is some of the worst AI programming I have ever seen. When the only way to even let the AI have a chance is massive obvious cheats and it still loses due to stupid choice trees is pretty lame.

[Jolly Joker]

A movie is predictable, right?
So is a campaign AI because things enfold like a movie.

[DaemianLucifer]

A movie is predictable, right?

Oh yes,fight club was so predictable,and so was memento,and the dogma,and starship troopers,and devils adocate,and the godfather,.....You want me to go on?

[Elvin]

(...)

[winterfate]

Jolly Joker wrote:

A movie is predictable, right?
So is a campaign AI because things enfold like a movie.

Depends on the movie, and I've yet to see an excellent movie that's predictable. Just when you think you know what's going on, the movie throws you a plot twist.

As for the campaign AI, um...not really, IMO.
The story was a nearly direct rip from WarCraft 3, but that's beside the point.

[DaemianLucifer]

Depends on the movie, and I've yet to see an excellent movie that's predictable.

Hmmm...Try naked gun or hot shots then.But Im not sure parodies can count as predictable...How about der untertag?An excelent movie,even though everything seen there is well known and documented.

[winterfate]

*casts Resurrection on the thread*

There we go! Much better.

Ok, it is just me or do Necromancers make heroic a bit TOO easy? I beat No Man's Land with necro today...those skeleton archers, combined with Dark Magic hindering your enemy are deadly.

[DaemianLucifer]

What version?Try necro on pre 1.3 and you are almost inbeatable.Master of life+vlad+mark of the necromancer=unlimited skeleton archers.And necro is very map dependant.If neutrals are scarce,that means youll fall behind.

The main flaw of necromancy imo is that it is only HP dependable.Killing 10 titans means you get 10 skeletons,killing 100 peasants means you get 40 skeletons,but killing 10 titans and 100 peasants means you get 110 skeletons.Did they ever change this btw?

[winterfate]

I'm playing on 1.3. My main hero was Orson (the Zombie Lord).

The main flaw of necromancy imo is that it is only HP dependable.Killing 10 titans means you get 10 skeletons,killing 100 peasants means you get 40 skeletons,but killing 10 titans and 100 peasants means you get 110 skeletons.

I agree. But how would they rebalance it?

Did they ever change this btw?

As far as I know, not yet.

[DaemianLucifer]

I agree. But how would they rebalance it?

Same way as before:Raise a percent of both creatures and hp.So from 10 titans youd raise 4 skeletons,not 10.And from 100 peasants youd raise 40 peasants,but from 10 peasants youd raise 2,maybe 3.

[winterfate]

Oh ok, I see .

Have you beaten a map on Heroic?

[DaemianLucifer]

I didnt even play any of the SP maps because I got disappointed with the campaigns.Maybe when the second expansion comes,who knows.Untill then,Im still playing HIII and HIV,civ and gal civ.

[Gaidal Cain]

The main flaw of necromancy imo is that it is only HP dependable.Killing 10 titans means you get 10 skeletons,killing 100 peasants means you get 40 skeletons,but killing 10 titans and 100 peasants means you get 110 skeletons.

I agree. But how would they rebalance it?

Like they had it in H3, perhaps? I.e. calculate raised creatures from Hp and numbers separately, and raise whichever is lowest. Or calculate it independently for each creature type.