Interesting that this is on front page. I thought HOMM is quite niche right now. I'm still spending some time on HOMM3 and recently created mcp for Claude Code (with opus 5.5) to play it - you can look at short playthrough here:
https://www.youtube.com/live/uU-vvTOZFQY
It is not good at it! And very slow! I tried to improve, teach strategies, write some analysis scripts, etc. Did not help much.
I see a lot of people doing recreations of games (or simply new ones) but I've noticed that there's just some ick-factor (hard to explain), in addition to performance issues, and just plain old bugs.
It's amazing what can be built using vibe-coding, but it still feels far away from actually being able to create proper games with soul.
People have been re-creating old games for much longer than AI has been around. This project, fheroes2, started development over 7 years ago and published it's 1.0.0 release 4 years ago. So it's safe to say it's made by humans.
I'm not sure if meant to ask if this project is an ai-recreation. I just wanted to say that the fheroes2 project is much older than that. I was playing with it during covid (if I remember right).
I suppose it's possible that they have changed development practices recently, but I think you can probably enjoy it without guilt if that worries you!
Adding 2c, because I am building stuff with near zero previous game building experience.
I don't think you are that far off. It did get remarkably easy to get to a working prototype and get ideas you carried all those years in your head, but then you get to see how well they fare in reality. Worse, some combinations of those ideas don't mesh well with others.
Example, among other things, I am building a.. protest simulator based on Chicago riots ( not plopped anywhere.. it is still not balanced based on AI vs AI results so far). The basic location, sprites, system is in place, but the overall .. dunno.. there is something missing ( edit: though atm, violence is minimal by design and I wonder if that needs to be present even if it is rare ).
I guess what I am saying is: AI can you get pretty far with the initial build, but your guiding hand is still what counts for making it good.
I think a lot of it comes down to game design is hard.
There's a lot of moving parts that go into it, and even experienced teams have difficulty finding the fun. It's extremely common that games that are now popular and beloved weren't very fun until a few months before they released.
> It's extremely common that games that are now popular and beloved weren't very fun until a few months before they released.
I'm not sure that's so common, most games released that weren't fun before they were all completely finish, don't fare so well in the end. What tends to do well, is stuff that is fun/feels good early on, but might not be "visually/creatively complete" until just a few weeks before launch. That last part, is extremely common in game development, maybe you got the two confused.
In general terms, the machine can do the coding, you have to provide the soul. It's no different from the tons of human-made shovelware with no clear direction.
I can explain it. I spent six months trying to rewrite Bards Tale. For me it is pure nostalgia. I’ve played the game many times but rewriting it is another way to get play it. I think that’s it. Once you’ve sucked the marrow out of the bones, I guess we try to eat the bones.
Eventually my real job got too hectic and I shelved it and never returned. But I’m tempted to buy some Ai tokens and vibe transpile it with better 8bit graphics.
You’re not wrong but there are only so many game designers out there who have a distinct style and an art director who is a bit unhinged. Moreso when it’s a style that isn’t obviously trying to sell you something, which is unfortunately a pit many indie games fall into as they race to be a better version of what they saw other successful indie games do. It was already a saturated product before AI
The ick is the lack of passion for the game and making sure it is accurate and faithful.
Instead it is: hey I burned some tokens on this... I dodn't test it or give you any indication thereof, so it's probably a waste of your time, since it's legacy is well known enough, it'll get some attention.
I don't know what kind of answer you're looking for. There's no soul and games don't have them.
You have some kind of subjective feeling of nostalgia (or something else) and want to pin it on AI-assisted development. That's real for you, but not objective.
Okay, there's no soul. But clearly there's a massive difference between say, Monkey Island 1 and some remake sloppily vibe coded over a month. I just see a lot of that these days.
People taking the shortcut to build new games - and for what purpose, I wonder? To satisfy themselves or they really think that vibe coded games will be memorable for the next generations.
> People taking the shortcut to build new games - and for what purpose, I wonder?
It's a trend that started before the AI. Games went from niche to mass market. There're billions of devices - and potential customers. Mobile game publishers haven't been trying to make a "memorable" game for over a decade. It's all about A/B tests, retention, lifetime revenue, etc.
The same is true for many desktop studios. The recent Wolverine debacle is a prime example. Studios are making generic, "safe" games.
In summary, I think it's an industry problem that predates AI. The only difference between now and five years ago is that you don't have to rely on cheap human labor all that much to create a clone of a clone of a clone.
I know what you mean. Recreating the retro art style of Monkey Island is easy for AI but those little quirks in dialogue and comedic moments... Can an AI do that?
It's like, when I watch a movie and there's a married couple arguing or having a romantic date, it's believable. But if I spent a whole day with those two people still acting like they are married-- would I still be fooled into thinking that what they are doing and saying is the result of an organic relationship over years? I would hope not
HOMM3 is so alive that fan-make enhancements have not only released TWO new balanced factions, but they've been integrated into the official board game.
The Polish HoMM III Championship is happening right now, with last semi-finals match scheduled for 30/31 October, ahead of Grand Finals in Gdańsk 13-15 November.
May be niche but I actually bought HoMM2 last week on GOG to replace my aged CD, what a coincidence.
Heroes 2 in particular and Might and Magic 2 (Gates to Another World, the RPG blobber) are some of my favorite games of all time. I also discovered there was a Mega Drive (Genesis for you Americans) port of King's Bounty, the predecessor to HOMM.
I may want to try what you did too, I managed to have a local Qwen 3.8 instance play M&M2 fairly successfully using dosbox-x and MCP, so I may be able to have it play HOMM2 too. Thanks for the idea!
HOMM was niche when it came out. The genre was always petite and a few studios here and there is all that the market could ever support. This is why a well-executed game like HOMM 3 for example has such a high stickiness factor. The brand loyalty is almost built in given the type of audience it attracts.
The demo of HOMM2 came with a map Broken Alliance. Just the free demo was a top 5 of my all time favorite gaming experiences. I still play it every couple years.
I'm surprised that it's not very good. I'm working on an MCP for a Foss civ 5 reimplementation (unciv) and I find that 5.6 Luna is better than the built in AI, and costs about $2-4/player/game at API prices. I suspect Astra would blow humans out of the water, but I don't have $1k/player/game to spend on that lol
Maybe civ 5 is more reactionary than strategic, so simple strategy that doesn't look too much into future works reasonably well. Or maybe civ is more mainstream and models absorbed more knowledge about strategies that work reasonably well.
It's definitely not the first. I'm not convinced it's the second, either. I'm pretty sure if you Google for HOMM3 strategy, there should be enough hits! I guess what I'm trying to say is that you shouldn't give up :)
Also the speed is painful in the extreme. It's definitely not yet at the point where I can just play a game. But maybe if they get Astra on those cerebras chips and get the price down...
Interesting. I did the same but with Police Quest 1 (the AGI version). I vibe-coded an AGI interpreter which my harness could drive an had it run the game. I gave it a scan of the manual for initial information. It's silly how much it fumbles whereas 15 year old me could beat the game fairly quickly. https://youtu.be/JIs1QCcmQIY
I did manage to give it information via. screenshots, planning and several other things. Harness engineering is fun to do and a worthwhile exercise. It finally managed to score 93 of 245 points and then just choke out. I figured that a large language model would be able to burn through a 1987 primarily text based game with a primitive parser but it's not as easy as it sounds. My takeaway was that the LLM is definitely intelligent but not "generally" so.
VCMI is also available for Android. Mod support. New factions. Not that the depth or homm3 leaves anything to be desired, but it's really fun nonetheless to try new stuff :)
I also loved 3, "Songs of Conquest" has been an enjoyable spiritual successor for me recently even if the base game is quite limited on factions and a number of mechanics differ. I'm also keeping an eye on "Heroes of Might and Magic: Olden Era" when it eventually releases.
I know it'll sadly never happen under Ubisoft's watch, but I would absolutely kill for a new HoMM game that combined HoMM2's graphical aesthetic with the game mechanics from HoMM3.
There’s the mod homm3 succession wars that kinda do this. They recreate homm2 in the engine of 3, and thus also bring in a lot of the gameplay/qol improvements from 3 like the magic system and more tiers of creatures
I've actually never played Wesnoth, although I've heard good things about it. I want to say that combat plays out more like the Civ series rather than having a dedicated "battle screen" you'd see in HoMM - at least based on the gameplay footage I've seen [1].
OTOH, can never have enough turn-based strategy games though so I should give it a shot!
Wesnoth is great if you like games where you have to create and exploit situational advantages. Using advantageous terrain, the day/night cycle, and varied unit types makes it one of the most tactical games I’ve played.
Each time I recreate an old game I quickly realize that the gameplay which entertained me as a child/teenager is rather unbearable for me as an adult. HoMM is this type of game.
I still enjoy Morrowind. And I've replayed Ultima IV, although it does get grindy at one point. (For one of my play-throughs as a kid, I got the grindy part and simply boosted everybody to level 8 by editing the save file.)
But, as an adult, I don't really make the time for games anymore. I've been sitting on Baldur's Gate 3 (maybe an hour or two in) for years.
That's my point: as a child/teenager I could dive deep into those worlds, and fill the gaps in graphics/gameplay with my imagination, and just like yours my time was pretty much unlimited. Today I feel alergic towards any kind of game that feels like a job, and requires work and some effort.
HoMM from adult perspective is this type of game, it's not a hop-in hop-out type of experience, it's fully opposite.
I remember using a hex editor (Copy II+ Sector Editor?) to max out both characters and items in the Apple version of Ultima III & IV. The glory days of green and amber
It's fun reading about the differences they've made regarding the AI computer players, giving them a more modern/better tactics vs the also interesting way it was implemented before (computer players being always able to see the full map and have extra creature growth)
I don’t think you develop this kind of things to be useful and solve problems. It’s more of a fun hobby project/homage kind of things. Maybe not everything has to have a purpose yk
There is this HoMM3: The Succession War total conversion mod for HoMM3 in the style of Heroes 2 which happens to have new factions for what looks like Heroes of M&M II.
I've been working on a few of my own engines (which I'm also developing https://tanepiper.github.io/game-frame/ around - an agentic framework for game preservation)
Would be interesting to see how more accurate it could get with it, as this is along the lines of what I'm doing for POD: Planet of Death, Hardwar and Frontier: Elite II
Frontier!!! I will now write a small script that keeps track of your github to see when it lands. Good luck, I've spent countless hours as a kid playing and also modding the game in various ways.
Interesting that this is on front page. I thought HOMM is quite niche right now. I'm still spending some time on HOMM3 and recently created mcp for Claude Code (with opus 5.5) to play it - you can look at short playthrough here: https://www.youtube.com/live/uU-vvTOZFQY
It is not good at it! And very slow! I tried to improve, teach strategies, write some analysis scripts, etc. Did not help much.
I see a lot of people doing recreations of games (or simply new ones) but I've noticed that there's just some ick-factor (hard to explain), in addition to performance issues, and just plain old bugs.
It's amazing what can be built using vibe-coding, but it still feels far away from actually being able to create proper games with soul.
Or am I wrong?
People have been re-creating old games for much longer than AI has been around. This project, fheroes2, started development over 7 years ago and published it's 1.0.0 release 4 years ago. So it's safe to say it's made by humans.
I'm not sure if meant to ask if this project is an ai-recreation. I just wanted to say that the fheroes2 project is much older than that. I was playing with it during covid (if I remember right).
I suppose it's possible that they have changed development practices recently, but I think you can probably enjoy it without guilt if that worries you!
Adding 2c, because I am building stuff with near zero previous game building experience.
I don't think you are that far off. It did get remarkably easy to get to a working prototype and get ideas you carried all those years in your head, but then you get to see how well they fare in reality. Worse, some combinations of those ideas don't mesh well with others.
Example, among other things, I am building a.. protest simulator based on Chicago riots ( not plopped anywhere.. it is still not balanced based on AI vs AI results so far). The basic location, sprites, system is in place, but the overall .. dunno.. there is something missing ( edit: though atm, violence is minimal by design and I wonder if that needs to be present even if it is rare ).
I guess what I am saying is: AI can you get pretty far with the initial build, but your guiding hand is still what counts for making it good.
If anything, things are harder now.
I think a lot of it comes down to game design is hard.
There's a lot of moving parts that go into it, and even experienced teams have difficulty finding the fun. It's extremely common that games that are now popular and beloved weren't very fun until a few months before they released.
> It's extremely common that games that are now popular and beloved weren't very fun until a few months before they released.
I'm not sure that's so common, most games released that weren't fun before they were all completely finish, don't fare so well in the end. What tends to do well, is stuff that is fun/feels good early on, but might not be "visually/creatively complete" until just a few weeks before launch. That last part, is extremely common in game development, maybe you got the two confused.
In general terms, the machine can do the coding, you have to provide the soul. It's no different from the tons of human-made shovelware with no clear direction.
fheroes2 predates the era of vibed AI recreations which get abandoned in a week or two.
I can explain it. I spent six months trying to rewrite Bards Tale. For me it is pure nostalgia. I’ve played the game many times but rewriting it is another way to get play it. I think that’s it. Once you’ve sucked the marrow out of the bones, I guess we try to eat the bones.
Eventually my real job got too hectic and I shelved it and never returned. But I’m tempted to buy some Ai tokens and vibe transpile it with better 8bit graphics.
You’re not wrong but there are only so many game designers out there who have a distinct style and an art director who is a bit unhinged. Moreso when it’s a style that isn’t obviously trying to sell you something, which is unfortunately a pit many indie games fall into as they race to be a better version of what they saw other successful indie games do. It was already a saturated product before AI
The ick is the lack of passion for the game and making sure it is accurate and faithful.
Instead it is: hey I burned some tokens on this... I dodn't test it or give you any indication thereof, so it's probably a waste of your time, since it's legacy is well known enough, it'll get some attention.
Get used to it!
> Or am I wrong?
I don't know what kind of answer you're looking for. There's no soul and games don't have them.
You have some kind of subjective feeling of nostalgia (or something else) and want to pin it on AI-assisted development. That's real for you, but not objective.
Okay, there's no soul. But clearly there's a massive difference between say, Monkey Island 1 and some remake sloppily vibe coded over a month. I just see a lot of that these days.
People taking the shortcut to build new games - and for what purpose, I wonder? To satisfy themselves or they really think that vibe coded games will be memorable for the next generations.
> People taking the shortcut to build new games - and for what purpose, I wonder?
It's a trend that started before the AI. Games went from niche to mass market. There're billions of devices - and potential customers. Mobile game publishers haven't been trying to make a "memorable" game for over a decade. It's all about A/B tests, retention, lifetime revenue, etc.
The same is true for many desktop studios. The recent Wolverine debacle is a prime example. Studios are making generic, "safe" games.
In summary, I think it's an industry problem that predates AI. The only difference between now and five years ago is that you don't have to rely on cheap human labor all that much to create a clone of a clone of a clone.
I know what you mean. Recreating the retro art style of Monkey Island is easy for AI but those little quirks in dialogue and comedic moments... Can an AI do that?
It's like, when I watch a movie and there's a married couple arguing or having a romantic date, it's believable. But if I spent a whole day with those two people still acting like they are married-- would I still be fooled into thinking that what they are doing and saying is the result of an organic relationship over years? I would hope not
Weird times we are living in.
HOMM3 is so alive that fan-make enhancements have not only released TWO new balanced factions, but they've been integrated into the official board game.
https://archon-studio.com/homm3boardgame
https://h3hota.com/en/documentation
HotA also has a up-to-date VCMI port: https://vcmi.eu/Mod%20Repository/Expansion/Horn%20of%20the%2...
WOG is also available for VCMI: https://vcmi.eu/Mod%20Repository/Expansion/In%20The%20Wake%2...
I have tons of friends who know HoMM III, it’s the game of our youth
I think it’s also popular in eastern Europe
The Polish HoMM III Championship is happening right now, with last semi-finals match scheduled for 30/31 October, ahead of Grand Finals in Gdańsk 13-15 November.
h3.gg for anyone interested
May be niche but I actually bought HoMM2 last week on GOG to replace my aged CD, what a coincidence.
Heroes 2 in particular and Might and Magic 2 (Gates to Another World, the RPG blobber) are some of my favorite games of all time. I also discovered there was a Mega Drive (Genesis for you Americans) port of King's Bounty, the predecessor to HOMM.
I may want to try what you did too, I managed to have a local Qwen 3.8 instance play M&M2 fairly successfully using dosbox-x and MCP, so I may be able to have it play HOMM2 too. Thanks for the idea!
HOMM was niche when it came out. The genre was always petite and a few studios here and there is all that the market could ever support. This is why a well-executed game like HOMM 3 for example has such a high stickiness factor. The brand loyalty is almost built in given the type of audience it attracts.
I think HOMM2/3 is pretty well known to mid thirties.
The best games when you were around 10-12 years old.
They had a reveal for the remake on Gamescom Opening Night Live and it was pretty prominently marketed there as well
I just started live streaming another game - now played by Muse Spark 1.3 (much cheaper than Opus 5.5):
https://www.youtube.com/live/jRi9cdo95As
The demo of HOMM2 came with a map Broken Alliance. Just the free demo was a top 5 of my all time favorite gaming experiences. I still play it every couple years.
I'm surprised that it's not very good. I'm working on an MCP for a Foss civ 5 reimplementation (unciv) and I find that 5.6 Luna is better than the built in AI, and costs about $2-4/player/game at API prices. I suspect Astra would blow humans out of the water, but I don't have $1k/player/game to spend on that lol
Maybe civ 5 is more reactionary than strategic, so simple strategy that doesn't look too much into future works reasonably well. Or maybe civ is more mainstream and models absorbed more knowledge about strategies that work reasonably well.
It's definitely not the first. I'm not convinced it's the second, either. I'm pretty sure if you Google for HOMM3 strategy, there should be enough hits! I guess what I'm trying to say is that you shouldn't give up :)
Also the speed is painful in the extreme. It's definitely not yet at the point where I can just play a game. But maybe if they get Astra on those cerebras chips and get the price down...
Actually the AI in Civ games has always been known to be poor. Particularily after the 4th instalment.
Have you tried DeepSeek 4.1 Flash Pro? Should be cents per game. And probably 2x faster turns.
Interesting. I did the same but with Police Quest 1 (the AGI version). I vibe-coded an AGI interpreter which my harness could drive an had it run the game. I gave it a scan of the manual for initial information. It's silly how much it fumbles whereas 15 year old me could beat the game fairly quickly. https://youtu.be/JIs1QCcmQIY
I did manage to give it information via. screenshots, planning and several other things. Harness engineering is fun to do and a worthwhile exercise. It finally managed to score 93 of 245 points and then just choke out. I figured that a large language model would be able to burn through a 1987 primarily text based game with a primitive parser but it's not as easy as it sounds. My takeaway was that the LLM is definitely intelligent but not "generally" so.
There’s not one but two major HOMM games in late development right now.
I will check the video later, have you looked into decision models such as jev to improve performance?
Oh, I loved HOMAM 3 - and this made me google a bit and there is also an engine for it for Linux: https://github.com/vcmi/vcmi
For HOMAM3 there’s also the amazing HotA: https://h3hota.com/en/documentation
HotA also has a up-to-date VCMI port: https://vcmi.eu/Mod%20Repository/Expansion/Horn%20of%20the%2...
WOG is also available for VCMI: https://vcmi.eu/Mod%20Repository/Expansion/In%20The%20Wake%2...
VCMI is also available for Android. Mod support. New factions. Not that the depth or homm3 leaves anything to be desired, but it's really fun nonetheless to try new stuff :)
I also loved 3, "Songs of Conquest" has been an enjoyable spiritual successor for me recently even if the base game is quite limited on factions and a number of mechanics differ. I'm also keeping an eye on "Heroes of Might and Magic: Olden Era" when it eventually releases.
Oh yes. The number of hours burnt away by this...
I know it'll sadly never happen under Ubisoft's watch, but I would absolutely kill for a new HoMM game that combined HoMM2's graphical aesthetic with the game mechanics from HoMM3.
There’s the mod homm3 succession wars that kinda do this. They recreate homm2 in the engine of 3, and thus also bring in a lot of the gameplay/qol improvements from 3 like the magic system and more tiers of creatures
People have been praising https://en.wikipedia.org/wiki/Heroes_of_Might_and_Magic:_Old...
Released this year. Homm3 mechanics, aesthetic close to Homm2.
Could The Battle for Wesnoth fit into that description?
(https://www.wesnoth.org/)
I've actually never played Wesnoth, although I've heard good things about it. I want to say that combat plays out more like the Civ series rather than having a dedicated "battle screen" you'd see in HoMM - at least based on the gameplay footage I've seen [1].
OTOH, can never have enough turn-based strategy games though so I should give it a shot!
[1] - https://imgur.com/a/Kgqz2tx
Wesnoth is great if you like games where you have to create and exploit situational advantages. Using advantageous terrain, the day/night cycle, and varied unit types makes it one of the most tactical games I’ve played.
You are not alone!
Thank you (edit: and/or devs) for doing this! HOMM2 was amazing
Here is a great HoMM3 channel I've been enjoying, roughly one game per week with great commentary:
https://www.youtube.com/playlist?list=PLeEpTaU26pQgpMOAJZ-Eq...
Not quite HOMAM2 but.
I have such fond memories of HOMAM3 and playing it in hot seat mode as a kid.
Interestingly the game is still massively popular in ex-Warsaw Pact countries. Although I think it’s having a resurgence in the US too.
Each time I recreate an old game I quickly realize that the gameplay which entertained me as a child/teenager is rather unbearable for me as an adult. HoMM is this type of game.
I don't think this is universally true.
For instance, I still can play Masters of Orion 1. And simcity, to some extent; but specifically civilization 1.
Whereas those modern games with a 3D engine I can not play. It is always the same game. Just taking away my time.
I still enjoy Morrowind. And I've replayed Ultima IV, although it does get grindy at one point. (For one of my play-throughs as a kid, I got the grindy part and simply boosted everybody to level 8 by editing the save file.)
But, as an adult, I don't really make the time for games anymore. I've been sitting on Baldur's Gate 3 (maybe an hour or two in) for years.
That's my point: as a child/teenager I could dive deep into those worlds, and fill the gaps in graphics/gameplay with my imagination, and just like yours my time was pretty much unlimited. Today I feel alergic towards any kind of game that feels like a job, and requires work and some effort. HoMM from adult perspective is this type of game, it's not a hop-in hop-out type of experience, it's fully opposite.
I tried BG3 but found it strongly encourages savescumming and I got overwhelmed keeping track of the saved state tree
I remember using a hex editor (Copy II+ Sector Editor?) to max out both characters and items in the Apple version of Ultima III & IV. The glory days of green and amber
It's fun reading about the differences they've made regarding the AI computer players, giving them a more modern/better tactics vs the also interesting way it was implemented before (computer players being always able to see the full map and have extra creature growth)
Cheating AI players used to be the standard. Now they need to be hobbled.
Btw, there is HOMM Olden Era (google it) and it's great!
The added features look like a comprehensive back port of Heroes 3.
There's also VCMI [0] for Heroes 3!
[0] https://vcmi.eu/
It would be helpful to know how this is different than the other OSS rebuilds
I don’t think you develop this kind of things to be useful and solve problems. It’s more of a fun hobby project/homage kind of things. Maybe not everything has to have a purpose yk
What other OSS rebuilds?
I wish someone would create new faction.
There is this HoMM3: The Succession War total conversion mod for HoMM3 in the style of Heroes 2 which happens to have new factions for what looks like Heroes of M&M II.
I've looked into it before but it's just not on the same level of artistic quality and taste as the original.
If you are willing to consider Heroes 3, then Horn of Abyss has fan made factions that are as good as the originals.
how can i play multiplayer online?
https://heroes2online.com/heroes2/play
[flagged]
I've been working on a few of my own engines (which I'm also developing https://tanepiper.github.io/game-frame/ around - an agentic framework for game preservation)
Would be interesting to see how more accurate it could get with it, as this is along the lines of what I'm doing for POD: Planet of Death, Hardwar and Frontier: Elite II
Frontier!!! I will now write a small script that keeps track of your github to see when it lands. Good luck, I've spent countless hours as a kid playing and also modding the game in various ways.
[flagged]