I wanted to make my own version of HoMM myself one day, because I was really annoyed with the design flaws of the original and I doubt Van Caneghem will ever return to his games. But it is unlikely, that I will ever have a chance to make more than one game, so here is a proposition to improve the  core rules of HoMM type games:
1. Limit army size by hero's leadership ability, like in the original King's Bounty. Each unit would have leadership cost. That will favor using several heroes, instead of a single strong one. Therefore larger tactical depth on the strategic map.
2. Limit max city's unit population by player's controlled land, like i.e. monster layers. That will force players to fight over map structure.
3. Allow splitting unit stacks during battle (in particular during the beginning of a battle), while maintaining fixed 7 unit type slots. That would simplify strategic map interaction by moving it to tactical phase, and greatly improve the tactical element itself.
4. Allow building map structures, like fortifications in bottlenecks, like mountain passes. That ability should be limited by facilities provided by nearby city or other structures, to have a way to disable it in some maps and encourage more strategic uses of such feature, instead of just cluttering whole map with fortifications.
5. Use Final Fantasy Tactics style turn scheduling system, where fast creatures get more turns, than slower ones, with hint for user what creatures will move in the next few turns.
6. Make unit upgrades auto apply to all hired units to avoid forcing player to wait until upgrade structure is ready. Upgrade system should be a nice gimmick, instead of a way to annoy player or force him to wait.
7. Damage to units (and heroes, in case HoMM4 style heroes participating in battle) carry over to the next battle, unless healed. To emphasize use of healing structures, abilities, items and magic. It also make armies with high tier creatures less unstopable. A mandatory requirement for 1. There are several obvious way to implement stack splitting with damaged creatures.
8. Unit stack specific items, like say banners, giving that stack some bonuses.

One can probably easily implement these changes into say fheroes2 engine.

Sound like good improvements.

Altough I always hate games where the turn rotation is based on "speed" or similar. Its very hard to survay what will happen and how useful this stat is. Games normally generates a long line of "character turn order" that fills the screen. Really annoying!

Id rather have a simple, predictable turn order (best if all of a teams units move and then all of the enemy units move, to avoid hopping around of the game control) and let "fast" units (if you need that mechanic) get extra attacks/actions during their turn instead.

On 12/11/2018 at 4:49 PM, suliman said:

I always hate games where the turn rotation is based on "speed" or similar. Its very hard to survay what will happen and how useful this stat is.

That is why user interface must expose the queue of the units, which will act next. Original Heroes games used even less intuitive scheduling. But if game time is divided into discrete turns between players, where player moves all his units, during his turn, there will be much less tactical depth. Because fast creatures would stop offering first turn advantage and there will be no way to counter these pesky archers with your super fast phoenixes, while slow zombies at close ranger would attack just as fast as any other unit. I.e. it will stop being HoMM game, become more casual knock-off, with different target audience.

There is other ways to add tactics to a game. Im just saying i always dislike that feature when added to games. Many games dont use it and are better for it

On 12/14/2018 at 10:01 AM, suliman said:

There is other ways to add tactics to a game. Im just saying i always dislike that feature when added to games. Many games dont use it and are better for it

That is okay, not all people like complex games. But game design involves two fundamental concepts: gaming pieces and space, which are there even in card games. Time is part of the space concept. There are numerous ways to implement the notion of time, and velocities of objects moving through it. And the Final Fantasy Tactics style scheduling appears to be closest to model real time and without making game mechanics overly complex. In my game I'm using the "player moves all his units during his turn" mechanics, but I had to implement numerous complex hack to make fast units receive more actions per turn. In fact, the whole time mechanics takes several A4 pages to describe. Despite the original Final Fantasy Tactics having no multiplayer, its system actually favors multiplayer, because players don't have to wait long before having to move again, but play more like a chess exchanging smaller turns.

For me it's not about complexity but transparancy, but yeah if you like this mechanics you should have it in your game. I just dont like how it hops around the characters in an non-uniform manner.

21 hours ago, suliman said:

For me it's not about complexity but transparancy, but yeah if you like this mechanics you should have it in your game. I just dont like how it hops around the characters in an non-uniform manner.

Is there any other way to implement initiative? Say you have a fast unit, say a cat, while opponent has a turtle. Who would go first?

On 12/2/2018 at 1:16 AM, NikitaSadkov said:

1. Limit army size by hero's leadership ability, like in the original King's Bounty. Each unit would have leadership cost. That will favor using several heroes, instead of a single strong one. Therefore larger tactical depth on the strategic map.

Sounds to me like this would favor high leadership heroes over everything else, assuming a 1 hero per army limit.  And if leadership can be improved by leveling up your heroes, then it becomes all the more important to save all your leveling opportunities for your high leadership hero, to the point where all of your other heroes are more of a liability (because they use up leveling opportunities) than an asset.

12 minutes ago, a light breeze said:

Sounds to me like this would favor high leadership heroes over everything else, assuming a 1 hero per army limit.  And if leadership can be improved by leveling up your heroes, then it becomes all the more important to save all your leveling opportunities for your high leadership hero, to the point where all of your other heroes are more of a liability (because they use up leveling opportunities) than an asset.

There are a lot of ways to implement that. For example, limited gain per unit of investment. You can have 5 heroes with 1000 creatures each, but also with spells and some perks (like the one allowing rearranging units on the first turn), or single hero with 1500 creatures, without spells and skills. I.e. no single hero would be able to collect all creatures.

Ok. With everyone suddenly making remasters of older games, like Blizzard's Starcraft remaster. I found that Ubisoft has also remastered Heroes game. I expected them taking the original 3d models and rendering them in real-time for any resolution. But it was like WUT?!!! They just badly resized and edited old 3d rendered sprites for higher resolution. For example, look at the title screen archangel's sword fire animation:

In original flame looks like flame:
https://www.youtube.com/watch?v=GFbMLorQkIM

Original devs probably sampled real flame and put a lot of effort into editing it into game.

In a remaster we have an ugly out of place paintover, of a quality an 8 years old would do without using any reference.
https://www.youtube.com/watch?v=ymc6jaGyV7w

They also added some sparkles, and removed the muscles. Is that some cryptic message? And the icons got moved closer to center, leaving huge space on the right.

Is that what you get when people don't actually care about what they are doing?