They include many tips for making a map more enjoyable. Many tips are basic, some not-so-basic.

I encourage anyone who wants to make a map- whether you are new to it, or a master map designer- to read through these tips to see if any of them can help you. Although some true masters of design may not need the more basic tips, I think that even they might find an interesting aspect of the game mentioned.

Sometimes this can work, but not in all cases. For example, duplicating an event that includes an ambush. The problem is that without any variety in ambush events, the battles become more monotonous than interesting. Copying objects is usually better done when you want to balance a map, and so want each castle to have the same spell restrictions, or starting buildings. Triggered events, in my opinion, should be more varied. You can use a template to copy, but at least vary the attached text, if not the game-influencing portion of the events.

It also uses the "unreachable necropolis with cover of darkness trick" to keep several areas in shroud. You need to be certain when using this trick that there really is a need for it, as a recurring shroud is often more annoying than anything else. Also, the AI seems to often have trouble remembering things inside a reshrouded area.

Finally, an apparent trap in the map (an area of cursed ground that can't be left without magic) has a cleverly hidden way out. You should download the map and try it out before looking in the editor to see how the trap was made. The link above is to the map file at the Astral Wizard website.

Even though to a casual observer it would appear that giving each player 2 level 6 dwellings is balanced, the town strengths change this. First and foremost, remember that the Dungeon has that portal of summoning. If a player has only one type of dwelling flagged, the portal will ALWAYS produce that kind of troop. Until other areas are invaded (and even then, only if the dwellings there are captured), the dungeon player will get an extra 2 manticores per week in every town that has a portal built - a great advantage in the early game. Also, if you provide stables on the map, remember that they provide free upgrades to the Castle's level 6 troop (cavaliers become champions). The other towns have no equivelent bonus.

Second, different factions have advantages at different levels - although it is hard to reach a consensus on this. A few of the more important considerations are: At level 2, the slow walkers (dwarves, walking dead) are generally not worth the cost in speed, and should be noted as such in balancing. At level 3, elves and serpent flies are core units for their towns, particularly if they are close to town, and so can be upgraded. Level 4 poses the first really important consideration: Vampires. Except against the Conflux, vampire lords can dominate almost any other stack once they get to around 50 or so (particularly if the hero has counterstrike). Providing the necromancer with increased vampire growth is a tremendous benefit - 13 vampires per week (4 from the dwelling, 9 from the town - assuming a castle is built) means that the critical stack size can be achieved in one month. Level 5 isn't too bad, as most towns have good troops here. The lone necropolis shooter is here, as is the formidable mighty gorgon. In my opinion, only the Conflux really lacks at this level.

Finally, use level 7 dwellings VERY sparingly in general. There are two major notes here that are worth mentioning: first, the Tower absolutely LOVES level 7 dwellings - defeating 3 giants is very easy, and it increases titan growth. The only faction happier than Tower to see a level 7 dwelling is the Conflux - since the dwelling produces 2 firebirds each week (as opposed to only 1 level 7 from all other dwellings), it is a great improvement- especially since a firebird is NOT the weakest level 7.

Towns that produce their main requirement in a resource silo do not need their own mine, if they are not going to get a full set. Towns that produce wood and ore need the resource of the upgraded level 7. For the latter set, my list above says what they need. The other towns need their SECOND priority - rampart needs gems (for unicorns), tower needs crystal (for genies), dungeon needs mercury (for many upgrades), inferno needs sulfur (again, for upgrades). Note the pairings here - in cooperative play, the good factions support each-other, as do the evil factions. The Conflux, however has very minimal requirements beyond the mercury, and so this one probably does benefit most from a matching mine - although it can be argued that gems are more important. Also note that the other resources are needed to a lesser extent also (for example, gems for the labyrinth), but I do not think they are as important as the ones listed.

For variety, often random dwellings are used instead.
Random dwellings, especially when set to match particular towns, can create many problems when balancing a map. There are three cases to consider: fixed troop alignment, fixed level, and totally random.

If you set a fixed alignment, and allow the level to vary, you run the risk of one player getting a great dwelling- say a level 7- while the opponent gets a level 1 dwelling. Of course, if you limit it to one level you end up in the same situation as I discussed previously. It can usually be left to vary over a few levels (more level if lower powered) For instance, a dwelling of between levels 1 and 4 will give most factions a very desireable dwelling, a couple moderate ones, and one bad one in the potential choices. At higher levels, you most likely will want a small range (say, level 6 or 7). Be certain to consider the factions playing as well.

For a fixed level, then any position that is lucky enough to get a matching dwelling wil be happy. A possibility you may want to consider is to fix the player positions, and require the dwelling to be of a different alignment. Then, even if a good creature appears, it will cause a morale penalty.

First is the possibility of being ambushed by a patrolling enemy hero without warning. If the ambushing hero has a weak or moderate force, then there is no problem. Remember to guage the distance into the map when assessing likely hero strengths. If the ambush has too strong a force, especially in a later game position (say, at a point roughly 2 months in, you place a patrolling hero with 200 black dragons), the player may very well get annoyed or frustrated at having to back up weeks of game time, or even restarting.

There are three ways this could be remedied. First, the limits of the patrol area could be clearly marked, though this may hinder the storyline and may not be desired. A trivial example of this is placing signposts at the edge of the patrol distance. Something more likely to fit into your storyline would be a placed event, giving you the warning.
Second, a redwood observatory could be placed in such a location as to reveal the enemy hero without revealing unwanted map sections. This might take a little work to place properly, but would work.
Third, when nearing the patrol area, a well positioned hut/eye of the magi can serve the same purpose as the observatory. This is the easiest to control, but if you need the hut of the magi elsewhere, the option is not available.

First, is counterstrike, in conjunction with a necropolis player.
Once the skeleton stack gets large (as often happens), counterstrike becomes a very useful spell - especially against the AI. By walking skeletons to just outside a castle gate (to lure out monsters hiding inside), they are able to kill 4 stacks of enemy troops in one round of battle! If you have animate dead, this becomes even easier.
The larger problem is Vampire Lords. If you have a decent size stack of them, they will drain life from all those extra attacks, keeping their numbers at full strength.
Remember that expert counterstrike affects all of your stacks- and air magic is not a skill that most players would avoid (I seldom turn it down).
The reason I point this out is because necropolis/ counterstrike can eality make your well-planned major battle a simple matter. In most cases this won't occur, but it is good to note it.

Next is View Earth in a maze-map.
Most players don't care much for large, pointless mazes. However, a well-designed maze can be a lot of fun. Often the challenge in this kind of map is finding your way through before some major problem occurs. The question becomes moot if the player can view the whole map. One way to fix this problem is to make the passages twisted (and thus hard to follow on the map), but this increases the "annoyance" factor a lot. Restricting the spell (or, the earth magic skill- which also reduces the town portal problem) is a much easier thing to do.

First, in earlier versions of the game, there wasn't any way to restrict spells, artifacts, or heroes. You have several options to handle each of these.
For spells, you need to edit every town, and deactivate it. Then you need to go to spell shrines, and set them properly. Third, you need to remove pyramids from the map. There is no way to set their spells, and Dimension Door and Fly are both possible rewards. Next, you need to make certain you have no random scholars on the map. Finally, you need to isolate artifacts that can give your restricted spell in a secluded area that cannot be accessed, as well as heroes that start with them. This leads to the "prevent artifact note":
The random artifact generator won't give out artifacts a second time, until all of them are out there once. Try to keep the number of purely random artifacts low. Remember that in Shadow of death, some combination artifacts have components that can cast spells (such as the admiral's hat). A second method is to make the artifact the goal of a seer quest, and again lock the artifact away. If you have several such artifacts, set them in an event or pandora's box to conserve speace.
For heroes, place them in a jail somewhere. It isn't pretty, but it works.

If you have an editor that lets you prohibit artifacts from appearing, then that is the easiest thing to do. Here is a partial list of artifacts that you should prohibit from random appearance:
Combo artifacts:Angelic Alliance, Cloak of the Undead King, Elixir of Life, Armor of the Damned, Statue of Legion, Power of the Dragon Father, Titan's Thunder, Bow of the Sharpshooter, Wizard's Well, Cornucopia. The others are not generally unbalancing.
Assembly prevention- These artifacts may not be level 4, but you need to stop people from assembling combo artifacts by accident: Helm of Heavenly Enlightenment, any one of the necromancy boosting artifacts, any of the health-booting artifacts, any of the damned artifacts, Head of Legion, any of the dragon artifacts (I suggest the +1 morale/luck ring), any of the titan artifacts, any of the archery artifacts, any of the mana artifacts, any of the rare resource artifacts.
Other: Sword of Judgement, Lion's shield of Courage, Golden Bow, all rare-resource artifacts, all tomes of magic, Angel Wings, Boots of levitation, Armageddon's Blade. These artifacts can throw off balance too much- especially the ones that give 16+ statistic points. Yes, this severely limits the number of relic-class artifacts available, but most of them are overpowered for a random reward.

If you don't have the SoD expansion, then you cannot stop the combination artifacts from appearing- they are not even in the editor to be prevented. You then have two choices: either don't place any ARTs or ART4s on the map, or give a warning to players about them. In either case, you can prevent the assembly of a combination artifact, or any RoE artifact, by making it the goal of a seer quest. If you don't want the map validator to complain, place a pair of inaccessible pandora's box on the map next to each-other, and put the target artifacts inside. Use two so you don't get the error "the pandora's box at (x,y,z) is inaccessible".

First, it will tell you that the goal of a seer quest is not on the map. There are a few cases where it will not realize you have done so. One: it will not recognize the presence of all parts of a combination artifact. You will have to check manually that all the parts are there. It also may not realize an artifact is within an event. Note that the "find object" feature will not locate things inside an event, pandora's box, as a reward from a quest, or other similar thing. If you plan on using seer quests with combination artifacts, you are well advised to keep track of where the parts are.
Second, it marks as inaccessible any object that has no passable space adjacent to it. Use the passability feature to check this. If you want to make something inaccessible, it will report it a an error. You can either ignore it, or make a space next to the trigger square. remember that if you provide an empty space, a player can fly or dimension-door to that spot. Fill it with a resource pile to prevent that, or stick it out of viewing range in the underground.

Remember that a player can get resources as a starting bonus. If you want to drain all resources, you should remove more than just the amount you start with on 'easy.'. I usually just put a bunch of 9's in the fields (lose 99999 gold, 99 of each resource) It's easier than remembering how much of each resource you get at easy (although if you use drains often, you'll remember the number.
If you want to set a specific amount of each resource, drain everything, and then give some back. If you can make a reasonable reason for it, story-wise, so much the better. For example, event 1: Thieves break into your storehouse, taking all your wealth. Event 2: Fortunately, they missed your hidden stores.

• Decorations are often very tedious to add in- particularly in a large or huge map. The obstacle painting tool added in the later editors helps quite a bit here. Don't be afraid to place objects on a different terrain than the editor's categories state. Some mixtures look good.
• Blend terrains- don't have a line of one terrain against another, but rather put small splotches of of each one either side of the border. As with the previous point, sometimes an object from the decorations does well.
• I cannot stress this enough- and I have mentioned it elsewhere- make a day-1 event to describe how the situation arrived. Even on a battlemap.
• Customize hero names and histories. If any heroes are pre-placed on the map, then this adds a lot of flavor. Some craftier map designers may even put hints in there. If you are really ambitious, modify the histories of heroes in the hero pool also (although I would not expect that to be done)
• In a story map, the longer the story continues, the better. If you expect the typical player to finish the map in 3 months, then your story should take between 9 and 11 weeks to tell- assuming you are using mostly timed events. Have unimportant story events continue for at least a few weeks beyond the "expected" end.
• Story fragments are often better if they are included as site events, as you don't have to worry about the speed of the player to time them properly. Just be certain the AI cannot trigger pure story events.
• Remember that events can be attached to any artifact, resource, or monster stack as well as towns, sites and the calendar.

This problem is further compounded by external dwellings. Placing, for example, a level 7 dwelling near a starting town makes a tempting early conquest- but will likely involve very heavy losses if approached early. If a high-level dwelling is provided for free, however, then simply building a castle will give you the force you need to capture it. If the free dwelling happened to be a level 7 in the matching town, however, the player will have by day 8 a weekly production of 4 level 7 creatures (7 if they are pyres).

As if this weren't enough, this problem worsens even more if the map is intended for play at a high difficulty setting. At impossible, since the player gets no starting resources, a free high-level dwelling compensates for a lot of the difficulty that impossible creates. To a lesser extent, this can be said of expert and hard settings. For normal and easy, there are generally enough resources (both starting and lying around) that it will be only a couple days speed up if something is free.

Why do I say this? Because even though it is an obvious statement, it is very easy to say to yourself Hmmm... I never noticed that object before. Why not try it out?

Although you are most unlikely to find something like this in Heroes 2 or 3, since everyone knows about most of the features, there are several places where this trap can appear.

1. When a new game is released.
2. When a new expansion is released.
3. When a modifation comes out.

When a new game (say HoMM4) is released, many people, after playing a couple scenarios, will want to see what the new editor can do. This is good for personal use, but will likely make a poor map. You rarely want to throw anything in that doesn't have a specific purpose. How much worse it is if you throw in EVERYTHING!
More dangerous is the new expansion. For example, when Shadow of Death was released, I created a map containing almost everything from both the RoE release, and the AB and SoD expansions. Now, the purpose of this map was to be a replacement tutorial for the one that comes with the game, so it worked out well. However, if I billed the map as anything OTHER than a tutorial, it would have been a very bad map, at least in my opinion. As a tutorial, however, I think it does very well. If you want to check it out, use this link.
Finally, when a modification comes out, such as In the Wake of Gods, so much power is added into the editor that it is VERY easy to overdo things. While playing some of the early maps people have made for this expansion, I came into just this problem. There was so much stuff in there, that the maps became overwhelming.