Posts from Walkthroughs for custom maps.

Don't know. I wasn't even thinking about companions, just pumping up everything possible in Karasucha. They were basically just there, without participating in many battles. And she somehow managed to complete the 4th mission and all subsequent ones quite easily. There's a need to be careful in the second-to-last mission, so that the last one doesn't turn into a months-long ordeal of accumulating dozens of skeletons.

Wayfarer

Thanks for the review!

It would be good to write about the battles: which ones were difficult and which ones were not.
For me, during my first playthrough, Nela was incredibly difficult, and she killed a lot of my troops. During my third playthrough, I killed her almost without any problems, just in the field.

Regarding the -3 morale before the source, in my opinion, it was done intentionally because the author wanted to show that the enemy should and can be defeated with the available mana. Besides, the trip to get this doubled mana bonus takes only a couple of days, and those who are aiming for a speedrun playthrough won't go there anyway. So, I wouldn't deduct points from the map for this; I think it was done correctly.

Regarding Yogue's skills: earth + water + fire + wisdom. Intelligence is also essential, as well as accuracy. The other two skills are logistics + pathfinding (again, for faster playthroughs). Ballistics are not needed; I didn't need them anywhere in the game. In some difficult battles, you can simply take an extra ranged unit (e.g., cyclops), place them near the ballista to burn down the enemy ranged units and the ballista, or burn them with dragons or kill them with magic. So, in my opinion, ballistics are not needed at all.

You can add to the walkthrough that this map is truly one of the best, even though it was made a long time ago. After completing it, the player's experience in playing custom maps will undoubtedly increase. So, this map should definitely be added to the collection of completed maps for many reasons. I played the map three times: in 12, 9, and the last time, 6 months. Since most of the troops are given for quests (and not through accumulation), it is possible to speedrun and gather enough troops for the final battles by the end.

In the second-to-last mission, you need to be careful so that the last one doesn't turn into a months-long ordeal with accumulating dozens of skeletons.

And in what way should caution be exercised? In the last mission, before the keymasters' tents, there's an antimagic barrier guarded by 7,000 gold dragons (a thousand in each stack). Troops from the previous mission don't carry over, and I don't see any way to obtain flight or a portal in this or previous missions. This guard doesn't use any tricks; that is, it will have to be broken through with brute force, specifically with tens of thousands of skeletons, yes. Or with a vampire archer. What other option is possible?

("Wayfarer")

It would be good to write about the battles: who was difficult to fight, and who wasn't.
For me, during my first playthrough, Nella was incredibly difficult, and she killed a lot of my troops.

I published the "journal version" of the report: I wrote it quite detailed for myself - with tips and a walkthrough, but I didn't post everything, I shortened it so as not to tire anyone and to give future players the opportunity to put in some effort in searching and fighting.
Perhaps, it was difficult to fight until I had resurrection, and later, until I had enough mana. Here are some of my notes:

The battle with Nella (I had to fight her in the field) was difficult: I didn't have enough mana to resurrect everyone, so I had to sacrifice some of my troops. I replayed the battle many times, but the best result I stopped at was losing 160 ogres and some other creatures.

I remembered something else: with Nella, there was a specific difficulty that prevented me from conducting the battle as I wanted and as it would be more convenient. She sensed that I was going to defeat her and tried to hide in the Sanctuary. But here, I had to defeat her in order to continue: the guardian required it. To prevent her from hiding, I had to fight "on her terms."
In general, the difficulty of the battles here is relative and depends heavily on the timing: it's harder to fight earlier, and easier later. Let's remember the guy from the "Heroes 3 combats" channel... :)

-3 morale before the source, IMHO, was done specifically because the author wanted to show...

That's why I gave it a dislike. Because it's not right. :( Or maybe it was accidental...

Regarding Yogu's skills: ... Ballistics is not needed; I didn't need it anywhere in the game.

Well, that's a matter of taste. In my opinion, it's much easier when you know for sure that you will have to storm many cities, to use the advantage of controlling the catapult, rather than trying to avoid the premature destruction of walls.
Oh, I remembered something else about ballistics: in fact, it has another valuable property. As long as the catapult is intact, during a siege, your spellcasting in each round is always first, regardless of the speed of creatures, acceleration, deceleration, prayer, speed artifacts, native land, and so on. This is often very important.

And now, in "Uprising," I've reached some seriously "harmful" enemies: for example, Adel on cursed land, Xeron with the Sphere of Inhibition, and the Armor of the Cursed.

In the second-to-last mission, you can learn a lot of useful spells in the castles (if you're lucky), which will be very helpful in the final mission. But it's not easy to do – an enraged dream demon can defeat the computer-controlled heroes before you even reach them)).
Otherwise, yes – vampires are the best in the final mission.

In the second-to-last mission, you can learn a lot of useful magic in the Castle (if you're lucky), which will be very helpful in the last one. But it's not easy to do – an enraged Dream Demon can wipe out your units before you even reach them)).
But yes, in the last mission, vampires are the best.

Yes, that's true. You can also buy heroes in the Castle of an ally when he starts to repel the Ramparts. We defeat Karasuchka using peasants or a box (there are no special defeat conditions here) and visit all the Castles at the same time as the ally. But playing this entire campaign without Eagle Eye is quite masochistic :D

Added 5 minutes later

That's why I gave it a dislike. Because it's not right. :( Or maybe it's just a coincidence...

This reminded me of the Rise of the Sun King map. I looked at it very carefully, and there is a similar trick in front of the outer tavern and the fort on the hill: each visit threatens the loss of a significant amount of money.

This author has another campaign; it was mentioned here some time ago. There is a serious sticking point in mission 3—you need to simultaneously coordinate the Sphere of Vulnerability in Dragon Utopia and Berserkers in the AI's cities, which are very difficult to take. I struggled for a couple of days and gave up

Hello everyone! I haven't been here in a while (I've been practicing offline strategies in the online game), and I've also been doing some modding. I read a few pages of this thread, but the map discussions are changing faster and faster. What's currently popular, what are we playing?

Added 5 minutes later
[/U]
This reminded me of the map Rise of the Sun King. I examined it very carefully, and there's a similar trick near the outer tavern and the fortress on the hill: each visit risks losing a significant amount of money.

In Xiedu, it's similar: if you use hit-and-run tactics and retreat, you can only recover in two places: either in the castle, where you must visit the pond and spend the night there (i.e., lose a day), or in the outer tavern on unexplored land, with the same consequences.

Link: http://www.maps4heroes.com/heroes3/maps.php?keywords=The+Uprising&type=&size=&difficulty=&sort=3

Genre: RPG.
The map language is English, but for sophisticated connoisseurs ;) there is a Chinese version.
Plot: An evil spirit has bewitched the King and the kingdom soon follows a path to destruction. Discontented rulers, outraged loyal generals, and restless population rise to challenge the despotic king. Who will be victorious and emerge as the new King?
The story is based on a Chinese legend about a mad king who lost his mind over a beauty possessed by a demon. As a result, the kingdom fell into decay, some dissenters were executed, some managed to escape, and many began planning to overthrow the ruler.
Even before starting the Walkthrough, it is necessary to read the text file with the backstory included in the archive to understand who is who and who wants to kill whom for what. It will also be very useful to read the text file with the author's instructions and tips, especially regarding artifacts, bugs/glitches, and a possible problem with one of the enemy heroes.
I must warn you right away: there is a lot of text in the map, and it must be read! The messages contain not so much plot as information for the player. If the need to read and translate such a volume of text tires you, choose another map for now.
Topographically, the map represents a sea with many islands of various sizes. "To keep the crucian from dozing," enemy "pikes" of small and large "fatness" periodically swim out into the sea.

The review reflects my personal opinion, which may not coincide with yours.

"Classic" style of execution, where you are not required to use any game bugs or convoluted tricks available only to pro-players.
For speedrun enthusiasts, a very fast completion is possible (I read about such achievements in the map discussion).
This map can be considered a manual for authors on how to instruct the player on what, where, and when they need to do. Instructions are given in small portions at the beginning of certain days and in designated places.
You can choose one of 4 colors and one of 5 races for play:
Red — Rampart,
Blue — Castle,
Brown — Dungeon,
Green — Stronghold or Fortress.
As the author assures, the forces of all these 4 players are carefully balanced, and you can even use the map for Multiplayer game, though you need to change the victory condition.
Despite the RPG nature of the map, you can use as many heroes as you like, and there will be work for all of them.
You can recruit defeated enemy heroes (and two are even highly desirable, which the author himself will tell you).
You can learn the necessary schools of magic in the Conflux without being dependent on chance.
Although the main stages of completion are predetermined and you will perform them in a specified order, you have a fairly large degree of freedom here.
The presence of "spare" passage guards in a couple of places prevents impassability if the player suddenly fails to defeat designated enemy heroes (they will flee or disappear).
The location of the Grail and the absence of obelisks* exclude the trouble of it being dug up by an enemy.
A pleasant and useful author's solution: "acceleration zones" (see tips).

Although the author immediately states which skills are needed for the main and secondary heroes, the player must ensure they obtain them. In our lands, there are several scholars, witch huts, a university, and the author writes that necessary schools of magic can be learned in the Conflux, but it is important to do everything so as not to get unnecessary skills while obtaining the needed ones.
Of the many quests, most are optional. Generally, this cannot be unequivocally categorized as a disadvantage: depending on your situation and preferred plan of action, the map allows different options for obtaining some important things. You can exchange certain artifacts for spells or access to additional dwellings or other artifacts. It is only important not to rush the exchange until you have made your plan. If expressed in numbers, out of 42 seer quests, only 4 are mandatory. Out of 87 passage guard quests, 9 are mandatory (some "guard + scholar" pairs replace seers, the number of whom on the map is limited). And there are implicitly implemented quests: when approaching a temple, you are warned "want to learn the property skill for 2000 gold — go further." Therefore, the quest part of the map is, conversely, an example of how not to do it: without a Prospector, the player will simply "drown" in optional quests, not understanding what they need and where to get it.
Although the map is carefully debugged and tested and the author tells about possible difficulties and errors, the player still has the opportunity to use a known (non-dangerous) bug to keep the Sphere of Vulnerability for the final battle, which, by the author's design, should take place without it and should be very difficult.
And while studying one player's save, I came to the discovery of another (but not at all obvious) oversight by the author: a player can completely "legally" keep one Sphere by changing the order of some actions.
Regarding battles with enemy heroes: if "The Lone knight" had nearly a dozen unique and interesting battles where the right tactic had to be chosen for each opponent, here, given the necessary artifacts and spells, almost all enemy heroes are killed quite quickly and simply, and the last few opponents have almost the same artifacts: Cursed Armor, Sniper's Bow, and by the 3rd or 4th time you react in the vein of "this caviar again!".
Systematically recruiting trolls and thieves from dwellings scattered across the world is quite tedious, and this is a factor that prevents those wishing to complete the map very quickly (there are 5 troll dwellings in the surface and underground worlds and 1 thief dwelling in each).
Veils of darkness on enemies we cannot yet reach, primarily Orange in the center of both levels — that's quite an evil. Raids by Orange's heroes will become more frequent as time goes on, and they will tire you out on their own, plus these "upstarts" will inevitably activate veils when approaching Confluxes because that is how the author designed it: passage is only possible through them.

I chose Castle (Blue) for play. We play as two sisters, daughters of the king, who were forced to flee the palace after the execution of their mother, the queen, on a false charge of treason. The girls grew up, underwent military training and are now ready to reclaim power and punish their mother's killers.
At the start, we have 2 towns with dwellings of levels 1-3, but without forts, and a total income of 1000 gold. Yes, the snail's pace at the beginning of the game is depressing, but you need to pass this stage to prove your ability to develop and fight.
On 3-3-1 I beat Sepfinorof, captured the Dungeon and finished off Brown.
On 3-4-2 I began the invasion of the underground world.
On 4-1-6 I finished off the last hero of Red, and he disappeared.
On 4-3-4 I finished off Green.
On 4-3-7 I beat Tira.
Unfortunately, I was short a few days to reach the Academy Tower, but I will still be able to visit and obtain everything necessary near it. Moreover, I don't have the Alliance of Angels yet: missing Sandals and Necklace.
On 4-4-4 I captured the Emerald Tower and finished off Turquoise. Entering the garrison, I was happy and laughed at the enemy's overconfidence: the city was completely without defense, while huge forces awaited my arrival and all these troops became mine: 164 titans, 328 nag queens and other creatures! Also, the Tower gave me a view of Purple Cephiro's Castle and Gen. Ascot sitting in it.
On 4-4-6 I beat Nella, and Pink disappeared.
On 6-1-7 I beat Gen. Ascot.
Started digging in the tavern, striving to get this hero. A few minutes of mindless sorting and firing unnecessary heroes finally gave a result. Great: all fanatics and mages will become Wizards! Looked at his skills and decided out of joy to make him a second warrior hero, especially for beating Adele and Xeron. Unfortunately, he has no wisdom, but some high-level spells are written into his book by the author. This means that for battles with other enemies, he is, alas, unfit. But he will handle 2 battles where strength is important.
Unfortunately, Tira never appeared in my tavern, so I'll have to do without snipers.
On 6-4-2 I finally beat Adele and Xeron. Orange disappeared. Only Purple remains, occupying a small area in the northeast corner of the surface world.
On 7-1-2 I released Blue Leper.
On 7-1-5 Bloodruby passed through the monolith and dug in at Dark Sun. I had almost decided to use hit-n-run against her, but made several attempts to bring the villainess down in one battle. Finally, it worked, and I was able to resurrect everyone.
Giving 2 Spheres of Vulnerability to a seer, I received Angel Wings. Took the Tome of Earth Magic, got the 2nd set of Wings, freed Jass.
Dug up the Grail and installed it in my capital. Received azure dragons; captured azure dragon dwellings; exchanged extra (now) artifacts for additional level 7 troops. Jass received the Tome of Air Magic from a seer, making hit-n-run possible.
Trying to hit Wickedwig with Jass, I saw that she is now killable in one battle (without the Sphere of Vulnerability, I gave them all away).
On 7-2-6 Wickedwig was defeated, and I was declared the winner.
Result: 181 days, 494 points.
Out of curiosity, I looked at how a battle between Shana-Shana and Wickedwig would go with the same troops and artifacts. Although she bypassed all boosters and received skill bonuses from seers, her skills even with the Alliance of Angels were still 87/89/80/86. And Wickedwig crushed her.

Summarizing after completion, I can say this.
Although the map has enthusiastic fans and is indeed quite good, personally I have mixed feelings. I feel some disappointment: I expected more; I liked the map significantly less than "The Lone knight" by the same author. The quest part is weak, the combat part is average. The difficulty of battles depends more on when and with what troops you fight the enemies. I liked the stage from the very beginning up to the showdowns with Orange's main bad guys the most.
First and foremost, I would like to eliminate the ambiguous situation regarding the use of the Sphere of Vulnerability in the finale: if it's forbidden, then forbid it, and let all players be on equal terms! Surely, a speedrun without the Sphere will stall before the last battle. As far as I know, two players definitely used it against Wickedwig. It would also be better if one could sell unnecessary artifacts rather than exchange them for something with seers or sacrifice them at the Altar (the only Artifact Traders are in the Academy Tower, which not everyone reaches in time). And it's better to remove Veils of darkness from Orange.
Alas, contacting the author is currently impossible (Aeris already wrote about this in the comments).
I think that one day I will still try to start his "Eternal love" map to form some own impression of it. If the beginning doesn't repel me, maybe I'll complete it.

Map: "Jedi Strike Back" 绝地反击（铁人版）


1. Chinese map as is
There are 2 versions of the map that I uploaded to this forum on 19.02.2018: 绝地反击（大众版）- the popular version, and a second variant "Iron Man Edition" 绝地反击（铁人版）. I didn't perform an analysis of the differences between the maps, but superficially I see that the starting objects differ in favor of easier initial capture (possibly easier guards as well). I will say that I played the second, more hardcore version (see the archive file – the map, and the second file – the starting save). The map was downloaded randomly along with two dozen others from the well-known gamerhome DOT com. Effectively, it was a pig in a poke.
There are no links or descriptions, everything is in Chinese, but I really wanted to play through some Chinese same-old brutality, so to speak, on skill :) No interest appeared here on the forum (except for 13 people who viewed the map); the map is unknown. But that made it even more interesting to play.
Everything is in Chinese; I didn't do any translations: no events, no inscriptions, nothing. I very actively used Prospector and the map editor. The map is beatable without reading the texts; it's not that important.
Searching for the map or its translation on Google yielded nothing; there was nowhere to look for hints. But I don't remember where or when, but I saw a similar start on Chinese YouTube; I searched by the key name of the map and here it is: a guide for the first 6 weeks.
http:// v.youku. com/v_show /id_XMjk5ODcwMjc0NA==. html?spm=a2h0j. 11185381.listitem_page 1.5!19~A
For those who cannot manage on their own, you can see how it should be passed.

2. General impressions
The map was completed without cheats, without bug-using, with a loss of a great deal of IRL time, although the author for some reason provided 2 ways to pass it dishonestly:
1) in the second half of the 2nd month at point 125-20-1, you can open a pillar with any of your heroes, take a free air book, 1500 azure dragons, 1500 magic dragons and "tear everyone on the map apart," but I discarded this option immediately
2) closer to the second half of the game at point 17-40-0, you are given 1 azure dragon for free, but they give it an INFINITE NUMBER OF TIMES, so you can daily use 3-4 idle heroes to collect from 50 to 100 azures. I didn't use this either.
I recommend playing without this; it's more interesting.
I would give the map a solid 10/10, even despite a couple of flaws or misunderstandings of the author's idea.
It is almost a classic Xiedu or the recently discussed map 6424 Heroes. Everything is similar: ideas, implementation, claims that the player knows mechanics at an advanced level, and tight timing at the start.
I give the map a high rating for: 1) atypical start, everything needs to be thought through, really like in chess down to every movepoint 2) emphasis on completing 85% of the map with archdevils as the main striking force (which is not often found on author maps) 3) beautiful and complex final battles (different strategies) 4) complete ban of some spells: BERSERK IS MISSING (unless you manage to "steal" someone else's hero from prison with this spell – a speedrun is possible), blindness in a single copy in a scroll, clone appears only at the end of the map almost 5) absence of standard easy solutions for completion (there will be no ranged stacks until the end of the game, when we take titans for a couple of final battles, the arch stack forms only at the end). In short, gathering some archers behind a forcefield, doing blind/clone/resurrect - this is not that kind of map, definitely. Which is why I give it 10/10.
The map took a lot of time and effort, but brought a sea of pleasure, primarily from the decisions made.
If you played Xiedu and 6424 Heroes and liked them, feel free to play this map too.
You can play with HD+, but especially for the first 1-3 months of battles it is desirable to use either the SoD_SP utility or no HD+, so that battles take place on the original landscape. In general, play around with the landscape as you find most convenient
TIMING is very harsh until about 137, and important afterwards (for example, enemy heroes coming out to your respawn on 31X), but less and less.
Overall, a War Strategy for archdevils is something new for author maps for me, which is why the map clicked. Although Kreg clearly lacked knowledge and mana, all battles are based on some strategies.

3. Starting save
The start of the map implies that you will go through it like clockwork at a VERY FAST pace. Therefore, it is important to generate the correct starting save. I would even say: if it's wrong, you won't pass the map. In all seriousness, it will simply be impossible. I am attaching a file here with the correct starting save; use it, as the requirements for the save are as follows:
1) most important (!!!) in the castle 118-103-0, which you capture on 1-3-5, there must be "anti-magic" and "quicksand" spells. There is only one passage there, and killing Kirri's 1000 hellspawn with your 20 vampires and 5 ghosts without this magic is impossible by any other means.
2) a bunch of requirements for leveling in the save: main hero Kreg must catch "morale" immediately, and support must catch navigation immediately as well (and preferably earth)
3) Dragon Utopia at point 26-6-0 should be minimal or level two (stronger ones simply cannot be handled at the start)
An immeasurable amount of time was spent on this save; do not try to repeat it in real conditions :)

4. Beginning 111-137
The goal starts in the corner of the map and can free only one of 4 dungeons: from the fathers you can choose Orrin, Kreg, Tazik or Malekith. I took Tazik, progressed to 1000 hellspawn, but couldn't kill them. Then I replayed and realized that you must take only Kreg; he is the best candidate.
For Kreg, from 2 free slots we catch "morale" and "fire magic". Morale is mandatory; in the first month we won't win battles because our troops are from 4-5 castles, including undead. Morale will help more than once later in difficult battles. For support we catch navigation immediately + earth, this is also mandatory.
Timing is VERY TIGHT, as you need to kill Gurnisson on the starting respawn as quickly as possible (I killed him on 125), and Kreg must manage by 126 to open the pillar at point 47-14-0, and below via an event "eat" the event for +999 mana. Without this, it's impossible to kill Kirri the mage's 1000 demons later.
Count your moves here, think through, but on 126 you must take these 999 mana and then immediately leave the respawn, as an enemy is landing from the sea whom we cannot kill yet. On 136 we take the castle and learn anti-magic + quicksand; mana should be as high as possible (I had something around 650 left, which might not be enough).
On 137 we kill Kirri and her 1000 demons. For battle we take only 1-2 birds, 20 upgraded vampires, and ghosts to drain Kirri's mana. The battle took a couple of hours (20 vamps and Kreg's 22nd attack against Kirri's 1000 devilspawn with 68 defense!). One misclick – and you could burn your computer out of anger!

5. New respawn (141 – 247)
We reach the central Inferno castle 72-71-0, build gates, teleport to our castle in the underground, after which it is very important to free your resource heroes from prisons as quickly as possible. They should collect more money on the sea and resources; building needs to be faster. Kreg's task: take "water walking" at point 59-0-1 and "resurrect" at point 52-24-0 as quickly as possible. And then go very quickly to the Inferno zone in the northwest and take 2 castles there (and later with strength, a third).
Kreg's knowledge is meager (100 – 140 mana), so we fight without losing archdevils; not a single one can be lost.
After capturing the Inferno zone, you need to visit the green tent at point 136-48-1 as quickly as possible + take the last piece of cursed armor there – this is our key artifact until the end of the game. Just like the "mage's ring" artifact for 50+ turns of magic action (which I used very often). The faster you pull all this off, the better: soon enemies will open pillars, visit the green tent and come to visit you. Therefore, you must do it yourself and capture their castles with all unbought growth.

5. Trip for town portal (311-347).
From 247 (or better even earlier) you need to break the anti-magic garrison at point 122-55-1 and travel through the desert south, killing whoever you can along the way. The task at this stage is to kill Yoga; only after that can you get the town portal via quest, without which enemies will take all your castles. At this stage, instead of a fast breakthrough, I started saving money, reclaiming captured castles, and killed Yoga only on 333, although everything should be done and can be done earlier.
And meanwhile, enemies are pushing from all sides toward your central Inferno castle at point 72-71-0. While Kreg is traveling to settle scores, your pawns can make ransoms, transport and sell ransomed forces in Dragonspire, i.e., distract enemy forces. As soon as Kreg gets the town portal, Khalid retrains him in everything, and playing becomes easier.
MILLION DOLLAR TIP. Enemy heroes can be killed, just like in XIEDU, by finding a "safe point" for single-hex archdevils. This way you can kill almost any opponent with almost any forces. An archdevil, mage and pegasus go into battle. We enter the safe point with archdevils, cast anti-magic, kill the enemy's single-hex units, the enemy spends all their mana, we then slow everyone down and kill them one by one. This point is 80-71-0.

6. Killing legions, leveling Kreg, killing Torgrim, killing Nella and getting the medal of vulnerability (411-511)
You can level up Kreg well on the peninsula in the southwestern part of the underground, near Zeron's castle. If speedrunning, you can arrive very early, open the guards before Zeron and eat all the millions of gold and resources yourself. But I didn't arrive that fast.
Torgrim is a key kill because magic almost doesn't affect him; this is a quest kill and later we will finish the map with this hero (spoiler :)
Immediately after, Nella, because without the medal of vulnerability you cannot deal with titans (blindness, forgetfulness) and other dangerous and magic-resistant units.
Don't forget to use the "safe point", see above.

6. Capturing castles, leveling Kreg (511-711), killing enemy "fathers"
We kill one by one everyone who is still left on the map and whom we can kill.
A good strategy against Rion's and Sorcha's angel castles. For battle we take archdevils and a couple of azures. We cast forgetfulness on shooters, then loop the enemy into repeating casting slow / you wait and then healing. After healing, archdevils fly into the garrison, strike, at the start of the turn – retreat. And so 100 times.
Example battle with Sorcha: killed with stats 77-93-79-83, forces: 1857 archs, 1200 rust dragons, 3558 horses, 5091 zealot, 7126 knight, 5265 griffon, 8000 archers. Kreg's stats are 62-47-45-48 and army is 1 azure, 717 archdevils, 1 archmage, 1 pegasus and 3 dragonflies (for burning).
At this stage the hardest battle was with Lacus and his phoenixes and azure dragons (about 100 battle attempts); it was necessary to catch the right placement, the right fear of enemy units before your azure and double morale of your archdevils under frenzy. It takes a very long time to catch such a coincidence, and then loop Lacus into resurrecting his forces while being behind a forcefield (since almost always Lacus instead does either speed or healing for everyone, and our units cannot keep up with the phoenixes).

7. Game finale.
On the "trees" island we raise Kreg to level 88, open pillars, capture everything possible, and free 2 heroes in the only tavern: "broken" Dessa with negative experience (trinket) and Torgrim. The latter we pump up to level 88 for 100% magic resist, preparing Torgrim for the 2 final battles.
I can say that I haven't seen such exciting battles in a long time. I won't tell all the secrets, only the order of killing the final enemy heroes (where you can choose):
3 – Gundula,
2 – Gird (killed by Torgrim)
1 – Mephala (killed by Torgrim)
Map completed on 825 (although the last 2 months are just a trip to level-ups, it could be faster).

8. General recommendations and tips.
Start of the game, speedrun the first 1-2 months as much as you can; this will pay off a hundredfold. You cannot sit on your respawn or do everything slowly. Kreg doesn't necessarily have to visit all level-ups that are specifically placed inconveniently to waste time; it is more important to get key spells and intercept artifacts from the computer, because he will come out anyway and might eat your artifacts first.
It is desirable to combine troop buying and money accumulation so that the only treasury in your castle with +10 income starts bringing good money due to accumulation.
Do a full buy-out of all external dwellings (especially angels, devils, titans). Azure, Rust and Fairy dragons will be needed for final battles, but I recommend buying them only at the end when you have 5-6 million money (no less).
As soon as you dig up the Grail (which will happen almost immediately in the 2nd month), put it immediately in Kildar – the human castle for archangel growth and +2 morale. The grail cannot be placed in flux, and in other castles (Tower, for example), it is less preferable. I carried the grail in my bag until the 6th month because I didn't know which castle would need it more :)
ALL quests can and should be completed; there will be no losses from this. The only quest I didn't turn in was the titan's thunder collection in exchange for 100 archdevils and the "clone" spell. You get clone from Torgrim and Dessa bought in the tavern, and 100 archdevils are not as important as the helmet knowledge of +9 knowledge in battles. I didn't believe myself, but still exchanged 2000 archdevils via quest for 1000 titans. The exchange looks uneven, but devils in the final battle against Gird are useless because they die immediately from his magic and fire shield, all 2000, so ranged 1000 titans are better here. And against Mephala 2000 devils are unlikely to kill even 5-6 azures (there are 1500 of them), so it's better to blast Mephala with clones of your 500 fairy dragons (tactic used by me for the first time) – this is the key stack against Mephala, she cannot be killed otherwise, even with Armageddon's Blade, as there won't be enough mana to kill 1500 azures.
Use all pawns for collecting resources from external objects, for buying growth from external dwellings, and also for diversions against the computer: attack a castle, buy out growth, sell this growth in your own stronghold (no money lost, but computer is without growth). I leveled Sephinor to the rank of "father's deputy": with intelligence and over 1200 mana she could conduct medium-difficulty battles and was also raised to level 88.
A couple of computers can be lured into water and blocked for a couple of weeks/months so they don't interfere (lured Turis this way).
On the map I noticed that enemy heroes didn't always come out for quest pillars. For example, Lacus didn't want to give up 34 fire birds, but simply accumulated them, coming out only when I placed a pawn nearby as bait.

In general, thank you for your attention. The map is recommended to everyone who has time and patience! Liked the Chinese brutality :)
And a couple of strategic pictures of memorable battles

And here are the sophisticated connoisseurs of Chinese cards that I mentioned above. :)\ If you're interested, you can select Chinese cards using the search option on the website I mentioned earlier in this post. The names and descriptions of the cards are displayed correctly in all languages, so you can translate them using Google Translate.

I suspect this is an attempt to save both Dess and Orrin. Such a playthrough will end at the first freefix, if not sooner.

A couple of months (and a couple of dozen pages) ago, something like this was written about Metatax. The question is - how is it even possible to save both Dess and Orrin? I'm asking theoretically.

Rince, this post at this link has a working link to save files. There’s exactly that development path there.

The question is not entirely on topic – does anyone know what happened to the portal http://www.maps4heroes.com/? New maps for Homm3 stopped being uploaded; previously, maps appeared at least once every 1-2 weeks, so there was always something to play.
And the portal http://www.heroes3maps.com/ is even stranger; they uploaded a bunch of maps and then abandoned it. There isn't even a section for new maps, and there are no maps from 2018 :(