imo mountain ranges are fine, forests too, the issue I have is that trading paths and some connected villages and bound villages are ridiculously out of the way in a maze where neither you nor the AI Villager Parties manage to reach it in a feasible time frame. The worst zone's sturgian Varnovapol to it's extremely important fishing village, which was placed on the other side of a mountain range with a single path covered in forests, the village itself sits in a surrounded area of forests. That means Varnovapol's constantly if not permanently suffering from starvation almost every campaign. To add salt to injury, there are 2 bandit spawn fields in that forest + 2 more north of Sibir, making the entire pathway littered with bandits which often destroy NPC caravans, sometimes PC caravans, most AI lord parties that are on recruiting mode, and to finish with a grandeur of absurdity, nearly the entire area's a dead zone with 2 mountain ranges bottlenecks, when leading towards the fishing village and a castle bound village nearing Tyal, and the other passage leads from nowhere to nowhere and only the AI will use it on auto-pathing towards norther Khuzait lands and Tyal. What's so ridiculous about it one might ask, and I'll tell you, that's a simulation of a Taiga forest, Taiga forests were massively chopped for wood and also contained a fair amount of game, Sturgia virtually has no wood production (I think it's one village) when their territory spams in a aburdly massive taiga zone. What should be done? The fishing village should either be moved or a passage-way connecting it directly to Varnovapol should be there. Either way, a castle bounding the current village's spot + an extra village placed in the are for woodcutting - than another one north on the "nothing" zone holding the same, both a fishing village that should be on the northern shore + a woodcutting village in the taiga zone. Such a silly change would drastically change Sturgian economy, would allow for an extra clan + it would strengthen their capabilities of defending Tyal (which's lost 99% of the campaigns to either Khuzaits or, exceptionally rare, to the Northern Empire)
There are numerous dead-zones in the map that arbitrarily buff some factions and hinder others.The most notorious effects are observed on Husn Fulq, Tyal and Uriksala Castle, "forgot the name castle" nearl Husn Fulq that theoretically "belongs" to the Khuzaits, Than we have a horse village from Aserai that is literally on top of desert bandit spawn zones and 3 light years distance from any town or castle.
Meanwhile we get a packed battania which's arguably the best faction zone in the entire map because their economy's safe, their villagers and villages are safe except for 4 of those (as long as the AI patrols every so often), they have 3 advantageous choke points towards all factions except Vlandia where they are basically ripe for the taking. The natural geohraphical balance + the Auto-Calc imbalance comes into play from there and turns Vlandia into a massive power-house from the get go (too easy to invade Battania, Neviansk Castle is delivered to them in a platter by TW, and after pushing into Battania every entry point towards their territory becomes gated by endless chokepoints). That's why Vlandia dominates so often, Battania also will rarely be extinguished by anyone due to their packed layout. Sturgia will often lose 3 thirds of their slums territory to every other faction almost every game (Northern Empire, Western Empire, Battania, Vlandia and Khuzaits have no problem splitting it senseless taking Omor / Varcheg, all their nearby castles, Tyal, Uriksala, the castle behind Tyal, basically leaving them with Ryval (almost impossible to take), Balgard, Sibir, Varnovapol and their respective castles. Even than, if a game runs long enough Sibir will often be captured by someone.
Aserai's another that suffers a lot because it's too easy for them to lose Husn Fulq's entire region to SE, which basically rolls a dice, if Vlandia or WE decide to war them at the same time as SE or right away after they've lost that war, Aserai loses everything west of Sanala leaving the entire faction in shambles while spam generating endless immersion breaking named rebel clans in the process which'll often litter the game after 10 years, if that doesn't happen, they'll recover and eventually push into the castle west of Ortysia (than it's anyone's guess, which's how the game should operate 100% of the time)
Much of these details lie upon the geographical distribution of the map + massive dead-zones that create gaps that allow for bandit AI to destroy AI economy, and also prevents the AI from properly defending zones (I always said that most castle placements both in WB and in BL are ludicrous, they don't make any sense given they should be gating borders or shielding important resources, neither are true). What kills me on this regard's that some bound villages placement + geographical arbitrary shapes make for exceptionally non-viable "boundage". It makes no sense to bind a village behind a mountain range that has no direct pathway for a castle to defend it, you see that with Gersegos, with Varnovapol, the southern villages "outside of battania", etc. Another very bizarre thing are the couple castles that have a single village, never understood nor will understand why. In fact, wtf's Charas even a Town? It has 2 choke points (making for bad trading routes even the AI avoids it) littered with sea-raider spawns + it has 2 villages and a castle with a single village auto-pathed to it. It's the worst town in the game for any viable AI management.
When it comes down to it, I still think the original old map for BL was much better in terms of balancing the territory and actually making sense. Nearly no dead-zones, Sargot was in it's correct place, charas was much more significant in it's positioning (neighboring WE and Aserai). Idk, I don't really like this map, it could've been done better by filling the dead-zones and making more reasonable castle placements, but as is, it's really bad.
Finally, it would all be fine if TW applied actual logic towards why some settlements became towns and others didn't taking as example real life History itself, some basic anthropology and some basic geography. There's no logic at play here when it comes down to that. Some zones would never hold a Town for obvious reasons (like, ppl need to eat, otherwise it won't populate enough to form an actual town) and "handicapped" towns also make no sense as in therea aren't enough resources to justify why and how ppl would've amassed in said settlement.
Lesson time: How did Towns formed during the Middle Ages (and before that)
- They had plentiful access to natural resources
- They were geographically advantageous for safety from both Weather and life-risks like wild animals and "bandits", but only if the are offered enough food
- They had precious natural resources like metals, wood or stone - cheapening construction and making them paramount to any realm. As a consequence, virtual food placement would be made through the use of agrarian means, like pastures for anything that could survive the climate - be it farms with wheat or types of cattle, or both - They'd also inflate food production by introducing chickens.
- The settlement was geographically positioned turning it into a trading hub
Currently these were not followed, which begs the question, what was the logic applied to the settlement placement + geographical development of the current map? I really can't see anything that's even remotely reasonable. Placements make no sense for balancing, settlement specs and position neither, it also doesn't apply educated simulation... Was it at random? Throw a dice and bam, placed?
anyway, got carried away on that subject. Fact is that there are too many ludicrous decisions in the game design itself, countless counter-intuitive decisions that can't be explained by any logical pattern. That ranges from armor values, to weapons dmg, to settlements, to faction balancing, troop trees, etc... There's always this looming feeling that some blatant arbitrary favoritism was placed towards the balancing whenever you try to make sense of it.