Abolish the distinction between Arcanus and Myrran races

[Nelphine]

Sorry germanjoey - I have agreed with you many times before (especially about wave of despair - it Is too strong, but so is aether binding, maybe focus magic, heroism, sprites - I haven't found one in chaos, but maybe raise volcano), but about your suggestion, and seravy's dismissal of it, I think you're wrong.

He did use the numbers as a reason to reject it - but he also went on to explain that he didn't feel there was a good algorithm for deciding what island/continent a wizard would start on regardless of the numbers. He suggested a few possibilities and explained why he thought they wouldn't work - and I'm inclined to say he was right.

I've also thought about it since then, and I haven't found any other way to decide which island someone should start on.

I think catwalks idea to add an ore if a start is bad is interesting, but like seravy, I am unsure if a remotely decent algorithm could be found.

I would like to hear ideas from both you and catwalk with specifics - whether they get used or are doable at all for CoM, ideas in general are good and might be useable for other games or environments.

[Seravy]

Like, I just gave you a "precise" parameterized algorithm you could implement that should greatly normalize starting positions.

Numbers with ? on them being "precise". Anyway...

We most likely can't even use a constant there, but some sort of a formula based on Land Size settings and the number of players, two factors you completely ignored. (Note that "number of players" can be different from 4 even if 4 opponents are selected - Myrran retorts can result in pretty much any distribution of players)

Your "a." situation cannot happen, or it's more appropriate to say, covers about 0.0001% of cases. The maximal continent size is 200 out of whichever - even on Fair with 400 tiles that covers at most half the available tiles and I assume your intention was to cover the case where the "largest" continent takes up much more than that. On Large and Huge a continent can't be more than 1/3 or less of the tiles. On Small and Tiny it might happen in theory but as continents are smaller there, the chance is astronomically low, plus, this would put the fortresses closer than the allowed amount, as a total 120 tile continent is not nearly large enough for 3-4 fortresses.
Your "b." situation is almost the same, based on the "number" I assume the bigger continent covers at least two thirds of the land, as you suggest 3 out of 4 players. As 1/2 is the most we'll ever get, this also won't ever happen - and yes I'm ignoring the actual impossible number and trying to deduce what you meant by it instead and it still doesn't work.
c. is a reasonable scenario and the only part of the "algorithm" that can work - but only for exactly 4 players on the plane.
d. is not a new suggestion, it's just "otherwise do nothing".

The impression I get is that he doesn't want to understand what we're saying until he randomly stumbles onto the same thing himself.

That's a pretty much saying "He can't understand something unless he understands it on his own.". That's a tautology. Sorry to disappoint but I can only think with my own brain, like any other human being.
It's not randomly though - I prefer to play test games for issues that has been mentioned, but I can't play more than one game at a time. This isn't a problem that can easily be resolved by playing test games though.

...and now the post from yesterday I couldn't send because the forum was down.

I also find that it's difficult to convince him,

Indeed, I'm stubborn but even that serves a purpose. If people manage to convince me about a suggestion despite my stubbornness, it means it was a very good suggestion that improves the game. If I wasn't stubborn enough, we would have a game where every hero has wind walking with a base movement of 9 but can be oneshotted by any spell in the Death realm. (no, seriously, I have a friend who asks for stuff like those every day....ok, in reality he would be satisfied by 6 moves probably. lol)
It's also worth saying that this is somewhat of an endless well I can throw time into - If I wanted to do everything, it wouldn't be complete even 25 years from now. I have to filter out the most time consuming ideas that don't result in enough improvement to be worth it. The mod is already 1.5 years old and we are about to the point where the essentials are finally done, but in making software testing and fine-tuning takes several times more after that. (Especially in this game - the complexity is very high so many games are needed to judge any specific detail, and each game played takes several days.)

and I suspect that it's harming the AI who doesn't know that this is the case.

Fortunately with some earlier and the 3.2 updates this is no longer the case - The AI won't build swordsmen if the have a higher tier available and they are forced to unlock a higher tier as soon as they have 2 settlers in play (or under production). It certainly was true in the older versions.

Do you agree that my rough points above can form the starting point of a useful and easy algorithm?

I actually can't judge that without the actual algorithm. I'm tempted to say yes but in this sort of stuff the devil is always hiding in details.
One thing I'd ask though. Let's say we draw a line for what counts as "too bad" terrain and how much extra ores needs to be added to compensate. But what about "too good" terrain? And how big room do we want between the "worst" and "best" cases allowed? This is probably what worries me most - if the gap is tiny, games will be too uniform and boring. If it's too large, I waste a lot of time without achieving anything...and I'm not entirely sure there is something in between, as both of these are very subjective.

On a side note, it might be a good idea to put forth (or renew) your design vision for Caster of Magic. Like, what does balance mean to you? Do you want to promote or diminish the role of luck? What is your ideal balance for hero usage? Stuff like that makes it a lot easier for everybody to get on the same track.

-Every unit should be useful for at least some role. It doesn't have to be a big role, but it shouldn't be redundant in the race. So disposable cheap spearmen only used to allow you to cast 10 phantom warriors in that battle is OK. A race having two units that only do this is not. (Swordsmen do have a role - they are the units you make if you need something that can fight but have no Fighter's Guild yet, mostly in the early game only. They're also useful for countering bowmen with Large Shield although Cavalry might be better for it. I admit I don't use them much - bowmen usually work better and Sawmill unlocks them so they are always available. This might even need improvements but honestly I have more important things to worry about than the bottom tier unit(s).)
-Every spell should be more than just useful, and feel "impressive". They shouldn't be overpowered on the whole, but they should be great at their intended function. This was one of the greatest strength of the original MoM, except there it was only a few spells like this and 75% useless ones. I aim for 100% impressive spells. It's "Master of Magic" so Magic should play the central role.
-I want to diminish the role of luck in factors that decide the game without player interaction. I want to keep the luck factor on elements the player can plan for and on those that help/hinder a little but aren't significant - to keep the game more varied and replayable. (and I don't understand the comment about wars  being irrelevant - wars between AI players is currently the largest luck factor in the game as it eats up hundreds of their units and thousands of their mana crystals without any player interaction. It only depends on completely random factors. )
-The role of heroes is to be the ultimate unit that takes a lot of effort (and flawless play) to raise but when they reach it they are nearly unstoppable, trumped only by very rares summons backed up by powerful wizards, other heroes, and very few high rarity spells in certain situations. This should be kept in check by the fact that you have a limited number of heroes - you can't have a stack of 6 heroes fighting everywhere at once, so while they might conquer an enemy city easily, you still have to use other units to defend your entire empire everywhere else. This is probably the most questionable part but considering the costs of a good hero (artifacts, buffs, need to level), any weaker and they'd probably be a horrible deal. (at least compared to rare+ units which you can summon 4-10 of for the price of a single good artifact, and which don't need levels to work. Also, heroes are fun, and one of the main attractions of the game. If they are underpowered, that's much worse than if they are overpowered, so I'm trying to be extra cautious with hero related nerfing.)

I'm pretty sure I had these posted somewhere already though.

[GermanJoey]

Like, I just gave you a "precise" parameterized algorithm you could implement that should greatly normalize starting positions.

Numbers with ? on them being "precise". Anyway...

We most likely can't even use a constant there, but some sort of a formula based on Land Size settings and the number of players, two factors you completely ignored. (Note that "number of players" can be different from 4 even if 4 opponents are selected - Myrran retorts can result in pretty much any distribution of players)

Your "a." situation cannot happen, or it's more appropriate to say, covers about 0.0001% of cases. The maximal continent size is 200 out of whichever - even on Fair with 400 tiles that covers at most half the available tiles and I assume your intention was to cover the case where the "largest" continent takes up much more than that. On Large and Huge a continent can't be more than 1/3 or less of the tiles. On Small and Tiny it might happen in theory but as continents are smaller there, the chance is astronomically low, plus, this would put the fortresses closer than the allowed amount, as a total 120 tile continent is not nearly large enough for 3-4 fortresses.
Your "b." situation is almost the same, based on the "number" I assume the bigger continent covers at least two thirds of the land, as you suggest 3 out of 4 players. As 1/2 is the most we'll ever get, this also won't ever happen - and yes I'm ignoring the actual impossible number and trying to deduce what you meant by it instead and it still doesn't work.
c. is a reasonable scenario and the only part of the "algorithm" that can work - but only for exactly 4 players on the plane.
d. is not a new suggestion, it's just "otherwise do nothing".

You are so dense it borders on unbelievable. The numbers I used were parameters. You're a programmer, right? You know what a parameter is? If 200 is maximal, then substitute 150 instead of the 300 I used, and, say, 120 instead of the 225 example. 

The idea was to prevent AIs from getting cloistered on some tiny starting island due to the broken capital location assignment algorithm. This phenomena happens to at least one AI in every single game I play, and usually 2 or 3. In fact, I've had it happen to all 4 AIs in some of my games. 

That's a pretty much saying "He can't understand something unless he understands it on his own.". That's a tautology. Sorry to disappoint but I can only think with my own brain, like any other human being.
It's not randomly though - I prefer to play test games for issues that has been mentioned, but I can't play more than one game at a time. This isn't a problem that can easily be resolved by playing test games though.

...and now the post from yesterday I couldn't send because the forum was down.

It's only tautological if you start from the assumption that it is impossible to communicate with other human beings. Most of us don't think that way. For example, when I first started playing CoM I thought that FMS were very overpowered and never built Halberdiers. However, after reading your rebutal to my initial post in this forum, I tried and found that you were right in that Sprites by themselves are sufficient to succeed in many of the situations that I thought necessitated FM. After reading Nelphine's posts about Halber spam, I tried it and found that unit is extremely cost effective and powerful. 

Regarding the rest of your post - and again I say, your stubborness over your vision for the game isn't what bothers me. If you were to say something like, "I don't understand why you think this a problem so I don't see the need for this algorithm. Could you show me some examples where you think there's a problem? Oh, and 200 is the maximal continent size on fair-sized maps, so a continent of 300 tiles could never exist." - then there's no problem. But no, you grasp onto and isolate the "mistake" in my post as a way to dismiss it entirely, not even bothering to quote it in its entirety to show your disdain.

[Nelphine]

In case it matters, I currently think heroes take very little effort - warlord, tactician, and put 3 items (with combat stats, though it almost doesn't matter what) on them, endurance, heroism, win. If attacking a capital, you need to have items that grant defense and you should have as many units as possible with you to soak lightning bolts. Lizardmen spearman come to mind as good fodder, although obviously the higher tier units will be better - but all they need is HP and armor, since the hero will kill everything.

Nodes will need more effort (prayer spell, items with more combat stats) depending on the monsters, but it doesn't require interesting tactics - just charge in, bash them. Even if you are using zaldron. Better heroes just kill things faster, and can fight stronger nodes with less items.

Healing, exaltation, raise dead, resurrection are all bonuses that make it even easier, but are not needed.

[Seravy]

The numbers I used were parameters.

Erm no. Numbers are constants. A parameter is a variable given as an input to the function. Land size and Number of players on the plane are parameters in this case, as the desired output depends on them yet you ignored both. The word you were looking for is probably "examples" and not "parameters".

You are so dense it borders on unbelievable

I believe this is the point where moderators are supposed to warn people about not insulting others. Fortunately I happen to be one in this subforum so I'm warning you to stop insulting me. You are clearly not in the state of mind for a constructive debate at this point so I suggest staying away from this specific thread for a while until you cool yourself down.

The idea was to prevent AIs from getting cloistered on some tiny starting island due to the broken capital location assignment algorithm. This phenomena happens to at least one AI in every single game I play, and usually 2 or 3. In fact, I've had it happen to all 4 AIs in some of my games.

I would have plenty of questions about this but due to the above I'm going to restrain myself and not ask them from you.

Did anyone else experience this as a critical problem that happens way too often? (As I don't think I ever had this happen to multiple AI in a game, or any at all nowadays, the AI is good enough at making ships and even based on pure random chance, a smaller area of land should get picked less often than a larger one.)

[Nelphine]

I've only seen it cripple an AI twice - once where he got just one city and no shore, but no room for other cities, and once where there was room for other cities, but neutrals on the next continent prevented that, so no shore cities.

I've seen AI who have had bad but not crippling starts due to small islands (say 2-4 cities), but rarely would I say that was the cause of their demise. They normally still expand faster than me.

[GermanJoey]

The numbers I used were parameters.

Erm no. Numbers are constants. A parameter is a variable given as an input to the function. Land size and Number of players on the plane are parameters in this case, as the desired output depends on them yet you ignored both. The word you were looking for is probably "examples" and not "parameters".

You are so dense it borders on unbelievable

I believe this is the point where moderators are supposed to warn people about not insulting others. Fortunately I happen to be one in this subforum so I'm warning you to stop insulting me. You are clearly not in the state of mind for a constructive debate at this point so I suggest staying away from this specific thread for a while until you cool yourself down.

The idea was to prevent AIs from getting cloistered on some tiny starting island due to the broken capital location assignment algorithm. This phenomena happens to at least one AI in every single game I play, and usually 2 or 3. In fact, I've had it happen to all 4 AIs in some of my games.

I would have plenty of questions about this but due to the above I'm going to restrain myself and not ask them from you.

Did anyone else experience this as a critical problem that happens way too often? (As I don't think I ever had this happen to multiple AI in a game, or any at all nowadays, the AI is good enough at making ships and even based on pure random chance, a smaller area of land should get picked less often than a larger one.)

Oh for fucks sake. You are hopeless.

[Nelphine]

Seriously?

Uncool germanjoey. If you can't talk to him, at least explain to me (in pms).

[Catwalk]

Joey, I agree with your points (about starting locations) and I think you're a competent player with valuable input. I also agree with some of your criticism of how Seravy makes decisions, but you're shooting yourself in the foot right now. As another moderator, I encourage you to take a small break and calm down.  Seravy, I would like to request that you do not take moderator action against Joey at this time. I acknowledge your right to do so, of course. But we're a small community and we rely on participation and feedback of everyone. Especially highly competent players (who should, of course, behave).

[Seravy]

Joey, I agree with your points (about starting locations) and I think you're a competent player with valuable input. I also agree with some of your criticism of how Seravy makes decisions, but you're shooting yourself in the foot right now. As another moderator, I encourage you to take a small break and calm down.  Seravy, I would like to request that you do not take moderator action against Joey at this time. I acknowledge your right to do so, of course. But we're a small community and we rely on participation and feedback of everyone. Especially highly competent players (who should, of course, behave).

Don't worry, I'm not planning to take any action unless he ignores our warnings.
Meanwhile I realized we ended up completely offtopic somehow. This was originally a Myrran thread, how did it even turn into a starting continent quality discussion? I should have noticed this earlier!