A Shield Lying on the Water -- Moved to MBX

正在查看此主题的用户

状态
不接受进一步回复。
Hi, just thought I'd drop in to comment on the outstanding quality displayed in this mod, even though its not even complete yet. Great job!

And, I'd be honoured to help you with any texturing work you need done, and I suppose modelling as well to a lesser extent.. :smile:.
 
*sigh*  Sorry to bother you with more unit re-gearing, but foresters should use axes and wear leathers or furs.  Same thing with hunters. 
 
I'll start off with the usual compliments, that your mod has earned, it is absolutely amazing, and I am anxiously awaiting the storyline.

But, I found something strange.. I'm not quite sure if it affects the game that much, though I get frequent pauses (could be my computer, it's really just a turd inside a black box), but I always get these warnings that show up on the side of my screen, they only pop up in the map-mode:

Error Message 说:
Town: Khudan Merc's merchant is uninitialized

after the message pops up my icon (horse thing on the world map, whatever you like to call it) disappears, an other click on the map gets everything back to normal.. but after playing longer the message became more frequent, and at the point when 7 or so scrolled up at once I decided to quit...

An other problem I'm having is in battle mode, for the first 5 minutes or so, the damage inflicted, and speed bonuses, etc. (on the little ticker on the left hand side of the screen) don't show up, and when the battle's almost over all the sudden it all scrolls up... It happened a couple times, and I couldn't really notice any trend, aside form that every time that happened my men get slaughtered.
 
Patrick Blackbird 说:
I'll start off with the usual compliments, that your mod has earned, it is absolutely amazing, and I am anxiously awaiting the storyline.

But, I found something strange.. I'm not quite sure if it affects the game that much, though I get frequent pauses (could be my computer, it's really just a turd inside a black box), but I always get these warnings that show up on the side of my screen, they only pop up in the map-mode:

Error Message 说:
Town: Khudan Merc's merchant is uninitialized

after the message pops up my icon (horse thing on the world map, whatever you like to call it) disappears, an other click on the map gets everything back to normal.. but after playing longer the message became more frequent, and at the point when 7 or so scrolled up at once I decided to quit...

An other problem I'm having is in battle mode, for the first 5 minutes or so, the damage inflicted, and speed bonuses, etc. (on the little ticker on the left hand side of the screen) don't show up, and when the battle's almost over all the sudden it all scrolls up... It happened a couple times, and I couldn't really notice any trend, aside form that every time that happened my men get slaughtered.
Thats common.
 
Turin, which is common?  I've only seen either in this mod.  One difference-I have the "Reyvadin merc's merchant uninitialized" message.
 
Merentha 说:
Turin, which is common?  I've only seen either in this mod.  One difference-I have the "Reyvadin merc's merchant uninitialized" message.

yes mine is also with reyvadin's merc's merchant uninitialized. and i also have the battle damage and such not showing up, sometimes even outside of battle, when i save the saving game/save game complete message doesnt even show up. i've only seen these two problems in this mod and not any others.
 
First Citizen 说:
Hi, just thought I'd drop in to comment on the outstanding quality displayed in this mod, even though its not even complete yet. Great job!

And, I'd be honoured to help you with any texturing work you need done, and I suppose modelling as well to a lesser extent.. :smile:.

ALL RIGHT! :grin:

However, I'm currently going to try to engage in a vast alternative-placeholder-gathering operation; I'd like to take a raincheck on your texturing until I know what I've found, how much of it is usable in the game proper, and how much of what I'll want to re-do. One thing I do know I'd like to get -- though it's in modelling rather than texturing -- is modelling for 'civilian' Honseli clothing; I think that that could make a pretty large difference in immersion. (Also in the same category would be the feminine headgear -- just so that all female characters below level 25 aren't running around in hoods. The feathers, headbands etc. are a bit of an FE4 holdover, not what I'd ordinarily have thought of for the Honseli, but I think they're going to work...)


For other responses... Let's see...

Pinetor:

I'm afraid that I don't quite know what to do either, as yet... :???: I think I'm going to revise the taverns to stock some higher-quality troops. It's my experience that one *can* take out, say, tax collectors with the various Youth units, but it's not pretty, and it takes a while before one can gear up to hit enemy stacks in earnest.

Giving a town money changes how many soldiers and military stacks it can hire -- though there's a sort of upper limit for both.

I don't know *what* happened with your town battle. It might have just been a large tactical disadvantage -- towns are not supposed to be easily captured, and are currently strengthed to resist "Invasion Force" stacks (which, however, don't attack towns yet -- a limitation of M&Bscript). As to getting 'credit' for just one kill -- see my response to Blackbird and Turin...

Blackbird and Turin:

Here you've met my personal nemesis in this mod, a bug which, I *think*, is in M&Bscript proper and not in my own code. Apparently, every line of M&Bscript (note, the term is not intended to be derogatory, just short, expressive, and easy to type) counts as a message displayed in the buffer. There are four algorithms to restock merchants, one each for the merchant proper, armor and weapon vendors, and tavern; each of these involves a *huge* number of operations, to the tune of twenty or so script calls total in a loop that runs once for each town on the map (look at module_triggers.py sometime, even if you've never used the scripting language -- you'll get the idea).

The primary result of this is that it strangles the message buffer, so that no other messages display until the huge number of pseudo-messages finally finish scrolling up. This I've observed to take several minutes real-time on my Athlon 2600+... This shouldn't, however, have any effect on battles, apart from the fact that you're now 'playing in the dark' about effects of attacks... :???:

However, these pauses only occur four times total every day, pretty close to each other, so you get about 23 hours of game-time without these concerns.

I have to wonder, though, whether the sacrifice in playability is really worth it -- I'm pretty proud of the 'city-cultural' stuff, which ends up populating each city with culturally-appropriate arms and armor (well, most of them, Baheisir mostly gets two-handed swords -- let's just say that there may still be some bugs), but TLD's implementation, with a *much* simpler system of "if this town, this item has this rarity" (and no discernable overhead) may have been a better way to do it...

The "town reyvadin_mercs merchant uninitialized" problem is a wierd one. It *seems* to be produced by the restocking triggers firing too close to each other, resulting in two different triggers mucking with the same registers at once, but in that case, why is it *always* "reyvadin_mercs"? I had thought that spacing the merchant triggers decently far apart solved this problem, but evidently not, so...

The disappearance of the player's sprite is a *really* unusual bug in M&Bscript, but that's what it must be -- there is no way I know of to issue a "player vanishes from map" command. :smile:

Kaiser, Merentha:

This is the same problem described above. It's a major cosmetic issue, but not a gameplay one, at least not as far as I can tell. *knocks on wood*

I may want to slim this down considerably next version -- that, or bug Armagan to make M&Bscript a little faster -- that, or reduce the number of script_logarithmic_increase calls in the next version...
 
ex_ottoyuhr 说:
Here you've met my personal nemesis in this mod, a bug which, I *think*, is in M&Bscript proper and not in my own code. Apparently, every line of M&Bscript (note, the term is not intended to be derogatory, just short, expressive, and easy to type) counts as a message displayed in the buffer.

Are you sure?

What's much more likely is that you've got script errors. You know those "SCRIPT ERROR ON OPCODE 24323" messages. Yeah, those messages definitely count towards the message buffer and will totally scroll legitimate messages to never-never land.

But not non-error ones, otherwise we'd all be having this problem. I've got triggers with giant try-for loops that do cause noticeable lag in my Extra NPC mod but my messages show up perfectly immediately afterwards.
 
fisheye 说:
ex_ottoyuhr 说:
Here you've met my personal nemesis in this mod, a bug which, I *think*, is in M&Bscript proper and not in my own code. Apparently, every line of M&Bscript (note, the term is not intended to be derogatory, just short, expressive, and easy to type) counts as a message displayed in the buffer.

Are you sure?

What's much more likely is that you've got script errors. You know those "SCRIPT ERROR ON OPCODE 24323" messages. Yeah, those messages definitely count towards the message buffer and will totally scroll legitimate messages to never-never land.

But not non-error ones, otherwise we'd all be having this problem. I've got triggers with giant try-for loops that do cause noticeable lag in my Extra NPC mod but my messages show up perfectly immediately afterwards.

Hmm...

Thanks. I didn't realize that -- and it's a known bug that I tend to cut the Official Module System entirely too little slack. :oops: This might very well be it...

Edit:

I've found the problem: just as Fisheye predicted, a veritable boatload of error messages -- errors of my own creating, though, to the effect of "Error: Potentially illegal town khudan_mercs has cultural influence from potentially illegal culture Lunnic Honseli". Hmm...

Next release will hopefully see a solution to this -- for some reason, my towns_begin and towns_end are not behaving... (!? Likely register problem...)

Oh, yes, and I almost forgot: Thanks, Fisheye.
 
ex_ottoyuhr 说:
Next release will hopefully see a solution to this -- for some reason, my towns_begin and towns_end are not behaving... (!? Likely register problem...)

Glad to help, especially for a mod that shows so much promise, great work.

towns_begin and towns_end are defined in module_constants.py. They should point to the first town ID and the first non-town ID respectively. Every party (in module_parties.py) between these two pointers MUST be a valid town, of the sort that has the standard things - castle, merchant, tavern, town square, and mercs.
 
Another bug.  I am at war with Aimaren and allied with Ajaria.  When I try to attack an aimaren patrol engaged in combat with an ajaria patrol (i think, I forgot to post this last night when I remembered the details), instead of getting the usual "Its prince Arcos!"  I instead get the "its nice to see my allies campaigning so vigilantly."
 
fisheye 说:
ex_ottoyuhr 说:
Next release will hopefully see a solution to this -- for some reason, my towns_begin and towns_end are not behaving... (!? Likely register problem...)

Glad to help, especially for a mod that shows so much promise, great work.

towns_begin and towns_end are defined in module_constants.py. They should point to the first town ID and the first non-town ID respectively. Every party (in module_parties.py) between these two pointers MUST be a valid town, of the sort that has the standard things - castle, merchant, tavern, town square, and mercs.

The problem turned out to be something else: I had assumed that register state was preserved when scripts were invoked -- along the lines of C as opposed to assembly language -- and thus ended up stomping on register 0 and a whole host of others. The latest upload (0.1.35) fixes that -- the whole mod is significantly more stable now.

Thanks a heck of a lot, and I mean that completely positively -- I might never have found this otherwise...

Merentha:

Another known bug, that joining battle doesn't work -- you end up in dialogue with the friendly party, not the enemy.

I think, now that I'm writing this down, that there's probably a dialogue state that I'm not addressing -- I went through module_dialogs with fire and sword when removing the Swadian/Vaegir parties, and missed a couple of other things too. I actually thought that I had this bug solved, but it reared its ugly head once again...

Fisheye? :grin:

(Also, I'll get to work on the alternate hunter costumes and so on also. I had envisioned them wearing dark cloth, plus light armor given that they're, in this case, hunters pressed into militia/tracking duty rather than pure hunters...)
 
Joining battles works fine for me.

I think maybe otto is right, if the enemy force has nothing to say, then it defaults to the friendly force saying something.

Either that, or your wife snuck out last night and "made peace" with the Aimaren king without your knowledge.
 
I was having difficulty joining a battle... But then I found out that I was clicking on my allies, and when I clicked onto the enemy factions troops, I could join the battle.
 
fisheye 说:
Joining battles works fine for me.

I think maybe otto is right, if the enemy force has nothing to say, then it defaults to the friendly force saying something.

Hmm. That may be the case... I'm surprised you can do it, though -- and kind of annoyed; I hate Heisenbugs... :smile:

Either that, or your wife snuck out last night and "made peace" with the Aimaren king without your knowledge.

Not implemented... (Yet?)

SilentBob:

Does that do it? Hmm, the plot thickens -- how *does* the original game do it? ...

(No, don't answer that, everyone -- unless you really want to. :razz: I'll learn more if I find it out on my own...)
 
Previously, someone asked for a general tutorial.  I am beginning to think I could use one too, especially from a "tactical advisor" or someone in the Bahesier castle.  Some general information on what alarm actually means, and how cities can conquer cities, a bit about the economics and strategies within the game.  Right now, I am lovign trying to figure it out, but I look at the Siviac cities and see almost no garrison (like Bahesier), but then the Lunnic or Ajarian cities have massive garrisons. 

My question is this, I suppose.  How does combat between cities work, assuming the player was absent from the equation?

Another suggestion:  I noticed in Talchin that some archers on the walls and roofs could really have made my life hell.  Is there any way to get them to spawn up there so that invading troops come under a crossfire from all directions?
 
My question is this, I suppose.  How does combat between cities work, assuming the player was absent from the equation?

Another suggestion:  I noticed in Talchin that some archers on the walls and roofs could really have made my life hell.  Is there any way to get them to spawn up there so that invading troops come under a crossfire from all directions?

As far as I can tell they won't actually attack eachother's cities. It's possible to do, I just don't think it's in. All that's been done is the armies are triggered to patrol around that town and as such they generally fight other armies sent to patrol around that town. It's up to the player to do all city attacking.

Also, yes, it is possible to make any troop spawn anywhere you want in a seige. Notice 1066 where we've carefully placed all the defending troops. Especialy on the thin pallisade walls.
 
There is a bit of a problem attacking a city though, when you're not having a plains battle, due to clunky AI all of your soldiers get stuck outside running into the wall. (at least in Tetovo).
 
The AI has trouble in 1066l with the doors as well (where the code came from). You can fix it in two ways. You can either build walls that can be climbed (such as earthworks) or there is a way of triggering the AI to go cleanly through the gates. We were going to fix it so they would go through the gates, but the new accurate earthworks just made it pointless to do.
 
Merentha 说:
Previously, someone asked for a general tutorial.  I am beginning to think I could use one too, especially from a "tactical advisor" or someone in the Bahesier castle.  Some general information on what alarm actually means, and how cities can conquer cities, a bit about the economics and strategies within the game.
Right now, I am lovign trying to figure it out, but I look at the Siviac cities and see almost no garrison (like Bahesier), but then the Lunnic or Ajarian cities have massive garrisons. 

This garrison-size issue is a problem which I need to investigate further. Certain cities have a way of being seriously under-garrisoned in their first 'iteration' of hiring troops. This problem generally goes away eight hours in, with the second firing of the decision-making trigger, when they bring their garrisons up to par -- at least from what I've been able to see of it. I have a few theories about what might be causing this, but all of them point to something not really rectifiable...

I can also post a 'strategy guide' to ASLOW, if you like, but it may be a day or two before I have it up. There are substantial things in the works for the next iteration of the game...

(Also, I plan on removing the hard numbers from alarm in subsequent versions. It's gotten to be conventional wisdom in game development these days that the player shouldn't be exposed to numbers very often -- and, unlike most conventional wisdom in modern game development, I think this idea has merit.)

My question is this, I suppose.  How does combat between cities work, assuming the player was absent from the equation?

At present, when it comes to actually *attacking* cities, Colt is right -- it doesn't, and it can't. :smile:

The two governing phenomena of fights between cities are alarm level and treasury size; basically, the lesser of those two is the amount of money a city spends on military parties. Parties (excluding economic) are chosen at random; the city then pays a fixed price (in module_constants) for the military party in question. A defensive party is sent to patrol around either the purchasing city or another city that this one is trying to assist, with a chance of helping the other one based on the home city's level of alarm about it; an offensive party is dispatched to patrol around the city's currrent hostile target. The plan is for "invasion force" parties to be the only ones to directly attack cities, Sid Meier's Pirates-style, but if I can find a way for all parties to assault enemy cities, I'll do so.

(And come to think of it, assuming that the OMS has a way of iterating over all parties of a given template, that just might be possible...)

Another suggestion:  I noticed in Talchin that some archers on the walls and roofs could really have made my life hell.  Is there any way to get them to spawn up there so that invading troops come under a crossfire from all directions?

Yes, it's possible -- just a matter of placing spawn points, overriding equipment (especially forcing ranged weapons), that sort of thing. The current city maps are from Vanilla, and both they and the option to fight within a city are, at present, placeholders -- proofs of concept rather than full-fledged gameplay. I plan on doing new maps at least for Baheisir and Ambrolauri, probably for a few other cities or perhaps all of them; once these are done, I'll do much more thoughtful versions of city assaults.

In the meantime, there's always 1066. (Incidentally, I've noticed some debate on the matter, but I like the Norman archers holding shields -- it contributes to the 'archaic' ambience very nicely. The same goes for the font, even if it *is* illegible at times.. :smile:)

Colt:

I didn't take code from 1066 for any part of attacking cities -- if I had, I would certainly have asked your permission, or at least acknowledged you in the credits. I had figured out how to implement the basic (fighting on random map, plus occupying cities etc.) city attack on my own, though not without some trouble; as to attacking inside a city, well... there, I have to admit that I did it because I saw in 1066 that it could be done :smile:, and because some of these maps are begging to be fought over...

However, we do have a problem in common. It *might* be possible to hack together something that *resembled* AIs attacking cities, but it would be hideous, and wouldn't get along with player attacks or much of anything else very well. Hopefully now, with two mods begging to be able to do it, there'll be support for attacking cities in the next version of the game...

(My hypothesis is more or less thus. In addition to the conventional garrison of a city, one could specify a 'shadow garrison,' a normal party set to stationary and invisible that would contain the same number of soldiers as the garrison itself. An invasion force would march up and engage this shadow garrison; when the battle between them ended, the city's garrison would be changed to match the winning party, and its alliegance would be switched to the winning party's as well. As you can see, though, this wouldn't synch up very well with player attacks on the city -- and I'm not 100% sure that it could even be done with the current state of the Module System.

Of course, this means that Fisheye or Yoshiboy's probably going to pop in and prove me embarassingly wrong... :razz:)
 
状态
不接受进一步回复。
后退
顶部 底部