Re: Bug Thread: Discussion
 

What happens if you conquer C in magic phase and enemy units are trying to move from A to C through B? Will they still be able to attack C even though they didn't start out next to C.

PS: Sorry if this is the wrong thread for the question. It seemed kind of related to the discussion.

Re: Bug Thread: Discussion
 

That would be one test, one time.
But when it was said that another sanity check should occur after the magic phase then I was taking it to mean the full sanity check phase should occur again after the full magic phase. It sounded like it would involve checking all magically enacted movements again.

But if its just that each movement should be reality checked, and each magical action be checked when it occurs, then that would be minimal. Of course its never the ONE check. Its the additive of every additional check that comes up for discussion. And we have had many people praise Dom for running on systems they cant run other games on.

Actually I have no problem with long hostings. I dont tend to blitz much so everything is either hosted on a capable server, or its on my desktop and I appreciate the long hostings making me look up and realize that I should take care of other things. As far as I am personally concerned I wouldnt mind if every "hoggish code" (AI, checks, random events) that was cut-back due to host-time considerations was put back in full-force (or at least optioned).

Re: Bug Thread: Discussion
 

The thing is, the checks we are proposing are basically trivial. Most of the hosting time goes to the AI (in SP) and to battle resolution anyway.

For checks this simple I definitely disagree with the notion of making them optional--testing that option will take an appreciable percent of the time that the whole test would take.

Re: Bug Thread: Discussion
 

I agree. This one by itself would be more of a waste of programming overhead to make it optional than it would be itself. Only if it was a "recheck all moves" might it be considered. And even then probably not by itself.

I meant that the whole collection of "would be nice but takes too long" would have been good to pile into an option for servers and people at work.

For the original clarified description of this check, I withdraw my comment. :)

Re: Bug Thread: Discussion
 

I think there is a bug concerning underwater resources and land castles.
I've got two 50 admin land castles bordering an uw province I have taken and the uw province has 0 resources.

Re: Bug Thread: Discussion
 

And why is that a bug?

Re: Bug Thread: Discussion
 

Land castles don't get resources from underwater provinces, so it seems reasonable that the uw provinces shouldn't lose resources?

Re: Bug Thread: Discussion
 

Exactly.
I'm not sure though, that the land castles are responsible I just can't find another reason.

Re: Bug Thread: Discussion
 

I have that same issue in Chololera Mp game. I have two 50 admin forts bordering a water province and now the water province has 0 resources; just the same as the land province split between the two forts. The land one makes sense but not the water one.

Phoenix Immortal Bug
 

SOoooo..

Phoenix .. attacks some enemies...
Gets afflicted. (With disease) Dies. Blows up on battlefield.

Second incarnation. Has 1 hp. Blows up. dies.
Repeat, quite afew times.


Being immortal, and fighting in dominion, it returns to castle.
It has healed the disease affliction, but still only has 1 hp.
Kinda sucks being a god with 1 hp.

Re: Bug Thread: Discussion
 

I think if this happens in Late Winter you'll end up perma-dead. Phoenix Pyre is also buggy and you will end up perma-dead if you get hit with the bug where a Phoenix Pyre unit duplicates itself on the battlefield.

Phoenixes pretty much suck in combat because of this, even in CBM.

Re: Bug Thread: Discussion
 

I dunno, I've done quite a fair bit of testing with the Phoenix as an "awake, disposable, expansion aid" pretender, and overall results are fairly decent. I must say in probably 20 separate 2 year tests, I've not once had my Phoenix permanently die.

Re: Bug Thread: Discussion
 

You are correct. This has happened 2x. Best not attack with the phoenix in late winter= ).

Re: Bug Thread: Discussion
 

I suppose it depends on your paths and what you're doing with the Phoenix, and when you're doing it.

My Phoenix was A4F4, Dom 5 so no extra awe, CBM so built-in Phoenix Pyre, and using the Phoenix as a solo immortal suicide bomber to clear indies, before researching mistform. It's got so many death crosses in the HoF that I'd go blind trying to count them.

So, it worked pretty well for a bit, but I was planning on retiring the Phoenix from its bomber career in the midgame and using it for some specific battlefield buffs. At A3F3, those plans were pretty much shot.

Re: Bug Thread: Discussion
 

Sounds like just about exactly what I was doing. But I really considered it to be little more than an expansion aid, and then a booster forger and royalty summoner later. I think it's absolutely hilarious, really, and I don't know but out of all of my tests, I never suffered a catastrophic disease death. Was it just the one time that it happened to you? I'm not entirely sure, but I think it may be somewhat avoidable, even if it's imminent.

Re: Bug Thread: Discussion
 

I knew about the Late Winter bug, so I was very careful to avoid that. I had accumulated about 4 or 5 afflictions each fall, but usually had just a Never Healing Wound left by Late Winter.

The Phoenix Pyre bug is rarer, but when it hits, your Phoenix is a dead duck. That's what killed mine.

Re: Bug Thread: Discussion
 

Okay, not sure where this belongs, but as it is some kind of bug report, this cant be all wrong :)

Dont know if it has been reported earlier, I looked through Edi's bug list but did not find it. I am running version 3.21 + the CB mod version 1.3.

I think there is an issue with units that automatically attract other units (with the summonX command) and also have the stealthy attribute. I noted this when trying out the awesome Skavenblight nation mod by Sombre when my sneaking armies always triggered battles. Some additional testing proved that it was only while led by the Chieftain or Warlord commander that my armies attacked, despite that their move command clearly stated "Sneak", other stealthy commanders worked ok. Both these units automatically summon one unit per turn. I then deleted the summonX command from the .dm file, and then the sneak / stealth worked fine. Note that the summoned creatures in both cases have stealth +0, so the army should still be ok sneaking.

Is this only an issue with this mod or is this the case for all such units? I do not have that much experience with all nations / eras yet and do not know of another unit with these attributes.

Re: Bug Thread: Discussion
 

I would imagine it would be the case for any sneaking commander that autosummons units, since it is the summons that are causing the battles. That's a pretty unfortunate bug especially for sombre's skaven mod, it's hard to be sneaky and tricksy like a rat if you are forced to fight even when you sneak.

Re: Bug Thread: Discussion
 

I was under the impression only the autosummoned unit was involved in the battle when that happens and that if you had units under that commander, anything he summoned joined the first 'unit' of troops under him and wouldn't cause any problems.

Re: Bug Thread: Discussion
 

Yes. As long as the summoner has anyone in his units, then summons will go there.

If he does not have anyone already following him then it goes into the province queue for assignment.

Which of course means if he is not the owner of that province that they appear with the owners troops and have a battle.

Re: Bug Thread: Discussion
 

It also happens in other instances.

If you have a oreid with stealthy units, and she seduces the a commander, she leaves her troops behind and they fight. Which makes sense.

But even if she loses, and it goes to a fight, stealthy units in her command will trigger a province fight.

A Vampire with a black heart's auto summon has often triggered a fight. I think, when the vampire tries the assassination, his troops are removed from his slot.

Even if he succeeds, the troops then trigger a fight against the province.. which they always lose due to no commander.

Re: Bug Thread: Discussion
 

You can travel with units on your commander. Summoned units only go to the first of the five squads controlled by your commander.

1) Remove all units.
2) Add a dummy squad.
3) Add your other squads.
4) Remove dummy squad.

This should send the free-spawn to the province, not the commander.

Situation: I ran out of money while I still had a bunch of stuff queued. (I suddenly lost several territories to a just-declared war.)

I canceled the queued items and next turn I got the message "Inexplicable increase in wealth".

(SP game, Dominions 3000 mod)

Re: Bug Thread: Discussion
 

in a recent blitz, my warlock was attacked by the Lord of the hunt global that qm's pangaea had put up. He was in a forest province, but according to the description, that guy only goes after priests in forests, and my warlock was not prophetized or with a shroud or anything, so there's no reason he could get classified as a priest. It was the only time in that game that that oddity occurred, though it was near the end of the game.

Just reporting the bug.
Zlefin :)

Re: Bug Thread: Discussion
 

Don't know if this is on the shortlist. After a siege is lifted from a castle, the castle needs to be repaired to its full defense value by the troops inside it. But the damage to the walls is not shown after the siege is lifted, even when damage still exists. This can be misleading.

Re: Bug Thread: Discussion
 

WAD. The forts defenses are back to full strength as soon as there was one successful siege break in between. If the defender managed to do that, then you have to start getting the defenses down to zero again before you can storm the fort.

Re: Bug Thread: Discussion
 

I am fairly sure thats not actually true. I *believe* I have seen where subsequent sieges have continued from the value remaining siege value.

Aka Siege a gets for to 242/1200
Sieger A gets attack by sieger B
Fort repairs a little, depending on garrison.. say 300/1200
Siege B starts at 300...

Re: Bug Thread: Discussion
 

Forts are not immediately repaired when the siege is lifted. It is just shown like that, but actually the damage still exists.

The easiest way to try - siege an empty fortress. Then move away next turn. Then return on the third turn. The damage on the third turn will be the sum of damages on the first and third turn. But during the second turn it is shown as if there is no damage. The display during the second turn while the siege is lifted is misleading.

Re: Bug Thread: Discussion
 

"Sieger A gets attack by sieger B" is a different situation.
The defenders never had a turn to rebuild.

Re: Bug Thread: Discussion
 

Like has been said already, that's unrelated. For all practical purposes, the fort was under siege the whole time. If the sieging force was diminished in strength during some time, then the defenses may have been rebuilt. But once a successful siege break attempt happens, the fort's defenses are immediately back at full.

That's never been the case for me, and I don't think that this is true. Maybe there was some unit left that has been sieging the fort on its own - maybe without a commander.

An exploit against the AI
 

I discovered something: You can take provinces away from the AI without war. Sneak an army through and get caught. Of course this requires that you be able to win the battle with only a single commander and all stealthy troops.

Note that you can win the battle without even having a commander involved--my hidden commander was of a type that produced freespawn. One of those took an unoccupied province. Still no response and I'm up to 7 provinces taken from one AI and 2 from another.

Re: Bug Thread: Discussion
 

AI :bug::

In an 3.21 SP game (no mods), I observed AI-controlled EA Abysia attacking an indy province (Deer Tribe, 2 commanders, 35 troops) with a lone naked Slayer. Obviously, the battle didn't go so well for the Slayer.

Attachment 7545

Presumably we don't want the AI sending its commanders on suicide missions?

I have all associated files, if IW wants them.

And Edi: did you get the exploit PM i sent you?

Re: Bug Thread: Discussion
 

For the love of God, why do you argue about it? Just try it if you don't believe and you will see that it is true.

Re: Bug Thread: Discussion
 

He was probably planning to do an assassination and got caught. The AI and the Independents are all now susceptable to the same things we are.

Re: Bug Thread: Discussion
 

I've never seen a stealthy unit caught by independents.
(I'd don't actually recall ever seeing an enemy scout get caught by another enemies patrols. Do scouts let you see those battles?)

I have seen the AI send commanders on suicide missions all the time.

Re: Bug Thread: Discussion
 

That was my first thought too. But it was definitely in an independent province...can indys now catch sneakers?

Re: Bug Thread: Discussion
 

i believe the manual claims that the castle defense needs to be repaired by defending units, as Psycho says. The bug is that the real defense value of the fort is not reported, and so it looks like they are back to full repair. I have not carried out any testing though - if a castle is reduced to half defense, and has no defenders, will it still be at half defense after 5 turns of having been left alone?

Re: Bug Thread: Discussion
 

They definitely can in my games. But I purposely boost defense random amounts in indy provinces. Having some of them quite high has flown back in my face a few times because I cant get a scout across an indy province.

Im not sure what the game sets PD at in Indy provinces but Im guessing its at a point of occassionally catching sneakers.

Re: Bug Thread: Discussion
 

The short answer for me is - they never are in the experiments I tried. After five turns they seem back to full - but after 2-3 turns they often are still somewhat damaged.

I wonder how it recalculates.

Re: Bug Thread: Discussion
 

heh, another ineffable dominions mystery ;p

Re: Bug Thread: Discussion
 

I don't think indies default to having any PD. Their troops are just regular troops and can be whittled down. If they had any PD, the assassin technique wouldn't work. They probably can now get the PD boosting events though.

Regardless, I still suspect this is just the AI being stupid. I've certainly been attacked by lone commanders before.

Re: Bug Thread: Discussion
 

Its been a long time since Ive tried to figure the siege/rebuild thing out but... since there is a formula for how many and what size bodies it takes to tear down a castle, and a like formula applies for too many people inside rebuilding the castle for your siege to work at breaking down the walls....
does a formula along those lines affect the rebuild? Maybe some people see an instantaneus rebuild due to having enough forces after the seige, and others see a lag due to not enough forces/population?

Re: Bug Thread: Discussion
 

yes it does, but as the rebuild happens eventually even w/out troops, according to the information posted above, this means most likely that there is a base automatic value, like 50 def/turn, added.

The bug is that the true numbers for the castles' defense aren't showing during this... or that the castle is fully repaired instantaneously, contrary to the manual, and contrary to what some posters have reported.

Re: Bug Thread: Discussion
 

Id love to see the true numbers reflected. Especially during siege. I wonder if Kristoffer thinks its unrealistic.

Re: Bug Thread: Discussion
 

the true numbers are shown during the siege, to the defender... the bug involves the true numbers seemingly not being shown to the defender outside of siege.

Re: Bug Thread: Discussion
 

It seems that after two turns without siege the fort is fully repaired even if it's empty. I didn't bother waiting more than one turn in my earlier tests. In that case the manual is incorrect (from page 82):
"Fortresses which are damaged but not besieged are repaired normally - the repair value is simply added back to the defense each turn until the fort is all fixed up."

Re: Bug Thread: Discussion
 

Wouldnt the type of fort make a difference in how quickly it repairs also then?

Re: Bug Thread: Discussion
 

Its probably a value related to administration. Most anything the involves forts uses that value. It also makes a reasonable amount of sense.

Edit: Which might make swamp forts and LA Ermorian castles interesting test subjects.

Re: Bug Thread: Discussion
 

In my test it was a dark citadel raised by "three red seconds". It has an admin value of 20 and repaired all 600 defense after two turns.

Re: Bug Thread: Discussion
 

this sorta fits my observation... none of them were fixed the next turn, all of them were fixed by turn 5.