Re: Bug Thread: Discussion
 

That's the army movement bug, where you move an army to attack but it doesn't. But since you're in a province where you siege, instead of the defend province command, the commanders defalut to maintain siege.

Send the files if you want, but I don't know if it will help at all. It's not been reliably repeatable before, which is required for it to be solved unless it is discovered while fixing something else. See recent spate of fixed immortality bugs, they all have different times of addition to the list, but seem to have been linked and were taken care of at once.

Re: Bug Thread: Discussion
 

Hey cool! A bunch of the immortality bugs I sent in were fixed.
YAY! Thanks EDI, Johan, KO!

Re: Bug Thread: Discussion
 

LA Patala Nagarajas (1320) don't revert correctly into their second shape. When they fight in Gandharva shape and are almost dead, they turn into a Naga shape at 1 hp.

The other Naga mages (Nagarishi and Nagini) suffer the same fate when their Yaksha and Human shapes are near death. When I tested this, the nagas ALWAYS got hit for 1 point less than their total hp. E.g. Nagini in human shape has 12 hp, is hit for 11 hp of damage, and is now at 1 hp but naga-shaped. And then she dies.

They should have the full hps in their both shapes, just like other shape-changers.

Conceptual Balance was on during the most comprehensive test, but I noticed this without it.

Re: Bug Thread: Discussion
 

In a mp game on the llamaserver I have a battle replay that crashes partway in when I try to watch it. It goes probably 15+ turns before crashing, and I don't even get a crash error report.

Re: Bug Thread: Discussion
 

rdonj, check Illwinter's changelog. There have been a couple of bugfixes for problems that sound exactly like what you're having, and they will be fixed in the next patch.

Re: Bug Thread: Discussion
 

This is the only set of changes I see that talks about a crash, and I don't see anything about issues with battle replays. Am I missing something?

Re: Bug Thread: Discussion
 

Huh, another bug? I was just playing around with the Destroyer of Worlds as an SC chassis and I gave it a black steel hat, but the protection from it isn't showing up. He keeps the same (3) head prot he normally has.

Re: Bug Thread: Discussion
 

rdonj, I believe that the first fix addresses the battle replay problem. I may be wrong about it, but I have been inspecting turn files from the llamaserver from somebody, I think it was FungalReason, and confirmed that they led to crashes. The turn files were then sent to Illwinter, so I assumed that they are what the first fix is about. We'll see.

Regarding the head prot, it does change for me when I try to reproduce your problem. The protection values are only approximate numbers, though, as the protection mechanics are more nuanced than just head and body prot. JK or KO explained a little how they work, somewhere. Basically, every limb has it's own protection value, AFAIK.

Re: Bug Thread: Discussion
 

That would be very nice, thanks.

I was checking the specific number for the head. He had 14 or so body prot with the plate armor I gave him, but only 3 on the head with a 23 prot helmet. Maybe it has something to do with his afflictions? He has 3, lost a limb, never healing wound and reduced strength. I'll take a screenshot later when my computer is out of danger.

Re: Bug Thread: Discussion
 

missing limbs mess up helmets, that one has been around for a while.

Re: Bug Thread: Discussion
 

Good to know. That's really odd though. I'm not even 100% sure that the protection is actually messed up, it seems like it could easily be properly taken account of and just not updating the protection page properly.

Re: Bug Thread: Discussion
 

I am getting this recurrent problem with getting wrong turn numbers.

Even if I download the turn file again, put it in a new folder, and resubmit a turn - it still gives the wrong turn number.

I don't know if this is a function of my stupidity (probably), a problem with llama server, or a dominions bug but I thought I'd mention it here, as I know other people have had the same bug occur (and re-occur).

Re: Bug Thread: Discussion
 

I hate to say this as I might lose my edge, but is thunder bolts meant to work underwater?

Re: Bug Thread: Discussion
 

Orb Lightning and Lightning Bolt work underwater. Thunderstrike shouldn't. If you're casting it using Tempest it might work.

Re: Bug Thread: Discussion
 

Cool, thanks!

Re: Bug Thread: Discussion
 

edit: sorry, false report.

Re: Bug Thread: Discussion
 

There's one EA TC hero, the Master of Iron Clutch, who is Immortal and Plague carrier. In one game he got diseased. That means he is at 1HP constantly, dying and reappearing in the capital all the time, and his immortality does not yet wanted to heal the disease itself. Having a lot of TC cap-only mages in the capital, that's quite frustrating.

Re: Bug Thread: Discussion
 

I suppose that this disease is from battle wounds, though, or old age, not the plague carrying ability itself?

Re: Bug Thread: Discussion
 

Got one in battle I guess - was researching in his own lab, Vine men appeared, and he started to collect infections, and got into the Hall of Fame (Obesity, still 1HP max), I guess due to dying once every turn.

Then that would be another exploit. Send the immortal hero alone into a province just taken by the barbarians a few times to get into the HoF.

Re: Bug Thread: Discussion
 

Nope, the crutch guy has a plague cloud (like mummys) and everything in the plague cloud gets diseased, including the crutch guy. Then he dies, immortality kicks in, but he's diseased, so he ends up in the cap with 1HP.

He's still pretty useful IMO.

Re: Bug Thread: Discussion
 

Just remember to position him far from your other mages. The plague only spreads in battles, but the cloud can do real damage if you forget to move him away from the default.

Re: Bug Thread: Discussion
 

AI "bug": It seems to me that the AI only seduces with units that have the seduction effect of a succubus (fx number 210), and never with units that have the seduction effect of a Nagini (fx number 298). I suppose that the relevant code for assassinations/seductions hasn't been updated when the new seduction type was introduced. Thus the only units that the AI uses for seduction attempts are Succubus (811) and Lilot (2071).

Units affected:

• Nagini (1325)
• Oreiad (1650)
• Manushya Rakshasi (1740)
• Satyr (1881)
• Cloud Vila (1941)
• Mountain Vila (1942)
• Sirin (1945)
• Rusalka (1954)

Re: Bug Thread: Discussion
 

Added to shortlist. Thanks, lch. :)

Re: Bug Thread: Discussion
 

I'm also going to clean the shortlist up after the next official patch comes out, because for various reasons I didn't have time to do that when 3.21 came out.

Re: Bug Thread: Discussion
 

The seduction bug has been fixed. The Jotun Skratti problem has been fixed despite not showing on the progress page, so the shortlist has been updated. No more upkeep free Jotun werewolves or wolves.

Re: Bug Thread: Discussion
 

Hmmm...Has this one been reported?

If you give a commander with default Siege Bonus a siege booster, the Siege Bonus will not be accumulated. At least will not be displayed correctly in the status window. My case was the Siege Engineer of Ulm with Wall Shaker, not sure about other situation.

Re: Bug Thread: Discussion
 

Confirmed, siege bonuses don't appear to stack. Might be intentional though.

Re: Bug Thread: Discussion
 

Odd. I've got Siege Engineers with Gate Cleavers and the bonus is displayed there.

That's a CBM game, but I wouldn't expect that to make a difference.

Re: Bug Thread: Discussion
 

Siege Engineers with Gate Cleavers will see an improvement, yet that is because the bonus from the cleaver is 50. My tests show me that the bonus is 50 no matter who you put that cleaver on.

Re: Bug Thread: Discussion
 

Oh, duh. Yeah, that's bad.
Hopefully it's only a display bug? I haven't checked to see how long it actually takes to knock down a castle.

Re: Bug Thread: Discussion
 

Here's a bug similar to random Barbarian events occurring at a sieged fortress:

I was sieging a fort, and I wanted to get rid of pesky scouts, so I patrolled. It's probably irrelevant, but I also stormed the fort the same month with one unit, just to peek inside. Anyway, I got a message about finding a sneaking enemy, but strangely there was no battle that I could click on.

The next month I stormed the fort and took it ... only to find that I no longer controlled the province. It was now owned by the scout I had uncovered the previous month, when I patrolled.

I had armies moving into the province the month I stormed, but since they were on 'move' and not 'move and patrol', they just ended up inside the fort, instead of fighting the scout. Not that it's possible to 'move and patrol' to a fort you're sieging.

Next time I discover a sneaking enemy while sieging, I'll be sure to leave a small force still patrolling when I storm a fortress. That way the scout won't capture my zero-PD province.

Re: Bug Thread: Discussion
 

Actually vfb, I don't think it makes a difference.

I routinely leave units preaching, sieging, and patrolling when I storm forts just to make sure I don't lose the province if I lose the storm.

It has never seemed to matter- the province is still taken when I storm. Not all the time.. just sometimes.

Re: Bug Thread: Discussion
 

It probably does, because that sounds like it works exactly like a barbarian raid event, only it's a scout being caught. Looks to me like the bug triggers whenever a third party incursion into a sieged province comes about through something other than normal army movement. May be related to the no control/partial control/full control of province thing.

Re: Bug Thread: Discussion
 

The actual code in the mod is:

#selectspell "Hell Power"
#pathlevel 0 1
#end

That changes it from a B3 to a B1 spell. However, the mod does not change fatigue cost. So it still shows 3 slaves/300 fatigue.

If you add this line it fixes all the weirdness:

#selectspell "Hell Power"
#pathlevel 0 1
#fatiguecost 100
#end

Now it displays properly, is B1, and only costs one slave. I don't know if that's what QM wants though, it seems kind of drastic. But if the intention was to just reduce fatigue, then why not leave it at B3 and change the cost to 100?

Re: Bug Thread: Discussion
 

I dont know if this qualifies as a bug, but, shouldn't Jaguar Warriors get the skinchanger tag?

Re: Bug Thread: Discussion
 

Skinshifter is different from the transformation that Jaguars go through, though it is very similar. Honestly, I'm not exactly sure why they get their own special tag as there are dozens of units that change form.

Re: Bug Thread: Discussion
 

Here's a bug you don't see everyday. My guy blessing the enemy.

Re: Bug Thread: Discussion
 

That gives me an idea for a new unit.

Sleeper Agent - When you hire him he doesn't appear where you recruited him. Instead he kills another players recruit and takes his/her place. Then helps your side when they fight against you and causes unrest in the form of sabotage.

Re: Bug Thread: Discussion
 

I had a mod called death cult

All the units had gcost = 0, or negative. All had stealth.

Some of the units were - horsemen of the apocalypse. Death eaters. Eye collectors.

Some commanders even earned gems.

The entire idea was that they lived among us. And only when it was the final time of showdown (Ie., the end times).... would they appear.

They would have a capital, of course, A city of the damned. But as they would not own many provinces - most recruitment was done via a commander order, as well as some autospawns.

Many of the commanders had #onbattle spells - like darkness, terror etc - because research would be hard.

All in all it was a lot of fun to make. I got it about 87% finished before I quit....

And it had a really, really creepy feeling when playing against it - because you literally never would know when some of these units would pop up in your territories. And you would never know if casual PD would defeat them....

Re: Bug Thread: Discussion
 

Sounds very interesting. Any way, what exactly happens with a negative gold cost and upkeep?

Re: Bug Thread: Discussion
 

In one game I did cast a global [nature's bounty] successfully, but it didn't show up in the global list. Caster also does not have icon about owning a global. There are still 2 spots left for globals.

Did it happen to anyone before? It's a huge problem.

Re: Bug Thread: Discussion
 

Well, back then it generated you gold. No idea if it still works.

Re: Bug Thread: Discussion
 

Yep, happened recently to someone else, but not me. I remember it was the same as what you're saying: two slots open, global cast successfully, but global is missing. It was posted on the forums, but I can't remember if it was this thread, or its own thread.

Re: Bug Thread: Discussion
 

Apparently paralyzed units can still move in order to follow their commander if they have Guard Commander orders. :o

Sounds quite serious to me!

Re: Bug Thread: Discussion
 

If the developers are going to take a look at Guard commander,

I'd like to make a request as well, which particularly affects flying units, but affects other units as well, berserkers and tramplers for example.

Many times flyers will 'mill about' to no effective difference of placement.

XYZ
P1Q
RST

RST
Q1Z
ZYP

Each time they do so, they burn fatigue. So many times, by the time an enemy does make it to your commander - flying units will have 60-70 fatigue.

Another notice:

XYZ
P1Q.2
RST

In this circumtance countless times I've seen the troop q wait until 2 comes up to it and hit it - instead of taking one step and hitting 2 first.

And we'd all like archers to fire while guarding (but there are good balance reasons while that doesn't happen).

Here's an algorythm that would improve both these things.

Guards, like troops fill a box, the size of which depends on the number of troops. Ignoring size and the cmd donut, more or less 25 units make a 5x5 box, 16, a 4x4 box.

I presume that right now, guards 'mill' in order to allow them to encounter an enemy, to thus attack, and presumably to avoid blocking a commanders move all the time. My suggestion:

General movement
A. Each size guard box has a fill template (ie., what you do when you have a 5x5 box but only 20 units).

B. When a commander wishes to move, subject to ZOC have all of the units guarding execute an identical move. If the commander doesn't move, (subject to C) the guards execute the same move (ie., they don't).

Combat:

A. Record the size of the box at the start of combat.
B. Allow guards to attack enemy units within a box 2 units larger. So for example, 24 guards are in a box 5x5. Allowing them to move with in a 7x7 box would allow the guards to trigger if an enemy approached within 2.
C. An enemy withing the trigger area causes all guards to move to engage the triggering unit.

Several behaviors this would be good for:

Suppose this was the prefill for 18 guards, guarding (1).
A.B.CD
E.FGH.
IJ1.L.
.MN.oP
.QRS.T

The presence of # within the trigger zone would trigger all the guards to move to engage.


....AB
....ECD
..1.GH#
....IJL
....MN

Etc. ( I didn't put all 18 guards.. but you get the idea)

Now if you were really, REALLY slick, you would have 5-6 guards move to intercept the threat and the remaining guards move to a template position for fewer guards.. something more like


a..b.
.c.deKQ
f.1.gL#
.h.i.MR
k...p

As guards die, they would constantly still surround the commander and still send the others to attack the threat.

The routine would be something like:
(made easier for programming)

for an nxn box of guards with x actual guards
a). Always retain n guards on the commander. (Lowest unit #'s they've been with the commander the longest-).
b. For each triggering enemy, (z) send n guards to deal with it.
Reposition all x-zn guards within the a smaller box, but keep the original size box for purposes of triggering. (The guards fall back to protect the commander).

Re: Bug Thread: Discussion
 

I've seen troops with missile weapons (archers) fire when set on guard commander during an assassination attempt.

Re: Bug Thread: Discussion
 

Bodyguards behave differently during assassinations. Melee troops will charge the assassin, so I'm not surprised archers will fire.

Re: Bug Thread: Discussion
 

Regarding the recent gold cost fixes, there's still one unit which has a gold cost while its alternate shapechange form has a zero gold cost: The Serpent King's (653) serpent form (654). It's a pretender for Sauromatia, so it's not that much trouble...

Re: Bug Thread: Discussion
 

Lately in testing Patala the game (3.23) repeatedly crashed when Nagini tried to seduce...

Re: Bug Thread: Discussion
 

Not getting any crashes with a vanilla test game of Patala with 3.23b, so that seems to have been taken care of as well.