CA asks for Bug List - to be fixed in the expansion

[Myrddraal]

6/ When you direct a unit to "run" when the game is paused, and then unpause the game, the unit immediately reverts to walking. Also, ordering a unit to disable phalanx formation when the game is paused only works about 50% of the time, ie half the time they will stay in phalanx formation.

Are you sure of this, I've never experience any such problem.

Remember, we only want confirmed bugs. Don't put CA off with lots of minor issues. Stuff like "Problems with units ignoring orders generally." needs to be explained a bit more. They will probably want definate instances, like "units ignore orders when in group".

[Myrddraal]

1: Fire at will command enables a unit to fire at an enemy that is not engaged with one of yours which is fine since it reduces friendly fire losses. The problem is that sometimes you do want to fire at an unit that is overpowering one of yours. Now, with a hastaii or principii unit they will throw two spears then charge. Put on guard command they should remain in their position and fire until ammo is depleted and not charge afterwards. Am I making sense?

Um, not to me .

Could you explain again?

[Camp Freddie]

I think we could do with a quick list of trait bugs:

Of course, there's the double combat trait bug for starters. If you fight in manual combat, all trait triggers are checked twice.

GeneralLostHPRatio trigger is bugged so that it wors differently in auto and manual battles.

Heroic saviour and shieldbearer ancilliaries use GeneralLostHPRatio as a percentage (0-100) when it should be a ratio (0-1.0).

GeneralFoughtInCombat trigger doesn't work properly. Affects coward traits.

Good farmer traits have very high thresholds, making them impossible to gain without farming temples and extreme luck/micromanagement.
Bad farmer traits have very low thresholds, making them very easy to gain. Swapping the good/bad farmer thresholds would make it work sensibly.
Poor Assessor is perhaps a bit too easy to gain too which especially annoys newbies who don't understand exact triggers. Increasing the threshold or trigger chance would be a good idea.

And since no-one's mentioned it yet: The difficulty levels are bugged, making it easier on V.Hard for some reason (tested in 1-on-1 custom battles)!

[Viking]

2/ The pathfinding bug in cities is still extant. In my last siege battle for instance, I directed my missile cav to run away from an advancing phalanx, instead they charged straight INTO the phalanx. (Oh, and they had skirmish mode on too).

Skirmishers with skirmish mode turned on is somewhat uncontrollable when the enemy is that near that they (the skirmishers) consider to fall back.
Turn off skirmish mode and you will avoid this problem.

[SpencerH]

Issue: Retreating ships

1) They should be effected by ZOC.
2) Their pathfinding of the AI's retreating ships appears to be bizarre. If not, why does a decimated AI fleet often "retreat" through the players fleets ZOC (see point 1) and continue toward player controlled territory rather than back toward its own territory.
3) AI controlled retreating ships appear to be given extra movement points (I dont know how this could be tested so its just an observation). How else could one explain playing ping-pong with AI ships back and forth across the Adriatic 3 times when I can only move across once during my turn?

[Kraxis]

Archery bug/oddity:
Units in Fire at Will will not fire at a unit engaged, nice. But it will fire at a unit that is routing, even if it is pursued by a friendly unit. It is as if they ignore the change made with 1.2 when the enemy routs.
Not fun when the enemy unit consist of one man while your unit consists of lightly armoured troops such as light cavalry or mercenaries.
Also it can at times be hard to force units to fire into a melee (but that is way better than no being able to stop them).

Routbug on walls:
Why has this been ignored two times already. At least there should be a comment about others perhaps experiencing it.
Units still Fight to the Death on walls when they should be routing, despite supposedly fixed in 1.2. Not a good situation most often.

[SpencerH]

Issue: attacking ships in port

IMO, it should not be possible to always (or ever) attack ships in port (which often had their own highly effective defenses against attacking ships). OTOH, ZOC should prevent ships from leaving or entering without fighting.

[SpencerH]

Issue: Routing enemy

IMO, we need a "pursue" command that will allow all friendly units that are without existing commands to to engage the nearest routing enemy unit. Too often the end of battles consists of spotting units that have stopped dead and ordering them to attack a routing enemy unit that is passing near ,or even through, the players stopped unit.

[Myrddraal]

Hm, I dunno if we should include that. Its not a feature which doesn't work, its not really an issue, its more of a wishlist. Oppinions?

Sorry about that Kraxis, I've added it now.

[Myrddraal]

AI controlled retreating ships appear to be given extra movement points (I dont know how this could be tested so its just an observation). How else could one explain playing ping-pong with AI ships back and forth across the Adriatic 3 times when I can only move across once during my turn?

Don't retreating armies get extra movement points too? I think this is a feature.

[Lord Preston]

replay bug
when exiting a replay occationally it will start again

[Ahmose]

mybe it is not a bug .. but when I order alot of my troops to do movements while my general beeing chased by a chariot his horse halt and he do his imotion of raising his sword and then continue running but this result beeing cought alone with the deadly chariot and his boudygards away from him.
thx for all

[The_Mark]

Grouped cavalry speed bug - Grouping a quick cavalry unit with a slow unit results in both units moving at the speed of the fast unit Please could someone confirm this, the final list should only contain confirmed bugs

This might be, again, one of the issues that could be a "feature" instead of a bug, as the units seem to use something of an average speed for the group's units in order to keep form, and it affects infnatry as well.

I don't know if this qualifies as a confirmation, but here is a replay showing the issue. Two silver shield legions and two militia cavalry units are placed on line, with one silver shield grouped with other militia cav. Others are ungrouped. Single units and the group are ordered, at the same time, to run to a line some ways away. Grouped units stay in form, militia cav slows down and the silver shields accelerate. The ungrouped units travel at normal speed. As a result, ungrouped militia cav is the first one to arrive, grouped units are second, with the grouped silver legion leaving the ungrouped silver legion significantly behind.

[Myrddraal]

If the manual states otherwise its a bug. I'll add it.

I'll put the grouped unit speed on hold till I can find out more.


Has anyone else experience the difficulty level bug? Any research?

[SpencerH]

Don't retreating armies get extra movement points too? I think this is a feature.

I think they do but, it doesnt appear to be near the same extent and so doesnt cause the same (irritating) glitch. I've attacked fleets near the nile that retreated toward Crete/Cyprus, were attacked again and retreated toward Crete, were attacked again and ended up near Greece. Its too much especially when the movement ignores ZOC. Naval 'strategy' becomes meaningless.

[SpencerH]

Hm, I dunno if we should include that. Its not a feature which doesn't work, its not really an issue, its more of a wishlist. Oppinions?

Sorry about that Kraxis, I've added it now.

The issue is that player (but not AI) units will stop in the middle of a battle instead of pursuing an enemy that is moving close to or even through the stopped unit. I find it especially noticable with horse archer varieties and cav (since my inf units almost never catch the olympic caliber sprinting of the routing units).

[Red Harvest]

Myrddraal,

Please add the charge effects/charge bonus to the "issues" category. Those of us who have tested it agree that it doesn't work in a logical fashion. While it may be intentional, it doesn't work right and results in cataphract archers having more powerful charges than heavy lance armed units that were designed for charging.

The horse archer bug also afflicts chariot and elephant archer units. Also, the skirmish description for it is incorrect IIRC. The units fail to fire on the move, whether in skirmish or out of skirmish. They will fire once, then stop firing or go through th motions but fail to fire. Issuing a new set of orders will get them to fire once, then they get stuck again. As a result sometimes the AI can keep them firing this way since it continues modifying orders, but it is very challenging for a human.

Pila Animation Halting a Charge I would describe as a bug since its effect is very wierd and reverses what should happen. It should certainly make the issues list if it isn't on the bug list.

[Myrddraal]

K, I'm trying to stay on top of this. Any help from another mod would be appreciated...

[Shambles]

If the manual states otherwise its a bug. I'll add it.

I'll put the grouped unit speed on hold till I can find out more.


Has anyone else experience the difficulty level bug? Any research?

In real life they would usualy put slow mooving Units at the front
Then they would force faster mooving units to slow down When grouped in formation,

I think that should be the case for the game as well,

Honestly a slow mooving unit should not be able to moove faster becous you grouped it with a faster mooving unit,
Id say its a bug or glitch.

[Puzz3D]

In city streets, all AI units incur fighting fatigue even when not actually fighting
When I am defending a city street and many AI units come down that street to engage me, all the AI units are designated as "fighting", thereby incuring the fighting fatigue rate eventhough only the front unit is actually fighting my unit. The other AI units are not engaged, but are blocked from engaging by their own men. The effect of this is to cause all the AI units to become exhausted which lowers their morale, fighting ability and movement speed. Usually this results in all of the AI units in that street routing when the unit up front that is actually engaged routs. The AI has no chance unless its lead unit can break through fairly quickly. This is particularly noticable in RTW v1.2 because the fighting fatigue rate was increased.


In cities, the AI doesn't use ranged units effectively
The AI doesn't position ranged units at the head of a street and shoot down the street in an attempt to damage the enemy units blocking that street. It will often use these ranged units in melee when they would be more effective shooting.


The AI is too likely to use the general early in the fight
This is disastrous for the AI when it has a general with command stars leading the army. It's disastrous for the current tactical battle, and for the long term strategic campaign since the AI general might have gained a command star if he had stayed back and his army won the battle, but he is now eliminated entirely from the game. Even if the AI had lost the battle, it would retain its general possibly with one less command star which is still better than loosing him.


Phalanx is not effective enough against cavalry
There is too much frontal pentration of the Phalanx by cavalry forcing the phalanx to switch to its sword thereby loosing its anti-cav ability.


Phalanx game mechanic problems
The level pikes are held parallel with the angle of the feet, and this means they are somehow rendered ineffective when the phalanx is operating on a slope which is curving upwards. The pikes should be raised so that the points are high enough above the ground to be effective.

Sword infantry that is ordered to run into the front of a phalanx to a position behind the phalanx will pass between the pikes and force the pikemen to switch to their less effective sword. In other words, it's more effective to double click on a point behind a phalanx rather than to double click on the phalanx when attacking it frontally.

The tactical AI uses the phalanx as individual units and therefore targets individual enemy units. It exposes phalanx flanks when it does this, and doesn't seem to see the value of maintaning the integrety of the battleline.


Battle difficulty bonus is being added to both human and AI
The melee bonus for hard and very hard difficulty settings should go only to the AI not to both AI and human player. In addition, that bonus should not be entirely a bonus to the attack value. At least some of it, if not all of it, should go to the defend value. The fighting already goes too fast and adding bonues to attack makes it go even faster, and it's stuff like this that is driving away the people who used to play total war series.


Charge Bonus
There is something strange going on with the way charge bonus works, and I'll back Red Harvest that it should be looked into by CA. It's impossible for the community to figure out what's going on with this since no info on how the game calculations are made was ever released by CA.

[frogbeastegg]

The pause and speed altering hotkeys in battle are unreliable.
Many times when I tried to use them the effect would toggle twice, e.g. pausing and then instantly unpausing. It's not a case of holding the keys down too long; this happens even when the keys are only tapped lightly. The problem is not constant; about half the time the hotkeys do work as expected. The interface shortcuts work correctly, but in the minimal UI those buttons are missing. This isn't a frame rate relayed issue either; it happens in both large and small battles.

There was a thread about this a few months back; no solution was found.

[.Spartan]

@Myrddraal - good effort on your part for the community.

I suggest you also look at the CVP's readme for around 20 bugs as well as many issue fixes to add to your list as a well as the stuff we have on the thread at TWC. (personal note: with Therother already active with your support efforts I'm sure he will see to it that at least the stuff in the CVP is covered.)


-Spartan

[TF923]

Can we include specific AI bugs?

On the campaign map, the AI doesn't attack when it clearly has superior forces. For example, Scipii (AI) is at war with Numidia (AI) but instead of attacking they will continue to amass many large stacks of armies on the border that do not attack, or even enter, Numidian territory. This goes on for many years.

Another one--In RTW 1.0 the AI would over-build its naval forces to the extent that it would eventually bankrupt itself if it never lost any of those naval units in battle. This was fixed in 1.2 but now it has gone the other way. Now the AI rarely builds naval units, and when it does they are easily defeated because they won't stack them into large fleets like they used to. (This might be related to Scipii not expanding reliably.)

Brutii (AI) don't expand across the Aegean. Instead, once the Greek/Macedon region is conquered they expand northward into Thrace, Dacia and even Scythia. This is probably related to the AI naval deficiency bug above. Since the Romans never bothered with conquering Germania or "Scythia", I think this should be considered a bug on both a historical basis and the fact that the AI still obviously has trouble crossing bodies of water via ships, unless it is near the beginning of the game, (it appears to be "scripted" to do so.)

The AI never attacks Crete or Hibernia.

On the battlemap, when defending wooden walls that haven't been breached and entered yet, the AI will run its forces back and forth behind the walls, making them exhausted and allowing them to be picked off easily by missile troops. What it should do is attempt to conserve stamina. Run them to a destination allow them to rest up.

On the battlemap, again stamina should be a consideration for all possible courses of action. For instance, when the AI can't make up its mind what formation it should use to defend with, it will wear its units down by running them all over the place including up and down a hillside. No formation can be so good that it would be worth making your units exhausted trying to get into it. The AI should continue to attempt to change formations when it deems necessary, but if its troops are getting too tired it should be less likely to change.

Last one--On the battlemap, the AI will give up a hilltop position by maneuver. If facing an army with no apparent weaknesses, (i.e. the AI has decided it's not ready to charge yet), the AI will allow the player to march up the hill and take it, while the AI retreats off the hill. This maneuver involves the player keeping the lines facing the AI while moving troops sideways up the hill as if flanking. The AI is attempting to prevent a flanking maneuver, but in doing so it gives up its hilltop advantage to the player. Instead, the AI ought to take its height advantage into account while responding to an obvious flanking move. If the player tries to flank up the hill, instead of moving away as it would on flat ground, the AI should hold position and/or attack the player's flanking maneuver while it holds the height advantage, rather than rotate its formation allowing the player to take the hill. Rotating away from the player's flank attempt works only on flat ground. When holding a hill, the AI should take advantage of the hill while it still has it. It's as if the AI only values a height advantage during pre-battle, but during the battle it forgets that there is a hill there and it should attempt to hold that hill at all costs.

[Marcus Maxentius]

I've had two instances of my general freezing and getting slaughtered by other troops. It says he's marching, but he aint marching as far as I can tell. I keep double-clicking and nothing happens. He just stands there looking around. I think he's high or something.

Another thing with generals. I charged my greek general unit into the back of a carthaginian general and did zero damage. Infact his movement was stopped like he hit a stone wall. The other general turned around and killed mine. I guess some leaders are born with an iron ass or something.

[Marcus Maxentius]

Aw crap, I forgot I'm using 1.1. I don't want to waste space with something already fixed. Can anybody confirm, please?

[Simetrical]

I suggest two things:

1) Every single bug needs to either include a save that displays it clearly, or a method that shows it clearly. This is critical. It's not enough to say "it's been tested," you have to specify what the test is so CA can see for itself. There's no point otherwise.

2) Only the most egregious, consistent, and specific AI flaws (e.g., "loading a saved game severely inhibits AI aggressiveness in the first turn after the load") should be mentioned. AI is immensely complex, and you can't fix a problem like "the AI doesn't attack when its forces are clearly superior" the way you can fix a UI problem or an animation problem or whatever. There's no point in including fundamental AI inadequacies here—of course they know that the AI isn't top-notch. We don't need to tell them that.

-Simetrical

[Barbarossa82]

Marian Reforms Bugs

1. Praetorian Guard are trainable before the Marian Reforms take place. This occurs because they are trainable at the Imperial Palace, which no longer immediately triggers the Marian Reforms event. The effect is to improperly give the Romans access to a legionary-style unit before the reforms.

2. Pontic and Scythian generals are not upgraded to their armoured model. This is apparently due to a missing "general upgrade" attribute.


Marian Reforms Issue, Possible Bug

After Patch 1.2, non-Roman factions are completely dependent on the progress of the Roman factions for the upgrade of their generals, rather than being able to effect this themselves by building their highest level governor's building.

[Ab Urbe Condita]

Apparently nobody's mentioned this, but AI armies will sometimes exit the battlefield without a fight even when they don't have the option of retreat, so they end up losing all their units.

[Marcus Maxentius]

About Assasins not getting experience for sabotage. It could make sense since they need the spy's expertise to get information about the target. The spy should get more of the credit. Assassins, assassinate and get points for that. The spy should be the one to destroy buildings get a saboteur rating. How subtle is conan the destroyer with a spiked club?

[Kraxis]

Well since neither get any bonus from it, you can bet it is a misplaced feature. Either the spy was the old saboteur or a coder made a small mistake. Result: It doesn't work as it should, thus very much fixable.

There are many small inconsistencies in the stats of the units. We can supply a lengthy list of them if it is deemed worthy of mention. But none of them can be considered bugs and as such are less important.

About the grouped units' speed:
It is very much worthy of mention, even if it isn't a bug per se. The units act strangely and in a way we in general do not like. So I think it is perfectly ok to include it as a bug so that it has a chance of being looked into (even if only to be determined a working feature). If we let it sit on the sideline there is a chance it will be ignored.
Also we have had it confirmed, and personally I remember a lengthy discussion about it some time ago. So it is far from an unknown entity.