Matrix Games Forums

Hi All,

We have a new update for you to look over. It includes a number of improvements and fixes, including a couple crash fixes and improved AI for sieges. The update is in the Members Club. You can access it by registering Hannibal, if you haven't already. It will appear on your Members Home Page with a link under 'Downloads'. Please remember that this is a beta, and any feedback will be appreciated.


Change History:
v1.02 – October 11, 2010

• AI Changes[/b]
  • AI improved knowledge of capturing border cities to avoid Mountain Attrition. This occurs at the Normal and Hard level.
  • Improved decision-making by the AI for managing sally battles, relief battles, and defending a city against storm. The changes for this are not too dramatic.
  • Improved decision-making by the AI for judging the size of a relief force when Rome is besieged.
• Game Fixes and Improvements
  • Corrects a crash related to an obscure and apparently rare path-finding bug.
  • Corrected the crash reported when the game is run on a Japanese OD.
  • Some small play-balance improvements at the Hard level.
  • The number of Roman Option Cards currently held is now shown in the first panel in Rome’s city report dialog.
  • Theater permissions improved so that an army may travel from Africa to Spain when the current theater is “Reinforce Gaul” or “Send Forces to Gaul.”

Looks good, looking fwd to giving it a crack.

Hello Again,

This has been updated and the new beta, HannibalRC-UpdateRelease-102-Beta2.exe, has replaced the old and includes the following additions;

Fixes

• Fixes a rare crash that can occur when the Gallic Insurrection card is played under certain conditions.
• Fixes a rare crash that occurs when Roman leaders incorrectly get a duplicate army created. This bug is very obscure and the precise conditions leading to it are still being investigated.

AI Improvements

• Makes it easier for the AI to get its 5th Roman leader when things are going badly for Rome, especially at the “hard” level.
• Makes it easier for the AI to elect better Roman leaders when things are going badly for Rome, especially at the “hard” level.
• Some minor improvements in the decision-making logic for Roman generals.

Changes

• Reduces the base value of the leadership bonus from 20 to 15 percent; this brings it into line with the printed manual. The value for a unit’s Attack Bonus remains at 20 percent which is a change from the printed manual.
• The base for leadership bonuses in camp battles is now set at 10 percent. This makes the camp battles, in a sense, about halfway between a pitched battle and a city battle. This change is designed to make it harder for Hannibal to attack relatively small armies by virtue of his large command rating.
• Reduced somewhat Hannibal’s personal influence with the Carthaginian Senate. It now takes a greater number of victories for his influence to rise much above 10 percent. This only applies to the “Hard” level of play.
• Tweaked the recruiting pool in Zeugitania to be more in line with the printed manual. This means more Numidian Cavalry are likely to become available, and fewer naval squadrons and elephants will usually be added as the pool grows. An elephants unit has been added to the initial recruiting pool at the start of the game to compensate for this. (The Introductory level already started with an elephants unit available – now all levels do).
• Rule change: The Romans may recruit two units in Rome even if the Romans don’t control any of the major cities in Western Italy. They may only do this if Rome is not under siege, of course. This is a change from the printed manual.
• Increased slightly the terrain advantages for provinces outside of Africa. The makes it a little easier for leaders to decline pitched battle in those provinces.
• Reduced slightly the chance for pursuing cavalry to score hits. If the defenders do not have any defending cavalry at all, the pursuing cavalry will still get a combat bonus of 20 percent.

Unfortunately, I have found a simple but serious error with the 1.0.2 beta 2 patch. It manifests itself when Rome is under siege and it's quite likely to cause a crash. [:@]

Naturally I thought that I had tested this sufficiently before I released it, but clearly I did not. So I must advise people NOT to download this new beta patch.

I will get together with Andrew or Erik on Monday and we will make available a "beta 3" patch for those who are interested. That will correct this problem. I will have to check with Andrew and Erik to see when they can prepare the installer. We'll make it available as soon as possible.

BTW, I want to note here that the reason why the beta 2 patch was delayed until last Friday was MY schedule, not Andrew's or Erik's. So if anyone should be the target of your anger, it's me. [&:]

Edit: this particular error only affects game played at the "Hard" level.

This has been rebuilt and replaced for download in the Members Club, as Hannibal v102 Public Beta 3.

It corrects the Crash on Hard that can occur while there is a siege on Rome.
The base for leadership bonuses in pitched battles has been set back to 20 percent. The printed manual is incorrect and I will issue an addendum to the manual when 1.0.2 goes final. I decided that this one change didn’t make sense.
A couple of very minor improvements in the AI’s decision making logic for certain field operations.

I have now played a few games with the lastest beta patch and I thought I would give some feedback. First off I want to say again how much I'm enjoying this game, and how impressed I am with the work you've done on it, before and after release. Truly an excellent job.

The game is a lot tougher now, the hard level is really hard, which is a good thing. The AI seems more aggressive in terms of recruitement, in both land and naval units.

I especially enjoy the ability of the AI toward the late game to back me into a corner : I have a big enough army to beat the romans and then siege Rome, but the AI avoids combat, so I get stuck. I would get defeated one way or another if I tried to siege Rome, and if I were to split my force, the romans could just move whatever force they want in Rome through naval movement and crush my force sieging the city. I lost several games this way (because usually when that happens I'm runing behind in terms of victory points). Brute force or sheer numbers don't ensure victory, which is definitely a very good thing.

There is however one game where the AI made a very big mistake : I had an army with 33 units outside of Rome, and the romans had 2 armies : one of 15 units in Rome, and one just outside with 19 units. Usually the AI is pretty good in evaluating these kinds of situations (that's how I get backed into a corner), but in this instance it decided to move the units which were in Rome, to take back some city in Northern Italy, thereby leaving Rome empty. This is how the situation looked when my turn came :


So if I attack the 19 units army, it declines combat, as it should, but then I'm free to storm Rome, which I can do successfully. (I only have a save game right after the roman army declines combat, I don't know if it would be of any use). It seems the AI sometimes attaches too much importance to the retaking of cities in Italy, instead of concentrating units and confronting the threat to Rome. The situation in the screenshot is exceptionnal though.

Another rare occurence that should be adressed nonetheless IMHO : the election of new (and different) leaders can leave a force without leader. Most of the time the romans move their armies in a coastal city, so no problem, but I have seen a force of 7 units left in Corfinium whithout a leader. I could siege it with just 1 unit without any risk. I think that kind of thing shouldn't happen. (I don't have any saves illustrating that though, sorry).

I like the change made to the pursuit phase of the battle.
I'm not a big fan of this however : "Reduced somewhat Hannibal’s personal influence with the Carthaginian Senate. It now takes a greater number of victories for his influence to rise much above 10 percent. This only applies to the “Hard” level of play." I like that feature (the way the Senate works on hard level), and the additional dilemmas it creates, but the Senate felt independent enough before. And I would really like to see more coherence in its decrees. (If it allows me to send troops to Gaul, it should allow me to send troops to Italy in the subsequent turn(s), no sending troops in Sicily when I have full control of the island, or to Gaul when I control Cisalpine Gaul or at least Genua, etc.)

A final minor, very minor thing : when I'm on the defending side (during the ai turn) of a naval battle with more than 8 units, they are not placed automatically on the front line. Even though in this case I don't have an admiral, and therefore all units are interchangeable. It's really just a very small detail, just thought I'd mention it.

Basically, you are correct. [:)]

You're especially correct that the AI is sometimes tricked into leaving Rome insufficiently guarded. And sometimes it fools itself! You are also right about garrisons with large armies not always getting a leader after elections when they should.

I am trying to finalize this patch this week. When that is done I will start working on version 1.1 and I will be able to implement changes to address most of the things that you have mentioned here. I am going to reset the personal influence of Hannibal to something close to what it was in version 1.0.1. That will be in the final patch 1.0.2.

Now as for the Senate, yes it's sometimes silly. But that is meant to be a serious handicap. and when the Senate authorizes forces to Gaul that may be for recruiting. But some additional coherence or intelligence is something which I'll seriously consider for the future version 1.1. That business with automatic deployment in naval battles is spot on. It will have to wait for 1.1 but it's on the list.

Thanks for trying the "beta patch" 1.0.2 and providing such well-written feedback. It is greatly appreciated.

Just to clarify my thought about the Senate : the issue I have with certain decisions it makes, is more about immersion than gameplay. The way I understand the feature, it is supposed to recreate the situation of a political body disagreeing or interfering with military leaders. But when it makes silly decisions, the "illusion" is broken, I'm reminded that's it's a random thing, and not something that is thought out.

It's not that I want the chances of obtaining what I want increased ; but just that when the Senate makes a decision, some options are taken off the table. Maybe by capping the chance that my request goes through (without the use of an option card) at a certain percentage, and spread the rest between the remaining sensible options. For example, if the chances between different options were as follow : 25% for my request (Italy), 25% for Sicily, Spain, and Upgrading defenses, each ; they could be changes to : 25% for my request and 75% for Upgrading defenses (if I control Spain and Sicily). Something like that.

Although now that I think about it, I'm not sure this could be feasible. At any rate, I just wanted to clarifiy what I had in mind. This is really just about immersion, so it's definitely not a big deal, and even if it were doable (and provided I'm not the only one who feels that way), it certainly shouldn't take priority over anything.

Chiming in on the issue with the AI leaving Rome insufficiently protected when I have a big army in Italy.

It's an issue I run into frequently. Rome is often left insufficiently guarded when I have a monster (30 - 40+)Carthaginian army within striking distance.

I've played the Hard difficulty level 20+ times and have won almost every time...many of them due to this weakness in the AI. I just finished a game where I was in deep trouble: I had an army of 31 in Eastern Italy, and the Romans had taken Spain, Sicily, every city in Italy except the one I had just taken (an isolated garrison of 12 that they had left to die in the siege). They combined a few armies in Pisa that totaled 23 or so and left Rome with only the 8 defenders. I swooped in and assaulted the city with 31 units as my last best hope of winning (I had almost no hope of getting new ground troops anywhere, and surly would have been ground down through attrition.)

You both make good points. I can't change this in the 1.0.2 patch but I will definitely do something for version 1.1. The Carthaginian Senate is a more difficult challenge. Getting the AI to protect Rome better isn't that hard. I guess that I didn't anticipate how often this particular weakness would come up. In any case, these two things are on the list for improvements in the (near) future version 1.1.