The "New" Frontier of M2TW Modding: Memory Editing

[lolIsuck]

Content has been asked. Let's hope so.

Maybe give the Rome1 one a free plug, too.

The Rome 1 one is already on the front page.

[Drorin]

Another limitation that perhaps you can remove, are the portraits of dead characters at the beginning of the campaign, which currently depart randomly

[irishron]

The Rome 1 one is already on the front page.

I didn't look. Thanks.

[paradamed]

Single man per unit is ABSOLUTELLY NECESSARY, we are talking about balrog, sauron, treebeard, generic trolls, generic bigger orcs etc. Those guys will be even stronger than a unit with 150 man, in fact you need an entire army to defeat sauron itself.

Yes, I belive that after 3 or 4 broken limits its time to release a 1.0 version.

I agree with you. An initial version with the changes below would already be enough for a good start. We could have additional features via other versions of the launcher, just a suggestion. In so many years in the community I already had so many disappointments with cool mods never being released and Im so excited for this new mod I really want to see at least these changes happening.

Here is what I achieved so far:

• changed the EDU units limit from 500 to 1000 (just some memory allocation adjustments and pointer shiftings remain to be done);
• changed the soldiers limit from 4/100 to 1/150 (if I'll find the multiplier being used when you set the CFG option unit_size to a value different than normal we will finally be able to have real one-man units);
• made the executable aware of large memory addresses on 64 Bit systems with at least 6 Gb of RAM;
• optimized the game engine for single core execution.

[leo.civil.uefs]

Yep, I can see single units for the EDU for LOTR, but will the AI end up doing suicidal charges and having those single units die too easily? I could see single units to be spawned as a kind of grudge duel for standard MTW2 Kingdoms too, but using a single unit in a standard mod would just end up dying too easily since you'd have to micromanage their movements PLUS the AI probably couldn't do it effectively. Think of the duel as a tournament between two knights, who get spawned into some terrain and battle, but then you have to script setting the demeanor because it would put kingdoms at war. Then one probably survives and what do you do with that single unit then?

No. A single Balrog can kill an entire enemy army, remember his stats are brutal. Im talking about balrogs, sauron, trolls, not humans. Its not suicide, they will attack and kill everyone, suicidal is stading in his way.

Units with a single man were always present in med2 if you consider a unit of 150 man who got 149 of them killed and there is only 1 guy lasting, the AI can handle it pretty well. Also remember that such units would be very rare, so would be the possible bugs related with them.

Leo, I play a ton of custom battles. In many mods you can configure a custom battle on any spot as long as you can do a reasonable show_cursorstat, then plug that in as a custom battle location. I think the map heights in the Third Age mod are a bit off and that's why the custom battle locations don't match up with the map depiction versus what you see at that map coordinate point. I once sat down and literally found a 100 unique map coordinate points on the vanilla map, made an atlas of what you would see if fighting there (both defensive and offensive sides) and it matched up very well. There are some rather spectacular mountain battles in passes or on cliffs where cavalry can actually be momentarily aloft by charging down a peak. That doesn't work in the Third Age map though.

So regardless if a terrain is mapped as impassable or not, as long as you play with the show_cursorstat and plug this into the battle editor to take a peek, then you can configure it for a custom battle quite easily.

I said nothing about custom battle locations don't matching up with the map depiction versus what you see at that map coordinate point.

Im talking about a well known hardcoded limit that will flat the battlemap terrain if the battle is set on a coordianate located in any heigh above 80 on stratmap. So in my submod there is settlements above mountains and you can also walk on some mountains but any battle fought there will have its battlemap totally flat.

[Discord1996]

There is some weird sorcery that makes huge hills flat battlefields
I am really happy about your discovery i will rep you when i will get to the pc.

[RubiconDecision]

I said nothing about custom battle locations don't matching up with the map depiction versus what you see at that map coordinate point.

Im talking about a well known hardcoded limit that will flat the battlemap terrain if the battle is set on a coordianate located in any heigh above 80 on stratmap. So in my submod there is settlements above mountains and you can also walk on some mountains but any battle fought there will have its battlemap totally flat.

This last bit about the limit conclusively explains why I couldn't set up custom battle locations on those map coordinate points in Third Age. If that can't be fixed by the proposed memory editing launcher, then I hope that the Third Age folks reconsider the map heights. Why? Because you can fight on beautiful mountain peaks in the vanilla map, as well as fighting in around a deep valley lake in another part of the vanilla map in Anatolia as I recall. It sounds like purely a scaling error. Yep, I've definitely seen the odd plateau on custom battle map locations in Third Age.

I'm not a mapper, just a custom battles fanatic, and have explored several maps extensively looking for excellent and bizarre map locations. Places that are so defenseable they might as well be impregnable locations to try to assault. Such places would be excellent to fight upon normally while playing the campaign. There are a couple of issues for anyone not familiar with these kinds of rare locations. Usually they are inaccessible such that those mountains are set up to be impassable and so you skirt around the mountain. Next, it's a matter of the attack direction upon that point. Unless a named character is located upon that very spot and held in place with the locked trait (a feature utilized in Third Age) then the defenders might move to another location when attacked.

For those who don't know, the map is made up of a grid in effect like this:
1 2 3
4 X 5
6 7 8
Where X is the defending spot with eight different ways to attack that spot. As such, unless the named character is "locked" then the defenders tend to move to even higher ground in order to better defend. Since the mountains are often inaccessible, the defenders will tend to seek the highest possible ground based upon their remaining movement points.

An excellent mapper might be able to create mountain passes in which armies could retreat to in order to evade the attacking army. Since there are thousands of map coordinate points, it takes someone to explore a map looking for ideal places, to then suggest to the mapper to alter it slightly to make part of those mountains accessible. If the mapper makes it all accessible, you'd have lots of these kinds of battles, but probably people would bellyache when attacking defenders so high up. The attackers would lose a lot and if the defenders had cavalry, they literally can mow down the attackers as they ascend to the defending spot. The attackers thin out into a very weak inverted V formation as they ascend while the defenders with cavalry can charge into that weak formation and annihilate them. Making such mountain passes or peaks in which smaller kingdoms can retreat to would make them last longer than retreating to settlements. People who autoresolve would never see these situations.

Sure, the new memory editing launcher might be able to eliminate the map heights limit, what I'm saying is that it's entirely possible to fight upon mountain tops NOW. Besides these in the vanilla map, there's some excellent weird cliffs in the vanilla map where France borders Portugal (the Castile portion of the map) as well as in Burgundy.

Since a fort placed upon these artificially alters the terrain, then a lot of really great map points get ruined by the addition of a fort, for it flattens the terrain. A good mapper might actually explore those regions, and then make it a custom tile instead with a fort, so as not to ruin that map terrain.

Third Age's campaign map appearance and the way that it resolves when depicting terrain looks remarkably different at numerous places even river crossings. Maybe if the upper limit can be removed, then the terrain will be depicted more correctly?

[DavidtheDuke]

By the way, I believe this and the RTR version should have their own forums. It's obviously the best potential news in years regarding MTW2 and TW in general (especially since I don't have much faith in the upcoming Warhammer TW anymore).

[UndrState]

FYI Looks like our friends over at the Citadel (a.k.a. totalwar.org ) will be making an announcement, site wide.

[Wicked Moose]

This is GODLIKE. Best luck with the project!

Any chance of targeting ground with artillery? That would be pure gold for renaissance/pike&shot mods imho.

[RubiconDecision]

In the descr_walls, you have levels of them, and these are configurable to both alter the strength of the walls, portcullis, and door, but ALSO the levels of the towers in which either roboarchers, ballista, or cannon are shot. Now if the launcher could memory edit it such that you could have say 6-10 levels, then we could create some very interesting robotowers indeed. We could have basic arrows that are not very accurate, better regular arrows, bodkin arrows, crossbow bolts, ballista, mangonel shot, catapult shot, trebuchet shot, and then cannon shot. If we could vary it up, then taking a settlement would be much much harder. This most often would help the AI and it would mean planning with multiple armies in order to overwhelm the AI. I'd love to have that if possible.`

See a picture in this post for the way I configured the roboarcher main towers to fire trebuchet shot, which is deadly to the attackers siege towers. Meanwhile the other robotowers were firing ballista.
http://www.twcenter.net/forums/showt...1#post13851626

[AntonisTheGreek]

In the descr_walls, you have levels of them, and these are configurable to both alter the strength of the walls, portcullis, and door, but ALSO the levels of the towers in which either roboarchers, ballista, or cannon are shot. Now if the launcher could memory edit it such that you could have say 6-10 levels, then we could create some very interesting robotowers indeed. We could have basic arrows that are not very accurate, better regular arrows, bodkin arrows, crossbow bolts, ballista, mangonel shot, catapult shot, trebuchet shot, and then cannon shot. If we could vary it up, then taking a settlement would be much much harder. This most often would help the AI and it would mean planning with multiple armies in order to overwhelm the AI. I'd love to have that if possible.`

See a picture in this post for the way I configured the roboarcher main towers to fire trebuchet shot, which is deadly to the attackers siege towers. Meanwhile the other robotowers were firing ballista.
http://www.twcenter.net/forums/showt...1#post13851626

Well,just imagine a tower which shoots Greek Fire,or Super Banana...
Pity to the poor soldiers who attack a settlement with such towers...

[RubiconDecision]

Well,just imagine a tower which shoots Greek Fire,or Super Banana...
Pity to the poor soldiers who attack a settlement with such towers...

As I configured it, the rate of fire for the trebuchet is slow, and so instead of a head-on attack, you attack the corners and so avoid the trebuchet. But if there were many levels in this memory editting launcher, then yes, you could have all of the towers fire trebuchet as long as someone was manning the parapet around it, and as you say...Greek Fire. Why shouldn't the Romans (Byzantines) have some really fantastic defending artillery for Constantinople. Makes sense to me.

If we make the settlements harder and harder to attack, then we also need some really bright person to reconfigure the BAI too. So far we've had garrison scripts for settlements under siege. It might make sense to have a spawned script for the attackers when sieging the human player under certain circumstances.

EDIT1:
If it were possible to alter the size of the battle map, or to alter the distance the armies can deploy to initially for sieging, then attackers would have to inch their way up, and so the defenders might sally with cavalry, especially horse archers. It could be a game changer versus initially deploying so CLOSE when attacking. As it is now, the attackers have too much of a surprise attack, when in reality they should be way out at the edges of the battle map and work their way closer. If a clever person configured a lake around many of the basic settlements, then attackers would have to navigate this, which limits the approach, or even makes the attackers funnel in a column of troops, which of course makes the above altered roboarchers pummel the invaders even MORE.

[xHolyCrusader]

Changing deployement zones is possible with iwte afaik.

[wilddog]

Changing deployment zones is possible with iwte afaik.

In IWTE you can set the settlement defensive deployment zone. Its inbuilt into the settlements. As are the levels of tower upgrades for example. Also things like settlement level determines the size of ladder to bring to a siege (as some people have mentioned having same settlement levels but with differing wall heights so would probably wonder why their ladders would look out of place!).

[Garensterz]

So, is it now possible to implement Germanicu's Really Bad AI Battle System to Rome Total War? Or maybe mod the Battle AI in RTW like Germanicu did?

[RubiconDecision]

I think that the problem with deploying the attackers is the BAI will attempt to get them as close as possible. What might happen is maybe we could have a bigger zone around the periphery of the battlemap models in IWTE (that borderline where the attackers get to deploy) and so they would be further out without having to adjust the BAI. I believe that it measures from that borderline.

The further out from the settlement, then based upon the range of the artillery ammunition, then the robotowers will fire many more. As it stands now, the attackers are too close and so not as many arrows/ballista/cannon balls can actually hit them before they get too close. The further back the attackers, and the better the artillery ammunition, the more opportunities to burn up the siege equipment.

Now this is really great, for on the successively harder levels, then you would literally have to bring siege artillery and not count on your manufactured siege towers/rams/ladders to help you get within.

Naturally this will only come into play in battles you elect to fight in and not any kind of autoresolved battles. This only helps the AI defend much better.

Now think back to that scene with Saladin and Balian at the Battle of Jerusalem. Saladin's trebuchet cannot fire as far as Balian's since Balian's trebuchet are raised up higher. It's risky to bring up the attacker's trebuchet for they will come into range of the defender's trebuchet far sooner than they can fire upon the defender's walls.

But Saladin's trebuchet are moving targets while Balians's towers are fixed. This means that a lot of trebuchet fired from the defender's will miss, and they could expend too much ammunition without destroying enough UNTIL the attacker's trebuchet are also fixed into range.

A smart attacker might intentionally send fast cavalry in to get the defenders to launch from their robotowers and use up the artillery ammunition, and THEN move up their own trebuchet and take out the walls or the robotowers. Those cavalry will be too fast moving and so trailing shots almost all of the time, and use up the defenders robotower ammunition.

[leo.civil.uefs]

Just added the following item to my list on the second post of this thread:

- Ajust screen resolutions for widescreen monitors.

[Hexdragon]

Congratulations.
I am still looking for my jaw in the floor

This project is the best thing I have read in years.
M2TW is my favorite TW Game, and the scope of this project is amazing.



It would be great to have opportunity to help and bring my two cents.
My coding skills are incipient but I have some practice analysing and editing files with hex editor .
I have yet to learn how to use IDAPro and OllyDBG, however it is anyway on my agenda for personal project related to my all time favorite game: SSI Fantasy General

[xHolyCrusader]

Just added the following item to my list on the second post of this thread:

I dont know what u exactly mean with this, but can't you create custom resolutions in nvidia control pannel?