mod J & P Rebalance Mod by JonnyH13 and Pandakraut 05/06/2023 1.28.4

[pandakraut]

On 1/9/2021 at 5:50 AM, ug_might said:

1) Officer experience gain
As this is my first run with the JP mod, I don’t know if its intended, but officer promotions are glacial slow. I fought all the way through Malvern Hill and ended up with only about 10 Brigadier Generals and no Major Generals.
I can get officers to full bird Colonels reasonably fast but not beyond.

Officer experience gain was intentionally slowed down, so that there is actually a sense of progression throughout the campaign instead of having entire corps worth of brigades led by MGs. The mod changes the xp thresholds to hit various ranks. Getting to Colonel is very quick, but it will take a while to get to BG, and then take a similar amount of time to hit MG. Note by xp I mean an internal total xp for the officer, not the xp progress bar shown in the UI. That shows xp towards next rank rather than total xp.

Many players are getting their first MG around the 2nd bull Run mark, though the timing will vary depending on officer losses, how much focus is put on farming xp for officers, and if you get lucky and get a high xp BG in the barracks early on in the campaign. Depending on how close some of your BG are to MG and the various other factors, you might just be a bit behind schedule.

Corp command and division command give flat xp per battle, grand battles will give more. Corp command is relatively good xp but won't be as fast as what you can achieve on a unit where you're focusing on gaining xp. Officer xp goes up as a units stats go up. Stats go up slower the higher they are. So the fastest way to level an officer is put them on a recruit unit, and then get that unit lots of kills and time spent walking around. The xp bonus from multiple battles led stacks counteracts the xp slowdown as the stats go up a bit, so finding the optimal point of when to switch officers around will not be consistent.

On 1/9/2021 at 5:50 AM, ug_might said:

2) ‘Battles Led’ doesn’t work
I think this is a J&P addition? Been a while since I played the base game. I doesn’t work – nearly all my officers show only 1 or 2 led battles, even when I used them since the start of the game. I suspect this could be related to the perceived experience issue mentioned above.

Battles led is a completely custom mechanic to the mod. The battles led for a given officer is tied to a specific unit. The officer that ends the battle in command of the unit gains the led status for that battle. So if your officer is wounded, in their first battle commanding a given unit, they will still be at 0 after the battle ends. This is more a technical limitation of how we managed to get this to work than an intended restriction.

If you have an officer who has completed multiple battles with unit A, and then you assign them to unit B, they will start out at battles led 0 with unit B. If you swap them back to unit A they will retain their battles led for unit A. If that is not happening let me know and we can dig into what is going on further.

On 1/9/2021 at 5:50 AM, ug_might said:

Random thought, could a reload screw this up? I usually save and reload quite often during a battles.  
Maybe the game doesn’t account for experience gained before a reload and only counts battles if there aren’t any reloads?

As far as I know, saving and loading doesn't break this system because it gets updated on battle end. Saving and loading does lose wounded officers as we are unable to add them into the save file currently. You'll still keep the replacement officer, but if this is happening repeatedly it can definitely be a major hit to your officer pool.

On 1/9/2021 at 5:50 AM, ug_might said:

However, on the third day of the battle the enemy kinda just respawns again and I’m facing another full blown CSA army, including many units wiped out the day before.

This is a somewhat rare bug caused by in battle save/loads. What seems to happen is that all units that were shattered get restored to full strength when you go to the following day of a multiday battle. The only solution I've seen is to restart the battle, though perhaps going to an earlier save might work. I don't save during battles very often myself, so I've only run into it with single saves when testing other things. Otherwise you're stuck killing them all again.

Hope the above provides some context. Glad to hear you're enjoying the mod overall, if you have any follow up questions feel free to ask.

 

[ug_might]

Thank you for the response, some more feedback.

Officer experience gains is just too slow for me. Limiting the number of higher ranked officers is a good idea, but with your settings, I’m just ending up with an army led by Colonels for most of the Campaign. I’d prefer brigades being led by, well, brigadiers, but it’s looking like I’d need to advance past Gettysburg to make that happen. Playstyles vary of course, so maybe I’m just not that into maximizing experience gains.

I take it the ‘Battles Led’ has no impact on anything? I do switch officers around quite a lot. At Chancellorsville, I have two guys with 5 Battles led. The vast majority has 0 or 1 battles. But the system does seem to work like you describe.

About the 2nd Bull Run bug, I’ve tried going back to earlier save but that didn’t work either. It wasn’t an issue in this specific case, since I just ended the battle on day 2 and 30k kills instead of 40k or 45k. Later on, this might be definitely game breaking though. I routinely play legendary, but I’d be hopelessly lost without saving.

 

So anyway, I progressed up to Chancellorsville and tbh I’m not sure I want to continue. The army is in good shape and I can field 100k and 500 guns, but playing is getting less and less fun as of late. Basically, it already feels like I’m playing the grind game post Chickamauga.

The major issue I’m facing is performance of all things. On day two at Chancellorsville, when the map opens up and I’m have 3 Corps and 120 brigades on the field the game slows down significantly and gets very sluggish and unresponsive. My PC is fine (5800X, RTX 2080, 64GB 3600MHz) - the game just doesn’t seem able to cope. Somehow this was way less of an issue playing without the mod, dunno whats up with that.
In any case, I just hate playing this way and with performance as it is I just can’t perform at a level good enough to beat the AI on this map.

But back to mod related issues I’m facing – I don’t really like how the combat is shaping up at this point in the game. This is entirely subjective of course. Basically, everything is fine but the line infantry. It’s both overpowered and severely lacking at the same time. It feels like the enemy infantry is either close to unstoppable when charging or can be ignored when just firing at me.

I don’t know if its true or not but the AI seems to charge every chance it gets in your mod. Way more than in the base game. And it’s hilariously effective. I’m sitting there in 100% tree cover with two brigades at 3k men total in good condition with interlocking fire and artillery support – get charged by a lonely 3-star (what else) 2k brigade over open field and my line just freaking gone if I cant stop them before melee.

On the other hand – put two units into even just semi decent cover and they can blast at each other for the remainder of the battle without any significant result. In the base game my units would suffer considerable casualties in sustained rifle combat, even in good cover. You could only hold on for so long before being grinded down. This just doesn’t seem to be the case in the mod. Unless you get charged, everything is fine. Just hold the line and blast them to pieces with your artillery.
What I frequently end up with is extremely high kill counts on the artillery while my infantry sits at meager 3 to 1 kill rates. The issue here is not that artillery is too effective (ok maybe it is) but that line infantry just can kill shit, especially on lower stats.  

And that’s before scaling hits you. On the first day at Chancellorsville, I’m facing 5k infantry 3-star brigades for some reason. 90% of my line infantry has 2k or less and at best 2 stars.
I can’t hold them. I mean I can – easily enough even, since they are completely ineffective when just firing at me – but when they charge there's not much stopping them.

Those 5k 3-star units charge through intense artillery fire and can easily take 1.5k to 2k casualties before finally routing. Scaling difficulty is fine and all but at some point, it’s either gaming the system to high heaven to achieve remotely manageable results, or facing odds way past the point of ridiculousness.

What to do about it? I don’t have any idea if that’s even possible, but here are some thoughts on how I would change combat:

1) Make the AI charge far less and reduce the effectiveness of a charge. Maybe double the time it take to get a unit ready to charge
2) Improve overall infantry rifle damage. Especially lower stat units should be considerably more effective. Really high stats units are probably fine but you only have very few of them anyway.
3) Improve cover effectiveness. Or rather – cover itself is probably fine. But units on open field with next to zero cover should be decimated by sustained enemy fire. Units caught in the open should retreat sooner. Advancing against prepared positions over open ground should result in extreme casualties.
4) Reduce Artillery effectiveness while increasing range by at least half. I could write pages on artillery but an increase of damage of the infantry should be met with a nerf of the artillery to keep casualty counts manageable. To counter this nerf, range should be increased.
5) Cap brigade sized at 4k at most. Alternatively give the player the funds and rifles to build up similar sized units or reduce the number of deployable brigades in battles. Cap Skirmishers and Cav similarly. Higher numbers doesn’t equal a better game.
6) Reduce condition/morale buildup after combat all across the board, but especially after routing. It should take hours for a unit that suffered high casualty rates to regain combat effectiveness.

To be clear i do think the mod is great. The early game is much more interesting than the base game and especially the focus on condition/morale is awesome. But progress beyond a certain point and it just expands into a grind fest in which artillery slaughters everything. Thats true for the base game as well, but the mod amplifies it considerably.

Some other random points:

I don't really like that basically every rifle has the same range. They differ in effectivness of course but in reality you end up blasting at each other at max range anyway. That's just how the game mechanics work. It was considerably more interesting to have rifles with varying ranges and dps.

I seem to have way less money than usual in my Union game. Granted, i put only started investing in Politics after 2nd Bull Run, but usually i'm drowning in money at this point in the game. This time im limited to buying the best artillery pieces and some scoped rifles. Can't even afford all 24 pound howitzers. If it were up to me i'd make weapons considerable cheaper and but strictly limit their availbility.

The splitting units on spawn feature is great at times and sucks at others. Overall i'm not a big fan. I just end up rerolling spawns to get a decent enough result.

Edited January 16, 2021 by ug_might

[ug_might]

So yeah, a quick update from Chancellorsville - i decided to play it, but i run into the same bug as at 2nd Bull Run. This time its gamebreaking.

I pretty much annihilate the enemy on day 2. I doubt he has more than 20k infantry left on the map and most definitely no artillery. On day three however, the enemy respawns again with no less than 55k infantry and 300 (!) guns. Again including many units already killed before. Espeically the freaking 1.2k artillery units.

There's just no way this can be countered on Chancellorsville day 3. I'm still in decent shape - 82k infantry, 500 guns, down from 100k - but there just isn't enough ground to hold and have an army left when the dust settles.

 I'm out of options. I even pushed forward, taken the two vicotry points on day 2 but this doesnt give you the win, the game just proceeds to day 3 anyway. Did cost me some 8k, otherwise i would have been in very good shape going into day 3. Not that it would have made a difference.

So I guess this ends my run with the rebalance mod. Sucks but it was a great while it lasted. I have more than 1.000 hours on this game (kinda frightening) and this mod managed to provide a very novel experience. So thx @pandakraut

[pandakraut]

Sorry for not responding earlier, sometimes I miss the new message notifications.

On 1/16/2021 at 6:54 AM, ug_might said:

Playstyles vary of course, so maybe I’m just not that into maximizing experience gains.

These settings are somewhat configurable. In the /Mod/Rebalance/ConfigFile you can increase the values for brigadeOfficerExperienceMultiplier, divisionOfficerExperienceMultiplier, and corpsOfficerExperienceMultiplier

On 1/16/2021 at 6:54 AM, ug_might said:

I take it the ‘Battles Led’ has no impact on anything? I do switch officers around quite a lot. At Chancellorsville, I have two guys with 5 Battles led. The vast majority has 0 or 1 battles. But the system does seem to work like you describe.

Battles led affects officer xp gain rate, command stat, and the bonus the officer provides a unit towards their next perk. There is a penalty to these at 0, and the maximum bonus caps out at 3 battles led. I did some checking and all indications are that it is working as expected so maybe I'm just not explaining it well. 

If you select a unit, to the left of their name is a small box that contains an upward line. If you click on this you'll get a list of all battles the unit has fought in. If you hover over the battle you can see which officer ended the battle in command. Unless you have been going through officers at an incredible rate, or are frequently moving units around or disbanding them, these numbers are far lower than I would expect.

On 1/16/2021 at 6:54 AM, ug_might said:

The major issue I’m facing is performance of all things.

A patch with a variety of performance improvements will be out soon. While the engine itself will eventually experience a lot of slowdown no matter how good the computer, some of the recent changes in the mod dropped the FPS more than they need to be. The upcoming patch should restore base game FPS levels.

Regarding charges, we make the AI more able to recognize when it's units have an advantage in melee over the player. We also make the AI more likely to actually charge together instead of one at a time where they can be picked apart. 

The big piece of information I'm missing here is how you are perking and equipping your units. If you're full accuracy spec'd with low melee damage weapons and 20 melee stat you are going to be in for a bad time if a 3* unit impacts on a charge. Especially if that unit has a good melee weapon and charge perks(which would also make it much more likely to charge.) If you have melee spec'd units in your army you can counter these charges fairly effectively in melee.

On 1/16/2021 at 6:54 AM, ug_might said:

And that’s before scaling hits you. On the first day at Chancellorsville, I’m facing 5k infantry 3-star brigades for some reason. 90% of my line infantry has 2k or less and at best 2 stars.

I'm not sure what happened with scaling here, I'm completely unable to replicate those kind of results. Closest I can get with those kinds of numbers are 3k units, which would be much less of a threat. This would be a case where I would highly recommend using the scaling configuration values to just decrease them to a more reasonable size.

On 1/16/2021 at 6:54 AM, ug_might said:

5) Cap brigade sized at 4k at most. Alternatively give the player the funds and rifles to build up similar sized units or reduce the number of deployable brigades in battles. Cap Skirmishers and Cav similarly. Higher numbers doesn’t equal a better game.

Wanted to address this point specifically. Capping the sizes lower would make balancing easier, but it also removes the option for a playstyle that some players really like. We try to make scaling recognize which size units players are using and to adjust to that accordingly, but it doesn't always work out in every scenario.

There is also a config option that allows you to make this change if you want. In the AIConfigFile there are size caps for the main unit types

AIArtilleryMaxSize, 1250
AIInfantryMaxSize, 6000
AICavalryMaxSize, 2000
AISkirmisherMaxSize, 1000

On 1/16/2021 at 6:54 AM, ug_might said:

I don't really like that basically every rifle has the same range. They differ in effectivness of course but in reality you end up blasting at each other at max range anyway. That's just how the game mechanics work. It was considerably more interesting to have rifles with varying ranges and dps.

I think this is the base game tooltips causing confusion again. Every infantry weapon in the base game other than the henry had 300 range. The tooltips were just indicating a relative effectiveness at max range, if I'm being very generous in my interpretation. 

The only max range matters a lot of time is definitely an issue. Though I would argue that moving in closer as the player does have significant benefits, especially on certain weapons. Getting the AI to do that effectively is a whole different issue though. This is something we're continuing to look into.

On 1/16/2021 at 6:54 AM, ug_might said:

The splitting units on spawn feature is great at times and sucks at others. Overall i'm not a big fan. I just end up rerolling spawns to get a decent enough result.

This can be turned off in the AIConfigFile by setting duplicateRandomProbability to 0.

On 1/16/2021 at 6:23 PM, ug_might said:

So yeah, a quick update from Chancellorsville - i decided to play it, but i run into the same bug as at 2nd Bull Run. This time its gamebreaking.

Sorry you hit this twice, wish it was a bug that I could fix. 

Thanks for the feedback overall, sorry again that my responses are a bit late. Glad you got a bit more value out of the game, if you decide to pick it up again at some point we may even have a new version out by then

[JonnyH13]

Updated to 1.27.2

changelog:
Performance Improvements
- Various updates to restore improve FPS in larger battles. The largest battles will still cause notable drops but the mod should perform approximately as well as the base game again.

Balance Changes
- Reload speed for mounted carbine cavalry now has a penalty when not moving.
- Infantry firing arc is 25% wider.
- DN&C damage reduced slightly.
- Infantry weapon costs adjusted.
- Pistol range increased slightly.
- Henry range curve improved slightly
- Minimum xp gained by an officer participating in a battle now increased by battles led.

Text Changes
- Weapon descriptions updated. Thanks to TheSoldier.

Bug Fixes
- Union Gettysburg day 2 save issue fixed.
- CSA Gaines Mill, all supply wagons now deploy.
- DN&C reload speed fixed.
- Whitworth ammo cost fixed.
- Fixed timer at Siege of Suffolk.

[Corentin]

Hello,

Firstly, thank you very much for your mod ! It is really really good !

Secondly, I have two questions :

- Can we change rifle for historical battle (Antietam for example), as there are many Lorentz rifle currently (probably beacause of mod modififation) ?

- Do you think a drop of morale linear to deaths could be possible ? It may reduce number of death as objective couldn't be defend at all cost. Anyway it could be more realistic (I don't think a brigade reduced by 50% will go back to fire easily). But I have the feeling AI will have difficulty to manage this, so it may be a bad idea.

 

I didn't check all conversation, so I hope these questions have not been asked.

 

Thank you again for your amazing work,

Corentin

[pandakraut]

1 hour ago, Corentin said:

- Can we change rifle for historical battle (Antietam for example), as there are many Lorentz rifle currently (probably beacause of mod modififation) ?

Not easily. You would have to manually change the weapon in the assets file for all the units at Antietam. We basically don't look at the historical battles at all in the mod, so they may not play well with it installed.

1 hour ago, Corentin said:

- Do you think a drop of morale linear to deaths could be possible ? It may reduce number of death as objective couldn't be defend at all cost. Anyway it could be more realistic (I don't think a brigade reduced by 50% will go back to fire easily). But I have the feeling AI will have difficulty to manage this, so it may be a bad idea.

The maximum morale a unit can have during a battle does go down as the unit takes losses. Though this is not a 1:1 relationship. I don't know the exact numbers, but I think 50% casualties means the unit can get around 70% morale maximum. We could probably adjust that ratio at some point, though I think 1:1 is a bit too much.

The shatter at 0 morale system that the mod uses does make it much easier for units to shatter or surrender if they've taken that level of losses. So they should be relatively fragile if they get routed at that point.

Glad to hear you're enjoying the mod

Edited February 8, 2021 by pandakraut

[Corentin]

On 2/8/2021 at 11:53 PM, pandakraut said:

Not easily. You would have to manually change the weapon in the assets file for all the units at Antietam. We basically don't look at the historical battles at all in the mod, so they may not play well with it installed.

I try to understand assets file, but it is not easy. I have use Notepad, which is probably not a good idea. Anyway, I cannot find where weapon is assigned to a specific brigade. I found weapon, but for desciption, not for assignation to brigade. Do you have any tips to help me please ?

 

On 2/8/2021 at 11:53 PM, pandakraut said:

The maximum morale a unit can have during a battle does go down as the unit takes losses. Though this is not a 1:1 relationship. I don't know the exact numbers, but I think 50% casualties means the unit can get around 70% morale maximum. We could probably adjust that ratio at some point, though I think 1:1 is a bit too much.

The shatter at 0 morale system that the mod uses does make it much easier for units to shatter or surrender if they've taken that level of losses. So they should be relatively fragile if they get routed at that point.

I didn't pay attention to that before ! This is nice.

 

On 2/8/2021 at 11:53 PM, pandakraut said:

Glad to hear you're enjoying the mod

Yes, really. Thank you again !

[pandakraut]

14 hours ago, Corentin said:

I try to understand assets file, but it is not easy. I have use Notepad, which is probably not a good idea. Anyway, I cannot find where weapon is assigned to a specific brigade. I found weapon, but for desciption, not for assignation to brigade. Do you have any tips to help me please ?

You need to use a hex editor such as hxd to read and make changes to that file. Some examples of how to change weapon stats and perks can be found here https://forum.game-labs.net/topic/26225-weapon-and-perk-modding-guide/

 

Changing weapons for individual units is probably going to be pretty hard. The unit definition probably has a weapon id that would need to get swapped out with whatever you want instead. Finding the Id and then finding the actual unit definition would be a lot of trial and error. This is something we basically haven't done in the mod at all due to the difficulty and amount of work.

[ug_might]

@pandakrautlate reply but

Officer Experience:

Yea I already played with those multipliers. Worked like a charm.

Battles Led:

I definitely move officers and units around a lot. I checked and it seems to work as indicated. I guess it’s fine if you know what’s going on.

Charges:

I’m definitely guilty of pretty rigorously building for ranged combat. My average Melee stat was ~29 at Chancellorsville. Thing is, I don’t think I had much choice. My infantry already came across as pretty weak as explained previously, I felt like I had to push Accuracy and Reload Times come what may. In the end I broke charges with massed Artillery anyway. If you have 500 guns on the field you can break anything. Resulted in pretty lopsided kill counts. At the end of Day 2 at Chancellorsville, most of my main Artillery units have 3k to 6k kills while most of my line infantry has at best a 2-1 kill ratio. All while engaging the enemy in an open field form perfect cover. Very different from the base game and as said, this just result in a passive artillery grind.

Scaling:

https://imgur.com/e1RzL7c

They weren’t all this big of course but it came across as a bit extreme. But deploying all my 20pound artillery with 1^st Corps did the trick. This guy will break after sustaining 1000 casualties in one volley since he’s out of cover. It’s only a problem if more than two of them charge and get into the woods.

As you can see, most of my infantry is well below 2k. I have one meaty unit there to soak up damage. I had one 3k and one 3.3k going into Chancellorsville, resulting from Brigade merging in Camp. Couldn’t get them to unmerge. Everything else had 2k or less. I think the scaling is just the result from having a very high unit count. My line infantry is on the small side, but I fielded 100k men and 513 guns in 111 brigades. The enemy scaled from 76-81k pre battle screen to 127k and just 336 guns.

In the end it’s not that big of a deal if the artillery is well supplied. But is it fun? I’d much rather see two 2k units and smart AI gameplay… … honestly the only really challenging thing about it (apart from the game performance) was the duplicated 1.2k artillery units. Those take time and a lot of ammo to destroy.

Thinking maybe I’ll try turning off duplication and replay the battle. Should be doable, even if I run into the respawn bug again.

 

Edited February 12, 2021 by ug_might

[pandakraut]

12 hours ago, ug_might said:

I definitely move officers and units around a lot. I checked and it seems to work as indicated. I guess it’s fine if you know what’s going on.

It was very recently discovered how to add and increase the size of tooltips, so now we should be able to add more of this information in game so it's easier to tell what is going on and what the benefits/downsides are.

12 hours ago, ug_might said:

Thing is, I don’t think I had much choice. My infantry already came across as pretty weak as explained previously, I felt like I had to push Accuracy and Reload Times come what may.

I also tend to follow this fairly early on, but post Shiloh I try to start building up a few pure melee units. I try to get them extra kills towards the end of battles and in counter charges. A bit of work to get going, but having some solid counter chargers behind your firing units can be a huge help later on.

12 hours ago, ug_might said:

At the end of Day 2 at Chancellorsville, most of my main Artillery units have 3k to 6k kills while most of my line infantry has at best a 2-1 kill ratio. All while engaging the enemy in an open field form perfect cover. Very different from the base game and as said, this just result in a passive artillery grind.

Past a point there is only so much we can do here, if the player brings enough artillery the AI just isn't going to counter it properly. That said, a continued reduction in kill damage for various weapons and artillery in particular is something we're continuing to work on. Mass artillery barrages should absolutely be effective, but we would prefer if the player had to send in the infantry to clean up a bit more. 

Also the end game much past Chancellorsville is definitely a problem and always has been. We could make it harder, but I'm not sure it'd actually end up more interesting. Slogging through fortifications is always going to be slogging through fortifications, and that's kind of what the end of the war ended up as. Maybe we'll come up with something eventually.

12 hours ago, ug_might said:

but I fielded 100k men and 513 guns in 111 brigades.

You're probably correct that this is what is driving things up. It's an issue with several of the battles where the AI has relatively few units, but the player brings a very large force. The AI tries to scale into fewer slots and the numbers skyrocket. We probably need to try to come up with some other way to handle scaling in this scenario. Extra splits when the unit exceeds some ratio to the player's average size perhaps, though that kind of requires figuring out how to get them to show up on the 2nd day which is currently a technical limitation.

12 hours ago, ug_might said:

Thinking maybe I’ll try turning off duplication and replay the battle. Should be doable, even if I run into the respawn bug again.

Hopefully getting rid of the dupes and the new patch improves performance a bit for you. Also definitely look into capping the max AI infantry size if that makes things more enjoyable.

Thanks for the continued feedback, hope you can get around the bugs that have been causing your trouble.

[ug_might]

 

@pandakraut

https://imgur.com/aTRfAIK

1.27.2 duplications off

I'll share some of my experience later. What a battle.

 

Edited February 15, 2021 by ug_might

[pandakraut]

Looks like some nice work there. Would you be able to share your save? I want to try and figure out why scaling is double what I have on my saves. Save files are stored here C:\Users\Your User Name\AppData\LocalLow\Game Labs\Ultimate General Civil War\Save\CampaignBattle

AppData is a hidden folder if you aren't displaying those already. Easiest way to find the right save is to make a new save of what you want to transfer and then sort the folder by date.

[ug_might]

@pandakraut

Yeah sure.  Camp before battle.

https://drive.google.com/file/d/1ftaiZdMUC5zPuJ6nkoJMKzY6qge3-1qC/view?usp=sharing

Just ignore the Lt. Generals, as said i messed with the Multipliers.

--------------

Sorry for the long post.

Yeah so. You just can't beat Chancellorsville playing fair. But I beat it without duplications.

It's considerable easier, but after I played the battle three times in a row, you know what's what anyway. In my last attempt I destroyed the enemy completely on day 2. The only red units left on the map were Cavalry units. I'm still in near perfect condition, lost about 11k (and one whole gun - yay). When I push to day 3 on this save it get 83k and 171 guns against me.

Another save, with more enemy units left on day 2 had 78k and 193 guns. Yet another 68k and 302 guns. 78k, 182 guns. 76k, 171guns. It's actually beneficiary to not go all out on day 2 - if there are more enemy units left, those units just get teleported over and don't respawn at full strength. At least that's what I think is happening. Another factor is if you take the victory points and tbh, just reloading and letting the game a couple of minutes longer gives you a different result without anything happening on the field. Maybe it’s just random after all.

So onwards to day 3. It ultimately doesn't matter if you're facing 83k or 76k. There's just no time to set up properly. I can try and defend at the farm but even if I manage to hold a charge of 30k units spawning more or less directly on top of me, I get destroyed once their artillery joins in. Maybe I could beat it, but the losses would just be catastrophic. Can’t walk away from this in legendary.

So, I thought about alternatives. The timer is 3:26 on this battle and the thing is - it doesn't matter if you lose the victory point in between. You just need to hold on at the end. One obvious strategy (and done before, I just never needed to) is not defending the farm at all, set up in the dense woods to the east, destroy the enemy with artillery and move forward with ten minutes to spare to capture the point. This would work, only for three issues - first it's incredible hard to get all your units to the eastern woods without being overrun by charging infantry. They attack incredible fast. Just 5 more ingame minutes would do the trick here, but unless you get very creative with save/loading to stop them, it probably can't be done.

Second, there is a southern front problem. A fourth of my units is set up south of the farm and there definitely isn't time to move them up. Especially not the artillery. You could try and set up on the southwest hill on the map, but you just end up getting charged and overrun by 12k infantry sooner or later.

Third – ammo. Stocks are very low after two days with 500 guns on the field

So luckily there is a solution. Not about ammo but the rest. Its cheesy but I don’t give a damn at this point. Rearrange the whole Army. Not before battle, although that would be possible, but between day 2 and day 3. Long story short, I picked a save, made careful note of which division spawned were on day 3 and adjusted everything accordingly. Fast, fresh units west and south, everything else as far east as possible. Then run east like hell and use some unlucky units as tide breakers.

This is how it looks like setting up in the east.

https://imgur.com/iwX9DiD

I never had to do this before and I don’t really like it but I didn’t have any other ideas. I mean you could try some really weird strategy and play super, super passive on day one and two, essentially moving on to day 3 with an untouched army on both sides. At least you wouldn’t have to waste ammo and kill all units twice.

So how did it go? All this considering, it went great. I got very lucky, the AI decided to park all of its artillery in the southwest, completely away from the fight. Thus, almost no counter battery fire needed and I put every single gun I had on the infantry. Not 500, since you’re one Corps short and I didn’t have the ammo left, but enough.

The slaughter was intense, the enemy could barely hold on. I’m not sure if I’ve ever seen solid lines of 3-star 30k infantry to break apart that fast. At one point I had a 3 star 3.5k unit charging me completely destroyed barely into rile range. Destroyed - not just routed. That was fun. One trick you need to do is to set up additional threat vectors on the flanks for the AI to refocus on. Once he regroups, he won’t charge you en masse anymore and he is done.

But of course, it’s not just about killing the enemy, the goal here is not to beat him but to have an army left when the dust settles. And I manage that – total losses amounted to 17k and 8 guns.

After you held the farm when the time is up you can continue to play and just wipe out the enemy. You transition to phase 2 but he just spawns with whatever units survived day 2 and weren’t used in day 3 so far – no respawn bug there thank god. So just walk up to the victory points and you’re done.

So this was pretty difficult to find a solution for the respawn bug. Execution more so. If the enemy charges your line with more than two brigades at once you’re done. You need to be very careful about where you place your artillery. Also, lots of work with Cavalry and Skirmishers to go after Artillery behind the lines. Micromanagement hell and very poor game performance.

Took me an awful long time. Think I sunk 15hours into doing day 3 the conventional way alone. Lockdowns are fun I guess.

Could you do it without turning duplications off? Probably. This victory was far from perfect, I guess a run with under 10k infantry casualties wouldn’t be impossible. But I’m not good enough for that. With duplications on I guess I’d be looking at 20k casualties. If I had gone into this with more ammo at least.

So on to Gettysburg I guess.

 

[pandakraut]

 

4 hours ago, ug_might said:

It's actually beneficiary to not go all out on day 2 - if there are more enemy units left, those units just get teleported over and don't respawn at full strength.

You're analyzing this correctly. For whatever reason, when this bug occurs the shattered units respawn as if they had never taken any damage at all. So if you could manage to leave all enemy units intact but beaten down, day 3 would be much more reasonable.

5 hours ago, ug_might said:

it doesn't matter if you lose the victory point in between. You just need to hold on at the end.

I didn't actually realize this was the case, if it wasn't so difficult to get all your units back in time, giving up the point and letting them fight in the open seems like clearly the best strategy. Especially since this situation usually results in the AI not knowing what to do, such as leaving all their artillery in the SW like what happened to you.

5 hours ago, ug_might said:

But of course, it’s not just about killing the enemy, the goal here is not to beat him but to have an army left when the dust settles. And I manage that – total losses amounted to 17k and 8 guns.

Considering the bug forced you to fight nearly double what you should have, this is a very impressive result. Awesome work sticking it out and finding a way through.

5 hours ago, ug_might said:

Micromanagement hell and very poor game performance.

Did you notice any real difference between 1.27.1a and 1.27.2? Or was this battle big enough that the improvements didn't matter? There are almost certainly more improvements that could be made, but they would require a review of a huge amount of base game code. Maybe it'll happen one day, but it probably wont' be soon.

5 hours ago, ug_might said:

So on to Gettysburg I guess.

Part of me hopes you get the bug on the Pickett's Charge phase so the AI actually is any kind of threat there, but you've probably suffered enough from the bug already

I took a look at the scaling on your save and the main thing driving up the AI brigade size is the difference in total brigade numbers. The AI is bringing 49, you're bringing 110. Since you're bringing a lot of brigades and a decent amount of men, the AI tries to scale up a bit and this has an outsize affect when those numbers get stuffed into the smaller number of brigades. Without the bug, the total numbers are relatively even and I would consider the scaling fairly reasonable for legendary.

This is a problem that doesn't have a good solution. We could cause units to split over a certain size so you would face two 2k infantry units instead of one 4k, but this potentially runs into the unit cap and crashes the battle, or just degrades the performance even worse than it already is. 

[ug_might]

I’ve seen no real improvements on performance. Once you have say 150 to 200 units in total on the map performance degrades. Anything smaller is fine. Maybe the new version gave me 5% to 10% better performance on those bigger engagements, but it’s not a noticeable difference.

Yeah so Gettysburg. No changes there, the battle remains a joke on legendary. Potentially because the AI had bad reinforcement rolls after Chancellorsville and had maybe 70k on the Army Camp screen. I did a quick, unoptimized playthrough with duplications turned on again – the enemy deployed some 90k infantry total. Translates to 3.5k enemy infantry at most. No respawn bugs.

I destroyed the enemy on day 1 by defending in downtown Gettysburg and Cemetery Hill. Day 2 Union left was somewhat tricky since my units spawned in very unfavorable positions and I kinda had the wrong Artillery types deployed there. Didn’t try to defend Round Top Victory point but hold the line in the Woods North of the position. Couldn’t get to the enemy’s artillery, took maybe 3k too many casualties there. Lost too many guns too. But it really doesn’t matter since the rest of the battle is just running down the clock on max speed. My artillery kills everything on the Union right and Union center. No Pickett's Charge happening, just mop up duty. I had 11k infantry dead in the end, think I could get it to 8k without much effort.

Chancellorsville was way, way more difficult - no comparison at all.

[pandakraut]

2 hours ago, ug_might said:

I’ve seen no real improvements on performance. Once you have say 150 to 200 units in total on the map performance degrades. Anything smaller is fine. Maybe the new version gave me 5% to 10% better performance on those bigger engagements, but it’s not a noticeable difference.

That's unfortunately what I expected. The bigger battles just grind the engine down after a point.

4 hours ago, ug_might said:

Yeah so Gettysburg. No changes there, the battle remains a joke on legendary.

Unsurprising, the way battles are setup just leads to long multiday battles exhausting the AI. In some ways that battle might be better off if only day 1, day2, and pickett's charge were available. The AI would probably still have too little remaining by the third day, but the Culp's hill phases really are just free kills for the player.

Reducing kill rates even further and allowing shattered units to return on later days(not at full strength like when the bug happens) might help a bit, but currently that's not possible on a technical level.

[ug_might]

@pandakraut

The main issue with Gettysburg is the lack of AI reinforcements for Pickett's charge. If you know what you're doing you wipe out the enemy on day 1 and hold the Union left on day 2 without much trouble. After those engagements, the enemy is just spent when it tries to attack the center. What's needed there is another 10k fresh infantry.

I mean this happened in reality too. Two Divisions of Longstreets Corps arrived at the Third Day of the Battle ot attack the Union left and join the attack on in the Center.

Another factor is scaling. As discussed at length, the AI had 5k brigades at Chancellorsville, while at Gettsyburg i faced just 3k brigades. Probably in part to my crushing victory at Chancellorsville but it makes a huge difference if you need to rout brigades 3 or 5 times to make them combat ineffective.

 

Anyway, Chickamauga.

No bugs. Poor game performance again, actually worse than Chancellorsville when the map opens up at day 2. Luckily all you need to do is to put the infantry on the line, watch out for skirmishers and let the artillery do the work.

Final result was victory on Day 2 (as always, i have never advanced this battle to day 3). Confederate strength 97k infantry and 550guns vs Union 60k and 400 guns. 66k vs 11k (including starting units) losses.

Not difficult just annoying as hell to play. The dense woods are one thing, but the enemy 3star Skirmishers/Sharpshooters were somehting else this time. Usually i can handle them pretty well in a non moded game but this time i took horrendous casualties when trying to hunt them down. I settled for a very passive play style, sneaked in some recon units and blasted them with artillery. Eventually the AI gets very passive and just marches around the Brotherton victory point, so you can bombard the infantry to destruction.

I played with duplications off. First day is way easier if you get a lucky with your starting units. More and but smaller enemy Skirmishers are also easier to handle compared to 750 men units...

Edited February 19, 2021 by ug_might

[dixiePig]

Just a little update on gameplay:

Currently playing as Confederate on MG level

• Infantry : 1000 - 1300
• Artillery : ~14 cannons:  always advance to Horse Arty at first opportunity
• Cavalry : 500 max, slightly fewer for Melee Cav
• Army profile (roughly) : 3 Inf / 2 Arty / 1 Cav, 1 skirmisher unit for every Corps

This echoes historical Civil War profiles in terms of infantry unit size, but whole army is artillery- and cavalry-heavy

I find that the limited infantry unit size makes for winnable battles at the MG level, especially since cavalry and artillery are proportionally stronger.

[i64man]

@ dixiePig, the question I ask, is the AI scaling accordingly to your unit size? or you still facing units with over 2K union troops

[dixiePig]

8 hours ago, i64man said:

@ dixiePig, the question I ask, is the AI scaling accordingly to your unit size? or you still facing units with over 2K union troops

Reasonable question:  AI scaling is very modest at the 1K level - which seems to be an optimal number.  Although it is tempting to build big units, there's little advantage to it.  And there's some satisfaction in fighting battles that reflect historical accuracy ... imo

SIDE NOTE:  I fiddled with values (slightly) in config files, in fear that MG level would be too hard.  The real issue appears to be unit size - specifically for infantry.

The advantage is this:

• small infantry units reduce the AI scaling to manageable levels
• artillery and cavalry are proportionally a more powerful portion of your army

That's particularly important to me, as I use cavalry as my tactical advantage

Civil war tactics are based on maintaining 'lines' of contiguous infantry (supported by artillery) which can mass their fire on an objective in order to wear it down - and ugcw pretty much works that way.  If you can 'flank' the enemy - so much the better.

And that's the value of cavalry.

Cavalry moves fast in a slow-moving environment (all of my artillery is horse-drawn, for the same reason)

You can 'flank' enemy units and take advantage of gaps in the enemy's line

Cavalry really flusters the AI. You can cause major disruption of the enemy line simply through threats and movement.

Then use cavalry to destroy & capture exposed units

While the distraction of the AI allows you to maneuver slower infantry and artillery into positions of advantage

MeleeCav is over-powered, but that works to your advantage . And beware AI meleeCav, which can be particularly annoying in an 'openfield' situation, like First Bull Run.

MeleeCav is vulnerable,  You usually don't have big numbers to begin with.  They can get 'stuck' in attack mode (you can't convince them to disengage). They can tire and take heavy losses when overused.  You want to husband them and use them selectively.

The One-Two Punch:  MeleeCav attacking in tandem with CarbineCav can be devastating. I attack first with carbineCav, which hits and withdraws reflexively. The AI will focus on the carbineCav. Follow up with meleeCav immediately. The AI is still focused on the carbineCav, so you can hit with meleeCav, be more effective, and still disengage.  I often dismount the carbineCav - especially after the second attack - so that it continues to occupy the target unit.

My initial cavalry forays behind-the-lines focus on

1. capturing Supplies
2. destroying Artillery
3. destroying Skirmisher units

Threats to supplies really make the AI go nuts and disrupt their infantry lines ... which can then be effectively attacked by your combined units.

Massed Cavalry : Although the individual units are not that large a group of cavalry units working together can easily destroy larger units. By the time we get to  Malvern Hill my 2 confederate corps will each have a division-worth of cavalry brigades (about 5 units: 3 carbineCav (350 men apiece) and 2 meleeCav (250 men apiece).  One division slips into the enemy rear, destroying their artillery, capturing supplies, and generally causing havoc. One division is in support of the main infantry forces attacking the entrenched union forces frontally.  As the union line breaks from artillery/infantry barrage (and disruption in the rear), massed cavalry exploits the weakened AI positions.  A seeming advantage of massed cavalry attack is that AI is unable to mass its fire on any one cavalry unit - so there are fewer losses.

Cavalry will take substantial losses over the course of a battle simply because it is used so aggressively in my scenarios - but the payoff is huge. The weakness of cavalry is that the losses may be unsustainable ... if you aren't a little careful. 

General rule: Send the cavalry in for non-combat distraction, flanking, and then for the opportunistic mop-up.  If you try to use them for grinding attrition, then you will waste a valuable tool.  Infantry and artillery are good for that.  Cavalry isn't.

[i64man]

@dixiePig thanks for the detailed information... B4 I posted the question, I was on the 2nd BR and it was hard. So, I when back to the potomac fort, and basically restarted the campaign using your recommendations... It makes a BIG difference.. I am no longer facing Union troops in the 2.5 to 3K size units.... like you said, they are a little higher than my 1300 troops per Bde, but it is manageable... and my cav division is a thing of beauty.... 2x mele with 250 and 3x carabine of 400 each.... I am up to Shiloh.... had to save the game in order to go to work, but I am about to capture Pittsburgh landing on day one!

[KeithD]

I don't really see the point of playing MG when players spend the whole time trying not avoid scaling.  Why not just play BG and ramp up the AI experience / increase duplications to make the game a harder challenge?

[pandakraut]

52 minutes ago, KeithD said:

I don't really see the point of playing MG when players spend the whole time trying not avoid scaling.  Why not just play BG and ramp up the AI experience / increase duplications to make the game a harder challenge?

MG adds a few difficulty modifiers that can't be configured at the moment. Such as reduced rewards to the player, better weapons to the AI, increased costs in camp for veterans and officers, and higher reputation penalties for losses. So depending on what you're looking for adjusting size and xp down on MG instead of up on BG might work out better. 

It is possible to play MG and above without paying attention to scaling but it can require a pretty high level of skill to ignore it entirely. Part of this is because we try to support players who want more historical or base game sizes and those who just want ridiculous giant units smashing into each other. The scaling tries to detect what the player is doing so that it's fielding reasonably sized units, but there are certainly some hard to solve outliers. As expected, MG and especially legendary will experience the biggest swings due to the higher scaling modifiers.

Either way, it sounds like the config files are doing their job and letting people adjust as desired. Glad that's working out at least.

[dixiePig]

On 3/8/2021 at 1:58 AM, i64man said:

and my cav division is a thing of beauty..

Congrats on having games that are - hopefully - more satisfying, @i64man. Extend your thanks to @pandakraut, who provided valuable insight on the working dynamics of AI troop scaling, as well as the admission that meleeCav are systemically over-powered. 

Both of those work for my playstyle.  I admit that I prefer a more 'historical' game, as reflected in the 1-1.3K Infantry brigade unit sizing and modestly larger artillery units.  The upscaled cav is just a nice-to-have that makes the game more playably fun for me - and actually reflects earlier  "Napoleonic" tactics, where massed cavalry was often a major battlefield factor.  i don't really know why that changed so dramatically by the 1860's (firepower may've been an issue), but I believe that mounted units were - understandably - more of a battlefield factor in "the West" arena of the US Civil War.

It'd be nice - if unlikely - to have a game option which allowed you to set the game parameters as "Historic" at-the-push-of-a-button.  Unit sizes and army makeup would reflect historic levels, timelines, and profiles (with some wiggle room). This might also result in some subtlety in the Victory conditions.  I can appreciate that some others may wish to play a different style of game, but ...

Although tempted to try Adishee's historically accurate mod (it has some wonderful attributes), I found it iffy to manage - and really didn't appreciate its refusal to allow you to save or backtrack on battles. (Adishee - That's how we learn) Still hoping to follow up on it at a later point, as it has many nice features.

In the meantime am fine-tuning the config files in attempt to keep UGCW at the MG level competitive. Some thrilling battles, but my historically-scaled / cavalry-heavy armies are already overpowering the AI with regularity by Antietam and Fredericksburg (which is a 'turkey shoot' for Confederates).  It appears that I must now go back to 'up' the AI ...

@pandakraut: KUDOS : the mods that you and johnny have crafted add a lot of playability to UGCW - It appears that there is robust willingness in the UGCW audience to customize the base game for a range of player needs, and Adishee's nicely conceptualized mod opens the door to even broader customization.  A nice-to-have at this point (for me, at least) would be tools to more easily and effectively edit, manage, and save alterations to the config files.  Would it be too much to ask for a little app that handles the config file libraries?  As you've already noted, I may want to adjust-up or adjust-down in order to try several different game profiles ... Such an app would also serve as a repository for defining what the various config file attributes mean - and how to manipulate them effectively

On 3/9/2021 at 3:11 PM, pandakraut said:

So depending on what you're looking for adjusting size and xp down on MG instead of up on BG might work out better. 

Thanks again for really fine work, guys.

Edited March 15, 2021 by dixiePig
clarity