Alpha 17 Balancing Branch

[scythetwirler]

Hello everyone,

Due to some of the imbalances of gameplay in Alpha 16 (notably early cavalry rushes), I have started a re-balancing GitHub branch for experimental gameplay balancing purposes. (NOTE: These changes may or may not make it into Alpha 17 or future releases). If you would like to give input and test it out, I would appreciate it greatly. (Though please try to refrain from making balance criticisms until you've played a couple multiplayer games so that everyone is on the same page. )

The repository is located at https://github.com/scythetwirler/0ad/ (branch 'balance'). You can see the changes at the bottom of that page (though I may forget to update that with changes).

Installing the branch:

Windows

Download GitHub for Windows and open the application Git Shell. From that window, type in the following commands:

mkdir 0ad-balancecd 0ad-balancegit clone https://github.com/scythetwirler/0ad.git

The last command will take a while to execute, as it will download an approximately 4.47 GB repository.

When the git clone has finished, you can run the game using pyrogenesis.exe (most likely in your Documents/GitHub/0ad-balance/0ad/binaries/system/ folder).

Linux

Install git on your distribution of Linux and create a folder in which you want to download the repository. Run the following command to start the download:

git clone https://github.com/scythetwirler/0ad.git

The last command will take a while to execute, as it will download an approximately 4.47 GB repository.


When the git clone has finished, you will need to compile the game. See BuildInstructions for compiling instructions (SKIP the "Getting the code" section). After compiling, you should be able to run the game using pyrogenesis (in your 0ad/binaries/system/ directory).

Keeping up to date:

I will make adjustments to balance quite often.

Windows:

Open git shell and type in

cd 0ad-balance\0ad\git pull

Linux:

cd inside your 0ad/ folder and run

git pull

If there are source changes (usually if there's a merge commit), then you will have to recompile (BuildInstructions).

EDIT: The branch has been merged into SVN. (See BuildInstructions) Please throughly test gameplay and bugs in general!

Hope you enjoy it!

Thanks,

[Jagst3r21]

Alpha 16 is currently unplayable. This balance branch is a great move for the game (it will allow for funner, more competitive games, etc.). BIG +1

[auron2401]

gave up on the bonus - on - ageup did you?
Don't blame you. Maybe we can find another way (maybe something similar to aom + age3?)

Nvm, still there.

Edited June 28, 2014 by auron2401

[iNcog]

I haven't been able to get github to get a working clone up. corrupted something or something

[zzippy]

Hi!

Already playing this balanced branch for a few days, imho it is a big step forwards. At last fun is back in the game ..

For those interested:

Changelog - Differences from SVN--------------------------------Units- Cavalry have been made weaker to pierce damage, but more resilient against hack.- Rams can now attack all units. In compensation, all organic units get a default of 3 Crush armor.- Ballistas and Onager have a much longer range and vision.- Ballistas deal much more damage.- Healers now cost only food since metal is quite rare.- Elephants now are significantly stronger but more expensive.- The basic formations have been disabled for the time being (with a surprising performance benefit).- Female Citizens train much faster.- Elite Iberian Skirmishers have been nerfed.- All units attack faster.- Singers deal both crush and pierce damage.- All hard bonuses have been taken out.- Swordsmen cost less metal, but cost a little wood.Structures- All structures are now weaker against crush and hack damage.- Civic Centers have a better garrison multiplier in an attempt to combat 2v1 rushes.- Walls and Wall Towers take significantly longer to build.- Significantly nerfed the Roman Entrenched Army Camp.- Farms have increased HP and more armor against hack and pierce attacks.Technologies- Some technologies have been made expensive (wip)- Some pair technologies have been unpaired.- Added a powerful tech to the wonder.- Phasing now increases health to all citizen soldiers.

Edited June 28, 2014 by zzippy

[scythetwirler]

I haven't been able to get github to get a working clone up. corrupted something or something

Did you download the whole repository or did it stop midway?

[iNcog]

Pretty sure it stopped midway, saying something was corrupted.

[scythetwirler]

Can you try running it again?

[Sanguivorant]

Games in this branch are definitely more fun and unpredictable than in Alpha 16. 4/20 would play again.

[wraitii]

I've done my testings with "making units move faster" and the effect is 100% what I expected: gameplay feels nervous and quick. I had hardcoded into CCmpUnitMotion that all units moved twice as fast, which is a little fast, but I think changing templates so that all walking speed are between 1.5 and 2 times their current values will help the feel of the game greatly.

(Merchants OTOH shouldn't be changed, they were ludicrously fast).

However it also makes training times seem ludicrously long, which is bad. Training a batch of 5 women at the start of the game right now takes about 1:30 (game time), which is just a ton of idle time.

(testing done using svn, not your branch)

[hollth]

Some feedback

- Cavalry have been made weaker to pierce damage, but more resilient against hack

Aren't spearmen and spear cavalry supposed to counter horsemen? Since they are a problem shouldn't they have less hack to highlight the weakness?

- Healers now cost only food since metal is quite rare.

I think they should have another resource added to the cost.

- Female Citizens train much faster.

I think this is the wrong approach. From what I can gather people feel there is not enough to do during the early phases of the game? Most of thats due to batch training being available from the beginning. Although a more difficult way to fix this, I think making batch training a technology available in phase2 would be a better way to solve this.

- Swordsmen cost less metal, but cost a little wood.

I dislike that they have three materials needed to make.

[zzippy]

Your feedback is based on testing, means gaming vs humans? Have not seen you yet in the balance branch lobby. Whats your nick there?

[Prodigal Son]

Some feedback

- Cavalry have been made weaker to pierce damage, but more resilient against hackAren't spearmen and spear cavalry supposed to counter horsemen? Since they are a problem shouldn't they have less hack to highlight the weakness?- Healers now cost only food since metal is quite rare.I think they should have another resource added to the cost. - Female Citizens train much faster.I think this is the wrong approach. From what I can gather people feel there is not enough to do during the early phases of the game? Most of thats due to batch training being available from the beginning. Although a more difficult way to fix this, I think making batch training a technology available in phase2 would be a better way to solve this.- Swordsmen cost less metal, but cost a little wood.

I dislike that they have three materials needed to make.

I think spearmen do both pierce and hack. Also changing their bonus damage multiplier against cavalry could make it enough anyway, coupled with increasing hack armor would make cavalry stronger against swordmen and other melee troops but still weak against spears (if that's what's intended). I haven't played A16 much because I got annoyed by people using all forms of exploits in it, but before that ranged troops seemed too strong even vs cavalry so I'm not sure about making them weaker vs pierce (could be the lack of the ability to attack moving targets for melee troops though). Besides than I'm in favour of even tougher but costlier cavalry compared to foot units, much like in reality, as it's something that is balancable. Most ancient armies had ratios of 1:10 or even less cavalry to infantry due to cost/lack of horses, more training required etc. So a 50% tougher, 50% costlier cavalryman compared to infantryman isn't that accurate.

Healers could cost half metal/half food. Priests have to cost metal(;p) and wood or stone wouldn't make much sense.

Not sure about the other two.

This guide might be helpful (especially its parts about balancing and unit roles), even if more warcraft/startcraft gameplay styled.

Edited July 5, 2014 by Prodigal Son

[wowgetoffyourcellphone]

Long time lurker. First post. First thoughts.

What is the justification for making cavalry weaker against ranged attacks? Aren't cavalry (generally) supposed to counter ranged units?

Priests made to cost only food and not metal, for what reason? Are we supposed to be able to mass Priests? Shouldn't Priests be special support units, hence cost some Metal?

Also, why do swordsmen cost wood? Makes no sense to me. And rams "attacking" soldiers and women sounds pretty weird. Seems like a lot of the changes were made to just make changes? Experiment?

Look forward to some matches though.

Edited July 5, 2014 by wowgetoffyourcellphone

[auron2401]

The justification for making horses weaker to Pierce, not just ranged attacks, is because of the current strategy in most multiplayer games. which is to build 3-4 skirmcav (if you have 'em) and rush the enemies resource production. The problem Wasn't that it was powerful. It was Unbeatable. Anyone with a decent micro could kill 50% of your gatherers without losing a single jav-cav. So it's obvious cav needed a counter.

[thamlett]

Scythetwirler: can you add me to the GitHub so I can have some edit privileges?

[wowgetoffyourcellphone]

The justification for making horses weaker to Pierce, not just ranged attacks, is because of the current strategy in most multiplayer games. which is to build 3-4 skirmcav (if you have 'em) and rush the enemies resource production. The problem Wasn't that it was powerful. It was Unbeatable. Anyone with a decent micro could kill 50% of your gatherers without losing a single jav-cav. So it's obvious cav needed a counter.

Then nerf the skirmcav. Has that been tried?

[scythetwirler]

Scythetwirler: can you add me to the GitHub so I can have some edit privileges?

I'd rather not for the time being, at least until some things have been tested out and fixed. You are always free to fork it and submit pull request for fixes/suggestions.

What is the justification for making cavalry weaker against ranged attacks? Aren't cavalry (generally) supposed to counter ranged units?

It does seem that way in other RTSes, but I'd like to try out a more generic battling system that doesn't revolve around rock-paper-scissors combat. The problem with cavalry countering ranged units (imo) is that they'd be either immensely overpowered (due to their speed, they decide when they want to battle, and thus, usually only when they have an advantage), or vastly underpowered if we nerf them too hard. Right now, I'm giving them a sort of raiding niche to attack unprotected workers.

Priests made to cost only food and not metal, for what reason? Are we supposed to be able to mass Priests? Shouldn't Priests be special support units, hence cost some Metal?

Metal, imo, was too precious to spend 200 on each healer. Merely having 25 healers was enough to deplete your entire starting 5000 metal mine. Perhaps they should cost a little metal, but currently I want to increase the usage of healers to see how people use them and go from there.

Also, why do swordsmen cost wood? Makes no sense to me.

Swordsmen were almost never being used beyond early rushing, due to their high metal cost. I'm not sure where wood would be used for a swordsman, but right now, I'm just trying to balance it. If metal cost is decreased too much, they become too strong early game. Thus, I had to add another cost to make them on par with other soldiers.

And rams "attacking" soldiers and women sounds pretty weird. Seems like a lot of the changes were made to just make changes? Experiment?

Realistically, you would not stand in front of a ram because you'd be bowled over. Previously, you could just send any unit (healers included) to just block the ram. With its large obstruction size, rams became useless quite often - they couldn't reach a structure to attack.

I've done my testings with "making units move faster" and the effect is 100% what I expected: gameplay feels nervous and quick. I had hardcoded into CCmpUnitMotion that all units moved twice as fast, which is a little fast, but I think changing templates so that all walking speed are between 1.5 and 2 times their current values will help the feel of the game greatly.

(Merchants OTOH shouldn't be changed, they were ludicrously fast).

I rather like this idea. I'll give it a try sometime.

Another note: It is still in the experimental phase, many things are subject to change.

[wowgetoffyourcellphone]

It does seem that way in other RTSes, but I'd like to try out a more generic battling system that doesn't revolve around rock-paper-scissors combat. The problem with cavalry countering ranged units (imo) is that they'd be either immensely overpowered (due to their speed, they decide when they want to battle, and thus, usually only when they have an advantage)...

Yeah, but that's how cavalry was used in real life.

Metal, imo, was too precious to spend 200 on each healer. Merely having 25 healers was enough to deplete your entire starting 5000 metal mine. Perhaps they should cost a little metal, but currently I want to increase the usage of healers to see how people use them and go from there.

Yeah, but who trains 25 healers? Do you want players to train 25 healers? Maybe 100 Food and 100 Metal?

Swordsmen were almost never being used beyond early rushing, due to their high metal cost. I'm not sure where wood would be used for a swordsman, but right now, I'm just trying to balance it. If metal cost is decreased too much, they become too strong early game. Thus, I had to add another cost to make them on par with other soldiers.

Could just go 50F 60M and then use blacksmith techs to reduce their cost in late game.

Speaking of techs, I miss the wheelbarrow.

Realistically, you would not stand in front of a ram because you'd be bowled over. Previously, you could just send any unit (healers included) to just block the ram. With its large obstruction size, rams became useless quite often - they couldn't reach a structure to attack.

Sounds more like a pathfinding problem?

Another note: It is still in the experimental phase, many things are subject to change.

Of course. I don't understand why the other guy got mad.

I've done my testings with "making units move faster" and the effect is 100% what I expected: gameplay feels nervous and quick. I had hardcoded into CCmpUnitMotion that all units moved twice as fast, which is a little fast, but I think changing templates so that all walking speed are between 1.5 and 2 times their current values will help the feel of the game greatly.

(Merchants OTOH shouldn't be changed, they were ludicrously fast).

However it also makes training times seem ludicrously long, which is bad. Training a batch of 5 women at the start of the game right now takes about 1:30 (game time), which is just a ton of idle time.

(testing done using svn, not your branch)

Have you tried playing the game at 1.25 or 1.5 speed? In my opinion, using the game speeds is much better than changing a bunch of other stuff to get the same effect.

Edited July 5, 2014 by wowgetoffyourcellphone

[Prodigal Son]

Swordsmen wood cost could easily be justified as part of their shield materials. Cavalry should be given different roles, some being skirmishers/raiders/anti-ranged while others heavy shock cavalry that can cost effectively break most non-spear troops, not all of them being harassers. And besides the realism aspect of it, I agree that very easy to mass preachers isn't a good idea, so half or something of their cost being metal sounds good. Rams ramming units indeed sounds bad, perhaps if a pushing mechanic is easy to implement to have them slowly make their way through enemy units? It could also work for war elephants and to a lesser extend cavalry or even heavy infantry, pushing back the enemy formation.

Edited July 5, 2014 by Prodigal Son

[wowgetoffyourcellphone]

Swordsmen wood cost could easily be justified as part of their shield materials. Cavalry should be given different roles, some being skirmishers/raiders/anti-ranged while others heavy shock cavalry that can cost effectively break most non-spear troops, not all of them being harassers. And besides the realism aspect of it, I agree that very easy to mass preachers isn't a good idea, so half or something of their cost being metal sounds good. Rams ramming units indeed sounds bad, perhaps if a pushing mechanic is easy to implement to have them slowly make their way through enemy units? It could also work for war elephants and to a lesser extend cavalry or even heavy infantry, pushing back the enemy formation.

I can see ranged cavalry being the major harassers, with melee cav being used for hard hitting cavalry charges into the flank (see the Total War series for example). About the swordsmen wood cost, I just find it odd to make units cost more than 2 different resources, especially a default unit that you train at the beginning of a game. Pushing would be awesomeness for large units.

[WhiteTreePaladin]

Yeah, I don't think swordsmen should cost wood. Perhaps there can be an upgrade later in the game that lowers the cost. Early game they would be more expensive, then after researching some relevant tech, the cost would go down. The tech could have something to do more efficient sword making abilities like an improved smelting process or something.

I think the Mauryan worker elephant can get around the wheelbarrow tech, but the other civs might miss it. Not sure why it was chained to lumber abilities anyway.

[hhyloc]

For ram attacking infantry, how about we give it a slight damaging aura instead (unlockable tech?), this simulates the soldiers inside the ram retaliate against attackers, also with this aura the ram doesn't have to stop to attack enemy units and can still sort of push forward given enough time. This way we can silmulate the fact that ram can still be blocked by enemy but it can slowly bowl over opposition, but clearing the blocking enemy with your own units will be more effective though.

[iNcog]

Right, so I'm going to give a personal update (since I've been asked to play the balance branch) and I'm going to also give some personal thoughts.

First off, I have little free time as of right now, things irl are gobbling up all the time I have, I barely have time to just play games, much less 0 AD. This means that I don't have time to figure out how to install the balance branch and even less time to do actual play-testing. This unfortunately means that you have to scratch me off the play tester list. I'm going to continue to play a few games now and then though. I'll be doing it on A16, even if it's not as good as A17, At very least I can keep in touch with the game during that time.

As for the actual game itself, I have a few thoughts. I believe that the game itself needs to be broken down and rebuilt, at very least the units themselves need to. I PM'd alpha123 and he linked me a text document that explained some of the changes that were in store for the game. These changes would basically invalidate the counter-system completely, which is something I agree with. These changes also mean that we would no longer have need for multipliers. It also means that all the changes being tested would become irrelevant as well. Alpha123 is envisioning a total and complete rework of units. So trying to balance things out from the way units are right now becomes somewhat un-needed; everything is going to change anyway in the future.

You can PM alpha123 for the details. I'll just leave this here: https://docs.google.com/document/d/1PxeSahVf5J6h32jy2iqTHlnogtQSHPocXKqNSspyMUs/edit

In the mean time, I'm going to write an open document about how I believe units should be changed. I had started writing this but I don't have access to my laptop as of right now. I'm either going to try to find the document I wrote or re-write everything.

E: basically the changes I would so would be the following.

There are three big attributes that a unit can have:

Stats (attack/HP/rate of fire)

Range

Speed

My, very basic, idea is that units should have only two of these attributes as their strengths. No unit should have all three. For example, the way things work in the game currently, ranged units tend to have stats that are quite comparable to those of melee units. Skirm-cav have all three attributes, which is why they are broken. They have the attack, the range and the speed to make them very versatile units. Ranged infantry have good stats but they're also ranged. The attack/HP that melee infantry aren't good enough to compete with ranged units.

My idea is that every unit should have two of these attributes as strengths, the last one would be a weakness. For example:

Melee infantry:

Stats: Lots of HP, lots of dps (hack)

Range: 0

Speed: infantry

Ranged infantry:

Stats: Low HP, medium dps (less than melee infantry) (pierce)

Range: Medium to high

Speed: Infantry

Melee cav:

Stats: lots HP, lots of dps, higher cost to reflect that (hack)

Range: 0

Speed: Cav

Ranged Cav:

Stats: medium HP, low dps, medium to high cost (pierce)

Range: Low to medium

Speed: cav

You can then make a distinction between units. Spear infantry would get a mutliplier against cav, that's the most obvious and needed attribute they need. they get lower dps than swordsmen to reflect their cav mutliplier, as well. however spear men and swordsmen, if they catch ranged infantry with a melee attack, they should win. ranged infantry should have their range be their winning attribute, not their dps. e.g. ranged infantry can use their range to attack melee infantry without being attacked themselves. this attribute makes them good if they're well-positioned (at range, under the cover of a CC, behind friendly melee units, etc), but bad if they're not (caught by a flank from melee infantry, caught in the open by cav without spear men protection, etc). similarly, ranged cav should get very low dps, to reflect their speed and range. they'll become scouting and harassing units. they'll be good at picking off reinforcements, attacking harvesters, scouting, etc. however ranged cav will then be bad in straight up fights.

you can also make distinctions between units themselves. e.g. archers will be medium dps but long range. skirmishers will have shorter range and medium dps. you also can give skirmishers very high attack, but low rate of fire. this means they're very good hit and run units, but they need to be micro'd to be really efficient. you can play on these attributes to make units feel and work differently.

spear men could give a range of 2 or 3 meters, this allows them to be able to fight with less surface area than swordsmen for example. so swordsmen that have a good surround do well, but if they don't have good surface area, spearmen with their small range have the upper hand.

^these are the kinds of changes that make units unique. units have a role. you can do all this without multipliers (except for spear infantry vs cav units). units will feel and act differently but there are no hard-counters in the game. units well used and well positioned come out on top. you can balance a unit through stats. e.g. you can give skirm-cav just enough dps that they're good harass units without being good in fights. same for other units. this is the best way to "fix" things, imo.

Edited July 6, 2014 by iNcog

[Prodigal Son]

Nice iNcog, in fact that's almost exactly how I balance my wc3 mod. I also agree that this discussion gets confusing/pointless since at points we talk for balance hotfixes while at others for a revamped combat system.

Three different resources for a unit might indeed be harsh in practice, so even if it makes sense the cost reduction tech way might be better. It can also simulate the increased cost effectiveness of roman swordsmen and the choice to follow up for other civs which reformed their infantry. Romans could even get a slightly stronger or cheaper version of this tech

The aura for the ram might be better that it attacking human soldiers, if no pushing can be implemented.