Why basepushing is so strong

aka, how i learned how to play the game

Discussion about the game and its default mods.
User avatar
SoScared
Posts: 1116
Joined: Wed Apr 11, 2012 6:16 pm
Location: Oslo
Contact:

Post by SoScared » Wed May 24, 2017 5:29 pm

Greath thread and OP!

@Doomsday: Don't forget the main tanks' damage boost vs heavy and +1 vision. Being damage soakers tanks are the first in line to grind down def.structures and punish lone expansions with only def.structures to defend them.

As mentioned above the line of sight is the main culprit for a wide variety of issues present in the RA mod today. I'll reproduce a former reply of mine on this issue regarding RA's balancing up until today:
One thing that is brought up from time to time but unfortunately missed in the discussion is the vision range of RA's units and structures. As opposed to TD the retreat factor for RA has been almost non-existant, meaning once you've found your opponents forces you stand face-to-face with them and retreating only means loosing your army without exchanging damage. The natural tendency is to build up an army, keep it immobile out on the field or safely behind structures. Some players get around it with good scouting and tech will help you make better decisions with your army movemet but the early- to mid-game is often pretty tense this way. The safe course of action is to stay home and grow. RA's balance on unit vision range could have been more agressive.
This by the way is one of the main motivations behind the creation of maps such as Sidestep, Warwind and to a more extreme degree, Agenda. Make the passages so wide and open that armies one way or another will be exposed on the battlefield.

This is also why I allowed the RAGL map pool to be more or less a test lab for the latest release balance changes. Forgive me!

OMnom
Posts: 255
Joined: Fri Oct 28, 2016 4:04 am

Post by OMnom » Wed May 24, 2017 8:49 pm

Doomsday wrote:
Orb wrote: Of these, I think long build times on defenses is the most relevant. They don't have to cost more, but simply take longer to build. This sort of works the same way as putting tech structures in the defense tab, as it limits the amount of defenses you can produce.
Defensive structure build time nerf is often mentioned when discussing base pushing. I want to point out it would affect both offensive and defensive player. It would also amplify "I was here" problem.

I tend to compensate scouting with defensive structures. If I have pillboxes or flame turrets and barracks everywhere, it will take some time for my opponent to destroy them - thus buying me some time to move my main army to defend or counter attack / attempt a base trade.

RA has very weak scouting tools for players because of most units having awful vision. I fear nerfing defensive structure production time could bring more annoying random elements to games because players don't have proper tools for reliably scouting flanking attacks from all directions.
Yes, all of this is true. Buildings are a way for you to buy time. So if someone decides to put a BuildRadius Delay or a static defense delay, you're taking away BOTH player's ability to buy time, which means that the fastest person there is the one who will benefit the most, aka, the "I was here first" / snowball problem. The second issue is that once someone gets there, there is no option to unroot them from that position barring game-situations. Base pushing comes into play as the only option to unroot someone from said position.

On a related note, the speed of basepushing is at least 50% quicker than using a mobile army. It's quicker to just make a new army at an new location than it is to split off your existing army and make them walk half-way across the map. This idea makes it easier to launch multi-pronged attacks with basepushes.
Wippie wrote: Did you explore the option of placing a deployed mcv in the building tab Omnom?
Yes, but not in-depth. I couldn't figure out a good tech level to put that option at. Radar dome made the most sense, but I think Lorry was the one who exploited mass MCVs + mass Harvesters to support production of 4 non-stop production queues...was very annoying to play against. I think finding a way to limit the amount of con yards you could build through the Defense queue was the main problem, without limiting the amount of deployed Con Yards you could have on the map.

On a slightly related note, people have long been supporting building limits, such as 7max barracks and 3/4max MCVs/conyards. It's a perfectly valid solution, but this creates a new problem of confining the game into a pre-ordained build progression. There's already a dearth of diverse build orders in this game (many new players complain about that already), so making those 3-4 build orders all lead into the same mid-game progression would simplify game to mind-numbing levels...players should be able to choose if they want to tech, not be forced to tech.

What would be nicer, is if each tech level unlocked 2 more barracks + 1 more MCV, but I haven't had the time to figure out how to implement the new conditions system to make that work. However, this still wouldn't fix basepushing -- it only indirectly addresses the "building-tank" problem, which could be easily circumvented by selling your prior barracks, and it doesn't address any of the other issues.
SoScared wrote: Greath thread and OP!

@Doomsday: Don't forget the main tanks' damage boost vs heavy and +1 vision. Being damage soakers tanks are the first in line to grind down def.structures and punish lone expansions with only def.structures to defend them.

As mentioned above the line of sight is the main culprit for a wide variety of issues present in the RA mod today.
There is no great way to address vision in this game. Making units see+shoot the same distance would make A-moving much stronger in the early game and slightly stronger in the mid-late game. This, however, would also make it easier for building-tanks to soak up the damage and A-move blobs. Getting rid of auto-target on unarmed buildings was also mentioned somewhere, but this would massively buff A-moving because every shot fired would be at something that can shoot back.

OMnom
Posts: 255
Joined: Fri Oct 28, 2016 4:04 am

Post by OMnom » Wed May 24, 2017 9:07 pm

double post

Blackened
Posts: 308
Joined: Sat May 21, 2016 6:27 pm

Post by Blackened » Wed May 24, 2017 10:27 pm

OMnom wrote: Getting rid of auto-target on unarmed buildings was also mentioned somewhere, but this would massively buff A-moving because every shot fired would be at something that can shoot back.


I feel this is one of the few options that addresses a lot of your 20 points. Have any more thoughts on it?

User avatar
WhoCares
Posts: 287
Joined: Mon Feb 13, 2017 11:28 pm

Post by WhoCares » Wed May 24, 2017 11:24 pm

I mentioned it some month ago, back when I played Total annihilation, unit were in auto ignore non armed building and it was up to you to put them in attack everything if you wanted to destroy everything in youir path. Those stances were separated from the classic hold/return/autoattack.

If someone put that up in a playtest, i'm very interested to participate as guenea pig to see the results.

edit : (And for a game where you're a commander/general, it's a bit frustrating to command an army of idiots who will prefere dying shooting at a tent than defending themself from the real threats slaughtering them. A buff in inteligence/logic of the units woundn't hurt)

User avatar
avalach21
Posts: 213
Joined: Mon Aug 10, 2015 8:01 pm

Post by avalach21 » Wed May 24, 2017 11:40 pm

OMnom wrote:
There is no great way to address vision in this game. Making units see+shoot the same distance would make A-moving much stronger in the early game and slightly stronger in the mid-late game. This, however, would also make it easier for building-tanks to soak up the damage and A-move blobs. Getting rid of auto-target on unarmed buildings was also mentioned somewhere, but this would massively buff A-moving because every shot fired would be at something that can shoot back.
I believe you are referencing discussion in the thread I started about enemy oil derricks being attacked, even when your units are in"hold fire" stance.

but anyways, so what? These are my 2 biggest gripes with the game currently. auto targeting unarmed buildings (particularly oil derricks and neutral civilian structures) and second the current vision system.

I appreciate good micro no doubt. I like having a micro centric game. But your units should be semi intelligent enough to have some very basic attack decision making logic. It makes sense they would be threatened by a machine gun blaring in their face and attack that rather than shooting some random nonthreatening structure, unless they are given an explicit direct command to do otherwise. Again, micro is nice and adds depth to the game, but having your units behave so idiotically without constant explicit guidance is more baby sitting than enjoyable tactical micro.

I think you keep saying that these changes would empower the "attack move" blob, which to me sounds like, yes... armies of units actively engaging in combat... rather than the constant basewalk, arty v2 spam dynamic every game falls into if it gets past the cheese beginning stage.. which I thought was the whole issue you were trying to address by creating this thread?

User avatar
avalach21
Posts: 213
Joined: Mon Aug 10, 2015 8:01 pm

Post by avalach21 » Wed May 24, 2017 11:49 pm

Blackened wrote:
OMnom wrote: Getting rid of auto-target on unarmed buildings was also mentioned somewhere, but this would massively buff A-moving because every shot fired would be at something that can shoot back.


I feel this is one of the few options that addresses a lot of your 20 points. Have any more thoughts on it?
WhoCares wrote: I mentioned it some month ago, back when I played Total annihilation, unit were in auto ignore non armed building and it was up to you to put them in attack everything if you wanted to destroy everything in youir path. Those stances were separated from the classic hold/return/autoattack.

If someone put that up in a playtest, i'm very interested to participate as guenea pig to see the results.

edit : (And for a game where you're a commander/general, it's a bit frustrating to command an army of idiots who will prefere dying shooting at a tent than defending themself from the real threats slaughtering them. A buff in inteligence/logic of the units woundn't hurt)
Wow you guys came to basically the same conclusions I did before I finished posting.. maybe we are onto something..

OMnom
Posts: 255
Joined: Fri Oct 28, 2016 4:04 am

Post by OMnom » Thu May 25, 2017 12:15 am

The list of "20 reasons why basepushing is strong" is not the point I'm trying to make. It's to show why balancing basepushing is not a simple task. All of the practical ideas require modding the base code (something I am not experienced with at all), and most of the work-around YAML solutions are bandages at best.

As far as giving unarmed buildings a neutral stance, it's technically possible to do this through YAML. My suspicions are that a change like this would buff the usage of V2/Arty, and it would also enhance the A-move snowball. It would throw off Smitty's triangle of balance and create something entirely new....i'd be interested in seeing what happens, but I'm not exactly hopeful that this will produce more positive changes than negative changes.
avalach21 wrote:
OMnom wrote:
There is no great way to address vision in this game. Making units see+shoot the same distance would make A-moving much stronger in the early game and slightly stronger in the mid-late game. This, however, would also make it easier for building-tanks to soak up the damage and A-move blobs. Getting rid of auto-target on unarmed buildings was also mentioned somewhere, but this would massively buff A-moving because every shot fired would be at something that can shoot back.
I believe you are referencing discussion in the thread I started about enemy oil derricks being attacked, even when your units are in"hold fire" stance.

but anyways, so what? These are my 2 biggest gripes with the game currently. auto targeting unarmed buildings (particularly oil derricks and neutral civilian structures) and second the current vision system.

I appreciate good micro no doubt. I like having a micro centric game. But your units should be semi intelligent enough to have some very basic attack decision making logic. It makes sense they would be threatened by a machine gun blaring in their face and attack that rather than shooting some random nonthreatening structure, unless they are given an explicit direct command to do otherwise. Again, micro is nice and adds depth to the game, but having your units behave so idiotically without constant explicit guidance is more baby sitting than enjoyable tactical micro.

I think you keep saying that these changes would empower the "attack move" blob, which to me sounds like, yes... armies of units actively engaging in combat... rather than the constant basewalk, arty v2 spam dynamic every game falls into if it gets past the cheese beginning stage.. which I thought was the whole issue you were trying to address by creating this thread?
Thats not the main point of this thread. If you read what I bolded, you'd know that I'm trying to get you people to look at the bigger picture. It's not "I can't see stuff, so lets give stuff more vision" or "my units are attacking a useless building, that doesn't make sense, so lets allow units to only attack stuff that shoot back." This was the pitfall of SC2 and why many programers are currently going back to the original Starcraft: Brood War. The entire SC2 balance team treated their game like it was a MOBA and constantly tweaked the game according to the meta. They never bothered to identify or address the root problems of the game and were content with just surface treatments. The "tank-buildings" are a surface problem. If you aim a targeted fix to only fix that one problem, I promise you, a different one will take its place. Same goes with extending vision.
Last edited by OMnom on Thu May 25, 2017 12:42 am, edited 1 time in total.

Lorrydriver
Posts: 64
Joined: Thu Dec 01, 2016 3:55 am

Post by Lorrydriver » Thu May 25, 2017 12:32 am

I'm afraid at some point we just need to become better players instead of blaming everything on balance or dumb AI. Recent changes have made mcv pushes significantly worse and they are not that hard to counter. This game is easy to play and it's especially easy to copy strategies/ build orders. Base pushing just happens to be one of the easier ways to win games right now. I've had a lot of success against base pushes on favourable maps for base pushing, by just adjusting to it and finding timings and strategies to counter that mcv heavy playstyle. I'd even go ahead and say that basepushing is not the best way to win in 1on1s. However, if you try to beat base pushes with strategies that you used before mass mcvs became so popular, you're going to have a problem for one simple reason: They counter the timings that used to be standard. By making small changes in your build order you can actually get ahead if your opponent is going for an mcv push. There's so much whining going on while even the best players of OpenRA aren't even close to playing this game optimally. Let's get good at this game first and then reconsider what needs changes.

User avatar
avalach21
Posts: 213
Joined: Mon Aug 10, 2015 8:01 pm

Post by avalach21 » Thu May 25, 2017 1:26 am

OMnom wrote: The list of "20 reasons why basepushing is strong" is not the point I'm trying to make. It's to show why balancing basepushing is not a simple task. All of the practical ideas require modding the base code (something I am not experienced with at all), and most of the work-around YAML solutions are bandages at best.

As far as giving unarmed buildings a neutral stance, it's technically possible to do this through YAML. My suspicions are that a change like this would buff the usage of V2/Arty, and it would also enhance the A-move snowball. It would throw off Smitty's triangle of balance and create something entirely new....i'd be interested in seeing what happens, but I'm not exactly hopeful that this will produce more positive changes than negative changes.
avalach21 wrote:
OMnom wrote:
There is no great way to address vision in this game. Making units see+shoot the same distance would make A-moving much stronger in the early game and slightly stronger in the mid-late game. This, however, would also make it easier for building-tanks to soak up the damage and A-move blobs. Getting rid of auto-target on unarmed buildings was also mentioned somewhere, but this would massively buff A-moving because every shot fired would be at something that can shoot back.
I believe you are referencing discussion in the thread I started about enemy oil derricks being attacked, even when your units are in"hold fire" stance.

but anyways, so what? These are my 2 biggest gripes with the game currently. auto targeting unarmed buildings (particularly oil derricks and neutral civilian structures) and second the current vision system.

I appreciate good micro no doubt. I like having a micro centric game. But your units should be semi intelligent enough to have some very basic attack decision making logic. It makes sense they would be threatened by a machine gun blaring in their face and attack that rather than shooting some random nonthreatening structure, unless they are given an explicit direct command to do otherwise. Again, micro is nice and adds depth to the game, but having your units behave so idiotically without constant explicit guidance is more baby sitting than enjoyable tactical micro.

I think you keep saying that these changes would empower the "attack move" blob, which to me sounds like, yes... armies of units actively engaging in combat... rather than the constant basewalk, arty v2 spam dynamic every game falls into if it gets past the cheese beginning stage.. which I thought was the whole issue you were trying to address by creating this thread?
Thats not the main point of this thread. If you read what I bolded, you'd know that I'm trying to get you people to look at the bigger picture. It's not "I can't see stuff, so lets give stuff more vision" or "my units are attacking a useless building, that doesn't make sense, so lets allow units to only attack stuff that shoot back." This was the pitfall of SC2 and why many programers are currently going back to the original Starcraft: Brood War. The entire SC2 balance team treated their game like it was a MOBA and constantly tweaked the game according to the meta. They never bothered to identify or address the root problems of the game and were content with just surface treatments. The "tank-buildings" are a surface problem. If you aim a targeted fix to only fix that one problem, I promise you, a different one will take its place. Same goes with extending vision.
I agree on a lot of things. I dont think SC2 is very fun to play. I dont really enjoy MOBAs. I really like this game (OpenRA). Some other RTS I really like are Company of Heroes, Warcraft III, and Starcraft 1 etc. I don't think the current Meta is completely broken or anything, I love where this game is at and in my opinion it vastly improves on the original game that we all fell in love with 20 years ago.

I agree that my criticism of vision system is not very fleshed out at this point. I don't have a fleshed out specific replacement system to recommend that would be an improvement, and it is definitely a secondary discussion on my list.

I agree with you that there is no easy fix to address basewalking. I agree that all of your hypothetical adjustments end up exacerbating the issue, or create other issues of their own.

What I will definitely stick to and I think most agree is that units should have attack priority logic implemented. I feel that's what most agreed on in the other thread on this topic and seemingly many are agreeing with it again here. I think others went on to mention even that it was how all the original Westwood engines behaved, in particular someone mentioned RA2 I think as a model as it was the only one to implement neutral civilian structures.

I don't think it would break the game dynamic or over simplify it any way. There is a lot of depth to the game : base placement, economy management, scouting, producing proper counters, engaging into battle at the proper place and time, deciding when to tech, Navys and airforces to watch over... etc etc... this game has a lot to it ! Compared to your average MOBA... I think that all those aspects of the game are fun and I never feel there is a shortage of things to focus my attention on. I think that having to constantly give explicit mind numbing babysitting orders constantly over and over again detracts from your ability to focus on the more interesting and enjoyable parts of the game.
Last edited by avalach21 on Thu May 25, 2017 1:40 am, edited 1 time in total.

User avatar
WhoCares
Posts: 287
Joined: Mon Feb 13, 2017 11:28 pm

Post by WhoCares » Thu May 25, 2017 1:29 am

As far as i'm concerned, i learn the game the way it is. As long as people propose playtesting changes, i might as well suggest the ones wich make sense to me. I don't suggest a change because i don't handle the way it is, but because the curent way feels dumb (even if i can exploit it or be annoyed by it). I don't personally have a problem with basepushing, it's a strat like another, but the flexibiity to have more differents behaviour on units interests me to dig the possibilities.

I'm 100% on the playtest concept to test what could work and what makes things worst. I basicaly play the game this way by testing things everygame to make as much mistake as possible and try to learn from them. Getting good is a long term process when you know you lack the talent. ;)

Printer
Posts: 43
Joined: Wed Apr 19, 2017 3:53 am

Post by Printer » Thu May 25, 2017 4:41 am

Read thread and really like simplicity in OP. Mainly in response to it:

I don't understand #2 (The " Houdini Tank-Building"), 6 (new Flame Turrets), 7 (Teslas on a cliff) and also why not remove #3 vetrancy from Pillboxes? Even if you don't it's not a huge deal imo.

What your opinions are on:

1. Scarcity applied to buildings (but maybe not units). - Aims to fix#5, 10, 11, 12, 15, 18,
Would be interesting if buildings began to increase more in cost as they were built. First (2) MCV/CY could standard cost, then go up $500.00 or so for everyone after that. You could also see a 20-30% increase on other buildings as well, including pill boxes and base defenses. Age of Empires II had this and it curbed SimCity play effectively. The game already tracks if you have a CY/MCV or not so, I suspect the number of these a player owns is already tracked without new coding needed.

2. Simulated Shockwave/Shrapnel/Fire Spread damaging to Buildings* - Aims to buff units over buildings placed offensively.
Can a building's destruction create an aoe effect (like the effect of Oil's Pumps exploding) that only effects Concrete and Wood armor types(this is not unplausible as explosions damage structural integrity, windows, etc in real life)? It wouldn't destroy the nearby building but take a few points down, based on the HP of the building and/or contents. The key thing here tho, is that this type of AOE would not affect the armor types of units, except perhaps Arty and V2 armor (to deal with #17). Like the other idea, I am not certain this would be hard to code as all the main AOE effects already have a diminishing damage radius.

Otherwise, I don't believe all of these need to be 'fixed' or responded to. The game is fun with them in it, and the reason I play Soviet is because I think Allies are OP- so it's worth the struggle.
1. The "I was here first" problem (Not a problem imo, part of the game)
4. Shorter supply line, quicker reinforcements (Not a problem, good dynamic with forward bases)
8. Abusing the AI targeting with the MCV deployment/redeployment (Ai play makes up for less than 10% of the game play imo, don't ballance for single player)
9. Concrete Walls (Recent game release ballanced this imo)
13. Ability to instantly profit from killing your opponent's base (Isn't that a reasonable goal in the game?)
14. Sacrificing your MCV/buildings to preserve your Army (Again, a tactical part of the game)
16. Basepushing helps maintain map control. (It has to though, bases are the only way to solidify progress)
19. Pocket Pillboxes (You can only build one, I don't believe this is a problem but part of the meta)
20. Getting base pushed by someone who knows what they're doing. (....that just makes me want to get good though.)

OMnom
Posts: 255
Joined: Fri Oct 28, 2016 4:04 am

Post by OMnom » Thu May 25, 2017 6:27 am

Ugh. I should just delete that stupid list.

The purpose of that list wasn't to enumerate every single strength of basepushing. It was supposed to show people that one change aimed at basepushing poses a risk of creating more problems than it solves. Basepushing isn't a direct problem like "people spam too many pillboxes." This is manifested in multiple aspects of the game...changing one facet of it will most likely have an effect on another.

Stop thinking in terms of, "how do I fix basepushing?" It's not going to work. Think more like this:
Doomsday wrote: From another thread.
Smitty wrote: I view RA gameplay as a triangle, with the three points being Base Crawling, Artillery/V2 duels, and mobile armies. Ideally these elements would all have their place within a game, with base crawling being strong vs. mobile armies, artillery being strong vs. base crawling, and mobile armies being strong vs. artillery. Base crawling in particular has been skewing the balance of these three sections, which is why the competitive community has focused balancing efforts on the MCV.
SoScared wrote:
One thing that is brought up from time to time but unfortunately missed in the discussion is the vision range of RA's units and structures. As opposed to TD the retreat factor for RA has been almost non-existant, meaning once you've found your opponents forces you stand face-to-face with them and retreating only means loosing your army without exchanging damage. The natural tendency is to build up an army, keep it immobile out on the field or safely behind structures. Some players get around it with good scouting and tech will help you make better decisions with your army movemet but the early- to mid-game is often pretty tense this way. The safe course of action is to stay home and grow. RA's balance on unit vision range could have been more agressive.
Look at the bigger picture and the underlying problems, not what's on the surface. Once you guys start thinking on those terms, you can start putting together real balance solutions, not "think tank ideas."

User avatar
Doomsday
Posts: 199
Joined: Thu Nov 10, 2016 11:45 am
Location: Helsinki

Post by Doomsday » Thu May 25, 2017 9:20 am

Printer wrote: 1. Scarcity applied to buildings (but maybe not units). - Aims to fix#5, 10, 11, 12, 15, 18,
Would be interesting if buildings began to increase more in cost as they were built. First (2) MCV/CY could standard cost, then go up $500.00 or so for everyone after that. You could also see a 20-30% increase on other buildings as well, including pill boxes and base defenses. Age of Empires II had this and it curbed SimCity play effectively.
I believe I know a fair bit about AoE2 and this is news to me. Are you sure about this? In Rise of Nations each additional unit and building of the same type costs more.
The supreme art of war is to subdue the enemy without fighting.
-Sun Tzu

User avatar
AoAGeneral1
Posts: 597
Joined: Wed Apr 29, 2015 6:11 pm

Post by AoAGeneral1 » Thu May 25, 2017 9:37 am

You can fix the base pushing. Its actually easy. Problem is its become the base gameplay of RA.

Post Reply