Playing To Win

Request new features or present your ideas.
Post Reply
User avatar
Ishq
Project Head
Posts: 1145
Joined: Tue Mar 06, 2012 8:32 pm UTC

Playing To Win

Post by Ishq »

Background (of the most likely culprits)
We've tried many gameplay changes in the last two years. Our initial primary focus was to obliterate camping as a strategy and force teams to explore, expand, and attack bases. This change diminished the effect of feeding, which first led to "The Dretch Rush" which first emerged in the Pk vs get match in the summer tournament. That was promptly “fixed” by decreasing dretch HP, but the tactic of suicide rushing early and often gave aliens a significant advantage throughout most of unv. This was further exacerbated by the introduction of minimum wage, which allowed spam of leeches and mara rushing until the humans died.

Current State:
The culmination of all these changes results in two mindsets: “play for fun” (not in a LOL sense, where you just suicide and otherwise don't play the game as it is supposed to be played) and “play to win.” The play for fun mindset is the type of gameplay we want. Players rush, players fight. They don't stall the game and they don't take deaths/feeding/negatives too seriously. Players generally play in a fashion that's fun for them to play and fun for their opponents to play against. The flip side is the play to win mentality: this is where you take the most rational decisions to ensure you win the game. In unv's case, that is to use your strengths to your advantage. For instance, humans know that aliens have no chance of winning against turrets unless they have at least a mara, so they will slowly turtle their base out, or to avoid fighting other players and instead attempting to just attack empty bases. This isn't especially fun and generally leaves the other players with a somewhat empty, anticlimactic feeling.

The Goal:
For me, the goal is to create gameplay in which the “play for fun” and the “play to win” mindsets are the same. By playing normally and by doing the right things you will move closer to winning and things that generally make the game less fun (such as avoiding all players to simply attack bases or camping all game) are discouraged either by making them exponentially more difficult (perhaps this makes them more fulfilling. Maybe. That isn't the goal of doing this though), so they aren't the defacto strategy in order to win.

The Problems:
Building
Building has become less strategic and more spammy. Players don't need to carefully position buildables or even use their buildables intelligently. As long as they clump their buildables together in some fashion, they will easily dispatch most aliens. Furthermore, with the advent of drills, there is generally a large surplus of build points, so any defenses you kill, are sure to reappear by the time you get back to try to kill another. This is boring. Bases don't require skill anymore to build and to kill. There are now two ways to kill a base: hit the base while the enemy is out (boring), or just use pure brute force and kill each and every structure with your team's overwhelming map control/money/skill advantage (too long. Boring).

Possible Solutions:
Make drills/leeches not only mine, but store resources.
Losing a drill will result in losing all unused BP currently in the drill.
A big problem is that drills/leeches are rather important structures, but they are not treated as such. They are cheap to build. And there is almost no penalty if one is destroyed. In fact, spamming them around the map is beneficial because unless it's late in the game, even if the drill stays alive for a minute, the drill will most likely already pay back its worth. By making them structures that can harm you significantly if lost, will encourage teams to guard them more carefully, and opponents to target them. If a drill is killed, the team potentially loses the ability to rebuild from future attacks.

Make defense structures harder to build, harder to kill
A defense structure can be made harder to build by making it more expensive, take longer to build (this might make things less fun, so I'm less enthused by this. I quite like fast build). This will make it more difficult to spam buildables and make each loss matter more. The downside of this is we also want to encourage forwarding, but this change is likely to have an adverse effect on forwarding. On the upside, forwards aren't used much except for ways to increase the opponent’s momentum. This brings me to my next point...

Forwarding (and base safety)
It doesn't work! In reality, this is an important thing for humans. Humans need place to heal and restore ammo otherwise, they will rarely get to the alien base. Meeting an dretch with some poison is enough to take away half of your HP. However, currently, it's tough to get a forward up and once it is up, it's tough to keep up. Even if you keep it, then you tend to leave your main base open to attack. The second you leave it, it basically becomes free momentum for the other team, so, often having a forward ends up harming you more than it helps.
Possible Solutions:
Separate the notion of bases and forward
There is a key difference between bases and forwards: bases tend to be more permanent whereas forwards are more transient. Forwards help you control an area for a short period of time, but then you will probably want to push the forward closer to the enemy as you start winning or if you start losing, you'll want to start moving the forward back to avoid giving free momentum to the enemy. There needs to be more advanced marking techniques, like prioritizing which buildables are replaced first, and the buildables that are marked should always be used first. Alternatively, since decon is now free, bases can be moved almost instantly. The downside to this is that a builder must come to the base which exposes the builder to attack. In this regards, the queuing is going to be safer for the builder.

Variable strength defenses (and variable momentum decay/growth rate)
Ironically, the more forwards a team has, the more vulnerable that team is. Each forward must be defended and any that are not defended allow the enemy to gain momentum. Instead, as a team gains momentum, the defenses gain strength (Defenses close to your base are stronger than those closer to the enemy). Furthermore, a team loses momentum at the same rate the team earned that momentum. Basically each point of momentum lasts for X amount of time, where X varies with the threshold it is in. (ie, if you're at 200 momentum, you'll lose momentum faster than if you were at 100). This ensures that the top really powerful weapons are hard to keep and if a team fails to make an impact with them, they will lose them. This is also very fun because stopping a player with a huge weapon becomes a large accomplishment since unlike now, they might not be able to go back and buy it again. Anyways, by having variable strength defenses, the opponents are forced to decrease a team's confidence before they can make an attack. This forces skirmishes by stopping the enemy from killing your structures or your players in order to make a decent stand instead of relying on a lucky sneak attack on the main base.

A New Gameplay Dynamic:
I think implementing the above proposals can lead to a new gameplay dynamic, not unlike those of the now popular MOBA games. Image
There are several paths from one base to the other. Let's call these lanes. The point will be to fortify these lanes in order to secure access to the enemy base. You fortify these lanes with buildables. The enemy will do the same. You will then battle each other for control of these routes. The team that kills the other team more, establishes control of the lanes will gain more momentum and be able to make pushes to beat the other team. It'll be hard to run by the defenses because the base defense will be too strong if the momentum is high enough, therefore, the team will be forced to skirmish and build up their momentum at the cost of the enemy's momentum. As one team gains a clear momentum advantage it'll be easier for them to obliterate the other team with their superior weapons. However, if the losing team manages to stop a rush, suddenly they have the momentum and they can make a counter-push. The greater the momentum difference the less likely a counter push will be. I think his concept is called “annealing” where the game gets less chaotic and more stable as time goes on. This new dynamic will make games more fun by preventing camping but having strong bases such that you don't have to constantly be worried about a sneak attack. Emphasizes skirmishes instead of base killing. This game, at its core, as a strong deathmatch element and this will strengthen that.

Thoughts? Comments? Questions?

User avatar
Viech
Project Head
Posts: 2139
Joined: Fri Aug 03, 2012 11:50 pm UTC
Location: Berlin

Re: Playing To Win

Post by Viech »

So, let me pivot my screen on edge and get started…

Ishq wrote:

Background (of the most likely culprits)
[…] Our initial primary focus was to obliterate camping as a strategy and force teams to explore, expand, and attack bases. This change diminished the effect of feeding, which first led to "The Dretch Rush" which first emerged in the Pk vs get match in the summer tournament. That was promptly “fixed” by decreasing dretch HP, but the tactic of suicide rushing early and often gave aliens a significant advantage throughout most of unv. This was further exacerbated by the introduction of minimum wage, which allowed spam of leeches and mara rushing until the humans died.

Just to clarify this: We fought camping by

  1. Introducing a new building resource system that would make losing a buildable matter, since build points would no longer enter a queue and be given back to a team for free.

  2. Making the income of this resource depend on the size of the area that is controlled via buildables. Except for open areas and huge rooms, camping is inefficient when there is a lot of surface area to defend.

  3. Introducing a new team progress resource (momentum), that allows teams to advance by destroying structures and expanding over the map with buildables, in addition to the player kills that were formerly the only way to stage up and were generally easier to archive for campers with the support of base defenses.

"The Dretch Rush" is not directly linked to our anti-camping measures. It happend when we first increased dretch health to 35 hitpoints and still had the old GPP style turrets with a lock-on delay that was reset whenever their line of sight to the target was blocked. The old layout of Parpax allowed dretches to enter the human base quickly and the pillars there prevented the turrets from locking onto them. Since the dretches could bring the fight into the human base from the start of the match and get away with it, humans were unable to fortify sufficiently in order to leave their base. This is the point were our anti-camping measures made the life for humans even harder, but it was not the root of the issue. Decreasing health again was only a hotfix: We later implemented turrets with no lock-on delay to make early human bases dretch proof.

Minimum wage was an attempt to give a team with great map control (via buildables) a tool to end the match against a turtling team. On big maps, the system could easily be used to circumvent a team's personal resource economy, even when the enemy wasn't camping. I know that you still want teams to be able to generate personal resources when the enemy doesn't give them a chance to get any kills and I support the general idea. I'm still unsure (1) what kind of behaviour we want to reward and why being able to behave like this is a clear sign that the enemy turtles and (2) how exactly we go about connecting aspects of this behaviour to income without breaking the simplicity of the personal credit resource system.

The momentum system doesn't entirely remove the concept of feeding but it lets us control the magnitude of its effect. The key design mentality should be to create simple rules, that reward good behaviour without punishing bad behaviour. I will leave the definitions of good and bad extra vague here but in the specific case of feeding, killing an enemy would be considered good behaviour. Now the issue is that dying to an enemy is another player action that is involved in every kill. We don't really want to punish it, but we can't help the fact that the minimum punishment will always be that the enemy team gets a reward. So the only way to make feeding less of a negative game experience is to lower the reward given to the killer. By splitting the income for the team progress resource over multiple sources, the momentum system does exactly this. The result is that there are now many ways to feed the enemy, but ideally none of them is so noticeable that players will get angry at their teammates for it. This is not easy to achieve given that most "good" things we reward introduce some sort of "bad behaviour" twin. In the case of building, I found that giving a reward for every completed structure would balance the reward (and thus punishment) associated with the destruction of a (badly placed) building. If this doesn't work out, there are three ways to fix it: Adjust the momentum knobs until no one notices the effect anymore, make it harder to place buildings in a way that it can backfire strongly or limit the amount it can backfire in a short time.

Speaking of the momentum system, it is often misunderstood in my opinion. By design, it has only two rather abstract goals: Segment the game into a sequence of phases by controling the availability of certain items and allow us to fine tune the amount of match stability by introducing a parameterized positive feedback loop. The positive feedback loop is exactly the system of rewards I wrote about in a previous paragraph. Why would we want one, anyway?

The winning condition of Unvanquished is destroying the enemy base and killing all enemies. There is no point system that would make you win the match when you had more points than the enemy team after a fixed match duration. This means there is not just a single main goal for players (i.e. score more points), but there are two: Win and don't lose or in other words attack and defend. A certain balance is necessary between the two: If attacking is too easy, games are chaotic, very short and lack the depth of ongoing strategic decisions. If defense is too easy, games stall and player actions become meaningless as no team would be able to win.

Now it should become apparant that the two goals of the momentum system boild down to one: Control the balance between the core features of the game over time. By limiting the access to weapons that are efficient for base destruction in the beginning of the match and unlocking stronger ones as the time goes by, games will be long enough to be complex but will end with a winner eventually. Now from a developer's perspective, the easiest approach to do so is unlocking item one at minute five, item two at minute ten and so on. The fun part is that the momentum system seems to do exactly that when you average over a number of matches (cf. the dev game analyses). However, enforcing this rigid scheme in every match takes away great opportunities to allow players to do meaningful actions. If the game ends in minute 20, why bother giving your best in minute five? This is where the positive feedback loop kicks in and that's exactly why we are free to define the vague term of "good behaviour" to anything that we consider a meaningful player action (or in other words any behaviour that can bring players in a flow state).

There is one question left, which is why momentum controls match stability. The feedback loop I just explained doesn't only give players the mere impression that they do meaningful actions, it also actually allows them to put their team into a better position. In this context, instability means that a successful strike against the opponent will make any further strike stronger or easier. The team that just suffered a loss is facing an even bigger threat than before. Strong instability can lead to a slippery slope, where a single strike can turn the game so strongly in a team's favor that they will be able to end it. We want a certain degree of instability because only the option to generate a significant advantage makes player actions throughout the entire game matter. However, we want those advantages to be limited in time, to avoid the frustration of losing to a mistake or lost struggle that happened a while ago. For this reason, (1) the momentum resource is subject to an exponential decay and (2) momentum rewards start small and become bigger over the course of the game. (1) means that any resource peak will be lost in case the team can't hold their level of successful activity (despite the unlocked items). (2) means that rewards in the beginning of the game are relatively small and thus won't have a big influence on the late game. At the same time it makes sure that the late game won't stall by making strong weapons easier to access and maintain.

Ishq wrote:

Current State:
The culmination of all these changes results in two mindsets: “play for fun” […] and “play to win.”

[…]

The Goal:
For me, the goal is to create gameplay in which the “play for fun” and the “play to win” mindsets are the same. By playing normally and by doing the right things you will move closer to winning […].

I can agree fully here. So let's move on to the problems you see.

Ishq wrote:

Building
Building has become less strategic and more spammy. Players don't need to carefully position buildables or even use their buildables intelligently. As long as they clump their buildables together in some fashion, they will easily dispatch most aliens. Furthermore, with the advent of drills, there is generally a large surplus of build points, so any defenses you kill, are sure to reappear by the time you get back to try to kill another. This is boring. Bases don't require skill anymore to build and to kill. There are now two ways to kill a base: hit the base while the enemy is out (boring), or just use pure brute force and kill each and every structure with your team's overwhelming map control/money/skill advantage (too long. Boring).

I agree that this kind of base building happens, but let's be more specific here: It happens mostly to human bases which, unlike aliens, (1) can't persist without a sufficiently strong setup of defense structures and (2) have a single defense structure to choose from. We intentionally made this single structure so strong that it can fulfill all of the various defense tasks that need to be done, which means that it will essentially not need much skill to be used correctly as every placement will do something for your base. More importantly the only way to improve your base's defense is to add more turrets, what else would you do? In Tremulous, you couldn't build this many turrets, so the only viable defense was not leaving the base as long as the aliens had enough evos to hurt it. Now humans can leave their base even in that case (just like aliens can). The downside is what you describe as the limited and boring attack vectors that are left to aliens. (However, they are also limited in themselves. The only reliable and efficient way to take down a strong human base is tied to one alien class and its special attack. I'm not telling which one, we all know it by heart. I will get back to this after I analyzed your solutions.)

Ishq wrote:

Make drills/leeches not only mine, but store resources.
Losing a drill will result in losing all unused BP currently in the drill. A big problem is that drills/leeches are rather important structures, but they are not treated as such. They are cheap to build. And there is almost no penalty if one is destroyed. In fact, spamming them around the map is beneficial because unless it's late in the game, even if the drill stays alive for a minute, the drill will most likely already pay back its worth. By making them structures that can harm you significantly if lost, will encourage teams to guard them more carefully, and opponents to target them. If a drill is killed, the team potentially loses the ability to rebuild from future attacks.

The fun part is that I had the very same idea for totally unrelated resons. (I had like, the most awkward flash of inspiration: I was watching a Starcraft match – I started doing so in order to undertstand our legacy better and maybe get some inspiration from it but ended up enjoying to watch them even though I never played the game – and when I learned those small square terran structures were called "supply depots" I literally paused the game, spoke it out loud and realized this is the exact thing we need to prevent teams from mining a lot early and living on those BP throughout the entire game.)

So, I already made some plans how to implement it to solve the greatest amount of problems possible. This really deserves another thread, but here's the mechanics in short:

Resource generating structures (RGS) …

  • Have a maximum BP capacity.

  • Store mined resources locally, using their own capacity.

  • Stop to mine when they're full. (I did not yet decide whether it was better if they still influenced other RGS or shut down entirely.)

  • Lose their stored resources when destroyed.

The act of building …

  • Takes resources from one virtual global pool.

  • Really substracts resources from all RGS equally, regardless how full they are. (Given that they have enough to pay their share, otherwise they will be emptied.)

Expected effects are …

  • RGS that (1) live long or (2) have a high efficiency (= low interference) will generate and store more resources.

  • Destroying these will have a greater impact.

  • RGS spamming is only viable if the BP are used right away, since destroying the spammed RGS takes their share of BP out of the pool.

In addition, there needs to be another BP storage that stores the initial build points as well as the minimum BP income (currently 50% of level mine rate when the main structure is up). It would also be the place where build points are transfered to when RGS are desconstructed (we would muliply the transfered amount with the RGS health fraction to prevent saving BP when the RGS gets attacked). It needs a maximum capacity, too, I'd guess up to twice the initial build point amount works. The place where all this happens could be the OM/RC. This has significant side effects, as destroying the main building as well as all RGS would end the game. We should keep in mind that this option exists. I'm thinking that the most elegant solution for a start would be making it look like the main structure stores these resources, but add them back whenever it gets rebuilt. This allows us to show builders exactly where their resources are stored.

I could start implementing this right away if it sounds good to you.

Ishq wrote:

Make defense structures harder to build, harder to kill
A defense structure can be made harder to build by making it more expensive, take longer to build (this might make things less fun, so I'm less enthused by this. I quite like fast build). This will make it more difficult to spam buildables and make each loss matter more.

I doubt this fixes the problem. If turrets are stronger and more expensive, this doesn't change the fact that your only chance to improve your base's defense is to build more of them. It will look less spammy and placement will matter more, but once your base deals so much damage that only the advanced dragoon can hurt it reliably, the alien attack vectors will be as limited and boring as before. In addition, it would make suicide attacks less efficient and while I know that you are not quite fond of them, this would limit the available attacks even more. I think what we need instead is more variety in both defense structures and base attack weapons. Placement and building skill would matter as you would need to carefully balance your set of defenses so that the weaknesses of one type are evened out by the strengthes of another. When the rocket pod is introduced, we can make the turret less universal and make a good mix of the two a requirement for a strong base.

So, let's get to the next issue.

Ishq wrote:

Forwarding (and base safety)
It doesn't work! In reality, this is an important thing for humans. Humans need place to heal and restore ammo otherwise, they will rarely get to the alien base. Meeting an dretch with some poison is enough to take away half of your HP. However, currently, it's tough to get a forward up and once it is up, it's tough to keep up. Even if you keep it, then you tend to leave your main base open to attack. The second you leave it, it basically becomes free momentum for the other team, so, often having a forward ends up harming you more than it helps.

This sounds like a on of individual issues combined:

  1. Humans are more reliant on a base than aliens because there is no other way for them to refill ammo and health.

  2. The alien guerilla tactics make this matter worse. Even with good equipment you can easily be stopped by an inexpensive alien.

  3. Human forwards are hard to keep alive. This is probably because they are so dependent on just the right number of defense structures.

  4. When forwards get destroyed, this happens very quickly and this leads to a momentum peak. This depends on the previous point but it is also true for aliens that have a weak spot in all their forwards (the egg). I think this is fine for aliens as useful forwards are rather inexpensive and thus don't lead to a high momentum peak (except whenever a newbe makes a lot of structures depend on a single egg).

Ishq wrote:

Separate the notion of bases and forward
[…] Forwards help you control an area for a short period of time, but then you will probably want to push the forward closer to the enemy as you start winning or if you start losing, you'll want to start moving the forward back to avoid giving free momentum to the enemy. There needs to be more advanced marking techniques, like prioritizing which buildables are replaced first, and the buildables that are marked should always be used first.

I'm not against this, but it will save your structures only in a subset of all relevant scenarios. This would certainly work for scrims but not too well in a public match, where a lot of inexperienced builders can take a great risk for their team.

Ishq wrote:

Variable strength defenses (and variable momentum decay/growth rate)
Ironically, the more forwards a team has, the more vulnerable that team is. Each forward must be defended and any that are not defended allow the enemy to gain momentum.

This is true, though not an issue in my opinion as long as the forwards are also a significant help in winning the match.

Ishq wrote:

Instead, as a team gains momentum, the defenses gain strength (Defenses close to your base are stronger than those closer to the enemy).

I'm strictly against invisible buffs. This takes the notion of skill and experience out of the game as players never know what they are fighting against. Furthermore, if this strengthens the main base and not the forwards, it does little to counter any of the issues you outlined. What we can do however, is introducing unlockable technologies, activated team powerups or structure upgrades – all of them with a visible effect and a clear outcome! I already commited a "hook" that makes the turret skip every forth shot by default. The aforementioned unlockable could revert this behaviour (temporarily or (semi-)permanently) which would lead to a 33% damage boost for the affected turrets that alien players would notice both by vision and audio after facing them for just 375 ms, when the usual fire pattern has turned into continuous fire. I can add smiliar hooks to all the other defense structures (e.g. have the rocket pod fire a bigger volley, make the spiker regrow faster, make the acid tube do more damage and change the gas color).

Ishq wrote:

Furthermore, a team loses momentum at the same rate the team earned that momentum. Basically each point of momentum lasts for X amount of time, where X varies with the threshold it is in. (ie, if you're at 200 momentum, you'll lose momentum faster than if you were at 100). This ensures that the top really powerful weapons are hard to keep and if a team fails to make an impact with them, they will lose them.

My response to this should be contained in my essay on the momentum system. TL;DR: This is exactly what the system already does, though it is configured not to overdo it. In the late game, access to big weapons should not be hard to maintain since this would make games stall (we had this situation quite a lot in early momentum dev games, which is why I implemented growing rewards in the first place). In the earlier game, you can only keep your momentum if you are able to turn the newly unlocked weapons into action quickly, as rewards are small and momentum decays exponentially. If you have a closer look you will notice that players already do that. Once the firebomb gets unlocked, everything starts burning out of a sudden. As soon as the goon is available, you are facing quite a lot of them and your turrets start to be eaten. The lucifer cannon? Players abort their attacks and run back to base to get it, shortly after you will lose parts of your base you deemed safe. So I guess this works.

I don't have one magic receipt to fix this problem. I think we should just constantly keep it in mind and do small changes until forwards are as strong as they should be. With regards to the momentum peaks, I would be open to reducing momentum gains for killing structures based on the number of structures that were destroyed shortly before that. Sounds good?

I will leave your gameplay dynamic proposal mostly uncommented since I feel this is a good motivation for individual design decisions but we won't be able to implement it in its entirety or even predict how the game would be if we did. Keep in mind that Tremulous UBP matches on certain maps (with distinct pathces between the bases) played very much like that. You turtle-expanded slowly from your base outwards (in your proposal, this would also be the most rational move) until you hit the enemy front, then you started to look for the path of least resistance and push harder than your enemy there, while at the same time trying to maintain your defense on the other pathes. The rules of such a game mode are certainly simpler to control than those of the current one, but we might also be losing strategic depth and match variance in exchange.

Responsible for: Arch Linux package & torrent distribution, Parpax (map), Chameleon (map texture editor), Sloth (material file generator), gameplay design & programming, artistic direction

User avatar
Ishq
Project Head
Posts: 1145
Joined: Tue Mar 06, 2012 8:32 pm UTC

Re: Playing To Win

Post by Ishq »

"The Dretch Rush" is not directly linked to our anti-camping measures. It happend when we first increased dretch health to 35 hitpoints and still had the old GPP style turrets with a lock-on delay that was reset whenever their line of sight to the target was blocked. The old layout of Parpax allowed dretches to enter the human base quickly and the pillars there prevented the turrets from locking onto them. Since the dretches could bring the fight into the human base from the start of the match and get away with it, humans were unable to fortify sufficiently in order to leave their base. This is the point were our anti-camping measures made the life for humans even harder, but it was not the root of the issue. Decreasing health again was only a hotfix: We later implemented turrets with no lock-on delay to make early human bases dretch proof.

I was merely illustrating an example of the side effect of our camping measures. By the 3 points you mentioned, we also reduced the effects of feeding, which, coupled with the 35 hp dretch, made it a nearly a guaranteed win strategy. This is part of the play to win mindset, as this isn't necessarily fun.

<Minimum wage + momentum explainations>

Thanks for clarifying.

I agree that this kind of base building happens, but let's be more specific here: It happens mostly to human bases which, unlike aliens, (1) can't persist without a sufficiently strong setup of defense structures and (2) have a single defense structure to choose from. We intentionally made this single structure so strong that it can fulfill all of the various defense tasks that need to be done, which means that it will essentially not need much skill to be used correctly as every placement will dosomething for your base. More importantly the only way to improve your base's defense is to add more turrets, what else would you do? In Tremulous, you couldn't build this many turrets, so the only viable defense was not leaving the base as long as the aliens had enough evos to hurt it. Now humans can leave their base even in that case (just like aliens can). The downside is what you describe as the limited and boring attack vectors that are left to aliens. (However, they are also limited in themselves. The only reliable and efficient way to take down a strong human base is tied to one alien class and its special attack. I'm not telling which one, we all know it by heart. I will get back to this after I analyzed your solutions.)

Yes, this is true. I can agree fully here. Now, the point is to find ways to fix this. Addition of new structs may help, but really, even then the main way to strengthen the base will be simply to build more of them, except now, new structs will be built in conjunction with turrets. We've discussed the possibilties of additional alien attacks that can counter this, but then this will result in humans not wanting to leave their base again. This is certainly a hard problem to crack. How can we make bases fun to attack, fun to build, strong enough to withstand an attack, yet weak enough to be destroyed with the proper rushes as opposed to being slowly picked off? This is the fun of balancing a game as complex as this :) You are correct in your analysis of my proposed solution, and it seems unlikely to actually fix the issue. We'll probably need a combination of new attacks and new defenses that will hopefully yield the balance we desire.

I'm not against this, but it will save your structures only in a subset of all relevant scenarios. This would certainly work for scrims but not too well in a public match, where a lot of inexperienced builders can take a great risk for their team.

Perhaps, but I honestly expect a lot of the general gameplay and bases used in scrims to spill over into pubgames. Obviously not the tactics and the teamwork, but bases and such since scrimmers will also play public games and build in them, teaching players the ropes. You'll always get new players doing stuff they probably shouldn't, and I think that if we try to factor them into our balance we'll never really be able to achieve decent balance. I'm still strongly in favor of some level of matchmaking which will keep games between similarly skilled players together. Of course this is a long term goal, and in the short term (especially during devgames), we can try our best to help new players by teaching them more of the basics before they begin and then let them improve more as they play. I think its possible, especially when you see some of our current new players who have embraced building and gotten rapidly better at building.

I'm strictly against invisible buffs. This takes the notion of skill and experience out of the game as players never know what they are fighting against. Furthermore, if this strengthens the main base and not the forwards, it does little to counter any of the issues you outlined. What we can do however, is introducing unlockable technologies, activated team powerups or structure upgrades – all of them with a visible effect and a clear outcome! I already commited a "hook" that makes the turret skip every forth shot by default. The aforementioned unlockable could revert this behaviour (temporarily or (semi-)permanently) which would lead to a 33% damage boost for the affected turrets that alien players would notice both by vision and audio after facing them for just 375 ms, when the usual fire pattern has turned into continuous fire. I can add smiliar hooks to all the other defense structures (e.g. have the rocket pod fire a bigger volley, make the spiker regrow faster, make the acid tube do more damage and change the gas color).

Yes, these buffs should not be invisible as that would be confusing. However, I disagree that these do not address the issues above. By forcing teams to first destroy forwards and take back control of the map instead of rushing the main base when its empty, it allows both teams to rush without filling their base with large amounts of defenses and focus their energy on the front line of the battle instead of in base. The game should get increasingly unstable with momentum. I've mentioned this before, but really OP weapons at the higher momentum tiers would allow the winning team to wreak havoc, and if they fail, allow the other team a chance for a comeback.

Lastly, for the gameplay dynamic thing, it isn't quite the same as uBP or a turtle base simply because of the magnitude of the number buildings is not the same. In uBP (and, basically in our current gameplay), players build a large number of buildables, and these must be destroyed before the enemy can move on. This takes a long time. Instead, I'm stressing fewer buildables that are stronger so there is less to kill to move on. This will hopefully not take as long, and therefore increase the pace of the game.

Post Reply