The Slippery Slope. How steep should it be?
Roobubba
Who you gonna call? Join Date: 2003-01-06 Member: 11930Members, Reinforced - Shadow, WC 2013 - Shadow
Who you gonna call? Join Date: 2003-01-06 Member: 11930Members, Reinforced - Shadow, WC 2013 - Shadow
NS2, like NS1 before it, has a slippery slope. Any game in which the collection of resources makes it easier to collect more resources (and simultaneously easier to deny the opposing team's resources) has this same positive feedback mechanism. The better you do now, the even better you will do later.
This doesn't mean NS2 can't be amazingly good fun, and it doesn't preclude some awesome, close games. But it does mean that every encounter won has an increased likelihood of an easier-to-win encounter later. This is the essence of the slippery slope.
The question I'm posing in this thread is this: How steep should that slippery slope be?
I'll put this note here... this thread is not about concede. There are lots of threads about that, please do NOT discuss that here.
There are several possibilities, and I'll take a bit of time to go through them.
1. Linear slippery slopes
a) Less steep than present
The winning team gets an advantage, but it's not so dramatic as at present. One way to make this happen is (as I've posted elsewhere) to recover a portion of resources which would otherwise be lost in some way. This might be linked to structures destroyed: so when a building is destroyed, 25-50% of its initial build cost is recouped over the next 10-30 seconds. I like this idea, because it lessens the impact of losing map control a bit, but doesn't detract from the victory that the team taking down the buildings just gained.
Another way to achieve this might be to have diminishing returns on multiple resource towers, either directly or by some sort of upkeep system (a percentage of resources removed to maintain powered structures). The drawback with this system is that the most important structures are those first ones that you get, and that later structures are comparatively less important. Therefore, losing one of your first 3 RTs would be *more* of a big deal than it is now, so I think this might not help with the slippery slope, but just make it a bit harder in the late game than it currently is.
A different way than upkeep might be to simply tweak the costs of certain upgrades. If early upgrades were more expensive, and a little less powerful, but later game upgrades (those linearly dependent on earlier ones, especially, like w3/a3 and jetpacks, or tier 3 upgrades) more powerful, this might have the same effect. Not sure about this bit.
b) The same as present
No change needed to achieve this. Why are you still reading?
c) More steep than at present
The winning team gets more of an advantage than they currently do. An example might be to charge more to place a building if a building of the same type has been destroyed before. I can't see any case for making the slope steeper than it currently is, but feel free to comment if you think I'm wrong about this!
2. Non-linear slippery slopes
a) Slope is shallow at the start, and gets increasingly steeper
The idea here is to slow down the early game a bit, make early game encounters less important than they currently are (ie a little more forgiving). A team has to put more small victories together in order to gain an appreciable advantage over their opponents. Then as the game progresses, the team that's ahead can build on that and get more ahead, and the rate at which they gain the upper hand increases as they do so. This should allow for less painful end games, as any team getting that far ahead will have significantly better tech than their opponents and be able to break turtles better/finish final hives faster. This would mean less easy comebacks in the lategame, but more chance of not being so far behind in the early game (therefore overall a closer gaming experience).
There could be several ways to achieve this, and these options are far from comprehensive:
One method to achieve this would be to tie the output of a resource tower to the length of time it has been constructed. The longer you hold onto your RTs, the more res you get.
Another would be to tie the res output to number of RTs your team holds. Your first few extractors get you some res, but not a lot. If you get up to 6 or 7, they really start churning out a lot of res. This might not be a good way of keeping the early game slope shallow - would need to think hard about numbers...
A third idea is to have an efficiency upgrade for res collection, or allow each tech upgraded to also improve efficiency of res towers. The idea here being that if you have researched more tech, you can gain more res. Exactly how this might work in practice isn't clear to me now, I'm just trying to think of ways to achieve the slippery slope effect...
b) Slope is steep to begin with, but becomes shallower
I can think of an example of how to achieve this, but I honestly can't see this being a useful way forwards so will omit it from this discussion.
Having a slippery slope which starts out shallow (early killing of extractors or harvesters doesn't kill a team as much as at present) and gets steeper as one team draws ahead of the other is, I think, a good goal to have. I don't know how to achieve that just yet, and hopefully this discussion will generate some good ideas here!
The most important aspect of this, from my perspective, is reducing the gradient of the slippery slope in the early game, regardless of what happens later on. One way to achieve this would be to have a mod that implements one of my suggestions in section 1a (like refunding a portion of resources when buildings are destroyed), so we can see how/whether this actually works in practice. I would do this if I had any clue how to, but I'm sorry - I don't!
So, what do you think? Do you think there is a need for early game disadvantages to be less severe than they currently are? Is this all guff and you don't agree with any of it? Should small victories in the beginning lead to smaller, or even bigger victories later in the game?
I personally think that recouping a portion of the cost of structures that have been destroyed is a good way to reduce the severity of the slippery slope. I'm keen to see how this plays out. It would be easy to come up with a lore reason for this (eg nanites recycling broken material, bacteria digesting destroyed structures), but frankly that is the least of my considerations just now.
Thanks for reading, and please discuss!
Roo
This doesn't mean NS2 can't be amazingly good fun, and it doesn't preclude some awesome, close games. But it does mean that every encounter won has an increased likelihood of an easier-to-win encounter later. This is the essence of the slippery slope.
The question I'm posing in this thread is this: How steep should that slippery slope be?
I'll put this note here... this thread is not about concede. There are lots of threads about that, please do NOT discuss that here.
There are several possibilities, and I'll take a bit of time to go through them.
1. Linear slippery slopes
a) Less steep than present
The winning team gets an advantage, but it's not so dramatic as at present. One way to make this happen is (as I've posted elsewhere) to recover a portion of resources which would otherwise be lost in some way. This might be linked to structures destroyed: so when a building is destroyed, 25-50% of its initial build cost is recouped over the next 10-30 seconds. I like this idea, because it lessens the impact of losing map control a bit, but doesn't detract from the victory that the team taking down the buildings just gained.
Another way to achieve this might be to have diminishing returns on multiple resource towers, either directly or by some sort of upkeep system (a percentage of resources removed to maintain powered structures). The drawback with this system is that the most important structures are those first ones that you get, and that later structures are comparatively less important. Therefore, losing one of your first 3 RTs would be *more* of a big deal than it is now, so I think this might not help with the slippery slope, but just make it a bit harder in the late game than it currently is.
A different way than upkeep might be to simply tweak the costs of certain upgrades. If early upgrades were more expensive, and a little less powerful, but later game upgrades (those linearly dependent on earlier ones, especially, like w3/a3 and jetpacks, or tier 3 upgrades) more powerful, this might have the same effect. Not sure about this bit.
b) The same as present
No change needed to achieve this. Why are you still reading?
c) More steep than at present
The winning team gets more of an advantage than they currently do. An example might be to charge more to place a building if a building of the same type has been destroyed before. I can't see any case for making the slope steeper than it currently is, but feel free to comment if you think I'm wrong about this!
2. Non-linear slippery slopes
a) Slope is shallow at the start, and gets increasingly steeper
The idea here is to slow down the early game a bit, make early game encounters less important than they currently are (ie a little more forgiving). A team has to put more small victories together in order to gain an appreciable advantage over their opponents. Then as the game progresses, the team that's ahead can build on that and get more ahead, and the rate at which they gain the upper hand increases as they do so. This should allow for less painful end games, as any team getting that far ahead will have significantly better tech than their opponents and be able to break turtles better/finish final hives faster. This would mean less easy comebacks in the lategame, but more chance of not being so far behind in the early game (therefore overall a closer gaming experience).
There could be several ways to achieve this, and these options are far from comprehensive:
One method to achieve this would be to tie the output of a resource tower to the length of time it has been constructed. The longer you hold onto your RTs, the more res you get.
Another would be to tie the res output to number of RTs your team holds. Your first few extractors get you some res, but not a lot. If you get up to 6 or 7, they really start churning out a lot of res. This might not be a good way of keeping the early game slope shallow - would need to think hard about numbers...
A third idea is to have an efficiency upgrade for res collection, or allow each tech upgraded to also improve efficiency of res towers. The idea here being that if you have researched more tech, you can gain more res. Exactly how this might work in practice isn't clear to me now, I'm just trying to think of ways to achieve the slippery slope effect...
b) Slope is steep to begin with, but becomes shallower
I can think of an example of how to achieve this, but I honestly can't see this being a useful way forwards so will omit it from this discussion.
Having a slippery slope which starts out shallow (early killing of extractors or harvesters doesn't kill a team as much as at present) and gets steeper as one team draws ahead of the other is, I think, a good goal to have. I don't know how to achieve that just yet, and hopefully this discussion will generate some good ideas here!
The most important aspect of this, from my perspective, is reducing the gradient of the slippery slope in the early game, regardless of what happens later on. One way to achieve this would be to have a mod that implements one of my suggestions in section 1a (like refunding a portion of resources when buildings are destroyed), so we can see how/whether this actually works in practice. I would do this if I had any clue how to, but I'm sorry - I don't!
So, what do you think? Do you think there is a need for early game disadvantages to be less severe than they currently are? Is this all guff and you don't agree with any of it? Should small victories in the beginning lead to smaller, or even bigger victories later in the game?
I personally think that recouping a portion of the cost of structures that have been destroyed is a good way to reduce the severity of the slippery slope. I'm keen to see how this plays out. It would be easy to come up with a lore reason for this (eg nanites recycling broken material, bacteria digesting destroyed structures), but frankly that is the least of my considerations just now.
Thanks for reading, and please discuss!
Roo
Comments
That is how I feel in NS2 when talking about the slippery slope and how it should be (in blue; it is your point "2a" I think). You can see base-rushs and turtles in the red graph.
I simply think, that close games where both teams struggle for the victory are much more fun. Stomping the other team or getting stomped is not fun.
Some people think, that a game should end fast when one team is much more skilled than the other. I agree partially. But only think that for the extreme cases. In the normal case where one team is only marginal better, the better team should not snowball into getting more and more advantages. The game mechanic should automatically try to even the playing field. Or at least don't let the slightly better team get horrendously better, only because they won early fights.
It is a difficult task to not devalue early successes, but remain the game fair over the cause so the outcome doesn't get predictable.
The same goes for aliens vs phase gates. The amount of rooms Marines hold for the whole game (mostly) comes down to how far they can push with your first life until phase tech.
I think the sweet spot for claimed RTs is 3.5 aliens and the rest claimed and contested by Marines. If Aliens have 4 from the start of the game they will probably win. If Marines can keep aliens on 2.5 average RT (getting RT kills) then they will win, assuming they hold the rest.
Think about it, if you lose an RT 8 minutes in does it even matter? Not really, you expect to lose buildings at that point and you've already gotten past the crucial setup phase. The later the game gets, the less each individual move matters, which is why comebacks are so neigh impossible to pull off short of a total failure on your opponent's part.
I think I agree that slower start would be most desirable. As it is unless your comm knows exactly what he's doing at the very beginning, your team better be amazing to get a good shot at winning.
If you flatten the curve it becomes like....well it doesn't really matter what strategy you do or how well you execute the early game because it all just comes down to a big fight at the end of the game.
The curve is already reasonably flat in that if you lose all your Pres stupidly, even if you were way ahead, you can be back on even footing.
I feel this would be terrible, the team ahead would then have 3 "powered up" nodes, the team thats losing would have less because they are probably defending more if they are losing. At this point diminishing returns work better and do what this wants to achieve.
Example
1st RT - 10
2nd RT - 9
3rd RT - 7
4th RT - 5
5th-10th RT - 4
The problem with all these systems is they add more complexity to an already complex game. This makes it harder for new players to start playing. I mean for the first 10 hours of this game I was not sure what end of the gun the bullets came out of.
You're presuming strategy already has a large role in determining victory (it doesn't). Making the FPS portion work better at the expense of the RTS side is a pretty beneficial trade-off.
If a restower doesnt start giving res untill its been active for a minute orso it would force the need for defense of the res a lot more then it is now.
Also I am for 2a.
Making mistakes early game, which becomes increasingly unforgiving later on. No clue how to get that result though.
in the late game, almost everything falls into obsolescence if you have enough coin. i wonder about having more specific counters, e.g. make grenade launcher especially useful against onos, flamethrower especially useful against fade, onos charge especially useful against exosuit, vortex especially useful against forward phase gate spamming etc.
currently we have a huge amount of asymmetry and almost zero 'counter strategies'. ergo the game balance is currently hinged on the state of marine versus skulks.
Strategy in terms of build orders and tech progression doesn't but strategy in terms of tactical decisions and responses to your opponents tactics are game breaking.
The power/infestation creates much more of a defined front line in the combat. It isn't nearly as easy to set up bases in a territory behind the enemy. Marines can't quickly relocate because a) they can only drop a chair in a tech point, b) they can't build on infestation and most alien territory is infested, and c) they need to power their structures, and the power may very well be off in alien territory to prevent that from happening. Likewise, aliens aren't able to set up forward bases as easily as they could in NS1 because they need infestation to be there - it's no longer as simple as dropping a couple OCs and DCs. Because of this, you don't get the feeling that you can come back nearly as often. In NS1 it always felt like you at least had a bit of a shot.
Not saying that one is necessarily better than the other, though. There are ups and downs to both. There was nothing more fun than relocating marine base to some random location in the map (especially the vents), but at the same time I like how the new mechanics give the game a bit more of a streamlined sense of direction - there's much more of a sense of "alien territory" vs. "marine territory" in NS2.
Then there's the fact that there's so much more tech in NS2. It costs more resources, it takes longer to fill up, and it requires that you control more of the map. Marines could get shotguns as soon as an armory was dropped. They could have HA and JPs even if they were backed up into just one room. Likewise, the aliens could get lerks and fades pretty quickly even with only a couple RTs, and their tech was less dependent of having multiple hives - fades still had blink even with one hive, lerks had spores, etc. In NS2 they changed it so that it takes longer to get tech up. You could realistically have fades or shotguns within a couple minutes in NS1 if that's how you wanted to play, now shotguns won't show up until generally 5 min or so and fades don't drop until 10. You need much more resources to fully tech up (marines need to research all of their weapons, aliens need to research every ability and upgrade) so controlling RTs becomes significantly more important, and teams are much more gimped when they only have 1 hive or 1 CC (moreso for aliens than marines) so it becomes more of an impossibility to come back.
Once again, there are ups and downs to both. Having tech take more time and cost more resources puts more importance in map control and the early game, but at the same time it makes the late game feel more hopeless if you don't have map control. So it's really up to you to decide which model you like more, but as long as the infestation/power mechanics and slow tech remain in NS2 I think it will inherently have a much steeper and more slippery slope than its predecessor.
EDIT: Also thought of another thing... RFK in NS1 made a big difference in keeping a team with less map control in the game. Even if you held less territory, if you were winning engagements you could stay in the game. In NS2 if your team fails to expand enough in the opening minutes, you can win every engagement from then on and still not be able to get back in it. Just another interesting change worth thinking about.
RFK assists the winning team significantly more than it helps the losing team.
Hand Grenades would help so marines have a chance at breaking alien turtles without the insane time/res investment of GLs and possibly reducing the maximum HP of cysts, increase recede rate, reducing max Drifter count, reducing scan cost.
I think the red line on the graph is an accurate representation of the game flow at the moment. Do we generally agree on that? Okay. I think that red line is exactly as it should be. The game doesn't HAVE a simple slippery slope- it has a slippery bumpy slope.
Teams have multiple opportunities to potentiate a comeback throughout the game, and if you stop to think for one second, your blue line of a game would just create lots of games where it's clear who has the higher base skill level at the start- but the teams are then forced to play out almost the entirety of a game that in their minds is already decided.
I like the flow of the game at the moment, and I try to stage my strategies around the progression of the alien tech tree- each time a new lifeform or skill (bilebomb) comes out, you have effectively missed a chance to end the game with a victory for your team (- for marine, or just gained one if you are alien) , and now have to tech up to deal with the opposition and even out the arms race. This is how it should work. Goddamnit.
In most cases, it absolutely does. I was mainly just talking about a scenario where a team has a lot of talented players but was unable to branch out quickly enough in the early minutes for whatever reason. In NS2 that team wouldn't be able to come back (I've been on plenty of teams that are more skilled than the other team but are forced to concede because they can't reclaim map control), in NS1 they could if they were able to win engagement against players with slightly higher tech.
But yes, in a situation where both teams are of roughly equal skill, RFK favors the team that has more map control.
In general, I think forcing teams to spread around the map is a vital comeback mechanic in melee vs ranged gameplay. The more map movement you have going on, the more openings both teams have for getting good engagements and the more possibilities the losing team has for making game changing plays.
If the positions are solid and the engagements happen in predictable locations, the guy with the bigger gun or beefier lifeform is going to have an upper hand big time.
NS2 certainly would be.
Ideally, in a game, there are numerous decision points spread throughout the playing time, and ideally, every decision is important and every time period has important decisions. Severe early snowballing means that the early decisions are vital but the later decisions are either trivial or futile. By contrast, in the other extreme, where comebacks are always equally possible at any time (i.e. the "slippery slope" is perfectly flat) then the early decisions are unimportant because no matter how well or poorly you play early on your chance of winning later hasn't changed.
So for a good game, you want to reward good early play with an enduring advantage and punish early mistakes with enduring disadvantages. But the advantages and disadvantages should not be so substantial that they determine the course of the game from the outset. If in the early game Team A plays well and Team B makes mistakes, it should still be possible for Team B to win IF they play well the rest of the game AND if Team A makes mistakes.
The other important thing is to raise the stakes over time. Early mistakes should set you back, but not too badly, so that you can still come back if you play well after that. Through the midgame mistakes should be more costly; comebacks should still be possible, but too many mistakes unmatched by the opponent should effectively put you out of the running. And at the endgame, the stakes should be so high that a single mistake will end the game right there.
This is the basic recipe for an exciting and engaging strategy game. NS2 isn't quite there, but it could get there with a little more work - work which UWE is still doing even now.
However, using the NS1 example, RFK also enabled endless Marine turtle holdouts. I just do not see a place for RFK in NS2. It only furthers the ability for a single skilled player to control the whole game. Best example is a Pro Rockstar Fade, who after getting sufficient res to go fade shuts down the whole marine team. And, should someone actually kill him, he certainly has the res to go fade again or worse Onos. All this well before anyone else on the server has the means to adequately stop them. It greases the slope way too much.
Just pointing out some of the differences that make it seem like comebacks/long close matches were more common in NS1.
Meanwhile if the RFK just flows to tres pool, it doesn't really make such a big deal. If both teams have similar RFK mechanics, the winning team is probably getting more benefit out of the RFK.
when people say they can't recover from an early mistake it's rarely because they actually made a strategic error. usually they can't recover because they are playing against a team with superior combat skills who also happened to win the early fights.
The only reason NS2 functions as designed is because some relatively rare games happen to get fair teams and actually go beyond having the winner decided at the 5 minute mark. In my opinion, NS2 is at its best in the mid game, and I think the game would improve if more rounds got to that stage. Right now the only way that happens is if the teams are very close in skill.
Changing the game to reduce the importance of early fights would increase the length of some of the games with unfair teams, but that's fine if there was also something that allowed a team to actually end the game and defeat a turtle.
Basically I think the game would improve overall if decisively winning by 5 minutes was harder as long as actually forcing the game to end by 10 or 15 minutes was easier. I think the players who are skilled in the early game and want that to be rewarded are also skilled in the mid-game, so it could theoretically work without upsetting too many people...