Kouji_SanSr. Hινε UÏкεεÏεг - EUPT DeputyThe NetherlandsJoin Date: 2003-05-13Member: 16271Members, NS2 Playtester, Squad Five Blue
edited March 2013
Like I said JAMESEARLJONOS, use the pistol and slow down the video before slapping it on Youtube. Freezeframe the frames when you shoot and hit/miss. Without freezing the frames we're just guessing, you'd be surprised how many of the "hits" are actually not on the model
Also the spread of the rifle is simply too big to make accurate assessments regarding hitreg...
Unfortunately the pistol also has a degree (as in 1) of spread, but its far more reliable as a testing mechanism than the rifle. Personally I dont think that recording videos of hitregistration issues is super useful unless its raw 60fps footage with little to no compression. There are other ways of testing which were used in the past, I havent run any tests since ~239 however so I cannot comment on how those work now - but they are more useful for debugging issues.
invTempestJoin Date: 2003-03-02Member: 14223Members, Constellation, Squad Five Blue
I have spent many hours over the past couple months analyzing random pub games that I recorded in various patches but I have yet to find anything conclusive. The times when I am so apt to blame hit reg or lag and I go back to review the encounter frame by frame I see that I missed because I either failed to effectively track the skulk (crosshair was just off the skulk model) or that I shot between the skulks legs or that the crosshair was on the skulk but the bullet spread caused the shot to miss.
More often than not I am actually amazed at how spot on the hit registry is in the game which I think actually causes people to perceive that they should be hitting when if fact they missed.
Regardless, if there is an issue then it is very rare as I have gone over hundreds of marine vs skulk encounters and have yet to find anything conclusive that points to a hit reg issue.
At this point I am more worried about the 100ms interp than anything else as it causes about 10-15 extra shots into an already dead skulk which adds to the whole disappearing bullets issue that people complain about.
Speaking purely theoretically, let's say that technical details like hitreg, animations, etc are all perfect, and skulks have their 239 acceleration back. Suppose we find that this is still too unforgiving for marines, and we have the following choice:
- Reduce skulk mobility to make them easier to aim at(as was done in 240/241)
- Make the skulk hitbox more forgiving(i.e. slightly larger than their real model)
Which do you guys think is the better choice? Clearly reducing skulk mobility makes them less fun to play, though it's not awful right now. I'm wondering if it would really be that bad to cheat the hitbox a little to make them easier targets while leaving the gameplay on the skulk's side intact. I'm talking something minor enough to be more or less imperceptible, i.e. those very ambiguous near-misses would now be hits. Marine aim now feels more consistent, skulks have their full movement back - everybody's happy, right?
If Skulks were easy mode pre 240, then what are Marines now? Walk in the park mode? Autowin mode?
No amount of "other" balance changes will fix that Skulk vs. Marine is a joke against any non-shitty Marine player. Not just early game (it's not like it gets better with A3 or non 1on1 fights).
It's not that it's "frustrating to be used to be able to do things that are now suddenly basically impossible". It's that the Marine team has to be absolutely horrible compared to Aliens to lose, everything else is an easy stomp for Marines.
Unlike pre 240, where half of Marine games were lost because everyone liked shooting Clog Walls instead of doing anything useful, there's just nothing Aliens can do with the broken Skulk.
Being stomped in the first 2 minutes, waiting other 8 minutes until finally concede is possible? (whoever decided to increase concede time to 10 mins???)
Barely being able to hold 2 Hive locations and the RT in between whereas Marines not only have everything else, but Aliens can't even harass due to having problem holding the few things they have?
Reducing the Marine team down to one base with superior play, only to be stomped when they come out of their one base because not everyone could save for an Onos as lifeforms were needed to survive to that point?
This is how games play these days.
Every Alien win I see is either the Marine team being plain horrible compared to Aliens, or the Marine being passive and doing nothing while Aliens saved for 3+ Onos.
Marines decide to be shitty? Aliens win. Everything else? Marines win.
That's the problem. When no matter how you play as Aliens, your only hope is the enemy team's incompetence, and you can do nothing except hope you are somehow vastly better than Marines.
@zek: neither
The problem with mobility and Marine hopping is that it's not continuous. Either you are fast enough to keep up with a jumping Marine, or you are not. You land a bite, or you are shot before. A middle ground barely exists, so "decreased mobility" would be hard or even impossible to balance properly.
And I don't see why Skulk hitboxes should be magically bigger. We already have the "being shot around corners" thing, and being killed much more often before even reaching the target.
If Skulks were too strong, they simply need to be less stronger. That is, just give them less HP. LMG damage acts pretty much continouously, so that's the easiest way to balance Skulk vs. Marine.
Speaking purely theoretically, let's say that technical details like hitreg, animations, etc are all perfect, and skulks have their 239 acceleration back. Suppose we find that this is still too unforgiving for marines, and we have the following choice:
- Reduce skulk mobility to make them easier to aim at(as was done in 240/241)
- Make the skulk hitbox more forgiving(i.e. slightly larger than their real model)
Which do you guys think is the better choice? Clearly reducing skulk mobility makes them less fun to play, though it's not awful right now. I'm wondering if it would really be that bad to cheat the hitbox a little to make them easier targets while leaving the gameplay on the skulk's side intact. I'm talking something minor enough to be more or less imperceptible, i.e. those very ambiguous near-misses would now be hits. Marine aim now feels more consistent, skulks have their full movement back - everybody's happy, right?
I think skulks should be given less health, so if seen at range they stand much less chance, but be more powerful in close quarter combat so if they can get in close they pose more of a threat. I would also reduce the spawn time for skulks, so you can get right back in to playing: I always feel skulks are supposed to be like expendable, masses of ants, but the spawn time--and all the negative aspects involved in being unable to play for any duration of time--add a sense of value to a skulk's life I don't think it deserves.
Maybe let cara give them a larger armour upgrade so they can remain sorta relevant mid-late game.
There were some hitreg problems remaining after 236, which was finally cleaned up in 240. So its hard to know if it is the lower skulk manuverability or the hitbox fix that matters the most to the effect.
Fortunately, we can begin to say how much the skulk-as-brick thing mattered, by comparing data from 239, 240 and 241. Using ns2stats and splitting pub/comp games, we get
... which looks pretty reasonable especially for comp games. For what's ns2stats is worth, at least.
Note that ns2stats is very influenced by KKG games which are 24 player servers. Isolating them and comparing them to the rest, you get an idea of just how much teamsize matters:
A plausible explanation for why teamsizes didn't matter in 239 and matters so much in 240/241 is just that the biggest influence is the sub-optimal egg spawning algorithm; when skulks starts to die a lot more, alien egg spawning starts to dominate the win/loss ratio, as the aliens can't afford to buy enough eggs to offset their death rate anymore.
Something funky is going on with the data you pulled from ns2stats, as I'm having a hard time reproducing the values you get for the B239/240 alien winrates. Trying to reproduce your search, I'm getting (with 95% confidence intervals in the brackets):
All other servers except KKG (both pub and comp)
239 = 60% (2270/3784 games) [58-62%]
240 = 49% (515/1043 games) [46-52%]
241 = 52% (475/922 games) [48-55%]
I suspect you might have some of the other filter options unchecked (e.g. mods not on select all, or the data range not going far back enough to capture all of the B239 matches).
Unfortunately, it looks like the skulk changes have particularly hurt public alien players (especially on the KKG servers). Comp players have seem to adapted much better, though its hard to see how much is due to smaller playercounts (6v6 vs 12v12) rather than the skulk changes.
There were some hitreg problems remaining after 236, which was finally cleaned up in 240. So its hard to know if it is the lower skulk manuverability or the hitbox fix that matters the most to the effect.
Fortunately, we can begin to say how much the skulk-as-brick thing mattered, by comparing data from 239, 240 and 241. Using ns2stats and splitting pub/comp games, we get
... which looks pretty reasonable especially for comp games. For what's ns2stats is worth, at least.
Note that ns2stats is very influenced by KKG games which are 24 player servers. Isolating them and comparing them to the rest, you get an idea of just how much teamsize matters:
A plausible explanation for why teamsizes didn't matter in 239 and matters so much in 240/241 is just that the biggest influence is the sub-optimal egg spawning algorithm; when skulks starts to die a lot more, alien egg spawning starts to dominate the win/loss ratio, as the aliens can't afford to buy enough eggs to offset their death rate anymore.
Something funky is going on with the data you pulled from ns2stats, as I'm having a hard time reproducing the values you get for the B239/240 alien winrates. Trying to reproduce your search, I'm getting (with 95% confidence intervals in the brackets):
All other servers except KKG (both pub and comp)
239 = 60% (2270/3784 games) [58-62%]
240 = 49% (515/1043 games) [46-52%]
241 = 52% (475/922 games) [48-55%]
I suspect you might have some of the other filter options unchecked (e.g. mods not on select all, or the data range not going far back enough to capture all of the B239 matches).
Unfortunately, it looks like the skulk changes have particularly hurt public alien players (especially on the KKG servers). Comp players have seem to adapted much better, though its hard to see how much is due to smaller playercounts (6v6 vs 12v12) rather than the skulk changes.
Comp servers tell us very little...few servers run 6v6...8v8 is most common though I always see the 18+ server being filled the most.
So comp games just skew any data analysis as there are almost no 12 player servers that are populated.
They certianly dont represent the majority of games played....as such they really should be mostly ignored (sorry comp players but if you want to foster new talent you need to understand they come into the game through pub games...so poor balance at pub lievel will lead to fewer comp players and the death of comp play).
What are you people crying about again!? That you cannot run up to a marine in a straight line or hopping around and expect to kill one or more marines? Are you insane? Haven't you yet realized that you have to ambush as a skulk? Try that sometime - it's the easiest thing in the world.
What are you people crying about again!? That you cannot run up to a marine in a straight line or hopping around and expect to kill one or more marines? Are you insane? Haven't you yet realized that you have to ambush as a skulk? Try that sometime - it's the easiest thing in the world.
Play a round of combat to get the gist of it. The first 2-3 minutes of the game are simply hilarious.
Last night I played a game where the teams were legitimately 3v1 for some time, and I was winning quite easily as the lone marine. This, against good players who would normally be able to 1v1 me quite competently.
Skulks are just pathetic at the moment.
As with NS1, it seems that the advanced movement mechanic is slowly becoming a necessity. I didn't take kindly to it then, and I won't take kindly to it now. At least then there were other aspects to hold my attention.
What are you people crying about again!? That you cannot run up to a marine in a straight line or hopping around and expect to kill one or more marines? Are you insane? Haven't you yet realized that you have to ambush as a skulk? Try that sometime - it's the easiest thing in the world.
This is the lamest excuse I've ever seen.
You can't just camp on walls and wait for someone fool enough to not look up to get a kill.
By doing that, you are also not contesting areas, which you should be.
I've said it once and I'll say it again: skulks got nerfed because your random average pub marine can't aim, how is it that good for the game?
If the skulk is too strong, we can try other options like removing glancing bites, not nerfing something that makes the skulk fun to play: speed.
How do you expect skulks to gain area advantage, when it's only possible to ambush marines?
Should I attack another point on the map?
Well then skulks would be really lame to play, I would call them sknails, because letting youself fall from a wall on a marine's head isn't much different.
If you think, the skulk- marine mechanic is working as intended, you have never played against good marines,
they know every hiding spot (and have the neon alien skins).
Skulk is fine now, I'm able to win encounters now even with two, it's all about movement and not even fast movement, get be-hide them, get above them, people find it very hard to shoot above them. Don't stay in one position, bite once, then wall-jump or get more elevation.
I believe people are using fallacies to compensate for a lack of skill, also the Skulks is an ambush/Zerg class works better on surprise and team attacks, and that's where we come to main issue, quite a few people don't like to work as a team.
AsranielJoin Date: 2002-06-03Member: 724Members, Playtest Lead, Forum Moderators, NS2 Playtester, Squad Five Blue, Reinforced - Shadow, WC 2013 - Shadow, Subnautica Playtester, Retired Community Developer
The playstyle really changed with better performance, better animations and fixed hitreg. You need to change the way you play. Let the marines come to you, dont run towards them if they are in a hallway or in the middle of the room. You could do that before because animations where broken and made you twitch around and hitreg made you run trough the fire without taking any hits. You are no longer that immortal ninja warrior. Adapt. (not saying everything is perfect, but i tried to adapt and dont have any serious problems now).
How do you expect skulks to gain area advantage, when it's only possible to ambush marines?
Should I attack another point on the map?
Well then skulks would be really lame to play, I would call them sknails, because letting youself fall from a wall on a marine's head isn't much different.
If you think, the skulk- marine mechanic is working as intended, you have never played against good marines,
they know every hiding spot (and have the neon alien skins).
Ambush, small distances, walls, ceilings - key to beat a marine.
If you, for example, run straight into nano grid and expect to kill a marine in there, well, good luck - you'll be dead at the entrance.
How do you expect skulks to gain area advantage, when it's only possible to ambush marines?
Should I attack another point on the map?
Well then skulks would be really lame to play, I would call them sknails, because letting youself fall from a wall on a marine's head isn't much different.
If you think, the skulk- marine mechanic is working as intended, you have never played against good marines,
they know every hiding spot (and have the neon alien skins).
Ambush, small distances, walls, ceilings - key to beat a marine.
If you, for example, run straight into nano grid and expect to kill a marine in there, well, good luck - you'll be dead at the entrance.
Then what do you propose? That we let marines have two free rts until we have fades?
The playstyle really changed with better performance, better animations and fixed hitreg. You need to change the way you play. Let the marines come to you, dont run towards them if they are in a hallway or in the middle of the room. You could do that before because animations where broken and made you twitch around and hitreg made you run trough the fire without taking any hits. You are no longer that immortal ninja warrior. Adapt. (not saying everything is perfect, but i tried to adapt and dont have any serious problems now).
You missed the part that wall jumping feels really bad now, sometimes it's just better to not continue wall jumping because past a certain point you somehow lose speed when touching walls.
Comp servers tell us very little...few servers run 6v6...8v8 is most common though I always see the 18+ server being filled the most.
So comp games just skew any data analysis as there are almost no 12 player servers that are populated.
They certianly dont represent the majority of games played....as such they really should be mostly ignored (sorry comp players but if you want to foster new talent you need to understand they come into the game through pub games...so poor balance at pub lievel will lead to fewer comp players and the death of comp play).
Comp matches are played as needed between clans or as gathers/PUGs, such that you won't see 12p 'comp' servers that anyone can join. They certainly aren't the majority of matches, but as the above data shows, there are enough of them that you just can't ignore them.
Pub balance is important, but ns2stats seems to suggest that its mostly an issue of playercount (e.g. the higher the playercount, the lower the alien winrate). Not surprising, as my experience is that egglock becomes a much bigger issue on larger servers.
There were some hitreg problems remaining after 236, which was finally cleaned up in 240. So its hard to know if it is the lower skulk manuverability or the hitbox fix that matters the most to the effect.
Fortunately, we can begin to say how much the skulk-as-brick thing mattered, by comparing data from 239, 240 and 241. Using ns2stats and splitting pub/comp games, we get
... which looks pretty reasonable especially for comp games. For what's ns2stats is worth, at least.
Note that ns2stats is very influenced by KKG games which are 24 player servers. Isolating them and comparing them to the rest, you get an idea of just how much teamsize matters:
A plausible explanation for why teamsizes didn't matter in 239 and matters so much in 240/241 is just that the biggest influence is the sub-optimal egg spawning algorithm; when skulks starts to die a lot more, alien egg spawning starts to dominate the win/loss ratio, as the aliens can't afford to buy enough eggs to offset their death rate anymore.
Although the brick skulk caused changes in the number of wins for aliens, given that 241 had A LOT of changes, how can you possibly be certain that the alien losses were entirely due to the skulk change? I hated the skulk of 240. The skulk in 241 is a lot better, but I do miss the skulk of 239. Too much has changed in the game for us to say what caused it.
You can't just camp on walls and wait for someone fool enough to not look up to get a kill.
By doing that, you are also not contesting areas, which you should be.
Part of the skulk's strength is its ability to out maneuver marines and attack places where the marines aren't defending far behind the front lines. Skulks should always be using hit and run tactics to undermine the marine economy rather than attacking where the marines are strong. Come on people- this is basic art of war! Attack where the enemy is weak, avoid where they are strong.
This also happens to set up situations for skulks to use ambush tactics. Once you start hitting the rt, the marines will come running and you will be in place to ambush. Use ambushes offensively- sometimes you have to create the conditions under which your strengths can be used against the enemy's weaknesses.
Ambushing does NOT mean waiting in some obscure corner of the map waiting for god knows who to stumble across your kill zone, it means being in a marine's face when he least expects it.
Comp servers tell us very little...few servers run 6v6...8v8 is most common though I always see the 18+ server being filled the most.
So comp games just skew any data analysis as there are almost no 12 player servers that are populated.
They certianly dont represent the majority of games played....as such they really should be mostly ignored (sorry comp players but if you want to foster new talent you need to understand they come into the game through pub games...so poor balance at pub lievel will lead to fewer comp players and the death of comp play).
Comp matches are played as needed between clans or as gathers/PUGs, such that you won't see 12p 'comp' servers that anyone can join. They certainly aren't the majority of matches, but as the above data shows, there are enough of them that you just can't ignore them.
Pub balance is important, but ns2stats seems to suggest that its mostly an issue of playercount (e.g. the higher the playercount, the lower the alien winrate). Not surprising, as my experience is that egglock becomes a much bigger issue on larger servers.
But we all agree marines scale better with larger numbers (which is what the new player base play on...and where balance should be focused).
Comp games having only 6 a side means aliens are stronger as at those number marines do struggle more to hold map control.
The fact is that most people play on servers of 16+ (mostly see 18+ populated) and comp games are a small component with a server size far removed from normal game size for joe public.
Its like trying to balance BF2/3 for 6v6....its impossible to do.
Most people who play this game are casual gamers who never play comp games (I have real life responsibilities that mean I cant dedicate the time to play comp).
As such the balance they see is not in a 6v6 game but 9v9 or greater.
There was much talk through the beta of perhaps having a different comp server set of balance parameters...this would make sense given the comp scene wants 6v6 yet joe public like 18-24 player servers.
matsoMaster of PatchesJoin Date: 2002-11-05Member: 7000Members, Forum Moderators, NS2 Developer, Constellation, NS2 Playtester, Squad Five Blue, Squad Five Silver, Squad Five Gold, Reinforced - Shadow, NS2 Community Developer
edited March 2013
Comp games are useful because the signal from them are a lot less noisy than pub games. For example, comp games all have the same teamsize, they usually swap sides and replay on the same map, they have dedicated commanders, skill level are usually at a fairly even level etc.
So there is a lot to be learned about balance by looking at comp games.
Now, how useful is comp balance for pub balance? The way to go about this isn't to point out that they are not likely to be identical. The idea is to find a way to predict pub balance given the comp balance numbers, or in other words, find a function of comp balance that gives you pub balance numbers...
A fairly simple problem if you accept a bit of sloppyness.
C = comp balance
P = pub balance
239: C = 63, P = 55
240: C = 51, P = 45
241: C = 55, P = 47
Assuming a simple linear relationship, ie a f(x) = kx + m, we solve it to something like
k = 0.6
m = 15
ending up with predictions of 53, 46, 48, which are close enough.
So for what its worth, comp game balance will tell you something about pub game balance.
Note that a variation of 1% in comp games only changes pub game outcome by 0.6%. This is because pub games are noisy compared to comp games or in other words, there are more random factors in pub games compared to comp games.
Played many matches as skulk last night (I play marine 98% of the time)... it was more fun for me because it is actually challenging now. Before it was was too easy for me to kill marines and get around. Last night I had to be more stealthy, sneak around, work with others and be more tactful. Camo works great and I didn't feel like I was using a cheap hack like I did in the past, yet at the same time I was able to use it extremely effectively and to my benefit. I like having to think a bit and have consequences to the choices I make as a skulk now. It was very enjoyable to have more team work needed on the alien side. Because of this I'll probably play alien team more.
I am very pleased with the latest patch and skulk adjustments.
BTW, Aliens won 50% of the 12 games I played last night on a pub server and I never felt like the Aliens were nerfed to death like a lot of people are saying.
But we all agree marines scale better with larger numbers (which is what the new player base play on...and where balance should be focused).
Comp games having only 6 a side means aliens are stronger as at those number marines do struggle more to hold map control.
The fact is that most people play on servers of 16+ (mostly see 18+ populated) and comp games are a small component with a server size far removed from normal game size for joe public.
Its like trying to balance BF2/3 for 6v6....its impossible to do.
Most people who play this game are casual gamers who never play comp games (I have real life responsibilities that mean I cant dedicate the time to play comp).
As such the balance they see is not in a 6v6 game but 9v9 or greater.
There was much talk through the beta of perhaps having a different comp server set of balance parameters...this would make sense given the comp scene wants 6v6 yet joe public like 18-24 player servers.
Ideally, balance should be the same regardless of playercount, but in practice NS2 falls short. I'd rather UWE focus on trying to eliminate or minimize those difference, rather than just trying to get NS2 balanced for a specific playercount (which was the problem in NS1).
Comp games are useful because the signal from them are a lot less noisy than pub games. For example, comp games all have the same teamsize, they usually swap sides and replay on the same map, they have dedicated commanders, skill level are usually at a fairly even level etc.
So there is a lot to be learned about balance by looking at comp games.
Now, how useful is comp balance for pub balance? The way to go about this isn't to point out that they are not likely to be identical. The idea is to find a way to predict pub balance given the comp balance numbers, or in other words, find a function of comp balance that gives you pub balance numbers...
A fairly simple problem if you accept a bit of sloppyness.
C = comp balance
P = pub balance
239: C = 63, P = 55
240: C = 51, P = 45
241: C = 55, P = 47
Assuming a simple linear relationship, ie a f(x) = kx + m, we solve it to something like
k = 0.6
m = 15
ending up with predictions of 53, 46, 48, which are close enough.
So for what its worth, comp game balance will tell you something about pub game balance.
Note that a variation of 1% in comp games only changes pub game outcome by 0.6%. This is because pub games are noisy compared to comp games or in other words, there are more random factors in pub games compared to comp games.
As is my nature, I wanted to get a more robust picture of how alien winrate correlates between pub and comp play, so I pulled all the ns2stats data available (starts with B228 back in November). I combined the win/loss numbers together for builds that lasted less than 3 days (e.g. B234 was only out for 2 days before B235 released, so I treat them as one build). Data is here for anyone interested.
Bottom line is that alien winrate is highly correlated across the builds between pub and comp play (roughly equal slopes), but offset such that comp alien teams appear to be able to win at higher rates than public alien teams. I suspect much of this is due to the playercount scaling issues (e.g. 6v6 vs 10v10+), but its hard to tell.
Comments
Also the spread of the rifle is simply too big to make accurate assessments regarding hitreg...
More often than not I am actually amazed at how spot on the hit registry is in the game which I think actually causes people to perceive that they should be hitting when if fact they missed.
Regardless, if there is an issue then it is very rare as I have gone over hundreds of marine vs skulk encounters and have yet to find anything conclusive that points to a hit reg issue.
At this point I am more worried about the 100ms interp than anything else as it causes about 10-15 extra shots into an already dead skulk which adds to the whole disappearing bullets issue that people complain about.
- Reduce skulk mobility to make them easier to aim at(as was done in 240/241)
- Make the skulk hitbox more forgiving(i.e. slightly larger than their real model)
Which do you guys think is the better choice? Clearly reducing skulk mobility makes them less fun to play, though it's not awful right now. I'm wondering if it would really be that bad to cheat the hitbox a little to make them easier targets while leaving the gameplay on the skulk's side intact. I'm talking something minor enough to be more or less imperceptible, i.e. those very ambiguous near-misses would now be hits. Marine aim now feels more consistent, skulks have their full movement back - everybody's happy, right?
No amount of "other" balance changes will fix that Skulk vs. Marine is a joke against any non-shitty Marine player. Not just early game (it's not like it gets better with A3 or non 1on1 fights).
It's not that it's "frustrating to be used to be able to do things that are now suddenly basically impossible". It's that the Marine team has to be absolutely horrible compared to Aliens to lose, everything else is an easy stomp for Marines.
Unlike pre 240, where half of Marine games were lost because everyone liked shooting Clog Walls instead of doing anything useful, there's just nothing Aliens can do with the broken Skulk.
Being stomped in the first 2 minutes, waiting other 8 minutes until finally concede is possible? (whoever decided to increase concede time to 10 mins???)
Barely being able to hold 2 Hive locations and the RT in between whereas Marines not only have everything else, but Aliens can't even harass due to having problem holding the few things they have?
Reducing the Marine team down to one base with superior play, only to be stomped when they come out of their one base because not everyone could save for an Onos as lifeforms were needed to survive to that point?
This is how games play these days.
Every Alien win I see is either the Marine team being plain horrible compared to Aliens, or the Marine being passive and doing nothing while Aliens saved for 3+ Onos.
Marines decide to be shitty? Aliens win. Everything else? Marines win.
That's the problem. When no matter how you play as Aliens, your only hope is the enemy team's incompetence, and you can do nothing except hope you are somehow vastly better than Marines.
@zek: neither
The problem with mobility and Marine hopping is that it's not continuous. Either you are fast enough to keep up with a jumping Marine, or you are not. You land a bite, or you are shot before. A middle ground barely exists, so "decreased mobility" would be hard or even impossible to balance properly.
And I don't see why Skulk hitboxes should be magically bigger. We already have the "being shot around corners" thing, and being killed much more often before even reaching the target.
If Skulks were too strong, they simply need to be less stronger. That is, just give them less HP. LMG damage acts pretty much continouously, so that's the easiest way to balance Skulk vs. Marine.
I think skulks should be given less health, so if seen at range they stand much less chance, but be more powerful in close quarter combat so if they can get in close they pose more of a threat. I would also reduce the spawn time for skulks, so you can get right back in to playing: I always feel skulks are supposed to be like expendable, masses of ants, but the spawn time--and all the negative aspects involved in being unable to play for any duration of time--add a sense of value to a skulk's life I don't think it deserves.
Maybe let cara give them a larger armour upgrade so they can remain sorta relevant mid-late game.
Public
239 = 55% (1857/3382 games) [53-57%]
240 = 45% (473/1050 games) [42-48%]
241 = 45% (379/838 games) [42-49%]
Competitive
239 = 66% (1161/1751 games) [64-69%]
240 = 51% (226/442 games) [46-56%]
241 = 55% (200/361 games) [50-61%]
Only KKG servers (both pub and comp)
239 = 55% (748/1349 games) [53-58%]
240 = 41% (184/449 games) [36-46%]
241 = 37% (104/279 games) [32-43%]
All other servers except KKG (both pub and comp)
239 = 60% (2270/3784 games) [58-62%]
240 = 49% (515/1043 games) [46-52%]
241 = 52% (475/922 games) [48-55%]
I suspect you might have some of the other filter options unchecked (e.g. mods not on select all, or the data range not going far back enough to capture all of the B239 matches).
Unfortunately, it looks like the skulk changes have particularly hurt public alien players (especially on the KKG servers). Comp players have seem to adapted much better, though its hard to see how much is due to smaller playercounts (6v6 vs 12v12) rather than the skulk changes.
Comp servers tell us very little...few servers run 6v6...8v8 is most common though I always see the 18+ server being filled the most.
So comp games just skew any data analysis as there are almost no 12 player servers that are populated.
They certianly dont represent the majority of games played....as such they really should be mostly ignored (sorry comp players but if you want to foster new talent you need to understand they come into the game through pub games...so poor balance at pub lievel will lead to fewer comp players and the death of comp play).
Play a round of combat to get the gist of it. The first 2-3 minutes of the game are simply hilarious.
Last night I played a game where the teams were legitimately 3v1 for some time, and I was winning quite easily as the lone marine. This, against good players who would normally be able to 1v1 me quite competently.
Skulks are just pathetic at the moment.
As with NS1, it seems that the advanced movement mechanic is slowly becoming a necessity. I didn't take kindly to it then, and I won't take kindly to it now. At least then there were other aspects to hold my attention.
This is the lamest excuse I've ever seen.
You can't just camp on walls and wait for someone fool enough to not look up to get a kill.
By doing that, you are also not contesting areas, which you should be.
I've said it once and I'll say it again: skulks got nerfed because your random average pub marine can't aim, how is it that good for the game?
If the skulk is too strong, we can try other options like removing glancing bites, not nerfing something that makes the skulk fun to play: speed.
Should I attack another point on the map?
Well then skulks would be really lame to play, I would call them sknails, because letting youself fall from a wall on a marine's head isn't much different.
If you think, the skulk- marine mechanic is working as intended, you have never played against good marines,
they know every hiding spot (and have the neon alien skins).
I believe people are using fallacies to compensate for a lack of skill, also the Skulks is an ambush/Zerg class works better on surprise and team attacks, and that's where we come to main issue, quite a few people don't like to work as a team.
Ambush, small distances, walls, ceilings - key to beat a marine.
If you, for example, run straight into nano grid and expect to kill a marine in there, well, good luck - you'll be dead at the entrance.
Then what do you propose? That we let marines have two free rts until we have fades?
JK we'll lose in the next five minutes.
You missed the part that wall jumping feels really bad now, sometimes it's just better to not continue wall jumping because past a certain point you somehow lose speed when touching walls.
Pub balance is important, but ns2stats seems to suggest that its mostly an issue of playercount (e.g. the higher the playercount, the lower the alien winrate). Not surprising, as my experience is that egglock becomes a much bigger issue on larger servers.
Although the brick skulk caused changes in the number of wins for aliens, given that 241 had A LOT of changes, how can you possibly be certain that the alien losses were entirely due to the skulk change? I hated the skulk of 240. The skulk in 241 is a lot better, but I do miss the skulk of 239. Too much has changed in the game for us to say what caused it.
Part of the skulk's strength is its ability to out maneuver marines and attack places where the marines aren't defending far behind the front lines. Skulks should always be using hit and run tactics to undermine the marine economy rather than attacking where the marines are strong. Come on people- this is basic art of war! Attack where the enemy is weak, avoid where they are strong.
This also happens to set up situations for skulks to use ambush tactics. Once you start hitting the rt, the marines will come running and you will be in place to ambush. Use ambushes offensively- sometimes you have to create the conditions under which your strengths can be used against the enemy's weaknesses.
Ambushing does NOT mean waiting in some obscure corner of the map waiting for god knows who to stumble across your kill zone, it means being in a marine's face when he least expects it.
But we all agree marines scale better with larger numbers (which is what the new player base play on...and where balance should be focused).
Comp games having only 6 a side means aliens are stronger as at those number marines do struggle more to hold map control.
The fact is that most people play on servers of 16+ (mostly see 18+ populated) and comp games are a small component with a server size far removed from normal game size for joe public.
Its like trying to balance BF2/3 for 6v6....its impossible to do.
Most people who play this game are casual gamers who never play comp games (I have real life responsibilities that mean I cant dedicate the time to play comp).
As such the balance they see is not in a 6v6 game but 9v9 or greater.
There was much talk through the beta of perhaps having a different comp server set of balance parameters...this would make sense given the comp scene wants 6v6 yet joe public like 18-24 player servers.
So there is a lot to be learned about balance by looking at comp games.
Now, how useful is comp balance for pub balance? The way to go about this isn't to point out that they are not likely to be identical. The idea is to find a way to predict pub balance given the comp balance numbers, or in other words, find a function of comp balance that gives you pub balance numbers...
A fairly simple problem if you accept a bit of sloppyness.
C = comp balance
P = pub balance
239: C = 63, P = 55
240: C = 51, P = 45
241: C = 55, P = 47
Assuming a simple linear relationship, ie a f(x) = kx + m, we solve it to something like
k = 0.6
m = 15
ending up with predictions of 53, 46, 48, which are close enough.
So for what its worth, comp game balance will tell you something about pub game balance.
Note that a variation of 1% in comp games only changes pub game outcome by 0.6%. This is because pub games are noisy compared to comp games or in other words, there are more random factors in pub games compared to comp games.
I am very pleased with the latest patch and skulk adjustments.
BTW, Aliens won 50% of the 12 games I played last night on a pub server and I never felt like the Aliens were nerfed to death like a lot of people are saying.
As is my nature, I wanted to get a more robust picture of how alien winrate correlates between pub and comp play, so I pulled all the ns2stats data available (starts with B228 back in November). I combined the win/loss numbers together for builds that lasted less than 3 days (e.g. B234 was only out for 2 days before B235 released, so I treat them as one build). Data is here for anyone interested.
Bottom line is that alien winrate is highly correlated across the builds between pub and comp play (roughly equal slopes), but offset such that comp alien teams appear to be able to win at higher rates than public alien teams. I suspect much of this is due to the playercount scaling issues (e.g. 6v6 vs 10v10+), but its hard to tell.