No, because there is no clear definition of what the game was designed for.
The span from 12 to 24 is too high to have a definite balance.
I dont follow you here.
Do you disagree with the following:
I show 45%/55% stats to a statistician, knowing nothing about ns2,
and tell them that is stats from a server that had more players than
what the game was designed/game tested/tweaked for.
Also telling that the game was designed/game tested/tweaked to have very balanced factions.
Would that statistician not estimate that servers without that many player
would have at least as good a balance?
Where are you going with this? Are you trying to make a point or are you really just curious?
NS2 is not faction balanced 50/50 on any slot count (12-24). It is absolutely irrelevevant how balanced the game is on a server that does not run NS2.
You practically get a curve in balance on multiple levels. This is why server admins are adjusting to these curves with mods. 6vs6 environments usually have competivtive mod installed, while large ns2 servers (12vs12) usually have a second infantry portal.
Both of those mods change the gameplay in order to be more faction balanced.
The elo shuffle has nothing to do with the games balance, but with how interesting the matches are. The system creates unbalanced matches on either side.
Wut? I believe the elo shuffle has everything to do with how balanced games are, but not necessarily how interesting they are.
Or do I misunderstand you?
Ahh, I did not make that distinction. (I would probably have called it faction balance).
Slightly off topic: Can we agree on the following:
In all (board-/pc-/)games that are not pure luck, will require some skill/experience match up mechanism for a fair match?
Nearly any system is better than nothing?
Anyhow back to what this came from (I have replaced game with match)
Not necessarily. Someone just might know someone else who might be of equal skill and he invites him to a fair match.
The actual problem is however, that my ability to play football is rated at how I play football, basketball, american football and golf.
A 6vs6 is not a 12vs12 just like football is not handball.
NS2 even has a balance level deeper than that with its asymmetrical factions. Someone might be good as lerk on a 12vs12 slot server, but be bad as a lerk on a 6vs6 server, but at the same time be good as a marine on a 6vs6, but bad on a 12vs12 slot server.
The system could work, if everyone was playing the same type of ns2 and if that system would assign you to a server, instead of the team within your current servers player pool.
Do you stats for that?
I mean I think stats tells the exact opposite; overall the matches are very even.
My assumption here is of cause that all matches started with teams that were created base on (some approximation) of even hive score.
I have no recollection of matches that were not shuffled based on hive score unless they already were nearly even.
The second ip is given at 9v9 and is not a mod. It's vanilla
It was a band aid solution mod that was first introduced on 24 slot servers. It was simply taken over into vanilla. It starts at 9vs9 because it is is the middle ground between 12 and 24 slots.
It is band aid, because 8vs8 is more unfair towards marines, while 10vs10 is more unfair towards aliens, since you can not have 1.67 / 2.33 infantry portals.
You couldn't even say that 6vs6 and 12vs12 are then equal in terms of balance, because if I managed to sneak into the marine base on a 12vs12 server, I would have to destroy 2 ips in order to keep the marines from spawning, instead of just 1. Losing 1 ip on a 12vs12 early game is a set back, while losing an ip in an early game 6vs6 is potentially game ending.
Now you could say: "yeah, well, but on a 24 slot server you could have potentially 2 skulks who could sneak into the marine base", but that case is less likely, because lane blocking and scouting becomes easier the more marines you have on the map.
Why can't people just let people play on the servers they want? If it's a higher pop server then so be it? There's already a warning that comes up if you join them, I never understood people that advocate removing them.
You can argue that people play on them because smaller pop servers are full, but clearly SOME people prefer them, so there's no reason to remove them.
For another similar pvp game, say tf2, if someone suggested getting rid of the high play count servers because the game wasn't designed with so many players in mind, they would be laughed at?
Why can't people just let people play on the servers they want? If it's a higher pop server then so be it?
... clearly SOME people prefer them, so there's no reason to remove them.
What he said.
And we know that changing that context (# of players) changes the game, how maps work and balance, therefore how you play it.
It's one more way to play NS2 and it's fine.
Why can't people just let people play on the servers they want? If it's a higher pop server then so be it?
... clearly SOME people prefer them, so there's no reason to remove them.
That is not the point. The point is that >24 slot servers are not ns2 and never will be ns2, but are presented as such in the server browser. New players get a false impression of the game.
What he said.
And we know that changing that context (# of players) changes the game, how maps work and balance, therefore how you play it.
It's one more way to play NS2 and it's fine.
That is also not the point. I was talking about how the ELO system is fundamentally broken, because it throws differing slot counts all into one final rating, even though game play and therefore game balance differs with every possible slot count.
The question was: "Which PUBLIC SERVER SIZE do you like playing on?" and not "which one do you prefer?"
You had the possibility to tick multiple answers.
6vs6 is not even up for debate in this poll, since 6vs6 public doesn't work without matchmaking, but quite a lot of people like playing 6vs6.
This brings back memories. Tried to come a clusterfuck on *snip*. Organized people for a sg rush with w1 a1 and saw around 20 ppl with die to 14 skulks.
In retrospect I should have meded the hive twice to represent my tears streamin down my face in this moment.
According to past surveys about 40% of the player base preferred 21v21 large servers.
Are you saying the player base of regular 6v6/7v7/8v8 matches could be increased by 67% if we remove the 21v21 large servers?
Great! Sounds like a good plan, just like the movie industry say they lose billions of revenue and jobs due to piracy because everyone would buy originals if there was better law enforcement.
IronHorseDeveloper, QA Manager, Technical Support & contributorJoin Date: 2010-05-08Member: 71669Members, Super Administrators, Forum Admins, Forum Moderators, NS2 Developer, NS2 Playtester, Squad Five Blue, Subnautica Playtester, Subnautica PT Lead, Pistachionauts
@NovoRei I get your point, but that's a poor analogy, really. We could enforce it if we really wanted to and no one would be able to over ride it or "pirate it" like your example.
@NovoRei I get your point, but that's a poor analogy, really. We could enforce it if we really wanted to and no one would be able to over ride it or "pirate it" like your example.
The main point was that someone could suggest to remove large servers in other to increase player base of regular servers based on Nordic's numbers. After all, some folks here stated that regular servers provide the "right" and "true" experience where other type of servers do not provide it.
The sarcasm is that this line of thinking is widespread (several analogies) and it's a known fallacy due to shallow reasoning. Those players of large servers are there not because of "NS2"/"true experience" but because of the gameplay experience that only a large server can provide.
And the movie piracy analogy is perfect. If you cant get access to the content/product/thing/experience, you don't seek it.
If you cant get access to the content/product/thing/experience, you don't seek it.
Uhh.. that's exactly why one pirates in a lot of cases?
Can't stream a movie because the producers wan't to sell DVDs for 9 months first? Just download it, then.
It's a poor analogy on all fronts.
As far as the subjective argument over what experience a player prefers :
1) It's a moot point if all the customer knows is lesser quality from day one
2) Any argument over the many factors that determine the quality of the experience for those that enjoy large servers (like less responsibility, more forgiving, more to do etc) is entirely irrelevant if the experience is fundamentally broken due to performance issues. The game already struggles enough with the occasional hitreg issue or high interp values - compounding this with degraded server and client performance levels that were never intended to be played, just gives an even worse representation. Which is a shame, and is worth highlighting when we're talking about the the subject of the quality of experience to certain players.
Just think: if we all played on 8v8 servers instead of 21v21, your server browsers would look a lot busier, and you'd have considerably more choice on where to play.
IMHO the main purpose of hive skill is to help make balanced teams so the policy should be designed to further this goal (and only this goal.)
If I were to write the policy, the criteria would be two questions.
1. Does the skill a player displays on this server translate directly to skill in a vanilla server?
2. Are the rounds of gameplay on this server similar in duration and complexity to games on a vanilla server?
If the answer to both questions is yes, I'd include it. This includes compmod, large servers, or really any modifications that primarily change the meta. So long as the basic gameplay is intact and rounds play out in a similar way, it should contribute to hive. The only things I'd exclude are things like faded and last stand.
Here's a thought experiment that might change people's mind. Suppose someone has demonstrated incredible skill such that rounds have to be balanced around them, but has only played on heavily modified servers. Should that player be allowed to stomp with impunity if they join a vanilla server?
A high skill value should be properly thought of as a handicap rather than a benefit because it reduces the quality of your teammates.
Just think: if we all played on 8v8 servers instead of 21v21, your server browsers would look a lot busier, and you'd have considerably more choice on where to play.
I love 8v8 ns2. I almost exclusively play on 8v8 ns2 and I refuse to play on servers 10v10 or higher if I can avoid it. Sure, there would be more servers to choose from but there would be less choice in how to play. If wooza has shown us anything, it is that there is a large part of ns2 that actively enjoys ns2large.
This seems to spiral quite fast out of an other tangent than what I actually intended.
So let me explain where I am coming from and what I wanted.
I have made an observation X.
I asked if somebody has stats that explains X/shows its a fluke.
Nobody seems to have (I am not talking who has opinions, there are hordes of those).
The second best to actual observations of reality, i.e. stats, is put forth testable hypothesis of the reality.
In this case we are (un)lucky that we have, what seems, 2 segregated clusters in the same domain.
So you can make a hypothesis in 1 cluster and see if that predict anything in the 2nd cluster.
If you are unable or unwilling to do that, it is worth nothing, it is even worth negative, as it clutters the discussion.
In that case your opinion of what makes one faction win over another does not matter.
I can have an opinion that the winning side's commander is always visited by the invisible unicorn first.
So every commander should have invisible carrots.
There are several problems with that; most ppl will say it is not their experience.
Secondly, once I made my opinion of that, confirmation bias will make me virtually blind for anything that does not fit that opinion.
It is for those 2 reasons alone you must be forced to step out of your comfort zone, and start predicting what that would mean
if something changes, instead of continued getting your (wrong) opinion confirmed.
I think I have seen only 1 testable opinion/prediction (besides those testable opinions, that I believe are already contradicted by existing stats):
The second best to actual observations of reality, i.e. stats, is put forth testable hypothesis of the reality.
In this case we are (un)lucky that we have, what seems, 2 segregated clusters in the same domain.
So you can make a hypothesis in 1 cluster and see if that predict anything in the 2nd cluster.
If you are unable or unwilling to do that, it is worth nothing, it is even worth negative, as it clutters the discussion.
In that case your opinion of what makes one faction win over another does not matter.
As stated very early first we have no new data to test any hypothesis though older data (that is gone by now) did not show any significant inbalance for long periods of patches, second there's no connection between the two clusters. Thus this discussion can only continue with opinnions and experiences.
1. Does the skill a player displays on this server translate directly to skill in a vanilla server?
2. Are the rounds of gameplay on this server similar in duration and complexity to games on a vanilla server?
This definition of Hive skill is just too lax and still not fitting to ns2large:
- Well yeah, aiming, shooting and moving translate over to vanilla, but similarly would playing Counter Strike. NS2Large would show you some basics like Craig = heal, but so would a bot game. Numbers are still very messed up.
- Some vanilla essentials do not translate well or have very different effects, like laneblocking or rush / sneak strategies.
=> Game modes that teach bad behaviour should be excluded from Hive.
=> Definition of "similar in complexity". I find games on vanilla servers fairly different in complexity.
Now the problem is that in its current implemention, hive calculation doesn't properly scale with playernumbers. It does not adapt accordingly to changed balance and playstyle.
Also still doesn't answer why it would be bad to exlude ns2large.
Compmod is another story. It perserves all the essentials of vanilla and barely if any bad behaviour translates over to vanilla.
It's certainly no less complex than vanilla.
If you compare which maps were won by marines and aliens that day, you'll notice, that aliens were winning custom maps more often.
ns2_combi, ns2_colosseum, ns2_caged, ns2_spaceship were exclusively won by aliens.
Maybe it was custom map wednesday that day, starting at 7pm?
I can't tell you much more without more detailed informations about those matches themselves.
As far as the subjective argument over what experience a player prefers :
1) It's a moot point if all the customer knows is lesser quality from day one
2) Any argument over the many factors that determine the quality of the experience for those that enjoy large servers (like less responsibility, more forgiving, more to do etc) is entirely irrelevant if the experience is fundamentally broken due to performance issues. The game already struggles enough with the occasional hitreg issue or high interp values - compounding this with degraded server and client performance levels that were never intended to be played, just gives an even worse representation. Which is a shame, and is worth highlighting when we're talking about the the subject of the quality of experience to certain players.
And yet nothing is forcing the players of large serves to play on large servers despite all the "fundamentally broken experience". They do on their own accord.
1. The flagship server of ns2large with 46 slots enjoys a full server at the prime time of different regions (Eastern/Western Europe and Eastern US). -> Players like it.
2. It has a recurrent/loyal player base of rookies and seasoned players. -> Players continue to like it over time.
3. It competes with both Euro and US regular servers at prime time. -> It has share of player's choice/preference.
If it's a bad or not rewarding experience, players would just not play it. An example of this is the latest updates to the siege mod that effectively killed it despite enjoying those 3 features above before the recent updates.
It's a moot point to: not to take ns2large seriously; not give it the same importance comp mod has; not give it any kind of support; not enjoy Hive skill balancing offer.
If you cant get access to the content/product/thing/experience, you don't seek it.
Uhh.. that's exactly why one pirates in a lot of cases?
Can't stream a movie because the producers wan't to sell DVDs for 9 months first? Just download it, then.
It's a poor analogy on all fronts.
"Access" is not option exclusive. An analogy may be poor when one does not understand it.
1. The flagship server of ns2large with 46 slots enjoys a full server at the prime time of different regions (Eastern/Western Europe and Eastern US). -> Players like it.
2. It has a recurrent/loyal player base of rookies and seasoned players. -> Players continue to like it over time.
3. It competes with both Euro and US regular servers at prime time. -> It has share of player's choice/preference.
Don't forget that the size of the server is a factor, though. Larger servers don't die nearly as easily as bigger ones. (This is the reason why there aren't many 6v6 pub servers. Among others.) Additionally the likelihood of slots being open are also higher because of the higher player throughput. I imagine this convenience is a big reason for at least some players to play on huge servers.
edit: I would like to add that the few times I actually played on that server was for these specific reasons; and if I weren't actively avoiding it/rather closing the game than join the server, I imagine I would be playing there a lot more, since it's often the only server with open slots.
IronHorseDeveloper, QA Manager, Technical Support & contributorJoin Date: 2010-05-08Member: 71669Members, Super Administrators, Forum Admins, Forum Moderators, NS2 Developer, NS2 Playtester, Squad Five Blue, Subnautica Playtester, Subnautica PT Lead, Pistachionauts
Convenience definitely is part of it.
Any seasoned player should be able to notice the obvious difference in performance though, and not just clientside but server side. Every now and then and I check out these larger servers and when I do the difference is obvious to me.
On average, the server update rate is 1/3rd less (it never gets better than this, btw, as it's permanently limited) and occasionally it dips to 1/2 the default server rate.
This isn't anecdotal, you can see it yourself
The game is less responsive and more prone to issues at those rates (increased hitreg errors, increased deaths around corners, collision issues, delays)
So if your argument is that a seasoned player noticed these things and still decides to play... ok.. but then how can they make any claims as to the "quality" of their experience being better there?
Comments
Is that a railgun? Concede girl is right.
Where are you going with this? Are you trying to make a point or are you really just curious?
NS2 is not faction balanced 50/50 on any slot count (12-24). It is absolutely irrelevevant how balanced the game is on a server that does not run NS2.
You practically get a curve in balance on multiple levels. This is why server admins are adjusting to these curves with mods. 6vs6 environments usually have competivtive mod installed, while large ns2 servers (12vs12) usually have a second infantry portal.
Both of those mods change the gameplay in order to be more faction balanced.
Not necessarily. Someone just might know someone else who might be of equal skill and he invites him to a fair match.
The actual problem is however, that my ability to play football is rated at how I play football, basketball, american football and golf.
A 6vs6 is not a 12vs12 just like football is not handball.
NS2 even has a balance level deeper than that with its asymmetrical factions. Someone might be good as lerk on a 12vs12 slot server, but be bad as a lerk on a 6vs6 server, but at the same time be good as a marine on a 6vs6, but bad on a 12vs12 slot server.
The system could work, if everyone was playing the same type of ns2 and if that system would assign you to a server, instead of the team within your current servers player pool.
No, I still don't have stats, sorry.
Not 8v8?
It was a band aid solution mod that was first introduced on 24 slot servers. It was simply taken over into vanilla. It starts at 9vs9 because it is is the middle ground between 12 and 24 slots.
It is band aid, because 8vs8 is more unfair towards marines, while 10vs10 is more unfair towards aliens, since you can not have 1.67 / 2.33 infantry portals.
You couldn't even say that 6vs6 and 12vs12 are then equal in terms of balance, because if I managed to sneak into the marine base on a 12vs12 server, I would have to destroy 2 ips in order to keep the marines from spawning, instead of just 1. Losing 1 ip on a 12vs12 early game is a set back, while losing an ip in an early game 6vs6 is potentially game ending.
Now you could say: "yeah, well, but on a 24 slot server you could have potentially 2 skulks who could sneak into the marine base", but that case is less likely, because lane blocking and scouting becomes easier the more marines you have on the map.
You can argue that people play on them because smaller pop servers are full, but clearly SOME people prefer them, so there's no reason to remove them.
For another similar pvp game, say tf2, if someone suggested getting rid of the high play count servers because the game wasn't designed with so many players in mind, they would be laughed at?
What he said.
And we know that changing that context (# of players) changes the game, how maps work and balance, therefore how you play it.
It's one more way to play NS2 and it's fine.
That is not the point. The point is that >24 slot servers are not ns2 and never will be ns2, but are presented as such in the server browser. New players get a false impression of the game.
That is also not the point. I was talking about how the ELO system is fundamentally broken, because it throws differing slot counts all into one final rating, even though game play and therefore game balance differs with every possible slot count.
Are you talking about this one?
https://docs.google.com/forms/d/1oOSek56ulYKkYghmy9I5UmwZA-hame8-cvtcc8BNCmQ/viewanalytics
The question was: "Which PUBLIC SERVER SIZE do you like playing on?" and not "which one do you prefer?"
You had the possibility to tick multiple answers.
6vs6 is not even up for debate in this poll, since 6vs6 public doesn't work without matchmaking, but quite a lot of people like playing 6vs6.
edit: oups, sorry @Wake.
This brings back memories. Tried to come a clusterfuck on *snip*. Organized people for a sg rush with w1 a1 and saw around 20 ppl with die to 14 skulks.
In retrospect I should have meded the hive twice to represent my tears streamin down my face in this moment.
Are you saying the player base of regular 6v6/7v7/8v8 matches could be increased by 67% if we remove the 21v21 large servers?
Great! Sounds like a good plan, just like the movie industry say they lose billions of revenue and jobs due to piracy because everyone would buy originals if there was better law enforcement.
but the majority of players who play on 21+ pop counts arent veterans who have been on the forums ages and likely dont even have forum accounts.
Afaik, these servers host a significant part of players (understatement?), so what population is ranking going to measure ?
Not white listing them is a bad idea.
The main point was that someone could suggest to remove large servers in other to increase player base of regular servers based on Nordic's numbers. After all, some folks here stated that regular servers provide the "right" and "true" experience where other type of servers do not provide it.
The sarcasm is that this line of thinking is widespread (several analogies) and it's a known fallacy due to shallow reasoning. Those players of large servers are there not because of "NS2"/"true experience" but because of the gameplay experience that only a large server can provide.
And the movie piracy analogy is perfect. If you cant get access to the content/product/thing/experience, you don't seek it.
Can't stream a movie because the producers wan't to sell DVDs for 9 months first? Just download it, then.
It's a poor analogy on all fronts.
As far as the subjective argument over what experience a player prefers :
1) It's a moot point if all the customer knows is lesser quality from day one
2) Any argument over the many factors that determine the quality of the experience for those that enjoy large servers (like less responsibility, more forgiving, more to do etc) is entirely irrelevant if the experience is fundamentally broken due to performance issues. The game already struggles enough with the occasional hitreg issue or high interp values - compounding this with degraded server and client performance levels that were never intended to be played, just gives an even worse representation. Which is a shame, and is worth highlighting when we're talking about the the subject of the quality of experience to certain players.
In what way would 'not white listing' be a bad idea?
Here is what Moultano, creator of the hive skill system said about it in the last thread debating this. I love 8v8 ns2. I almost exclusively play on 8v8 ns2 and I refuse to play on servers 10v10 or higher if I can avoid it. Sure, there would be more servers to choose from but there would be less choice in how to play. If wooza has shown us anything, it is that there is a large part of ns2 that actively enjoys ns2large.
So let me explain where I am coming from and what I wanted.
I have made an observation X.
I asked if somebody has stats that explains X/shows its a fluke.
Nobody seems to have (I am not talking who has opinions, there are hordes of those).
The second best to actual observations of reality, i.e. stats, is put forth testable hypothesis of the reality.
In this case we are (un)lucky that we have, what seems, 2 segregated clusters in the same domain.
So you can make a hypothesis in 1 cluster and see if that predict anything in the 2nd cluster.
If you are unable or unwilling to do that, it is worth nothing, it is even worth negative, as it clutters the discussion.
In that case your opinion of what makes one faction win over another does not matter.
I can have an opinion that the winning side's commander is always visited by the invisible unicorn first.
So every commander should have invisible carrots.
There are several problems with that; most ppl will say it is not their experience.
Secondly, once I made my opinion of that, confirmation bias will make me virtually blind for anything that does not fit that opinion.
It is for those 2 reasons alone you must be forced to step out of your comfort zone, and start predicting what that would mean
if something changes, instead of continued getting your (wrong) opinion confirmed.
I think I have seen only 1 testable opinion/prediction (besides those testable opinions, that I believe are already contradicted by existing stats): It is testable, but it seems no stats record grenades/bile usage.
This definition of Hive skill is just too lax and still not fitting to ns2large:
- Well yeah, aiming, shooting and moving translate over to vanilla, but similarly would playing Counter Strike. NS2Large would show you some basics like Craig = heal, but so would a bot game. Numbers are still very messed up.
- Some vanilla essentials do not translate well or have very different effects, like laneblocking or rush / sneak strategies.
=> Game modes that teach bad behaviour should be excluded from Hive.
=> Definition of "similar in complexity". I find games on vanilla servers fairly different in complexity.
Now the problem is that in its current implemention, hive calculation doesn't properly scale with playernumbers. It does not adapt accordingly to changed balance and playstyle.
Also still doesn't answer why it would be bad to exlude ns2large.
Compmod is another story. It perserves all the essentials of vanilla and barely if any bad behaviour translates over to vanilla.
It's certainly no less complex than vanilla.
Maybe I've found an answer to your question:
http://apheriox.com/wonitor/configurator.html#x=winner&y=count&t=map&plotType=bar&tSort=desc&time_gt=2016-04-06&time_lt=2016-04-07
If you compare which maps were won by marines and aliens that day, you'll notice, that aliens were winning custom maps more often.
ns2_combi, ns2_colosseum, ns2_caged, ns2_spaceship were exclusively won by aliens.
Maybe it was custom map wednesday that day, starting at 7pm?
I can't tell you much more without more detailed informations about those matches themselves.
1. The flagship server of ns2large with 46 slots enjoys a full server at the prime time of different regions (Eastern/Western Europe and Eastern US). -> Players like it.
2. It has a recurrent/loyal player base of rookies and seasoned players. -> Players continue to like it over time.
3. It competes with both Euro and US regular servers at prime time. -> It has share of player's choice/preference.
If it's a bad or not rewarding experience, players would just not play it. An example of this is the latest updates to the siege mod that effectively killed it despite enjoying those 3 features above before the recent updates.
It's a moot point to: not to take ns2large seriously; not give it the same importance comp mod has; not give it any kind of support; not enjoy Hive skill balancing offer.
"Access" is not option exclusive. An analogy may be poor when one does not understand it.
Don't forget that the size of the server is a factor, though. Larger servers don't die nearly as easily as bigger ones. (This is the reason why there aren't many 6v6 pub servers. Among others.) Additionally the likelihood of slots being open are also higher because of the higher player throughput. I imagine this convenience is a big reason for at least some players to play on huge servers.
edit: I would like to add that the few times I actually played on that server was for these specific reasons; and if I weren't actively avoiding it/rather closing the game than join the server, I imagine I would be playing there a lot more, since it's often the only server with open slots.
Any seasoned player should be able to notice the obvious difference in performance though, and not just clientside but server side. Every now and then and I check out these larger servers and when I do the difference is obvious to me.
On average, the server update rate is 1/3rd less (it never gets better than this, btw, as it's permanently limited) and occasionally it dips to 1/2 the default server rate.
This isn't anecdotal, you can see it yourself
The game is less responsive and more prone to issues at those rates (increased hitreg errors, increased deaths around corners, collision issues, delays)
So if your argument is that a seasoned player noticed these things and still decides to play... ok.. but then how can they make any claims as to the "quality" of their experience being better there?