Natural Selection 2's Unsatisfaction and Player Retention
ZdrytchX
Australia Join Date: 2016-02-06 Member: 212662Members
Australia Join Date: 2016-02-06 Member: 212662Members
yah di dah, I know. I understand that most people prefer to keep the gameplay as-is with very minor changes like the lerk biomass/starting health change two patches ago. But there is a lot of dissatisfaction with this game in terms of vanilla Natural Selection 2 gameplay from my opinion, and from many former NS1 players that I've asked if they liked NS2. Also WARNING: Lots of references to Tremulous
It has been roughly 1600 days since Natural Selection 2's release (1599). I decided to do a statistics comparison of players per day in Tremulous to Natural Selection 2's.
Here's a chart showing the history of Tremulous and Natural Selection 2's player count after 1600 days of release:
TOP: Tremulous player per-day moving average count since statistic gathering started after 1.1's standalone release.
Middle: NS2 average players per day in a month.
BOTTOM: NS2 players per day
Yes I know that it isn't a "good comparison" because Tremulous has been out for so long and is a free and open sourced shooter. But the thing is Tremulous didn't get as much publicity as Natural Selection 2 did, and NS1 despite seemingly having a larger current playerbase than NS2: Combat and NS1 and AusNS2 combined. Heck, even NS2 got TV ads in my country, and I was not even expecting that to happen.
The way Tremulous was distributed was often, and usually, through either through LAN parties starting from a USB transfer (since the game was only 90 MB with pure vanilla assets), or ModDB's "game/mod of the year" attention.
Looking at these graphs I realised that the free weekend was pretty much a disappointing failure. It's not about the sudden player increase, it's more about the players gained after the free weekend. I doubt that more than 10 more people than the usual rate even bought the game in the next 7 days after that free weekend.
Below are two major subjects that I have pointed out, but I may add some later. Apparently the "draft" feature of this forum doesn't work
1. "Concede Simulator" - Players don't like the feeling of losing. When compared to Tremulous, it seems as though Tremulous did pretty much everything better in this subject. Tremulous does have a 'concede' function without any server enhancements but it is rarely used because there simply isn't a need for it most of the time due to "sudden death" function and just the way the game functions overall.
a) It is very hard to turn the tables once your team is losing, especially if your team has poor co-ordination and/or skill.
Let's say you have two teams in a n 8v8 (since that is what pubs seem to be at least) of 1500 hive skill. Team 1 has two players with hiveskills at 3000 or above while the rest have hiveskills below
1002. Team 2 has around 1500 +/- 500. Which team will win most of the games? Team 2.
The reason for this is simply because the game is soo focused on team play that they forget to reward players for doing well. Instead they reward a team for doing well. A direct reference would be the lack of "kill for resources" but that isn't all of it.
Another reason would be because a team that starts to lose will generally start losing players. Just because Team 1 has two players with aimbot reflexes, if their team doesn't even do something as simple as capturing and securing resource nozzles then they've already lost the resource game.
b) The game tends to reward larger teams. A few months back I created a simple slightly game changing mod which apparently only Wooza's server uses in the form of a mod pack combining various mods into one. All it does is simply scale the period (wait time) between the resource 'ticks' that you get your resources based off the ratio of team sizes. If team 1 has twice as many players as team 2, they will only need to wait 4 seconds while team 2 needs to wait 8 seconds for every resource 'tick.' It isn't much, but it is something.
The way Tremulous had solved this problem is something that not even NS1 has from what I know, and that is the fact that the "resource required for research is scaled up by team size" (in Tremulous vanilla rules, you require 4 frags [kills] per player for a stage up meaning that a large team will take a long time massacre-ing a small team to get their upgrades while the small team, although may be suffering for a player shortage, is rewarded strongly for their successful efforts.)
c) Because of the above two, I feel that Veterans will not stick too long. The current "veterans to newbies ratio" is something like 6:1 right now in Tremulous where I define "veteran" as a player who has played the game for more than 2 years while "newbie" is a player who has not player the game for more than 2 years. I feel that NS2 has a veterans:newbies ratio of about 2:3 currently
I'm actually surprised that the elo that I often see are usually under my level, considering that I generally have aiming skill worthy of an elo of 1000 and that I really spent only like 300 hours or less actually playing against players with vanilla balance.
2. [/u]Requirement to get a "decent" game going.[/u]
a) The number of players to get a decent public game is somewhat high.
By this I mean that a server will generally need around 8 players for a seeding to be successful. Natural selection 2 is quite poorly balanced below the 6v6 "standard" for competitive games especially for public games. Tremulous is fine most of the time (though still slightly biased to the larger team) whether it be 4 vs 6, 2v2 or even 32vs32.
Natural Selection 2 on the other hand is heavily marine biased at 8v8 with an average hive skill of 2000 on each team while somewhat alien biased at 12v12 with average 1000 hive skill on each team. Tremulous does have a somewhat similar issue but it is still nevertheless, fun for even the team at a disadvantage because they know that they can change the tide of the game easily, for the opportunity is provided at the right time. A human with just 170 credits (which can be obtained by killing one dretch, or skulk equivalent) can in some base layouts, flank solo and take out the 'overmind' (hive) at the cost of losing one player on the front line battlefield. All it takes, is 3.75 seconds of touching the overmind with a painsaw, and the alien team defences are all paralised until the alien builder rebuilds the overmind and cannot build anything until the overmind is built, meaning humans are then provided with the opportunity to end a possible stalemate by simply destroying all the remaining eggs and builders.
b) Spawning balance. This has been discussed a lot on the forums and there is a mod which basically solves the problem to some extent.
The problem is basically that while Marines have to wait about 10 seconds after returning to spectator mode after dying to actually respawning, Aliens only need to spend about 3 seconds while when the situation for it happens, Aliens spend their time in the spawn queue in the form of spectator even while eggs are already spawned. The major thing that the eggs respawning mod fixed was basically not to spawn all eggs at the hive closest to danger, and that players spend more time in their eggs before spawning.
Tremulous' approach to this is instant spawning for both team, but 10 second spawning cooldown per spawn structure (eggs in Trem are fixed like infantry portals, but they and the human telenode spawns can be built just about anywhere flat while alien eggs can be hidden on ceilings in dark areas - Humans can find them with their own personal purchasable helmet-radars available at first stage up)
c) Auto building structures
Unlike Q2 Gloom and Tremulous and the Khaara team, marine structures in Natural Selection 2 don't build themselves without a MAC or player to build them. The reason why this was not a problem in Tremulous while Alien structures are generally fast-constructing compared to human buildables, is that a player doesn't have to sit next to and practically hug the structure and heal/repair/build it. Players can also "deconstruct" structures (like Marine Commander's ability to recycle stuff but at a full refund) or "mark" in some versions (structure will still exist until the team lacks the resources for new structures or if the player wants to build in a spot intersecting with the marked structure to be built by being replaced)
c) Independance
Unlike Aliens in NS2, Marines are heavily dependant on the marine commander. Some reasons for this include the fact that welding friendly players does not provide health, structures can only be placed by commanders and that marines cannot "solo" an alien well. In vanilla balance, an Onos fighting against a marine without a jetpack (such as using an exosuit) is pretty much guaranteed to die. But a problem with jetpackers is that the weapons they carry are too weak to deal with an alien class like the Onos which basically means you need at least two players to deal with an onos. But well, overpricing the Onos is a way to balance it.
I think NS2: Combat balanced the Onos somewhat well for 1v1 fights excluding the ability to 'devour' marine players. But marines still take waay too long to independantly destroy even something like shade. I think overall structure health is just excessive in this game to the point that a skulk can bite res in ns2_tram's Containment and a marine could basically run all the way from Ore and shoot the skulk before it can finish killing the extractor.
to be continued..
It has been roughly 1600 days since Natural Selection 2's release (1599). I decided to do a statistics comparison of players per day in Tremulous to Natural Selection 2's.
Here's a chart showing the history of Tremulous and Natural Selection 2's player count after 1600 days of release:
Middle: NS2 average players per day in a month.
BOTTOM: NS2 players per day
Yes I know that it isn't a "good comparison" because Tremulous has been out for so long and is a free and open sourced shooter. But the thing is Tremulous didn't get as much publicity as Natural Selection 2 did, and NS1 despite seemingly having a larger current playerbase than NS2: Combat and NS1 and AusNS2 combined. Heck, even NS2 got TV ads in my country, and I was not even expecting that to happen.
The way Tremulous was distributed was often, and usually, through either through LAN parties starting from a USB transfer (since the game was only 90 MB with pure vanilla assets), or ModDB's "game/mod of the year" attention.
Looking at these graphs I realised that the free weekend was pretty much a disappointing failure. It's not about the sudden player increase, it's more about the players gained after the free weekend. I doubt that more than 10 more people than the usual rate even bought the game in the next 7 days after that free weekend.
Below are two major subjects that I have pointed out, but I may add some later. Apparently the "draft" feature of this forum doesn't work
1. "Concede Simulator" - Players don't like the feeling of losing. When compared to Tremulous, it seems as though Tremulous did pretty much everything better in this subject. Tremulous does have a 'concede' function without any server enhancements but it is rarely used because there simply isn't a need for it most of the time due to "sudden death" function and just the way the game functions overall.
a) It is very hard to turn the tables once your team is losing, especially if your team has poor co-ordination and/or skill.
Let's say you have two teams in a n 8v8 (since that is what pubs seem to be at least) of 1500 hive skill. Team 1 has two players with hiveskills at 3000 or above while the rest have hiveskills below
1002. Team 2 has around 1500 +/- 500. Which team will win most of the games? Team 2.
The reason for this is simply because the game is soo focused on team play that they forget to reward players for doing well. Instead they reward a team for doing well. A direct reference would be the lack of "kill for resources" but that isn't all of it.
Another reason would be because a team that starts to lose will generally start losing players. Just because Team 1 has two players with aimbot reflexes, if their team doesn't even do something as simple as capturing and securing resource nozzles then they've already lost the resource game.
b) The game tends to reward larger teams. A few months back I created a simple slightly game changing mod which apparently only Wooza's server uses in the form of a mod pack combining various mods into one. All it does is simply scale the period (wait time) between the resource 'ticks' that you get your resources based off the ratio of team sizes. If team 1 has twice as many players as team 2, they will only need to wait 4 seconds while team 2 needs to wait 8 seconds for every resource 'tick.' It isn't much, but it is something.
The way Tremulous had solved this problem is something that not even NS1 has from what I know, and that is the fact that the "resource required for research is scaled up by team size" (in Tremulous vanilla rules, you require 4 frags [kills] per player for a stage up meaning that a large team will take a long time massacre-ing a small team to get their upgrades while the small team, although may be suffering for a player shortage, is rewarded strongly for their successful efforts.)
c) Because of the above two, I feel that Veterans will not stick too long. The current "veterans to newbies ratio" is something like 6:1 right now in Tremulous where I define "veteran" as a player who has played the game for more than 2 years while "newbie" is a player who has not player the game for more than 2 years. I feel that NS2 has a veterans:newbies ratio of about 2:3 currently
I'm actually surprised that the elo that I often see are usually under my level, considering that I generally have aiming skill worthy of an elo of 1000 and that I really spent only like 300 hours or less actually playing against players with vanilla balance.
2. [/u]Requirement to get a "decent" game going.[/u]
a) The number of players to get a decent public game is somewhat high.
By this I mean that a server will generally need around 8 players for a seeding to be successful. Natural selection 2 is quite poorly balanced below the 6v6 "standard" for competitive games especially for public games. Tremulous is fine most of the time (though still slightly biased to the larger team) whether it be 4 vs 6, 2v2 or even 32vs32.
Natural Selection 2 on the other hand is heavily marine biased at 8v8 with an average hive skill of 2000 on each team while somewhat alien biased at 12v12 with average 1000 hive skill on each team. Tremulous does have a somewhat similar issue but it is still nevertheless, fun for even the team at a disadvantage because they know that they can change the tide of the game easily, for the opportunity is provided at the right time. A human with just 170 credits (which can be obtained by killing one dretch, or skulk equivalent) can in some base layouts, flank solo and take out the 'overmind' (hive) at the cost of losing one player on the front line battlefield. All it takes, is 3.75 seconds of touching the overmind with a painsaw, and the alien team defences are all paralised until the alien builder rebuilds the overmind and cannot build anything until the overmind is built, meaning humans are then provided with the opportunity to end a possible stalemate by simply destroying all the remaining eggs and builders.
b) Spawning balance. This has been discussed a lot on the forums and there is a mod which basically solves the problem to some extent.
The problem is basically that while Marines have to wait about 10 seconds after returning to spectator mode after dying to actually respawning, Aliens only need to spend about 3 seconds while when the situation for it happens, Aliens spend their time in the spawn queue in the form of spectator even while eggs are already spawned. The major thing that the eggs respawning mod fixed was basically not to spawn all eggs at the hive closest to danger, and that players spend more time in their eggs before spawning.
Tremulous' approach to this is instant spawning for both team, but 10 second spawning cooldown per spawn structure (eggs in Trem are fixed like infantry portals, but they and the human telenode spawns can be built just about anywhere flat while alien eggs can be hidden on ceilings in dark areas - Humans can find them with their own personal purchasable helmet-radars available at first stage up)
c) Auto building structures
Unlike Q2 Gloom and Tremulous and the Khaara team, marine structures in Natural Selection 2 don't build themselves without a MAC or player to build them. The reason why this was not a problem in Tremulous while Alien structures are generally fast-constructing compared to human buildables, is that a player doesn't have to sit next to and practically hug the structure and heal/repair/build it. Players can also "deconstruct" structures (like Marine Commander's ability to recycle stuff but at a full refund) or "mark" in some versions (structure will still exist until the team lacks the resources for new structures or if the player wants to build in a spot intersecting with the marked structure to be built by being replaced)
c) Independance
Unlike Aliens in NS2, Marines are heavily dependant on the marine commander. Some reasons for this include the fact that welding friendly players does not provide health, structures can only be placed by commanders and that marines cannot "solo" an alien well. In vanilla balance, an Onos fighting against a marine without a jetpack (such as using an exosuit) is pretty much guaranteed to die. But a problem with jetpackers is that the weapons they carry are too weak to deal with an alien class like the Onos which basically means you need at least two players to deal with an onos. But well, overpricing the Onos is a way to balance it.
I think NS2: Combat balanced the Onos somewhat well for 1v1 fights excluding the ability to 'devour' marine players. But marines still take waay too long to independantly destroy even something like shade. I think overall structure health is just excessive in this game to the point that a skulk can bite res in ns2_tram's Containment and a marine could basically run all the way from Ore and shoot the skulk before it can finish killing the extractor.
to be continued..