• Please make sure you are familiar with the forum rules. You can find them here: https://forums.tripwireinteractive.com/index.php?threads/forum-rules.2334636/

Poll About Max Player Counts And Server Performance

Jump to latest Follow Reply

Poll About Max Player Counts And Server Performance

  • Reduce the player counts to 50 across the board

    Votes: 83 19.9%

  • Reduce the player counts to 50 on ranked, allow 64 on unranked

    Votes: 125 30.0%

  • Leave the player counts alone, we'll deal with it

    Votes: 209 50.1%
  • Total voters
    417
Sort by date Sort by votes
#41
Nikita said:
50 player cap temporarily? :D
Click to expand...

It wouldn't be a bad idea to coordinate a 50 player cap during a free weekend, or a number of free weekends for that extra performance Ramm speaks about for new players-- and then go back up to a 64 player cap after some time.
It would be like fishing-- give some new players the bait to reel them in.
 
Upvote 0
#42
[TW]Ramm-Jaeger said:
Just to put a little more input in here. While I generally agree with the idea that players will play on the servers where they enjoy the game the most, and thus bad (i.e. CPU overloaded servers) will go away by attrition, there is another wrinkle to this.

Lets say we have a free weekend, and a half a million potential new RO2 players play the game for the first time over those couple of day. Lets also say that within the couple days a free weekend lasts, there may well be lots of poorly performing 64 player servers. Yes over the long term those servers would likely be weeded out, but over the short term they won't. So the experience that those free weekenders could end up with is a poor one. And you never get a second chance to make a first impression.

In other words, while YOU personally might be a savvy long time RO players who will choose where you want to play (after already owning the game), a free weekender potentially is not. They are there to check out the game, and if they don't have a positive experience they may well just move on.
Click to expand...

OK then, drop the player count down for the mean while (I can deal with that).
Let it be known that 64 player servers are "as the game should be played".

I may be barking up the wrong tree (I asked this already) but why can't 50 & 64 servers both be ranked ?
You know 64/100x50xco-eff : stats = ?

Enjoying the beta btw.
Thanks.
 
Upvote 0
#43
Excellent. I am glad you guys have been tackling this issue.

It's a tricky one too. Players seem to choose servers with higher player counts, despite the performance issues that come along with it.

I think I personally would prefer the 50 player max. I think it results in better game play as well as better latencies.

Is there any way to allow the player count to be set based on polling or bench-marking server hardware? it would be a shame to limit servers to 50 players today, and then have that limit still be present in two years time when faster server CPU's are available...

The only downside to this change I think is that many areas only have one or two servers within decent pingable range that fill up with people. Hopefully the 14 to 28 people who no longer fit on those servers during high load, will populate another server in the area, and not just get disillusioned if they can't get in, and stop playing.

What would be really nice would be some sort of server script for server operators that - if the server is full upon joining - could place players in a temporary server while they wait in line. A sort of purgatory server :p and then dynamically switch them to the main server as spots open up, depending on how long they have been waiting.

You could even have player management based on honor level, moving the most skilled players (or at least players with most in game time) to the main server, and leaving those still learning in the purgatory server.

Unfortunately what happens now is that players seem to coalesce around one or two large servers in an area (personally I play on the 2.fjg server) and when that server is full, they tend to give up, or fragment across 20 or so mostly empty servers with bots. I feel part of this issue is due to bots being show in the player count in the server browser, so it is sometimes very difficult to see where other humans are playing.

I've definitely been there, joined server after server, only to find that there are mostly bots, and then moved on.

Anyway, thanks again for looking into this issue. Fixing/reducing the latency issue should be HUGE for the popularity of the game.
 
Upvote 0
#44
I have had no problems on the 64 player servers and reducing the max players to 50 would be taking a step back in my opinion. I understand the reasoning behind this for the weekend, so maybe you can top off the max player count to 50 during the free weekend.
Other than that, leave it to the server admins to decide afterwards. I think it would be a shame to reduce 16 players.

Cheers
 
Upvote 0
#45
Well, as others have pointed out-- players gravitate to servers inhabited by real human players-- and that there are only one or two human populated 64 player servers at any one moment, so the 50 player cap would create a bigger spread. Those 14 or 28 excess human players would most likely end up on the next servers of lower population. So, for example, if the next server lower in population had around 20 humans playing on it, the excess would bump into the 30s, etc. which is a very playable amount of people on a server.
 
Upvote 0
#46
PeteAtomic said:
Well, as others have pointed out-- players gravitate to servers inhabited by real human players-- and that there are only one or two human populated 64 player servers at any one moment, so the 50 player cap would create a bigger spread. Those 14 or 28 excess human players would most likely end up on the next servers of lower population. So, for example, if the next server lower in population had around 20 humans playing on it, the excess would bump into the 30s, etc. which is a very playable amount of people on a server.
Click to expand...

I hope this would happen, but in real life I find that those who can't fit on the main regional server tend to fragment or quit all together, which is not desirable.

I think a way to limit this from happening would be to not display bots in the server player total sin the server browser, and instead only show live human players in the player totals. This way players won't be frustrated jumping from server to server in order to find somewhere to play.
 
Upvote 0
#47
Personally I can only find one decent regularly populated server and thats Mord's 40-1 64 player valhalla. Please from the bottom of my heart....for the love of god...just leave well enough alone.

As a new player I was only just beginning to breath easy now that the horribly atavistic, pixel-sniping, tortoise-paced Classic mode seems to have finally satiated the vociferous rivet-counting anorak brigade, glad in the knowledge that dev time could now be spent on refinements and new content (Spawn on squad leader working now - YAY).

Please, please don't take away the opportunity to play ranked on 64 player servers. You listened to the fanatics re. Classic, now don't alienate the generality....
 
Upvote 0
#48
Lord_Lovat said:
I have had no problems on the 64 player servers and reducing the max players to 50 would be taking a step back in my opinion. I understand the reasoning behind this for the weekend, so maybe you can top off the max player count to 50 during the free weekend.
Other than that, leave it to the server admins to decide afterwards. I think it would be a shame to reduce 16 players.

Cheers
Click to expand...

I honestly don't feel the gameplay experience suffers with fewer players. A full 32 player server is every bit as enjoyable as a full 64 player server. (actually, in many cases more so, due to fewer latency problems) Get much below 32 players though, and it starts to become less fun.
 
Upvote 0
#52
Leave it as it is.

I don't see the minor performance increase as a good trade off for a large gameplay decrease.

In regards to the possible first impression on new players, I think possibly weeding out the 'bad' servers would be a better idea. Some restriction or minimums should be applied, but I'm not the one to decide what.
 
Upvote 0
#55
Garcia2100 said:
I remember there was this thing called an anti lag mutator or whatever. I remember playing on a 64 with it, it was freaking amazing. Didn't matter if it was 40, 50, 64, all the shots went where they were suppose to go.
Click to expand...

If the CPU of the server is overloaded due too more slots than it can handle, then the Antilag mutator won't solve the problem either. Even a top racing horse can't win a race if it has to run in deep waters.
 
Upvote 0
#56
Ducky said:
If the CPU of the server is overloaded due too more slots than it can handle, then the Antilag mutator won't solve the problem either. Even a top racing horse can't win a race if it has to run in deep waters.
Click to expand...
That's not actually true in this case. Antilag's design essentially ignores any variances in server performance. The delay before the shot takes effect will increase, but it -will- take effect...there's no special trick to it, that's just an inherent function of the design and the communication paths it uses with the server. The system will hold up to around 750ms of latency, and it only stops there because that's where I chose to impose a cutoff. By default it would function in anything short of infinite latency (aka, a crashed/disconnected server)

Then you factor in that the Antilag design also innately improves server performance by distributing shot physics across the clients, and it's a win-win.

If the server is really badly overloaded, you'll get the usual oddities of warping and the occasionally bizarre physics effect, but the shooting won't be significantly affected by any reasonable amount of delay or fluctuations in same. That was kind of the whole point of the mutator :)

Now if your *client* performance is bad, then you'll be in trouble...
 
Upvote 0
#58
Making 64 person servers "unranked" is the best of both worlds.
I've noticed very little to no difference in gameplay on 50 vs. 64 person servers. It's already a little annoying that all the players are being soaked into one server. 50 players is already pretty crowded and performance is always good.


Furthermore,
For both balance and performance, Platoon Commander cooldown should increase when more players are present.
It should scale. Scaling something like:
1-16 = 1.0
17-32= 1.25
33-50= 1.5
51-60= 1.75


As it is now, large player numbers make artillery not only "SUPER EFFECTIVE!" but also add stress to many systems both client and server-side. Can't remove it, might as well reduce the frequency.
 
Last edited:
Upvote 0
#59
[TW]Ramm-Jaeger said:
Just to put a little more input in here. While I generally agree with the idea that players will play on the servers where they enjoy the game the most, and thus bad (i.e. CPU overloaded servers) will go away by attrition, there is another wrinkle to this.

Lets say we have a free weekend, and a half a million potential new RO2 players play the game for the first time over those couple of day. Lets also say that within the couple days a free weekend lasts, there may well be lots of poorly performing 64 player servers. Yes over the long term those servers would likely be weeded out, but over the short term they won't. So the experience that those free weekenders could end up with is a poor one. And you never get a second chance to make a first impression.

In other words, while YOU personally might be a savvy long time RO players who will choose where you want to play (after already owning the game), a free weekender potentially is not. They are there to check out the game, and if they don't have a positive experience they may well just move on.
Click to expand...

Attrition will still play a very important role in the above described scenario. Further, even "new" players are savvy enough to be drawn to active and lively servers. They'll avoid the "stiff" servers simply because they see the majority of players are avoiding said servers.

If, for argument's sake, a half million players did jump in on a free weekend - it would make little or no difference "good or bad" server, because every server... from 12 man to 64 man would be full most all the time.

In any case, penalizing those Admins who've invested the time and money into providing top notch servers by making their 64 man servers unranked is not only unfair.... it is flat out cruel. Please, reconsider such a policy. Perhaps a load test for each of the 64 man Servers to pass?

In any case, attrition will work well and.... it will do so surprisingly fast when the Admins/Owners of the lame 64 man servers note their servers are pointed out by being dead servers.
 
Last edited:
Upvote 0
You must log in or register to reply here.
Top Bottom
Back
Jump to latest Follow Reply
Jump to latest Follow Reply
Community News
Forums
Merch
Menu