Higher bandwidths = less max players why

(iwound) #1

why is it that although higher bandwidths are avaiable for most people games particularly SDs are decreasing the max amount of players in game.
with W:ET there was 64, QW 24 down from an initial 32, brink 16. DB ?
please dont bother with the old argument more isnt necessarily better as it just doesnt make sense. i want more. if you like less and a game gives you more you just adjust the max.

my point is its nearly 2013 i have a very fast connection and many have much faster than i so the max should go up not down and give me the choice to choose how many i play against. if the levels cant accomodate higher numbers then at least have the choice for custom maps which could.

my suspicion is its to gear games towards competition play which is a big mistake as these players make up a tiny percentage of total players.
without the huge base of pub players on W:ET comp wouldnt last 2 secs as they can only survive if people are watching or take an interest. a game cannot survive on comp play alone. and is built on pub players playing for fun. the tragedy is to gear a game for comp you have to remove the fun.

with dirty bomb coming it will only be a worthy succesor to W:ET if the max players is high. maybe there is a max with unreal i dont know. but if its capped low for no good reason it just wont be succesful.

come on sd think big.

(GENETX) #2

Max should be unlocked imho as said. Altough, I don’t want the big servers as I see them now in W:ET. Hard to find a good server I like with a 20-24 player max. These 20v20 games in ET are useless. It is not what ET was designed for. So maybe it is good to help server admins by giving a bit more insight on bandwith usage and performance. Giving the right tools would give a bit more insight to tune the server. But also provide the target of a map, how much players it should support by design. For W:ET, the maps are all around 8v8 - 12v12 to have fun imho.

(murka) #3

Well, one thing you didn’t consider is performance. Just try recording a demo of full 16v16 action(bot game for ex) and then a solo run through the map. Now try both low and high settings and see how the timedemo goes. It wont be as bad for low settings, but will still take a noticeable dump and especially fps stability will go down. On high settings i bet the game becomes unplayable 1-30fps with nonexistant stability. To run a 64 player game in regular etqw you’d need a rather expensive rig.

Also the “fast” connection part is a bit iffy. Back in 2003 or so when fast broadband came, i had like 0-1ms ping to local servers and quite decent ping to neighboring countries. Nowadays internet has more bandwidth but on average the pings have gone up making online play less enjoyable so it’s not all about bandwidth.

I do have to admit that capping player limits make no sense. Especially if the game will be able to run mods. Etqw had an engine cap of 32, but some mods like Dusk would have been much more enjoyable with more. Just leave it up to the server admins to decide. I bet the game will balance itself out and everyone will find their own niche, be it 64 player spamfests or more organized near-comp 5v5.

(iwound) #4

[QUOTE=murka;417688]Well, one thing you didn’t consider is performance. Just try recording a demo of full 16v16 action(bot game for ex) and then a solo run through the map. Now try both low and high settings and see how the timedemo goes. It wont be as bad for low settings, but will still take a noticeable dump and especially fps stability will go down. On high settings i bet the game becomes unplayable 1-30fps with nonexistant stability. To run a 64 player game in regular etqw you’d need a rather expensive rig.
[/QUOTE]

firstly, running bots is a big hit on performance. secondly etqw had vehicles & deployables which a/ hits performance and b/ took a lot of extra bandwidth.
performance issues can be dealt with in map design as well as game design to optimise.
but also consider that cpus are getting faster/cheaper also.

il give you that. but there is an acceptable range that works. you dont need <10ms to have fun.

[QUOTE=murka;417688]
I do have to admit that capping player limits make no sense. Especially if the game will be able to run mods. Etqw had an engine cap of 32, but some mods like Dusk would have been much more enjoyable with more. Just leave it up to the server admins to decide. I bet the game will balance itself out and everyone will find their own niche, be it 64 player spamfests or more organized near-comp 5v5.[/QUOTE]

well said. to me to restrict the number is just putting 2 fingers up to those who want more and saying we dont want you to play this, go away.like it or lump it. unless there is a sound technical reason.

(jazevec) #5

Splash Damage have this fetish for clan matches. Trying to please clanners and leagues before making a good game is like trying to invent Olymphics before inventing running. Yet, that’s what SD does. Nevermind that clanners come from unusually devoted pubbers, and a more fun game means more clanners as well. Nevermind that vast majority of pubbers doesn’t give a crap about leagues and doesn’t watch replays.

Because it gets exponentially harder to organize groups of people with each extra member(“I have a job, cat, kids, blah blah”), most clans have settled for 6x6 matches and try to convince themselves it plays better because of the low player count.

(BioSnark) #6

More players works best with vehicles and similar mechanics geared to large scales. I never saw much point to 12 versus 12 on the infantry maps and segments in ET:QW. Cluster fk without the added value.

With regards to adjustable values, if there is a wide range then the game mechanics and map scale absolutely need to adjust dynamically. ET:QW has a lot of large scale oriented rock paper scissors mechanics that break down badly with small player numbers.

(.Chris.) #7

Well if I made a game balanced for a specific number I wouldn’t want there to be loads of servers with either way too many player slots or too few slots that offer a sub par experience that could potentially off put any new comers playing your game. If I joined a 64 man server for first time when first trying ET I doubt I’d be here today. You can’t easily cater for such a large variance in player numbers and relying on mods to do it for you is a bit sloppy no? By all means let mods try do something with different player numbers but for the base game, pick a sensible range and build a great game based on that. Be pretty hard to balance a game multiple times for multiple player numbers.

(Floris) #8

Personally I find games having small team sizes offer a much more intimate experience. In a 3vs3 ET match, when you die with a high remaining spawn time, your team is pretty much screwed for that stage. That’s clan play of course.

In a public match the most amazing moments I experienced was when your team was mostly down while you are defending an important objective with just a pair of people, with the majority of the opposing force still standing and coming for you. Take the crane control defence in supplydepot for example; you start out with a large perimeter, having people in the various towers of the complex. This might work for a few waves, but sooner or later the enemy will be pushing you back to the upper floor. At this stage combat is taking place in close corridors and even as a single person you can do a lot of damage by using your grenades and playing cat and mouse with the enemy.

My point is that as the team size grows you naturally play a less important role in the team, while the concept of spawn waves seems to fade away (in ET:QW and BRINK, to even get close to some objectives you’d travel longer than the maximum spawn time) and action gets more decentralized.

I won’t say having bigger team sizes doesn’t offer advantages of their own, because you will be part of a bigger whole, which actually feels good too. But it’s a different type of game, and personally I am more inclined to lean to the intimate experience offered by games with small teams. I mostly consider myself as a pubber who also plays (well, played) competitive matches.

Now the question is why SD like to build these more intimate games. I guess that’s because various of the staff members have a background playing competitively but also because SD used to be (and I guess in a lot of ways still is) a small, quite intimate, company.

(iwound) #9

with fewer players in a pub as soon as someone is afk or leaves the match or is new or just likes exploring it becomes lob sided.
ok if your a clan and you have to play on. but if i get a call or something else crops up i do it.
bigger player counts can absorb this with little effect.

but my point was a technical one. because if you like a smaller match then play in a smaller match. why stop the bigger ones.
why would you want to stop people playing the way they want to.

im not suggesting cram 64 players in a small level and force people to play that way.
some server owners would have the core game and others would go mental. whether mods, maps or player counts.
its down to sd how mental we can go.

i totally respect lockis own goals but you can have your cake and eat it. in other words we can all get what we want.

practically it could be locking down team sizes to a point but allowing bigger under certain circumstances.
ultimately having some control.

i miss a proper filled online battlefield, bodies exploding everywhere.
one day maybe…

** should just say my preference is 16v16 in a 40 slot server to allow for afks etc

all the best in 2013

(montheponies) #10

I understood this reduction in numbers was all to do with making the games more console friendly.

i would completely contest the idea that SD are a competition/clan minded developer - nothing they have done, apart from Brink (ironically), would be considered out of the box designed for competition. Look somewhere else for your answer - like I say my money is on the console limitations.

I fully expect dirty bomb to support at least 16v16 servers, which to me is a decent balance for pub play - sufficient to cater for larger maps or meatfests, without just becoming ridiculous.

(jazevec) #11

The argument that on pubs higher player count reduces the effect of luck is a very good one. Yes, on average the chance of getting bad players, AFKers or someone significant leaving is smaller when teams are bigger, because individual players mean less, and bad play tends to average on both sides.

The way I would try to handle player counts would be making maps that scale gracefully. When there are more players, a new passage would open up.

This is a general trend: despite computers getting much faster, programs aren’t getting proportionally better. They aren’t proportionally more responsive, for example - often less ! That’s because programmers cared more about optimization and performance in the past, now they think they can get away with a lot. Once upon a time I noticed Notrium savegames were around 50MB each, while zipped they were 10 times smaller. I mailed that to the developer, but he didn’t care and said people have big hard disks these days.

(Floris) #12

[QUOTE=jazevec;418114]The argument that on pubs higher player count reduces the effect of luck is a very good one. Yes, on average the chance of getting bad players, AFKers or someone significant leaving is smaller when teams are bigger, because individual players mean less, and bad play tends to average on both sides.

The way I would try to handle player counts would be making maps that scale gracefully. When there are more players, a new passage would open up.[/QUOTE]
At the same time, lower player count increases the effect of skill, which reduces the effect of luck. It seems increasing the player count to compensate for bad, AFK or leaving players can be solved in various ways:

  1. Match making so there are no bad players in the first place.
  2. Adding an AFK mode so that a player temporarily gets replaced by a bot and gets kicked after a time-out expires.
  3. Replacing a player by a bot when he leaves.

I think BRINK did the first and the third while certain ET mods had AFK time-out kicks. Then again, bots might not be the best solution considering many people seem to dislike having them in MP.

I really like your graceful degradation idea, do you have any examples of games already doing that? It would be a really cool feature and perhaps even a unique selling point for the game.

(jazevec) #13

[QUOTE=Florisjuh;418140]
I really like your graceful degradation idea, do you have any examples of games already doing that? It would be a really cool feature and perhaps even a unique selling point for the game.[/QUOTE]

Container City originally had an extra entrance from the first Security spawn, which (I imagine) made their spawn harder to camp. I also heard about this happening in Battlefield games, but I don’t think it’s a big deal there. Splash Damage maps are bottleneck-centric, and the limit is mostly due to passages getting overcrowded. Imagine the tunnel on Oasis (W:ET) getting opened based on player count rather than having an engineer. And another tunnel.

(attack) #14

hmm i would say for competition max 6v6 . pulic max 20

(tokamak) #15

[QUOTE=jazevec;417893]Splash Damage have this fetish for clan matches. Trying to please clanners and leagues before making a good game is like trying to invent Olymphics before inventing running. Yet, that’s what SD does. Nevermind that clanners come from unusually devoted pubbers, and a more fun game means more clanners as well. Nevermind that vast majority of pubbers doesn’t give a crap about leagues and doesn’t watch replays.

Because it gets exponentially harder to organize groups of people with each extra member(“I have a job, cat, kids, blah blah”), most clans have settled for 6x6 matches and try to convince themselves it plays better because of the low player count.[/QUOTE]
Nicely put.

(montheponies) #16

and completely inaccurate. :slight_smile:

(Kendle) #17

[QUOTE=montheponies;418029]I understood this reduction in numbers was all to do with making the games more console friendly.

i would completely contest the idea that SD are a competition/clan minded developer - nothing they have done, apart from Brink (ironically), would be considered out of the box designed for competition. Look somewhere else for your answer - like I say my money is on the console limitations.

I fully expect dirty bomb to support at least 16v16 servers, which to me is a decent balance for pub play - sufficient to cater for larger maps or meatfests, without just becoming ridiculous.[/QUOTE]

Totally and completely agree with this (don’t faint monthepoines!)

You can’t argue for or against server size based on “fun”, as that’s purely subjective. I find big servers are just mindless cluster-****s and always tend to filter them out. Even when I played BF3 I looked for 16 slot infantry only.

In DB I doubt I’d be interested in anything above 8-v-8, maybe 10-v-10 at a stretch, I’d certainly rather hammer rusty nails into my eyes than play on anything more than 24 players.

edit: Not sure why this thread showed up when I clicked “view new posts”, maybe I really am stalking montheponies !!

(montheponies) #18

i’m equal parts shocked and worried :wink: