(Dev Really Wants Info) State Of Multiplayer for 1.0

[keith.lamothe]

Yeah, that is one potential problem with the current model: the risk of total party kills when one character dies.  I'm not sure that can be resolved without opening up too many exploits, though.

Everything descending into chaos because somebody got careless is not a problem, it's a feature   If it does wind up as a problem then probably the simplest approach is removing vengeful ghost spawns altogether; it's just an orphan mechanic left over now that the rest of the whole structure that contained it is gone.  But it's kinda fun, at least from my perspective

There may also simply be a matter of perception here.  If players feel like "why even bother with MP if it has no advantage" thinking just in terms of raw numbers and not the new tactics that open up to them (seriously, healing touch is just ripe for all sorts of crazy stuff, in addition to the aforementioned flanking benefits), then perhaps we need to make at least a token benefit.  Instead of making the enemies 1x the number of players past one, make it 0.95x the number of players past one.  So one player is 1x strength in either scenario, but two players is 1.95x in the new model rather than 2x.

I think that would be helpful for perception, yes.

[x4000]

I don't think of the current vengeful ghosts mechanic as being an orphan feature, actually -- it's one of the defining things about the permadeath of the game, from the perspective of a lot of people we seem to bump into at PAX and such.  It creates risk/reward: you can't just make an easy grist-mill of characters into the overlord's room, or you reap what you sow with a ton of ghosts hanging around there.  That's wicked cool, and integrates so well into the rest of the strategic/tactical decisions of the game.

It's true it's orphaned from all the other features it was originally tied to, but it found a new anchor in my opinion.  But anyway, aside from that one quibble all agreed on the rest.

[TechSY730]

There may also simply be a matter of perception here.  If players feel like "why even bother with MP if it has no advantage" thinking just in terms of raw numbers and not the new tactics that open up to them (seriously, healing touch is just ripe for all sorts of crazy stuff, in addition to the aforementioned flanking benefits), then perhaps we need to make at least a token benefit.  Instead of making the enemies 1x the number of players past one, make it 0.95x the number of players past one.  So one player is 1x strength in either scenario, but two players is 1.95x in the new model rather than 2x. 

I could support that. Make it a sub-linear growth rate, but just barely sub-linear.

[zebramatt]

This is massively tangential but I actually don't agree that New Super Mario Brothers Wii got objectively harder with more people. The model they implemented was an amazing accomplishment: you had to cooperate.

It certainly felt harder with more people at first because whereas in single player (or with two very experienced players) you could rely on instinct and muscle memory, in multiplayer you had to communicate in order to achieve the same ends. I found if you played with a group you knew really well, you'd fall into a familiarity which was an extension of your friendship. You learned not just to all play the game in your own way, but actually a new style of playing together. Even I, the best Mario player I know well, had to re-learn my game to incorporate the notion of working as a team. In fact, it was arguably even harder for me to completely relearn how to play effectively.

But once you'd gotten into a rhythm with the people you played with, it became a joy - bouncing off of each other, picking each other up, grabbing a tricky shiny and ducking into a bubble, and just generally helping one another out - the likes of which I've never before or since experienced. And then it was not just easier than single player - because, let's face it, you relinquish as much control over your destiny as the number of people you're playing with - but actually more fun.

The time I spent on that game cooperatively will stay with me forever. Hats off to Nintendo for that. It was a remarkable achievement.

[x4000]

zebramatt: I disagree on NSMB Wii requiring cooperation in a good way for players of lower skill, however that didn't stop my wife and I from really enjoying it. Check out Donkey Kong Country Returns on the Wii, if you like platformers and have not played it -- it and NSMB Wii stand side by side as the best 2d platformers on that system in my opinion. DKC supports only two players, but I felt it was much better in that the two players are insubstantial to one another. That's literally the one change is have made to NSMB Wii to call it perfect.

Okay, end tangent.

[zebramatt]

zebramatt: I disagree on NSMB Wii requiring cooperation in a good way for players of lower skill, however that didn't stop my wife and I from really enjoying it. Check out Donkey Kong Country Returns on the Wii, if you like platformers and have not played it -- it and NSMB Wii stand side by side as the best 2d platformers on that system in my opinion. DKC supports only two players, but I felt it was much better in that the two players are insubstantial to one another. That's literally the one change is have made to NSMB Wii to call it perfect.

Okay, end tangent.

It's funny: I adored Donkey Kong Country Returns and played it absolutely to death in both single and coop, but the one thing it was lacking for me was the same sense of help/hinder cooperation which New Super Mario Bros Wii required when playing with a mix of skill levels! The fact that you could bump into one another in the latter was a total game changer for me - I'd argue that's what made it perfect!

Each to their own, I suppose. 

[x4000]

That's why I'm glad there's more than one game designer, and more than one type of game, in the world! Variety is the spice of life, to be sure.

[Wanderer]

And I meant to say in my last post: I'm also quite open to the idea that there may be factors in multiplayer that I simply don't know about that are contributing to an "X factor" making it harder than it otherwise would be just based on the numbers.  We haven't had enough concerted multiplayer testing with large enough groups to say for sure on that, to be honest.  With AI War I was playing the multiplayer weekly, but most of my play here has been solo.

Welp, guess I can discuss what I see and how it seems to work out and you can judge the merit of those observations for yourselves.

I'm quite familiar that issues of perception can cloud issues of reality -- another thing we had to deal with in AI War, and just part of human brain heuristics rather than an individual logical failing or something.

Which is quite possible.  So, without further ado:

Let's take Bob and Joe.  They both do ~200 DPS just getting going and not having gotten their upgrade stones yet.  They decide to go off together and raid some houses.  Hello Mr. Junkyard, we're here for your quickie-shacks.

Happily bouncing along they run into a blue ameoba.  Nothing extravagant.  Bob and Joe lay into it with a vengeance, each one avoiding most of the triple shots, each getting hit once.  So they've each lost 10 HP or whatever.  When the ameoba dies, 5 hp are released, just like normal.  Bob and Joe get a little back.  Because this health return doesn't change with multiple players, and both players are required to attack the piercing enemies, they have a net loss on their health return for the fight.

Now moving along they run into a landspeeder.  This thing is dropping double-strength miasma balls now so you can't pop them before they reach you typically, there just isn't enough screen real-estate.  While you're attacking, a simple lightning esper floats over you and starts whacking away from the other side.  You turn to take on the esper, which is double strength, so both of you have to turn and clock it.  Because of the heavier HP on the miasma balls, where one character could usually turn, pop the miasma, turn, put a few more into the esper, you've both got to be perfect if you're going to pull it off, or you're trapped.

After this nasty little scenario, which isn't that hard to envision, you have gotten your health orbs... which only can heal one of you.

Because of the nature of the current mechanism that is used to up the difficulty when there are multiple people, and since kill rewards stay the same, it becomes the equivalent of a net-loss in health.  Eventually, someone is going to die that wouldn't have if they both ran the area solo.

Add to that the healing spell is a net-0 spell.  Yes, in an emergency it could save someone's life... if you're on the ball and willing to give up your own HP to help them.  Considering one of my recent characters was taking on the overlord with 150 hp, that's not necessarily a good option all the time.  Sure, nice option, but it's not even in my primary bar, as it feels a bit silly to kill myself to help my partner, and if I'm running solo all it will do if I accidentally hit it is lower my hp for a time while I wait for the orbs to come back.

There are a few things multiplayers have available.  One, only one of you needs to have platforms going, so the other player can have two spells at their disposal.  Using different spell builds also helps to make sure that you won't run into a mob that's resistant to both of you on first hit.  But even then, there's not a lot of weakness in the game.  Example: Living Fire is not particularly suseptable to water.  This and a few other critters (blue/red ameobas, for example) seems odd to me, but that's a tangent.  How it applies here though is that in most cases your different spells won't be that impactful UNLESS you run into a resistant creature.

Another thing that they can do, as Keith pointed out we did against the OL, is divide a particular mob's attention, or their firepower.  This is of course very helpful, but in general you're running linearly through a cave/mazeroom/surfacechunk/whatever.  You're not, one hopes, spending a large amount of time where this is a critical concept.  Particularly in cases of our landspeeder above.  If one of you hangs out on the 'approach' side, and another doublejumps and storm-dashes over the landspeeder's head to 'split' fire... that second person has basically left their butt in the wind against 2x health enemies with no support while still taking fire from the primary problem.  You hope your buddy on the approach side doesn't get distracted and leave you out to die.

All of these things ARE influential in the gameplay, yes, helping to balance the two problems out, but they come with their own issues which almost balance themselves off.  Add to this if you're playing on an MP server ANYONE can enter your chunk while you're, say, halfway through it, and suddenly everything doubles in health and you've got no support yet.  Hope you weren't low on hp right there.  This is a VERY Pick Up Group kind of game by it's nature.  It doesn't need to be as heavily balanced, in my personal opinion, as an organized team playing AI War with a direct set of goals.  It's more like "Hey, buddy, Mind if I tag along?  I don't have upgrade stones yet and I'm getting whacked!"  "Sure, man."

THWACK THWACK.

"Um, dude?  Yeah, let's not EVER do that again, hm?"

[Terraziel]

As always noting first that I don't play multiplayer, but because options are always good, how about making the per player increase another difficulty setting?

Because as I interpret the problems mentioned by Wanderer that is the easiest way of doing it. my initial reaction to much of it is that if they find the game too hard with X players then they should just turn the combat difficulty down, but obviously that isn't ideal because it would make it too easy whilst not in the same chunk as another player, and wouldn't be taking into account how many players there are.

So I say add another option to the multiplayer difficulty totem, maybe make it an admin command, which controls the increase per player. As a command it could be more granular but on the totem it would be more obvious so people are more likely to change it, that way people can adjust it to suit the groups skill levels.