Recent PCgamer seems to have a special about Valve's next project: An online Zombie game with the developers from Turtle Rock! mmmm :biggrin:

http://hl2world.com/bbs/1-vt46502.html?postdays=0&postorder=asc
http://www.halflife2.net/forums/showthread.php?s=&threadid=116411

---

Why snark works.

You had me until Turtle Rock.

---

Blame it on Microsoft, God does.

You know, I first left it out but then felt like being honest.
But really, this looks like an awesome game. Valve seems to be invovled enough into this to add enough atmosphere and smartness. The idea alone sounds like it's going to be a great game.

---

Why snark works.

Been looking forward to a good zombie FPS for quite some while. Had a look at those screens and it looks promising. Hope they manage to catch a nice Romero/Resident Evil (the games!) feeling.

---

What the Snarkpitters listen to!

Dear Jesus, I hope they release this, like, right now. This is exactly the kind of HL2 game I'd want to map for! :biggrin:

ps: I lol'ed at one of the Valve members comments about how "Co-op is hard to implement."

---

Breaking the laws of mapping since 2003 and doing a damn fine job at it

That's funny how?

---

-- midkay

Because it's not hard.

---

Breaking the laws of mapping since 2003 and doing a damn fine job at it

How would you go about implementing it?

---

-- midkay

:insert rollseyesorgysmiely here:

This isn't complicated. The most primitive form of "co-op" is simply by having 6x the number of monsters you normally would (a,la, doom style.)

---

Breaking the laws of mapping since 2003 and doing a damn fine job at it

"Co-op is actually a really tough problem," says Valve's Erik Johnson, who is also there for the moment of unveiling. "Everything can be scripted in a singleplayer game because you have some idea of what one player will do. Having two or more players changes all that."

He doesn't say "it's hard", he explains why it's a problem. Sounds like a legitimate concern.. have you got a simple, definitive solution to this problem?

---

-- midkay

Thinking in consideration for more than one player sounds like a good place to start. Once again: It's really not hard to script sequences with multiple players in mind, you just have to make sure that the players can "move through" the map even after a sequence has executed. In a normal Single Player sequence, something might happen which would block off a previous path -- this for instance, you generally wouldn't do, unless you want to give players the ability to teleport around the "sequence result". Also, there's really no formula to any of this.

P.s.: He does actually say, "It's hard." The article I'm looking at directly says the following:

Co-op is actually a really tough problem.

source

---

Breaking the laws of mapping since 2003 and doing a damn fine job at it

"A good place to start"? Sounds a little harder in reality than you're making it out to be. It's a design challenge; like as you suggested, players can't ever become blocked off from each other.

So your "easy solution" is to simply toss, among other things, all path-blocking sequences? Man, you're a genius.

---

-- midkay

Uh, no. The point I'm trying to make is that there really isn't anything head-scratchingly-diffcult about making an FPS co-operative. Doom has done it, Sven-Coop has done it, Doom3 mods have done, Serious Sam has done it, -- we're not reinventing the wheel here.

---

Breaking the laws of mapping since 2003 and doing a damn fine job at it

I'd scratch my head to seriously think about it - why not?

You have a pretty hard design challenge - you have to reconsider every scripted event you'd find in single player. And no, skipping over any major scripted events is not really an option. It takes effort to come up with some ways around these restrictions - keeping players together, not leaving out any important points for guys falling behind...

I haven't played sven co-op (though that's a pretty darn elaborate mod, I wouldn't be surprised if their years of work has managed a good HL co-op mod). I've played a Doom3 co-op mod, honestly, my impression was that it rather sucked... it was way below-par - certainly something you'd never find on the official disc.

Bottom line is that you can easily write up some code for players to be in the same level together. Making that level actually work around the fact that you have someone else with you - maybe several more people - is a lot harder and more particular than simply designing levels, tossing in several player spawn points and reworking any major scripted scenes.

---

-- midkay

Ugh, arguing is turning my stomach. I'm going to leave a few thoughts floating in the air, and I'll come check back tomorrow:

you have to reconsider every scripted event you'd find in single player.

Wait -- reconsider what? I thought this game was co-op from the git-go, not single player. Based on my thought pattern, one shouldn't be designing the levels in a single-player sense to begin with.

And no, skipping over any major scripted events is not really an option.

I completely agree with that. To clarify, my original thought on this related to keeping the players together, and insuring that no one was stranded behind due to a sequence (which you also mentioned in your post 1 sentence later.)

Bottom line is that you can easily write up some code for players to be in the same level together. Making that level actually work around the fact that you have someone else with you - maybe several more people - is a lot harder and more particular than simply designing levels, tossing in several player spawn points and reworking any major scripted scenes.

That's only the nuts and bolts you mentioned. Most co-op stuff uses the same exact entities -- probably give or take some natural exceptions and changes, which are purely related to multiplayer -- that a single player map would use anyway. Entities are what make the map, and the body of the content of the game for that matter, what it is.

To put simply, I'm having trouble understanding why the concept of creating something co-operative is so difficult. I play co-op stuff on a daily basis, and like I said earlier: There's really nothing too hard about it.

---

Breaking the laws of mapping since 2003 and doing a damn fine job at it

Wait -- reconsider what? I thought this game was co-op from the git-go, not single player. Based on my thought pattern, one shouldn't be designing the levels in a single-player sense to begin with.

That's not really what I meant - while there is the possibility for only a single player on a team or so, you just have to stray from the "standard" set of scripted sequences and really be careful about making sure that absolutely everyone is together when something even remotely large happens. Just thinking of the first level of HL2, you'll have many problems with co-op:

• Stepping through a gate before being sealed in, photographed and met by Barney as a metrocop - what sort of way will you get all the players within this fenced area before the fence seals itself? You don't want players to really leave the area, but you need them all to be together in there before the gate is locked.

• Into the 'torture room' right afterwards - again, all the players need to be in here before the door shuts permanently, but you don't want players entering and then leaving the room either. It's silly if Barney stands there grinning for two minutes while everybody regroups, then the door magically shuts as the last player finally enters and then Barney starts his speech.

These are just two pretty simple examples from only the first segment of the first chapter. These ones aren't overly complex, but there are some really important sequences later on where it's just not that easy to make sure everybody's moving as a pack and nobody gets "left behind" or left out on some information.

The entities are of course the same - the problem isn't at all with coding. As I said, you can easily toss a bunch of spawn points in a level, but rethinking your entire logical outlook on scripted sequences and having to be absolutely careful not to do anything worth seeing without herding everyone together somehow.

---

-- midkay

Anyway, mapping for a professional coop-game with Valve's blessing sounds fun. They've produced nothing but sequels for the past years. Portal and this look very, very intersting.

---

Why snark works.

Crono said:
You had me until Turtle Rock.

What's wrong with TurtleRock? The Source remakes of CS maps they've released are top-notch IMHO (with the exception of inferno's reworked layout), and several of the CZ-only maps they made are still some of my favorite maps.

---

o

I didn't like the remakes. Most of them felt completely different from the originals. That doesn't need to be a bad thing but it wasn't my taste. But again, this preview looks fantastic beyond taste.

---

Why snark works.

I'm in. I'm assuming it uses hammer, so we won't have to
learn a completely new mapping program?

I'd say that's a fairly safe assumption.

---

o

sounds really suspiciously like Contagion-Theory

did valve rip off somebody or is is just a coincidence?

There are even a few zombie mods for HL1/2 that, of course, are very similar. It's hard to keep track. I don't like to see patents for "kill zombie" or "control AI characters", though. :biggrin:

---

Why snark works.