September 14, 2005

SWAT4: the best coop in 2005... so far

This year has not been -like- really rich in coop gaming. EXCEPT for Swat4. If you want hardcore, dead serious, coop gaming SWAT4 is there for you. The tension Swat4 puits in the players is awesome and the need for closeknit cooperation is a matter of life and death.

The game came with a 5 players limit originally, but people on the net figured out how to go round that and took the limit to 10, which was later acknowledged by the creators and inserted as a feature in a patch.

Unluckilly, even though the original maps were really good (and highly realistic, which i really enjoy) the community was far from the Joint Operations enthusiasm, and the game did not create such a flurry of new maps coming out.

Even then, any person that really wants to try a first class coop game should give SWAT4 a try. But PLEASE, complete the training tutorial before running and gunning. See, an additional twist in this game is that you are actually rewarded for catching the bad guys. And catching is not the same as killing. So you will learn to find and use the alternatives to brute force. A VERY nice change from all the slaughtering of the usual games.

Why is SWAT4 so good: The AI is also very robust. Bad guys run away, hunt you down, feint to drop their weapons... they're BAD, in a very very good way. For this alone, SWAT4 deserves a whole lot of credit. If you've read my blog carefully, you will have seen that SWAT3 is in my all time best of coop gaming hall, and SWAT4 was true to that inheritage.

Enjoy.

Best Coop of 2004

OK, in case you haven't noticed yet, i play mostly (almost exclusively, because i like them best, not that i am not willing to play other games) First Person Shooters (FPS).

Anyway the best coop action i had in 2004 was Joint Operations (from Novalogic) by far. The game's realism was great and the action was a lot of fun, the learning curve for my newbie friends just right. All in all Joint Operations (and Escalation, its' expansion) was by far and wide the best coop game of 2004.

I storngly recommend it. One of the interesting aspects of Joint Operations was that it allowed big groups to work cooperatively: squads of 20 players were not unheard of, which is unique as far as i know. And Joint Operations delivered a field of action that was big enough for everybody to have something to shoot at. Rare indeed.

Another fantastic feature of Joint Operations was the community: there are litterally TONS of maps out there to play (missions if you prefer) specifically for coop gaming. So many, in fact, that an -almost- addict like me have not the opportunity to play them all!

Just one site to check out if you want maps: www.dfbarracks.com The maps are rated and commented, and the creators are availbable for comments, and even betterments if you ask nicely...

Combined together, these factors made Joint Operations a worthwhile coop gaming year for me. Nope, i'm not exagerating.

July 03, 2004

The dirty secret of coop: the players

Today, i need to adress another issue that has been a strong deterrent for the appearance of a strong trend of coop gaming: The gamers.

I don't have enough worldwide experience with gaming to make any final judgement, and i do acknowledge readilly that making strong universal statements is normally pointless and too often misguided, but i'd like to mention that players can have hard times adjusting to the coop dynamics.

I'll be back with more on that.

June 28, 2004

network? could it be the networks? what about tools?

OK, this time i might give the publishers a break: The networks are for effective coop not the yet.
I have to agree with that. Living where i live, (in a country far away from any good pings and with lees bandwidth than a shoestring could hold) i have to accept that cooperating in an intensive game is, practically speaking, impossible. At the very least, not fun.

The networks aren't there to support good games that are usually dependant on good networks. Example: living in a Latin American country, except Chile, of course, all you get when you want to join in is LAG. Bad lag. And that's just to get together in the game.

BUT, there is a BUT, programmers can be pretty bad at netcoding: the programming that goes into the game to make it work on the network. Some games would not even run on LANs! Granted this doesn't happen anymore... or are those just memory leaks? (We’ll get on bugs' case at another opportunity). I will not go into detailing who did what wrong in which games, but players have suffered enough from this.

Now, coop requires more than just "being together" on the game field/map: It needs a lot of communication between players: people must give each other orders, make requests, report valuable information, get their action plans together. And this is only getting more and more complex: the virtual battlefields are increasing in sizes and possibilities: in some games (Battlefield 1942) you can have 2, 3, 4 or even more immediate goals to conquer/protect. And in Joint Operations, you can do that with very different vehicles, by land, air, sea... And effectiveness in the assaults or the defense can only be obtained by combining weapons and delivery of the assailants...

What we need, then, are games/tools that will support the enhanced requirements for active cooperation which includes, but is not limited to:
* typing stuff to each other (too cumbersome to be practical, but hey, if that's all you got...)
* pointing out stuff on maps that everybody can look at,
* Macros (special keys) that allow transmitting key/stereotypical information.
* talking to each other/being able to signal stuff vocally to other players,
* video feeding: showing stuff to other players in-game, such as what i currently see to others while the action is going on, sci-fi and fantasy settings allow easily for that possibility, and some well explained tech trick would do the same for most of the 20th Century, at least with a CCTV). At least transmit screenshots at quick speeds to allow good decision making.
* silenced movements that indicates stuff. (Commando stuff you see on most movies with those characters in them).

All these methods of communicating -linking people together- demand added bandwidth: It’s not just about the game anymore; it is an extra demand on the network to let that web of communication work on top of it. Therefore, we need networks and tools. In a LAN, all that stuff is usually possible, due to great network stability, physical closeness, and the good old “shout it out loud enough and people will hear you scream” backup methods. But to get to a good network cooperative experience, we will also need these other tools, and they have to be fully integrated to the game.

Of course, these tools have been integrated into different games to different extents and with different successes, and some other guys have actually created tool to supplement the need for them: Teamspeak being one of them.

My request to the game designers: get those tools aboard, and make them work with as small a netcode as you can. Taking bandwidth is not an option at this stage in the history of bandwidth, except for testing on LANs. But SOON, when the guys that wrote “Infinite bandwidth” turn out to be right, we can actually put the to “good use”.

April 24, 2004

Reason #8: It’s too hard to get good AI

he truth of the matter, it seems to me, is that programmers just don’t have the skills or the budget to get good AI. It is apparently a lot simpler to get a long series of scripted events right than getting individual and group behaviours set up so that they can act relatively independently from their environment, adapting to it and including it in their actions.

I must say that it is indeed a serious limitation.

However, bots have been around at least since Quake 2, if I remember correctly…
And I am positive that the bots from SWAT 3 know what they were doing (those bastards!). If anyone has tried Far Cry they also know that bots can be bad (in a good way) and do some pretty mean stuff, including flushing you out with a grenade, and flanking you, or finding back ways to get to you. That is the right stuff. In Far Cry, they go as far as shouting orders at each other, that makes it really real. (Even though, now that I think of it, good soldiers would use some sign language and trained special ops would use subvocal radios probably).

A complaint you see frequently on modders’ sites is that the games are not designed to support bots (MOHAA) and that sometimes getting the bots programmed is just too hard for the non-dedicated programmer (Hello Battlefield 1942!). That’s a shame, because it’s all about the replayability. And bots that go only one way are bound to be easily shot… boring.

By the way, bots with varying levels of intelligence and skills is a good idea (congrats on UT 2004)

To sum it up, get working on the AI, I can wait for graphics, now that we have reached Far Cry, but I want to see more and way better bots in action. And giving them perfect shooting capability and seeing through walls is NOT FUN.

Reason #9 for not including coop in a game: No one plays coop

No one plays coop, nor want to play coop.

That seems to be the excuse of a lot of publishers. Egg and hen problem? You bet. Most of my friends WANT coop, but can’t it on their games. Call of Duty coop? Medal of Honor coop? I could go on and on…

What we get from a lot of publishers is the possibility of getting coop when we play multiplayers versus computer. BUT THAT AIN’T REAL COOP, it’s just another match versus bots. No great goal is achieved. (It’s fun nonetheless and, if you’re not gonna do anything else about it, it’s a great beginning).

But my thoughts are towards something more than that: Being able to advance on a storyline together with my pals. And that’s hard to find. Duke Nukem coop? That worked: you could go through the entire game, all the levels fighting you way together against the aliens. That was good stuff. Even Doom 2 had coop.

If you look at Sven’s coop for Half-Life, or the Coopmap maker’s sites, you’ll see that coop is not only a possibility, it’s a requirement of a growing amount of players. And it makes so much sense: you want to play with friends before you actually go competitive.

Maybe we could get a series of pools started around the biggest gaming sites asking around What is best?
 Deathmatch,
 Team deathmacth
 Coop
I am positive the answers would not be as obvious as too many publishers like to think…

February 18, 2004

reason number 10 for not including cooperative gameplay in games

"It didn't occur to us". That seems to be an altogether straightforward explanation, and from some of my readings in forums and developer’s journals, it seems to be a fairly good explanation.

Let us go deeper into the human motivations and try to dig explanations for the phenomena:

Developers are not really familiar with the concept of cooperation. Given my experience on working close to software developers and engineers in general, I have a feeling that their training includes focusing on 1 problem at a time, and strictly avoiding social contact on all occasions (flame mail for this opinion will be warmly received). Programming, the way we do it these days anyway, seems to be a very lonely affair. I have not seen big companies do their programming, but I suspect that the pattern holds true: every programmer is king of his hill/island, and no-one can take that away from him. The result of this approach is Sid Meir’s example to the gaming industry a great game-maker that is totally incapable of thinking of multiple users.

Speaking of Sid Meirs, let me recall one of my frustrated cooperative experiences: When Alpha Centauri came out, I was visiting my father in-law. A programmer himself, he had greatly enjoyed Civ. I am not to hot on abstract strategy myself, but I thought it was a good way to link up and do stuff together. So, when I got the information about Alpha Centauri being a multiplayer game, I rushed and went for that game, set up the network between a couple of PII (I think) and convinced him to play. Of course, it was turn based, and you know that is. Anyway, we advanced more and more into the game, until it was time to start launching our armies and crush the rest of the factions of the planet (the bad guys). Because I am such an advocate for cooperative games, my father in law and me coordinated our attacks as much as we could, and sure enough we were crushing all those that dared to be on our way. Until the game started misbehaving… Soon enough we couldn’t get back into the game… I was confused, and could not understand what was going on, so I started searching for information, till I tripped on the mine that was waiting for me: Critics were saying that everyone had problems with the multiplayer at advanced stages of the game. CRAP!

The point is: Some programmers can’t think multiplayer. They can think of 1 person playing versus hordes of computer animated enemies and that’s about that.

On the other hand, we have programmers that certainly can: See Quake Arena, or Unreal Tournament made specifically for multiplayer mayhem. Those are the good game developers. And they do realize that too often the players for a 16-32 gamers match just aren’t there. What to do then? GET THE BOTS IN THERE IDIOTS!

Just today playing Black Hawk Down: Team Sabre, I am stuck in several spots in the game thinking: Imagine this bit played cooperatively, it would be greatness! But as I mentioned elsewhere in this blog: Novalogic just can’t get enough AI in its games.

Let me tell you though: muscle can sometime do the job of intelligence, even if it is not as rewarding: successive waves of enemies can do the trick for basic cooperation. Facing enemies coming from all sides does take coordination and gives a very good reason for tight cooperation. And it’s so movie like!


In synthesis, part of the problem or reasons for not including cooperative gameplay into the actual games seems to be that the developers themselves don’t think about it. It doesn’t cross their minds, because programmers have an exclusive relationship with their computers that prevents them from looking at the socializing that can be favored with the software they create.


February 11, 2004

Why won't game creators always include a coop mode?

Now that is a good question. Let's think of the top 10 reasons...

January 16, 2004

What is good cooperative gaming?

Enrolling newbies: First of all, the games have to be engaging and the basic ropes of roles should be very simple to learn, in order to engage new players. This is required so that the newbies get up to speed, and become functional for the team easily. Certain roles like being a grunt in a FPS or a grunt producer in RTS, should therefore be accessible. Getting into the details of carrying out bombing runs or producing advanced troops can be saved for later, as they become familiar with the game and actually DESIRE to increase their mastery of the game.

Be objective: Second of all: the game should provide clear-cut objective that are highly visible to all players. In SWAT 3, you had to kill every single bad guy and save the hostages. That was clear to everyone from the beginning to the end, and the tension was incredible.

Provide the means for cooperation: Some game just sit and demand cooperation. Castle Wolfenstein: Enemy Territory seems to be one of them, because of an aggressive class system that differentiates clearly each one’s function, and because of an aggressive map designing that demands use of the classes (a team without an engineer is as good as dead usually, and being undercover spy for the other army is just fun). Often, some situations are what the doctor prescribed: in SWAT 3, very often the realistic environments offered so many vulnerabilities for the team that was entering the location that we just HAD to cover every angle and hope to press the trigger at the right moment to protect the rest. Achieving that coordination becomes a goal, and is something that will be remembered after the game.

Motives for playing cooperatively:
1) Lack of participants: In my case it was very often because we were not enough to get a team deatchmatch started (just 3 or 4 of us).
2) The desire to work together: playing with my brothers and friends together in a team made so much more sense than shooting each other. [OK OK, from time to time, a little deathmatch is also enjoyable.] At some point finding out who is the fastest-baddest-greatest shooter is just not something desirable. It is being together that is attractive.
3) Sheer cooperating pleasure: The human being is naturally sociable, and wants to work together not just to ensure survival, but also to achieve something higher/bigger than himself. Working together and reaching common goals is naturally enjoyed.

Enemies: The enemy should be common: everyone attack! Sometimes situations happen where the enemy of one is not the enemy of the other (in RTS which have some form of diplomacy integrated), but workouts can often be made, such as delivering resources, or blocking the path. Computer animated bots, with good AI are a must here. [A good bad example is the Delta Force series, while I enjoyed the solo play, there was a cooperative mode that we tried to play with my team, and the result was unequivocally the same: a rush to see who got to kill most of the stupid enemies. Novalogic has a deserved terrible fame in creating AI, which is a shame because the rest of their engines –in my view- was very good.] Sometimes, when enough players are around, 2 teams with conflicting goals (i.e. terrorists-counter terrorists) can keep the cooperation flowing… What went wrong with Counter Strike?

AI: So AI becomes very important in cooperative games, as a team of players can get really creative and spread out on the map to go achieving their goals. Can the CPU react to that? Can the bots respond to different situations like throwing a grenade to a group that is walking in a tight formation, or snipe each individual team member on a loose formation? The AI is a big requirement, because otherwise you fall on using other restrictions to control the situations such as a restrictive map design that allows only a player at a time through a corridor. True, it could happen in reality, but reality usually also offers so many more possibilities. The deployment of a SWAT team would normally include at least 3 or 8 positions: 1 or 2 sniper positions with good views to the crime scene, 1 central command station, and 1 to 3 teams going in. And in order to keep the game challenging, the AI has to be able to counter the sometimes massive firepower of a coordinated team: 5 machine guns can do a lot of damage, and bringing them down will in turn require coordinated action from the AI. What i mean is: good AI is challenging and keeps the gaming team on their toes.
Now to the latest of my cooperative gaming experience: Battlefield 1942 (and specially Desert Combat):

Desert combat is a great fun game. It does most everything that Operation Flashpoint made before, but in a simple straightforward arcadish style that can get people hooked really fast. (Operation Flashpoint had an uneasy learning curve for those that were not self-motivated). And it comes with a native cooperative mode that genuinely fits the game. (Add inserted here: Why are the helicopters so damn hard to fly? They were so much better in Operation Flashpoint)

Of course, getting the game to work on a LAN can be difficult with certain "versions" of the game.

But, if you do manage to get the right patch-upgrades-add-ons-modifications-levels combination, you are in for a treat of a game.
Cooperation is required and soon becomes very focused: each one finds the weapon-vehicule combination that works best for them, and then can easily see what need to be done in the map: which flags need capturing-defending or where to drive the offensive. (Real battlefields are probably not as neat in organizing the tasks and goals, but who cares?) Therefore we achieve something beyond cooperation: self-organizing or "autopoiesis" as Maturana would put it. This is very good.
Of course, i should talk about RTS. Real Time Strategy games are usually pretty functional for providing some good cooperative gaming. The reason is simple: they almost always have AI ennemies available, and the most recent games offer even to set some characteristics of the AI (such as rusher behaviour etc.) and they allow for players to work together in a number of ways.

In Starcraft for instance, there was a mode where allied plyaers actually begun at the exact same spot and had to work together to make their cities grow, including the possibility of controlling each others' armies. This opened the way for some uneasy situations, but most of the time, on a LAN cooperative players can agree on what the plan is and what to do. That was very effective, and demanded a good deal of cooperation, even if, at the beginning of the game some players had to patiently wait for the other(s) to be set so that he in turn could get his race going.

Age of empires 2: Age of Kings was similarly fun, with the difference that the difficulty could be set (starcraft was set on "mean" and you couldn't do anything about it; Warcraft 3 suffers from the same problem that can really be nasty on new players that don't want to go through the single player campaigns).

Currantly, i am playing Rise of Nations, and here appears a new philosophy: the rules to win can be totally customized: for instance, you can decide that the game should be won totally based on economics results. Since my wife is on the peaceful side of games, she can now enjoy the game without having to wage war.

RTS, generallly speaking, are therefore pretty good for cooperation. Specially for the part where you can exchange ressources and send your army to protect your allies. I like also the idea of armies attacking from different angles a single ennemy... death by crushing is most likely.
So, SWAT 3 was a successful experience for cooperative gaming. Another came up on, curiously engouh, on the Xbox: Halo. From what i've read on the net lots of people had the same feeling: cooperation mode on that game made a lot of sense. [Except for having the share the screen, thank you very much.] Then the game comes on PC, and i am expecting to finally get back with my friends on a LAN and have some shared fun, but the developers decided against that. How could i begin to express my feelings regarding that decision??

Another successful experience was definitely Operation Flashpoint. That was a game that was just plain good. And this time around i got to play it on the internet (braodband finally got here). It was limited by the fact that many cooperative maps did not allow for re-spwan: Once you were dead, you were out. Some may advocate that made everyone serious and comitted to the ongoing action, but i think that it was just lame: it takes too much time to get together to be kicked out because of a hidden sniper with an aimbot. Anyway the game was a lot of fun because of real possibilities for cooperation: the tanks required several players to be fully functional, and combining weapons during an assault invariably brings some satisfying results. That game was good single player, multiplayer and could get just sweet in cooperative mode.
I really wish i could play fun cooperative games. As far as my experience goes, the best of the best i played in that sense was SWAT 3 Elite Edition.

That (version of the) game came with coop as one of the options, and it worked just fine. Add to that the fact that the computer animated bots (the terrorists/punks/bad persons) were quite credible in their behaviours and good graphics (specially when played on a screen projector: it looked like you weer in an action movie) and it was a game that me and a couple of friends played for nights and nights.