Name: Memento_Mori
Location: Zurich
Posts: 5115

A lot of people consider CA a broken/inferior mode. People playing CA competitively tend to disagree.

Although I’m far from being a top quaker, I’ve played different modes at an average competitive level. I used to compete at ra3 during 2003-2005, playing Clanbase div2 with DSQ. With QL I started playing Duel, and spectating quite a lot of the more “serious” modes.

My goal here is to discuss how CA works and what I think are its major problems. While in a game design discussion some things are a matter of taste, like for instance whether a game should involve computing numbers while under pressure or not, other things like having a consistent design are more objective arguments. I’ll try to focus on the latter.

The flaws of Competitive Clan Arena

In a nutshell, CA gameplay resolves into meeting up with your team in a convenient location and use numerical, positional, and personal-skill advantages to kill the other team.

This concept is in many ways similar to CS. That game may have bombs, hostages, a cash system, but still it is worth drawing a few parallels to it. One may find CS boring, but there has been a lot of polishing and refinement over the years to force on to admit that, to the least, in terms of creating a competitive game, that’s a fair product.

To me, the real flaws of CA are the (1.) poor spawn system, in conjunction with (2.) poor map design, i.e. maps that either were not designed with that gameplay in mind, or were simple poorly made.

CS has a clear map design. It solves most design difficulties by having a fixed paradigm. The two teams spawn at predefined locations at the edges of the map with two bombs locations more or less at equidistance. A map designer in CS can start by that, and then polish the various routes, trying to make maps less biased in terms of weapon balance, and as fair as possible for both teams.

In contrast, in Quake no effort was made to design a proper spawn system for CA. Rocket Arena was initially a fun mode to rocket jump in the air a try to hit mid-air crazy shots. There was no real design in mind, although people realized it could be fun playing it in teams, trying a variety of maps. These maps would have just standard spawn points, distributed evenly around the rooms and halways, and the spawn system would just be random. That was fine at the time. When people started to try and play more competitively, the game design was set in stone, and people just assumed that randomness would solve any unfairness issue.

Having seen many different games, one should try to distinguish between real design choices and excuses. Since there are no items or flags, the only thing in CA that makes a position important is its advantage in terms of fighting with your team.

Ignoring the general concerns of what creates advantage (true in any mode of Quake), there are CA specific questions that should immediately arise: how many of such points should be there in a map? How far should they be from the spawn points? How do we make it fair for both teams to reach these points? Etc. These are all important questions, and it is definitely clear, from looking at the maps (especially at the ra3 time), that a very poor job was done in answering those questions.

Competitive Ra3 has suffered immensely from this lack of design. It had a few decent maps (If I had to call out a few, ra3map7, ra3map9, ra3map3) but the map pools tended to be overreaching, resulting in teams picking the weirdest of the maps (ra3map4?!?) as their home, or compensating the ping/skill disadvantage taking cluster-fucks map (like ra3map2 and in a minor extent ra3map8) which would end in teams camping in adjacent rooms, depleting nades and rockets until someone finally managed to bully in. If you add the immense importance of ping, and how this would affect server search, the competitive side of ca, a setting where a team would do everything to win, tended to become super time consuming, frustrating and boring.

Over the years, CA evolved with sets of unwritten rules. In public, i.e. a non-competitive setting, people would start to play in a specific way, and even kick/ban from the servers (especially in Germany) those who would not follow. The most obvious case was being forced to meet at courtyard in ra3map11. People realized that by enforcing the rule of having a single meet up point, followed by carnage, the game would be a hell of a fun. Matches would be quick, action packed, and provided that adrenaline rush CA players learned to enjoy so much.

What was never written crystal clear, but that to me looks like a viable map design description is the following: Let’s make CA be centered in a single big room, with one strong and highly contested spot, and a few peripheral weaker spots directly connected for the losing team to fall back.

This would possibly be the case for the courtyard of Overkill, with the problem that that maps has large areas with marginal to no role, except for the fact they screw up the spawn system with a few players always being late to the party, or they allow +backers to make useless rail runs that make everybody in the server waste precious time. Recent maps, like Trinity and Asylum implement the concept much better, the former being arguably a bit too big, and the latter - in my opinion – being the most convincing CA map design-wise.


I think one reason why competitive CA never took off is because of lack of proper design. I think this could be fixed by being more explicit in the types of map that should be played, and by fixing the spawn system, which to date is just wrongly conceived.

To be more specific, closed rooms are a big problem and should be avoided; they generate deadlocks and prevent a good flow of players. The spawn system should be either changed to grouped team spawns, or maps should not be too wide and have good connections to make sure a match is not decided by the random seed.

Finally, I wish people involved with ra3 had more guts to try more things. There is a great range of design decision that could improve CA. Just as random comments, one could think of using short time-limits with a proper way of managing which team scores the point, so that people feel the urge to act. I cannot recall the name of that mod which had a circle appearing in the ground and killing you if you did not get out of it soon enough, but it was pretty damn fun, although possibly improper for competitive play.

In general, what makes CA good is the fact that Quake, even stripped off of additional strategic components based on items or flags, is still a fucking solid game, with awesome weapons and physics. I just think there could be more effort in designing game modes by conceiving interesting dynamics, and not by hacking together bits and pieces, as it was done with CA.

To conclude on a side note – not really related to competitive but to public CA - I’m very excited about the introduction of Red Rover. This is a mode that does exactly what it is supposed to do.. It provides an action paced Quake experience with all weapons and armors, it’s absolutely robust to whatever map, whatever spawn system, and whatever mix of people is on the server, and does not pretend to be serious in anyway. We will see what happens with that in QL.