Five Open Questions about Civilization 5


I typed up these thoughts in June of 2010, right around the time of the E3 interviews surrounding Civ5. I held off on publishing them until now, because I did not want to unfairly influence readers about Civ5 with limited information until the game itself released. With the game long out and clearly an underwhelming result, I decided to go ahead and post these thoughts rather than let them languish forever. They should be rather interesting to look back at now that we know how Civ5 ultimately turned out.

* * * * * * * * * *

Well folks, Civ4 is entering its fifth year on the market, and starting to get long in the tooth. Outside of us few direhards still playing new games of Civ4, most attention has turned to the upcoming arrival of its sequel sometime this Fall. The announcement of Civ5 came as somewhat of a surprise; even though it was logical to expect another game in the series, Firaxis Games (the developer) had been pretty tight-lipped about keeping a wraps on their project. Now that the first round of previews have been released to the public, we have at least some idea of what kind of game we'll be getting when the whole shebang releases later this year. As someone who worked intimately in the development of Civ4 - and who has no involvement in creating Civ5 - I thought it might be fun to provide an outsider's perspective on the info leaked out to the public so far. Since I worked on the last game, I have some idea of the kind of challenges that the current development team must be facing right now. I'm going to list five open questions that I hope are being addressed, so that we can look back at these later and see where I was right and where I was wrong.

I'll be concentrating on gameplay, the meat and potatoes of a turn-based strategy game, so I won't be wasting time on why civ or leader X is or isn't included, what the graphics look like, how many units can "realistically" fit onto one tile, or any of the nonsense that constitutes the majority of the discussion at CivFanatics. Now as such, a lot of this is going to sound negative and critical. But it's not intended as an attack on the Firaxians (who I'm sure know what they are doing); rather, it's the same analytical mindset that I used when doing QA testing for Civ4. Question everything, poke and prod and tweak the gameplay to make sure that there aren't any glaring errors. I can't resist approaching Civ5 the same way, knowing full well the incomplete and limited information available to the public at the moment!

Question #1: How will the new development team function?

This might seem like a silly question, but it's not. The creation of a game is a massive project that takes on the personality of its development team, and in particular the lead designer, who sets the tone for the whole thing. To demonstrate what I mean, let's do a quick recap of the past history of the Civilization series. The original Civilization was of course created by Sid Meier, who was already a veteran game developer with close to ten years of previous hits under his belt. Sid's game design was nothing short of brilliant, combining together the best aspects of Railroad Tycoon and Sim City with classic tabletop strategy games. It was one of those things that no one had seen coming, but which made perfect sense in retrospect, and made you wonder, "why didn't I think of that first?" Anyway, the first Civilization is Sid's game through and through; you can see elements of his past games (especially Railroad Tycoon) all over the place. It's no surprise at all that the game and series proved to be as popular as they did.

Civ2 has Sid Meier's name on the box too, but in reality the game was largely the work of Brian Reynolds, the lead designer. Civ2 did exactly what a sequel is supposed to do, build upon and expand the first game without breaking anything that fans loved about the original, or straying too far from the previous design. Indeed, Civ2 makes tons of improvements over the first game (especially in the realm of combat, which was a single dice roll of strength against strength in Civilization!) but it did very little to change up the gameplay. Instead, the focus was more on the atmosphere and the playing experience, what with the silly live action advisors and the animated wonder movies (hey, that was big stuff at the time!). You can once again see the fingerprint of the lead designer Brian Reynolds at work, as a lot of elements from Civ2 were later pulled over to Sid Meier's Alpha Centauri (which, despite the name, was also designed by Reynolds).

The reason why I bring this up is that there was a pretty serious shakeup of the design team at Firaxis during the development of Civ3. Brian Reynolds left (along with some of his staff) to form Big Huge Games, creator of the Rise of Nations series, leaving Civ3's development somewhat in limbo. Firaxis brought in some very capable new people, most notably AI Programmer Soren Johnson, but this clearly affected the development of the product. (There is no single lead designer credited for Civ3; instead, the game lists "Design by Jeff Briggs, Soren Johnson, and Members of Firaxis Games.") Some of you might remember the initial release of Civ3, which was plagued by many bugs and broken game elements. The development shakeup behind the scenes had resulted in a game that was not ready for release, but was sent out the door by the publisher anyway for the 2001 holiday shopping season. Now to the immense credit of Firaxis, they put out a steady stream of patches for the following *NINE* months that had Civ3 in very solid shape by the Summer of 2002, which earned them a lot of goodwill from the Civ community. Such a marvellous patching effort is pretty much unprecedented in this industry. But the Firaxians knew that Civ3 was not where they wanted it to be, and worked their tails off to get it in good shape. You could just see that the developers themselves didn't have a clear vision for what they wanted the game to be, and were figuring things out as they went along. That's why Civ3, much as I love the game, has always had a bit of a schitzophrenic feel to it, as if the various game mechanics were developed in isolation from one another and didn't necessarily fit together in the end product. I don't think that was an accident - there wasn't a clear director at the helm when the game was being built!

Fortunately Civ4's development was a much smoother process, which is undoubtedly the reason why Civ4 is a much more polished game overall than Civ3, and was in infinitely better shape on its release date. Soren Johnson took the reigns as Lead Designer, and provided the guidance and direction for the whole endeavor. Back in testing, someone made a mockup graphic of the title screen that replaced Sid's name with "Soren Johnson's Civilization 4", and that in truth was not an exaggeration. This was Soren's baby from start to finish, right down to something as minute as the foldout tech poster (which Soren did the initial sketchwork and planning for, and then I did the grunt work of typing up the details). I'd say more on this, but Soren himself wrote up a lengthy 13-page Afterword included in Civ4's manual where he explains at length his vision for designing the game, which focused on eliminating things that were unfun about previous Civilization games while creating new and interesting strategic choices for players. If you've never read it before, and you own Civ4, you really should check it out. It's a great read, and you can even see my name in there too!

So - that brings us back to Civ5. Soren Johnson has sadly departed Firaxis, and now works for Electronic Arts (EA Games). But there are new individuals in charge, many of whom I have the privilege of knowing from my summer spent working on-site with Firaxis. Ed Beach (Lead AI Programmer) and Dorian Newcomb (Lead Artist) both spent a fair amount of time jumping on the internal testing forums during Civ4's development, and they definitely knew their respective fields very well. Of course, most of the attention will rightly be placed on Jon Shafer, who has inherited the mantle of Lead Designer for Civ5. I've known Jon for quite a few years now, initially in his forum guise of Trip (still miss the Crono avatar!), and later had the chance to meet face-to-face at Firaxis a bunch of different times. Jon is a really cool and funny guy, and I have nothing but positive things to say about him on a personal level. Heck, we both have degrees in history from the University of Maryland school system! But on a professional level, as someone who is in position to take over for Soren and create the next iteration of the Civilization series... well, I'm a little less confident about that, to be perfectly honest. Those are some very big shoes to fill, I'm afraid.

Jon's previous big project was working on the development of Civ4: Colonization, the standalone game released by Firaxis back in 2008. This was essentially a remake of the original Colonization game from Microprose, using the Civ4 graphics engine. Opinion on this game is somewhat mixed; the general consensus is that it was fun to play for a while, but didn't have the same depth or lasting value as games like Civ4. The real problem with Colonization was that there was only one way to win the game: you had to build up revolutionary sentiment (symbolized in-game by something called "liberty bells"), declare your independence from the mother country, and then defeat a Royal Expeditionary Force (REF) to win the game. Compare this to the Civilization series, where there are multiple paths to victory that allow for different strategies; even in Civ1, you could win by spaceship or through military conquest. With only one way to win the game in Colonization, players quickly cracked the code and found the easiest path to victory: build up zero liberty bells for most of the game (because, due to a bizarre design decision, the size of the king's REF was directly tied to the number of liberty bells your colony had accumulated), then crash-build revolutionary sentiment in the span of a dozen turns, face a very weak REF, and win the game easily, even on the highest difficulty. What this really showed was that Colonization should have been a scenario, and not a full-fledged game on its own. I'm not sure exactly how much input Jon had on this game's design, but it certainly leaves some room for improvement.

I guess I'm a little bit worried because Jon comes from a modding background, and - perhaps unfairly - I've always been suspicious about putting modders in charge of game development. It seems like a natural fit, as modders are people who create their own scenarios; in theory, they should be the best at new development. However, the problem with most mods is that they tend to be very weak on game balance. That is, mods tend to be more about "adding cool stuff" than making sure that it all works correctly. Game balancing is exceptionally difficult to do, and represents some of the most unfun work of the whole design process. Modding offers somewhat of an easy way out, just changing around whatever the modders happens to dislike rather than understanding the entire complex system. Just to give you an example, trying to balance the civics system in Civ4 took 14 months of rigorous testing, with dozens of different attempted implementations, and I'm still not sure we really got the whole thing right! So when I see Civ4 mods that casually throw around 2 or 3 additional civics columns at the drop of a pin... well, color me suspicious, to say the least.

To summarize a long section, game development takes on the vision of the designer in charge of the process. My hope is that Jon will grow into his new role, and he and his team will be able to avoid some of the mistakes of the past as they go forward with Civ5. I have to be a little worried about a "modding" philosophy guiding Civ5's development though, as it did the expansions to Civ4. (Most people seem to disagree, but the expansions definitely watered down the core gameplay of Civ4 and ultimately did more harm than good.)

Question #2: What effect will the changes to combat have?

This has been probably the most controversial subject so far concerning Civ5. We know from the various previews and interviews released so far that Civ5 will take a radically different approach to combat and unit movement than any other game in the series. The development team has made it one of their goals to kill the "Stack of Doom" phenomenon, in which armies group together on a single tile and move together for protection. In Civ1 and Civ2, you could group units together on a tile, but if the defending unit was killed then ALL of the units would die at once. This was a bit of an overkill solution, which was removed in Civ3 and Civ4. I think it's fair to say that Civ3 was the king of the Stack of Doom tactic, as the AI would beeline simply gigantic collections of units (sometimes 100+ on a single tile) into the player's territory all at once. Civ4 combatted this issue by introducing collateral damage, whereby siege and air units would damage multiple units at the same time when grouped together into a stack. This didn't seem to remove the situation entirely, but it did offer the defender a way to punch back and inflict very heavy damage against grouped foes. Anyone who has played against humans in the Modern or Future eras and seen an invading stack shredded to pieces with artillery and bombers knows that collateral damage was nothing to sneeze at!

Civ5 takes a radical solution to this issue: only one unit may be placed on tiles. Ever. The Stack of Doom has been written out of the game completely. Instead, battles are to be fought and won through unit positioning, on the strange new hexagon-based map tiles, through a combination of melee and ranged units. Yes, units with ranged bombardment are back again, after they were removed from Civ4. (For a good reason, as ranged bombardment units were EXTREMELY overpowered in Civ3. Every Deity-level player from Civ3 days is nodding their head right now.) We've also been told that units will no longer necessarily be killed at the end of battle, suggesting that units will have to be worn down over time through a series of engagements. Furthermore, strategic resources like iron and horses will only enable a set number of units to be trained, so the number of swordsmen or whatever should be fairly limited. There's enough information to see what the developers are going for here: fewer units, battles fought in the open field instead of being concentrated around cities, tactical "chess-like" movements on both sides where unit positioning and correct mixing of melee/ranged pairs lead to victory. Jon Shafer has said that a lot of the inspiration came from the Panzer General series of games, and that indeed seems to be the case.

Now this is all well and good, and I have no doubt that many, maybe even most fans will be highly pleased with the changes. However... I have to be a little suspicious of all this. The thing is, the Civilization games are STRATEGY games, as opposed to tactical war games like the aforementioned Panzer General series. That might sound like a trivial difference, but it's not. Combat has traditionally been kept relatively simple in prior Civilization games, on purpose, so that the strategy elements of the game can remain in the foreground. You might say, why not have a rich tactical system of combat, with a great emphasis on unit deployment? Isn't that a good thing? Not necessarily. The more that a game dials up the tactical elements of combat, the weaker the strategic elements of the game become. Think about it: is it really a good thing if someone with 3 swords and 2 archers can defeat an army of 15 swords through superior use of terrain and movement? I would argue no. In that situation, you're playing a game where strategic elements (what to build, how to develop an economy, tech path to pursue, etc.) have been overshadowed by the tactical elements of unit movement and positioning. This is the flaw that wrecked many empire-building games in the past, such as Master of Magic (where I quickly found ways to defeat armies five times more powerful with broken spell combinations in battle) and the Total War series of games (where the AI is appallingly stupid in battle). These aren't bad games, very fun games in fact, but they can't really be called strategy games any more.

And I actually like tactical war games; I love chess (even though I'm a terrible player), and have a bunch of Advance Wars and Fire Emblem games which are based around the same concepts. Defeating a superior opponent through intelligent tactics is fun to pull off. But... these aren't true strategy games, and the whole model doesn't seem like a good fit for the Civilization series. Quite aside from the issues I already mentioned, combat is very slow in these games - and that's with the focus on nothing BUT combat! I'm not sure that having to micro every single unit each turn in battle is going to be fun, and scaling down the total number of units to alleviate this issue creates as many problems as it solves. I actually think that Civ4's combat system would benefit from less complexity, not more of it. What we've been reading about in the previews sounds like it would be really fun for a couple of games, then start to get tiresome.

I suppose that I'm unsure exactly why such a radical change in gameplay was desirable in the first place. I have no fault with the Stack of Doom, personally; it's one tactical choice out of many options, and by no means the best solution. The Civ3 AI's propensity to clump all its units into one big stack was one of its greatest weaknesses, and could be danced around or exploited in any number of ways. The Master of Orion AI does this too, and any expert player who's experienced the Negative Fleet Bug knows how to work around invincible fleets grouped into one giant mass. Civ4's solution worked well enough for me, adding collateral damage and forcing a choice between grouping units together for safety, or spreading them out to avoid being hit with catapults. Remember, the Stack of Doom was very slow moving, and could only threaten one target at a time. Maybe that doesn't matter against the AI, but it's a different story against human players. Sorry, but I have to question this whole decision. It's a revolution designed to overturn something that wasn't really broken in the first place!

I won't get into the potential red flags raised by the ranged bombardment units, or the problems with tying specific number of unit builds to strategic resources. (Can anyone stop the guy who starts with three iron resources?) And let's assume for the moment that the AI will understand all of this, and be able to play as effectively as a human, and that there will be no ways to exploit combat so that the human player takes disproportionately small losses. (Heh, heh.) Hopefully there are answers to all of these topics already in place. But I remain concerned that Civ5 seems to want to put so much emphasis on the tactical side of combat, and downplay the strategic elements. As one of the previews says,

No longer will the winner be determined by the player who can pump out the most units in the shortest amount of time.

Are we really sure that's a good thing? In a strategy game, shouldn't the empire with the most cities and the most units in the shortest time be the winner?

Question #3: How will city and empire management be performed?

This isn't getting much attention from the community, but the preview articles have suggested some radical new changes to city and empire management which have the potential to overturn many of the basics of how Civilization games are played. Some of these changes sound fun and exciting, while others are throwing up dangerous red flags based on the limited information we have available right now. Let's dive in and go through some of the changes.

First of all, cities will now work hexagon tiles instead of the standard square ones, which completely eliminates the familiar "fat cross" of 21 tiles which has been in existence since the original Civilization. This alone would be a huge change, but Civ5 further changes things up by allowing cities to work tiles that extend out THREE rings from the center, instead of the standard two rings. No one seems to be talking about this, which is extraordinary since city and worker tile management are the basis of all successful Civ play. Let's look at what one of these new cities will look like at full extention:

First Ring (Yellow) = 6 tiles
Second Ring (Blue) = 12 tiles
Third Ring (Green) = 18 tiles

That adds up to a total of 37 tiles, including the center one, much much higher than the 21 tiles that a city could work in previous Civ games. This immediately brings a couple questions to mind. If cities can now grow to vastly larger sizes before filling up all their workable tiles, does that mean that cities will grow faster to compensate? In Civ4, it takes an egregiously long time to grow up beyond size 20; I can only imagine how long it would take to reach size 37! In order for this system to work, and not have players overlapping cities like crazy to avoid wasting tiles that cities will never use, the game's food dynamics will have to be fundamentally different from those used in Civ4. Furthermore, it looks from the previews that Civ5 will use some kind of health and happiness system akin to that employed in Civ4. Not to repeat overmuch here, but is that same system going to work if cities can grow an additional 16 sizes? You'd need another dozen happiness and health resources to have any realistic chance of hitting those top sizes. Don't get me wrong, this is pure speculation here because we've heard little on city management, it's just that I really wonder if everyone has thought this system through on some of the implications of expanding things out to a third ring of tiles.

Next, the way in which cultural borders expand outwards appears to be going through some dramatic revisions. According to the previews, borders will only expand out one tile at a time, and more difficult terrain will take more time to claim. You can pay gold to speed up the process (in some fashion, the details are vague) but cities all seem to have random-shaped borders which look nothing like the classic rings. You can see some of that in the pictures above, and here's an even better picture of two neighboring cities:

This picture, and the one at the very top of this page, together serve as confirmation that city borders will indeed extend out to three tiles away from the center tile. Now as for the process in which tiles get picked for border expansion... ummm, let's just say that I am concerned. All of the previews suggest that city borders grow organically, one tile at a time, with the actual tile being chosen by the game itself. If that's the case - and I dearly hope that it is not - then it makes for some seriously bad mojo. Compare this to the first two Civilization games, where all 21 tiles were always available immediately (there were no borders in Civ1 and Civ2), or to the most recent two games, where amassing ten culture would expand any city's borders from the starting 9 tiles to the full 21. The concept was easy to understand, and a solid game mechanic.

But in Civ5, cities just seem to grab tiles in random fashion, producing bizarre shapes like the ones seen above. Remember, unless this game is totally different from the others, you can only work tiles inside your cultural radius. If your city doesn't grab the tile you want, you can't work it and start benefitting from the tile yield. What if you want to expand into high-food tiles, and the game keeps picking desert and mountains? (See the picture above, with its abundance of plains tiles.) Or what about the opposite, if a city has multiple food bonuses and I want hills or forests for production? I have absolutely zero faith that the game will be able to select tiles on its own in intelligent fashion, knowing full well the idiot governors from every previous game. Look at the picture all the way at the top of this page again; I've never played Civ5, but I guarantee you that the tundra region 3 tiles northeast is a much less desirable one than the grassland forests just outside the borders to the north. Of course the previews also say that you can "speed up" this process by paying gold (maybe even getting to pick which tiles you want?), but that's kind of a poor solution, and I doubt that there will be gold enough to pick each tile that you want. And what if you have a city that's growing like a weed, and simply outrunning the border expansions to the point that there are no spare tiles to work? Outside of a simple cultural model, this could be a real issue. Taking control away from the player is just a bad step in a game like this; I really hope that I'm misinterpreting how this whole system will work.

So if the cultural border system looks something like what I've outlined here, how are players going to respond? Good players will quickly work to eliminate the vagaries of one tile at a time expansion, and/or having to pay gold to do the same thing. We're really getting into speculation territory now, but here's what I suspect would happen. Let's look at that city diagram again:

I'm certain that each city will get its initial 6 tile ring when planted; every city in the screenshots has at least those 6 tiles, and it would be consistent with past Civilization games. So how then to maximize workable tiles and avoid getting screwed over by slow or poor tile expansion? Start cramming in cities close together in classic Infinite City Sprawl (ICS) fashion. Let's place additional cities on tiles B4 and K4 and see what happens:

If we place cities four tiles apart, it opens up the possibility for massive tile sharing between the cities, thanks to the new third-ring extention. The blue tiles that I've highlighted are each sharable by a pair of cities - theoretically a city could share first-ring tiles from the neighboring city even before EITHER of them expanded borders! Only the gray tiles cannot be shared in the above diagram. The benefits seem obvious, as grabbing any of those blue tiles would allow them to be worked by two cities depending on need, essentially getting a 2 for 1 deal. (Note that this can be taken even further by placing cities 4 tiles north and south of the central red dot, which I've avoided showing for purposes of clutter.) And unless the Civ5 gameplay has been radically changed, it's unlikely that most cities will need all 37 of those possible tiles, so packing them in tightly offers a lot of benefits with few losses. Pretty much the way things were in Civ3, unfortunately.

Naturally this is all speculation; there could be gameplay rules in place to prevent these measures. One preview states that there will be "game-specific disadvantages" from placing cities too close together, whatever THAT means! Guess we'll see, but even on Civ4 maps with very high maintenance costs I still found it advantageous to put cities four tiles apart on many occasions for purposes of tile sharing. I have difficulty seeing why that wouldn't still be the case. I suppose that the game could be hard-coded so that cities must be placed at least 4 or 5 tiles apart, but that would probably be even worse in terms of restricting gameplay... Anyway, I dunno about this one. Based on what I've read thus far, this whole system sounds like it heavily promotes an old-school ICS approach, which I truly hope is not the case. And the organic, one tile at a time border system sounds cool in theory, but it'll be nothing but frustration if the player lacks sufficient control over the mechanic. Firaxis, be very careful - do NOT screw this up. There's a reason why no past game has messed too much with basic city mechanics!

Question #4: What will diplomacy and inter-empire relationships look like?

The previews released thus far have concentrated very heavily on diplomacy and the upcoming AI leaders, so I have a lot to say in this section. First of all, each of the gaming websites (and most of the online fanbase) is falling all over themselves based on the promise of better and improved AI in Civ5. The word is that the AI will operate on four different levels, ranging from tactical up through operations, strategy, and grand strategy, each with its own objectives. Opposing civilizations will pool together these four levels of AI control into a superior whole... in theory, anyway. Forgive me for being skeptical, but every single strategy game promises before release that it will deliver an unprecedented new level of AI behavior, and these claims always fall way short of expectations. I actually think that the AI will be better in this game, simply because the AI has improved in every single Civilization game as technology got better, but I doubt there will be any noticeable leap in performance. Give it six weeks from release, and expert players will be running circles around the AI and rolling their eyes at its inept weediness. Just wait and see, I guarantee it.

Next, word is that each AI leader will have distinct personalities based upon rankings from 1 to 10 in different AI behavior "flavors". There will be a base value for each leader in each flavor, which will vary by up to two points in each direction, making for a different experience in each game. I thought that sounded great, until I read this quote about the flavors: "The 25 flavors are grouped into several categories including Wide Strategy, Military preferences, Recon, Naval recon, Naval growth, Expansion, Growth, and Development preferences." Seriously, are you kidding me?! TWENTY-FIVE flavors?!? You think that's enough? I'm no AI programmer, but that just seems ridiculous to score each AI leader in so many different categories. Do we need a flavor for Recon, Naval Recon, Naval Growth, and Growth? Excessive complexity tends to be a sign of feature bloat and poor design.

They've also stated that each leader will have their own unique bonuses, essentially doing away with the civ traits and going over to the Civ: Revolutions system of each civ being completely separate with no overlapping attributes. (There will only be one leader per civ, in a disappointing cutback from Civ4. This was probably due to art assets reasons.) I have to say that I'm worried about balance issues here; it took ages to balance the 8 civ traits in Civ4, and honestly even the finished product wasn't perfect, since Financial is pretty universally regarded as the best. Two more expansions actually made the balance worse - going up to 11 traits and adding the sucky Protective trait shows how difficult the whole process is. Now imagine having 18 civ traits to balance, which is essentially what Civ5 will be doing! Civ: Revolutions had 16 leaders with unique traits, and they were not at all balanced, with Rome crazily overpowered until patches reigned them in, and the final patched solution still has some civs vastly stronger than others. Well, we still know next to nothing about how all this will work, but I don't take any of this as a good omen.

For all that we've heard about new features in Civ5, there's actually a lot of stuff getting cut from the game. The biggest culprit is religion, which is simply gone for unclear reasons. Religion was very well implemented and well received in Civ4, so I have no idea what's going on there. The producers of Civ5 have been claiming in interviews that religion had to go because "it limited the diplomacy that could take place", which demonstrates a very poor understanding of how the religious mechanics worked in Civ4. Yes, shared faith bonuses or penalties were a key aspect of Civ4's diplomacy, but that was hardly a bad thing (often forcing tough choices between going with a self-founded religion or adopting that of a powerful neighbor) and overlooks the way that religion influenced happiness, culture, civics, shrines (monk economy), temples/monasteries/cathedrals, and the Apostolic Palace wonder. Cutting all that out is nothing short of ridiculous, and represents a huge loss. Also gone is Espionage, which was poorly implemented in the Beyond the Sword expansion, but a bit of a puzzling omission nonetheless. A better Espionage model would have seemed like a good addition to Civ5. Finally, we've also been told that there will be no Tech Trading in Civ5. I'm not a fan of rampant tech trading, but I certainly hope there's at least an option to turn tech trading back on. It's hard to believe that trading wouldn't appear in at least some form, after its uber importance in Civ3 and Civ4 Pitboss/Play by Email multiplayer stuff, but that's what the current previews are claiming.

So if there's no religion and no espionage and no tech trading... umm, what exactly are we supposed to be doing diplomatically, again? Instead of these things, we have the introduction of city states. These are AI-controlled individual cities placed on the map at the start of the game, which make no attempt to win the game and never expand outwards. Apparently the diplomacy in Civ5 is supposed to be based around these city-states, with other AI leaders liking you or disliking you based on how you treat the city states. To be perfectly honest on this, however... I just don't get this concept. At all. What's the point of having city states that don't try to win the game and don't bother to expand? Either replace them with an actual AI civilization, or remove them entirely. We don't need to manufacture artificial conflicts around these city states - conflicts will develop on their own between rival empires over the course of the game. Which is more interesting, a war started because Montezuma doesn't like the way I treated some neutral city elsewhere on the map, or a war started because Montezuma is coming to kill me and take my cities? The first one is artificially and gamey, the second one a genuine threat where you'd better get your ass in gear or get dead in a hurry. I don't want to have to worry about what AI leaders think about silly city states - just cut out the bullshit and let me deal directly with them!

City states remind me of the "minor races" from Galactic Civilization and Galactic Civ 2. I never understood the purpose of the minors in those games either; they were basically just there as a crutch for the player to exploit. Sirian summed this up perfectly in a long ago post at Apolyton:

Actually, having the Minors reach resources is a huge crutch for the player. Player can declare on minors with minimal consequences, on timing of his choosing. The Majors are slow to attack the Minors so you can always (in GC2, that I've seen) take away starbases or the home planet from any minor, whenever you are ready. Of course if you wait for them to build Capital buildings first, you get free extra Economic and Manufacturing capitals. The Minors are the biggest lever in the game. Use them to trade techs early and for cash and planets later, plus as "holding companies" on resources until you are ready to build a Constructor for each resource they are holding and go grab them. You need to scout really well first, though, or you may free up a guarded resource in the fog somewhere to hand to a Major -- which is a Major Oops.

Those poor Minors have big "Kick Me" signs on their back. They exist as food to nourish the human player's civ -- or the AIs, if you are too slow about your dinner plans. You can be incredibly evil to the Minors and still be a Good alignment. One of those things about the game. (I shouldn't describe how evil you can really be. These guys are living and breathing doormats!)

Hopefully this whole situation will be handled a bit better in Civ5. But as I said already, I just don't get the purpose of these city states. It feels like they are only present to add scenery and role-playing aspects to the game (one preview even described city states as acting like NPCs!) However, these elements don't have any purpose in a pure strategy game like the Civilization series, and are essentially fluff that get in the way of the real action between the competing civs. We already have the barbarians to put a check on early game expansion and provide minor harassment - what's the use of these city states? I don't understand.

So apparently the diplomacy in Civ5 will be based around how rival AI leaders "feel" about your treatment of various city states. And you'd better pay attention and take notes while playing, because the AIs aren't going to tell you why they like or dislike you any more! In possibly the most bizarre step backwards I can imagine for diplomacy, Pete Murray served up this quote in a video interview with Gamespot:

"We've removed at lot of the absolute information you had; in Civilization 4, you could see the pluses and minuses next to your interactions with them [AI leaders]. Now you'll have to sort of know a little bit about your leader and how your actions may be affecting him. He may be smiling and friendly on the surface but he might be plotting your demise behind that demeanor." - Pete Murray

Hoo boy, where do we even begin with this one? Civ4 had the strongest diplomacy of any game in the series, because for the first time you could actually see why AI leaders felt the way that they did. In all prior Civ games, the diplomatic system was a total black box with no feedback whatsoever for your actions. What made the AIs feel "Gracious" or "Annoyed" in Civ3, for example? I played that game for hundreds of hours, and I never figured it out. Everything took place under the hood where it was impossible to tell. Civ4 vastly improved this system by letting you know exactly why other leaders did and did not like your civilization. If they disliked your religion, you could see the penalty and make an intelligent decision about what to do. You could see how much value you received from Open Borders, trading resources, favorite civic bonuses, and mutual military struggle. The diplomatic pluses and minuses helped turn the United Nations diplomatic voting from a mystery machine into a scientific system where your decisions could and did make a real difference.

I... just cannot fathom why the design team would take a gigantic step backwards and turn diplomacy into a hidden system again, where you are left guessing what effect your actions have and how the AI leaders really feel about you. It doesn't do anything to improve the gameplay, and the expert players will start cracking the system using mods and cheat sheets to reveal how the whole thing works anyway. Overall, for all the talk about diplomacy, this whole mechanic seems to be going in the wrong direction. There are fewer things to do diplomatically (no tech trading, no religions, etc.) and much less feedback about what's going on. Research pacts are a good idea, but I'm not sure they make up for the exclusion of past diplomatic elements. And no, I don't want to have to babysit relations with a half-dozen "NPC" city states - that is not what the Civilization series should be about!

Question #5: How is Multiplayer going to be handled?

Multiplayer: the great question that no one is talking about in Civ5. The complete lack of information on the MP side of the game is a disturbing sign, and a clear departure from Civ4 where the MP aspects of the game were heavily featured in the early previews (so much so that the traditional SP community complained about them, I recall). The only word on MP so far is that it will exist in Civ5, with the same online options and Play By Email functions intact. There's been no discussion on how the new combat system will be streamlined to accomodate short MP matches, no mention of upgrades or improvements to the poor Gamespy lobby employed by Civ4, nothing on the effects of new diplomacy and empire management on player against player matches. In contrast, everything featured in the current previews are extremely SP-centric, focused on atmosphere and role-playing elements instead of online deathmatches.

It's very possible that this is no problem at all, and simply the case that Firaxis wanted to focus on SP in the first batch of information released. (Those previews are very carefully scripted, for those who are unfamiliar this process; when I worked at Firaxis, there were certain builds set aside for the media to see which often barely resembled the current build the internal testers were working on.) Nevertheless, it's not particularly reassuring either. More troubling is the lact of contacts with the elite Civ4 MP community, which stands in stark contrast to what took place with Civ4. Starting with Friedrich Psitalon, and eventually pulling in about two dozen highly skilled Civilization ladder players, Firaxis included the cream of the crop in Civ4's pre-release testing to improve the finished product from a MP perspective. These guys did amazing work with Civ4, especially Fried and his thousands upon thousands of posts in the private testing forums. MP pros were largely responsible for creating the unit balance in the early game (Jaguar Warriors were originally 6 strength... until the MP guys proved what you could do with a resourceless unit!), but they also made innumerable improvements to the functioning of the promotion system, balancing of the civics columns, how the Great Artist culture bomb functioned, and so on. The whole "score" system that Civ4 uses was developed almost entirely by the MP crew, in order to determine a winner in the many team games where no one was eliminated. These guys did tremendous work, and Civ4 would have been a far inferior game without their contributions.

Yet I have it on good authority that Firaxis is not pulling in these same individuals again for Civ5, and indeed has been rebuffing the ladder community's offers to help out. Firaxis would benefit from pulling in cutthroat players who will stress the game and do their best to break it at every turn. Firaxis' internal testing guys are very good at their job, and they find bugs like you wouldn't believe during the testing process. But there's no way they can play the game at a semi-professional way that the ladder community does, meaning that they can't catch the sort of high-level balance issues that elite MP guys will discover and exploit. The developers would be wise to follow the model that Blizzard uses: you'd better believe that there are hundreds of l33t Korean players currently playing the living daylights out of Starcraft 2 and getting it ready for release. They've gone through 9 major patches just since the public beta started, based on feedback from the elite online community!

So yes, I'm concerned that the development of Civ5 appears to be eschewing the approach that Civ4 used, what was assembling the best talent from the Civilization community in every area (SP, MP, modding, etc.) and inviting a small group to polish the game at the pre-Alpha stage, where they could make core gameplay suggestions before feature lock set in. Firaxis already moved away from this with the Warlords and Beyond the Sword betas, which were unquestionably BETA tests where we could only make balancing suggestions on a pre-decided feature list. My guess is that the approach used by Civ4 was associated with Soren Johnson, who unfortunately is no longer working with Firaxis. They have decided to let those ties with the Civilization community lapse in the past three years - I suppose we'll see how that turns out in the end. However, I look around me and see the elite talent of CivFanatics, Realms Beyond, Civ4 Players, and so on each doing their own thing and not working on Civ5. It feels like a wasted opportunity to me.

There's some rumors to the effect that Apolyton has some community members working on Civ5, which would make sense given that Jon Shafer has some strong ties there from his days as Trip in their PBEM games. However, I have little confidence in the skills of the Apolyton community to do a strong job as pre-release testers for Civ5. There were a number of long-time Apolyton forumgoers present in the Civ4 testing group, and I was unimpressed with their contributions to the process; most of them were much more interested in writing long theoretical forum posts and fighting out endless long-running factional arguments than, you know, actually playing the game in testing. And while many of the Apolyton guys were really polite and friendly, they just weren't that GOOD at Civ4, playing most of their games on Noble or Prince in a highly role-playing style. That sounds like a bold assertion to make, but I'll point to the empirical evidence of the Apolyton Demogame, where the three teams made up of longtime Apolyton forumgoers (Rabbits, Templars, original Banana) all turned in utterly pathetic and embarassing performances, while the non-Apolyton teams of Realms Beyond and PAL dominated the game. Look, testing a game isn't about being nice, it's about breaking the game and trying to find the most absurd, unfair, and ridiculous ways to screw over the AI (SP) or the other player (MP). You should not tab people who are more interested in the scenery, or in telling a fun story, or anything like that. I want to believe this isn't true, but all the stuff the previews are writing about city states and the like suggests otherwise to me.

Conclusion

Well, that was a lot to include, wasn't it? Much of it was highly critical, but I've tried to write this from the perspective of someone questioning and analyzing the design team, instead of the fawning praise that you will inevitably get from the gaming media. As I said before, testing isn't about being nice - it's about breaking the game. I'm trying to find all the holes in the game design, and question whether Firaxis has thought about the issues they will raise. I hope they have answers ready for all of this!

That's not to say everything is bad. The new graphical system looks beautiful, and the finished product will undoubtedly be even better once all the art assets are done. While I'm personally not a fan of the decision to use hexagonal tiles, they do represent something different, and figuring out new tactics to maximize their use will help innovate some of the old aspects of unit movement and city management. There are probably more features that haven't been announced yet, and good answers to half of the questions I've posed here which the design team can't respond to publically.

However, I do feel that the design team has decided to take Civ5 off in a separate direction; rather than competing with Civ4, and trying to surpass it, they zagged off on a different path entirely. While that's not the worst thing in the world, and probably even a smart move given how difficult topping Civ4 would be, it will be a bit of a personal disappointment for me if that turns out to be true. I guess we'll have to wait and see what happens, and how innaccurate these questions turn out to be when all is said and done.