Hacker News new | past | comments | ask | show | jobs | submit login
'Civilization' Creator Sid Meier: “I Didn't Really Expect to Be a Game Designer” (glixel.com)
176 points by danso on May 8, 2017 | hide | past | favorite | 77 comments



Oh man. Civ 1 was the first computer game my father purchased for my brother and me in our earliest teens. We got to choose ONE game from a store, based purely on its packaging, and obviously pending approval by our dad.

I vividly remember the magnetism of the pyramids depicted on the game's box - it had to be that one. Little did we know that we accidentally purchased what is probably the best game in computer game history.

Fast forward a year (and many hours of Civ) later in history class, when our teacher asked the class (somewhat rhethorically..) whether anyone had an explanation for why the Roman Empire became as successful as it was.

Before he could give us his explanation, I blurted out that it must have been their invention of the aqueduct, which eventually allowed the roman cities to grow as large as they did (which is, of course, the game mechanic of aqueducts in Civ). Regardless of whether that's true or not, let's just say that this wasn't the type of answer our teacher expected from a 6th grader!

As someone else noted in here, it's Paradox Interactive games that have somewhat ruined all advanced Civs for me (the exception being Civ 1!). If you're a grown up gamer looking for something to recreate that ol' Civ feeling, give Europa Universalis and Crusader Kings a try.


I can attest to your recommendation. Europa Universalis, Crusader Kings, Hearts of Iron, Stellaris, and Victoria are their primary grand strategy series these days. Each one of them specializes on a different time period, either in the past or future. They all have a considerable amount of depth, but after a few hours of playing around and accepting a few losses, they're a great step up for people with experience with strategy games.


I've heard rumors that Sid Meier is a foreign entity from some other universe. He sucks time away from your life & uses it to help him live forever.

If you don't believe me & you need proof (since this is Hacker News) I dare you to play any Civilization game from the series. I suggest Civ 3. Whichever you pick, you will find yourself randomly noticing the people around you are waking up for breakfast & you've been playing all night. You'll be upset with yourself for a minute, but then you'll realize you just need to make a couple more strategic moves to conqueror your neighbor. If you call in sick today, you might be able to take over your entire continent. Then again, now that you've got the hang of it, you could also restart & really perfect your opening moves. Dilemmas..


Currently at 504 hours on Civ V and I don't really even think I'm very good at it. It really eats my brain anytime I play it.


I remember playing Civ 1 back in the day. It was a great game, albeit easy to win with rapid-expansion strategies. Civ 2 was probably my favourite of the Civilization series (Alpha Centauri being my favourite overall).

As time went on, the designers worked harder and harder to block the best strategies in the game rules. They added more and more convoluted systems of upkeep, corruption, etc. in order to slow down a rapid expanding player. This has probably been my biggest beef with the series: redesigning the rules to counteract the player instead of making the AI opponents smarter. I haven't played Civ 6 but Civ 5 just bored me to tears with how much of that nonsense they did. That game felt like you had a million decisions to make and none of them had much consequence.


For me it was paradox games that ruined civ, they're so much more in depth, your decisions don't just have effects but the have ripple effects, an English Portuguese alliance could take down the Spanish fleet and allow Ottoman hegemony over the Mediterranean for instance. A large part of that is the sheer number of nations, but also some good modelling of their motivations.

They also feature some much better (but now largely gimped) late game features. An inheritance war can tear empires apart in CK2, culture and colonies may revolt in EU4, communists will start a revolution if they don't get social reforms in vicky and and soldiers will become more loyal to their general than you in rome.


6 is a decent rethink of the core rules and mechanics, that still doesn't stray too far from 5. I would recommend taking a look at it on youtube at the very least, it's definitely the better recent title.

Civ is one of the very rare strategy multiplayer games where AI mode is a lot of fun; different than PvP, but still a challenge and a mostly fresh experience. My main issue with it is the exceedingly annoying political behaviour of the ever-complaining AIs. (You're expanding too much! You're not expanding enough! Your army is too large! Your army is weak! One of your units stared at me wrong!)


I also find it a good thing that they try, with subsequent versions, to balance out the play so that you can't just use one strategy and win all the time. Anyone who remembers Nod Bikes knows how little fun that ends up being. Also [1]

The thing that's always been a bummer about Civ's single-player-vs-AI games is that, like most games, the difficulty level simply controlled the amount of handicapping rather than made the AI smarter or more cunning. The difference between hard and easy mode was that they'd just give the player more money, stronger units, better bonuses, etc.

1: I just searched Google for Zerg Rush and holy moly that is awesome. Great job whichever Googler thought of that.


I got incredibly annoyed with people denouncing me in Civ 5, to the point where I would invade and completely wipe out their nation just because they wouldn't stop whining about me (looking at you, Austria).

The only problem I have with 6 is a bit of an odd one; I don't really like the nation/leader options. They feel a bit stilted in comparison to 5, though there are a few exceptions.

The music in 6 is fantastic, however, and I love how it changes as your civ progresses.


My issue with the civ AI was that the tactic of paying them to declare war has always been too powerful. The civs only ever will by choice be in one war at a time, so if you bribe a warlike nation a (pitiful) about of gold to declare war on your closest competitor then you can keep them off your back and squabbling with each other to a cheap victory even on the hardest difficulties.


With V, Gods & Kings and Brave New World really completed that experience, though. One could argue both for and against the religion mechanic and the adjustment to culture, but expanded V is like night and day compared to vanilla V, and I like it better. I genuinely think Firaxis is treating vanilla releases as a somewhat "early access" period to observe how people play then tweaking accordingly, while slipping in more-complex mechanics that didn't make gold to keep the metagame interesting. They know to weather the shit reviews now, too, because V got them as well; people hated V against IV when it came out. Some still do, but I find most of them never tried fully expanded V.

I'm assuming VI will end up similar, and potentially diverge a bit more from V. I'm excited to see what the first, and maybe second, gameplay expansion do to the game, because there are some corner cases (such as the difficulty with tall strategies being discussed). Whenever I recommend V, I say "make sure you have the complete expansions," because it's really a different game. Vanilla VI didn't hook me like expanded V does, but I suspect expanded VI will, and I'm looking forward to that.

This article says it well, too, and also includes a bunch of stuff I hope they consider. The "break-up of nations/empires" point they make about Civ late-game in general is particularly salient. I've always liked the thought of the modern era looking more like the modern era, because late-game does get boring on occasion as everybody rushes toward their victory.

https://www.pcgamesn.com/civilization-vi/civilization-6-expa...


I am an OK Civ2 player (can expect to win on Deity without cheating unless I get quite unlucky).

Goto is hopelessly broken, probably an unfortunate limitation of making a game that can run on 1996-era computers.

City walls are too strong. Catapult, cannon, and artillery should be able to penetrate them. They are always literally the last thing diplomats will sabotage.

The computer gangs up on the player. I guess this makes the game fairer but it also makes diplomacy useless.

The simulation is nukes is hopelessly unrealistic and boring. I wouldn't mind it so much if the pollution system wasn't broken as well. I still got global warming even after making an effort to build solar plants and mass transits in all of my cities because of pollution caused by AI nukes. Rage quitted on that one ;-)

Nuclear non-proliferation in a later Civ was a very good idea IMO.


>I am an OK Civ2 player (can expect to win on Deity without cheating unless I get quite unlucky).

I think that makes you much better than an 'ok' player!


A slightly weird thing about global warming as a result of nuclear war in Civilization games is that nuclear war would likely make the Earth cooler.

https://en.wikipedia.org/wiki/Nuclear_winter


IMO, in Civ 6, going wide is the only way to really win. There are some penalties but they are minimal. In Civ 5 it was totally possible to win with one city, in Civ 6 that's almost impossible. So, if you like building large empires, I highly recommend it to you.


Yeah, that has certainly been my experience so far. In Civ 5 I'd generally take the culture/diplomacy route; I was able to beat one city challenge on immortal with a little bit of luck. In Civ 5, it just doesn't seem possible.

Part of the issue I'm finding is that real estate ends up being a real problem for buildings and wonders. You can't just indiscriminately rush and stack wonders - you'll simply run out of tiles. So, you're trying to place your cities in such a way to avoid land grabs by other civs, but not so much that they crowd out tiles for your other cities.

The barbarian spawning seems to be very unforgiving in Civ 6 as well.


I once saw a video presentation of Sid Meier talking about what he learned talking to playtesters.

It was mostly about people being angry about low odds of losing a fight actually prevailing, so the game shows fake probability to the point you'll lose 90%-chance-to-win fights way less than 10% of the time.

But he did tell one thing that really disincentivized him from pursuing a smarter AI: when the AI does something brilliant, people just assume it's cheating.


This way of thinking is way AI in games is so underdeveloped. Most players don't actually want smart AI, they want AI that makes them feel smart.


Fall from heaven 2 beyond the sword mod, if you like civ and never played it do yourself a huge favor and get it! I loved the civ games and this mod makes them all look boring.


It really was an amazing game. I loved Civ, but FFH2 was just such a good upgrade.

The main developer, Kael, is now a game designer at Stardock. His latest is Galactic Civilization 3. Have you tried it?


I lost many hours on galciv2...


My default statement is always that the best Civ is (and always will be) Alpha Centauri. I still have it installed and give it a playthrough every now and then.


Yeah, it's really disappointing that instead of actually improving the AI, they made it cheat to win. Really lazy game design, and very frustrating for someone who largely prefers playing against computer opponents.


Calling it "lazy game design" is not only offensive to me as a game designer, but it's plain wrong. Here's a talk about AI in Civ from Soren Johnson, Lead Designer/AI Programmer on multiple Civ games: https://www.youtube.com/watch?v=IJcuQQ1eWWI


It's absolutely lazy when they are on record saying they will let the mod community work on improving the base AI.


This certainly is. Do you have a source?


>is not only offensive to me as a game designer

http://www.goodreads.com/quotes/706825-it-s-now-very-common-...

Boo hoo, you're offended by people playing your games. What about us spending money on your game and then hours learning the mechanics only to get beaten by your AI violating the established rules and doing things we cannot.

I don't have have enough time with Civ to have experienced this there, but in Fifa watching defenders with speed and acceleration 10+ points below my strikers outrunning them or that magical switch flipping that says "AI will score now" and watching my own defenders actively run away from the ball or the ball even go right through their legs may not be result of "lazy" design but it's infuriating. Especially when I think of how much money I've laid down to have this experience.


I'm not offended by people playing "my" games - I'm offended by people who can't appreciate the amount of work which is channeled into making games. You can be certain that for every single feature within a (good) game, at least one person spent countless hours thinking about it.

You don't like a feature - that's totally fine. Talk about your grievances all you like, so we can learn and improve said feature. But don't call the people "lazy" if you don't know what you're talking about.


I think we all understand that making a game is a tremendous undertaking. But playing a game on a higher difficulty for the additional challenge should never result in that game cheating. That is lazy design. If you can't figure out how to make the game more of challenge without cheating then simply don't offer higher difficulty levels.


See I agree, but at the same time, Civ achieves being the only strategy game I can think of (I'm sure there's others, I just can't think of them) that actually has a solid AI mode. So is it possible we should heed the lesson we don't like?


> solid AI mode

Paradox games like Crusader Kings II!


I like CK2 a lot, but its AI is nothing to write home about. The game doesn't effectively develop provinces, its strategic war AI is flail-y, and its diplomatic AI is largely deterministic.


Of particular note to the HN crowd – the "typical" design process for a Meier game is he shows up at the office with a completely playable game prototype, and if people find it fun, then the whole thing gets reimplemented. Apparently he's been using the same engine for prototyping for some 20 years.

http://www.pcgamesn.com/civilization-vi/sid-meier-civ-6


That is actually really cool. I bet this strategy is super efficient for game design, you know for designers who do the work to maintain a custom game engine that can do everything they need.


If you want to listen to ~7 hours of Sid Meier being interviewed about Pirates, Railroad Tycoon, Civ, the move from Microprose to Firaxis, and more, these podcasts[1] are awesome. They're done by Soren Johnson, creator of Offworld Trading Company and lead designer of Civ IV.

https://www.idlethumbs.net/designernotes/episodes/sid-meier-...


I miss MicroProse and those fat game manuals! Looking inside for code words so you could land in Gunship...good times.

Edit-Airborne ranger was well ahead of its time IMO. Think that was first game I really felt like I was "sneaking".


I loved Gunship on the Atari ST with a full keyboard template, it was quite a rush to land a damaged chopper with autorotation. Got to work with one of the programmers later, which kind of made everything I did up to that point seem pretty inadequate since we were almost the same age.


Airborne Ranger - that's a title I haven't heard in a LONG while. I must have lost weeks to it though. :)

In fact they have so many good ones. https://en.wikipedia.org/wiki/List_of_MicroProse_games

Falcon 4 is still alive today!


Oh, man... I'd totally forgotten about that game! I played it a _ton_ as a kid :) Now that you mention it, the sights and sounds are coming back to my mind's eye.

Aaaand now I know what I'm doing with my free time later this week :)


Mmmm. I vaguely remember what it was like to have free time. :)

If I can find some again, I do want to get F4 BMS up and running with TrackIr on the 34".


Silent Service / Red Storm Rising - those were the days


I know everyone goes on and on about Civilization but all the simulators from F15/Gunship/etc were my first introduction to his games. I still think Pirates was the real breakthrough. It was an almost sand box game and of course Railroad Tycoon.

Oddly its perhaps my playing of traditional board games usually with others that kept Civilization from taking over my computer gaming at the time


Pirates was a fun little game. Another one that I loved growing up was Covert Action.

Only the electronics and breaking in mini-games were really any good, but the game was set up so that you could focus on those and pretty much ignore the others. I loved the whole mystery and trying to untangle the crime aspect with the different actors, as well as how if you took too long or looked in the wrong places you could miss capturing some of the bad guys, or even screw up the whole mission.


It's interesting to see how many classic/nostalgia-games were actually simulations and sandboxes, rather than scripted experiences.


It's no surprise at all. Look at MMORPGs. The best moments of these games were pulling high level monsters to city, crashing an in-game funeral in WoW and the drama between players in Eve Online. Not farming monsters for 20 hours to get enough cloth. Unfortunately in many cases the unscripted emergent gameplay was often connected with "toxic" behaviour, as a result most games nowadays reduce conflict between players to regulated arenas, and provide a themepark experience outside.


I think one explanation is that the sandbox/simulation games take more time, leave players with 'more' memories, and as a result are more linked with particular periods in their lives.

Looking at my gaming past, a number of scripted experiences have actually left the most vivid memories, but the sandbox/simulation games are the ones I associate more with friendships, periods in my life, developmental stages, etc.

Nostalgic thoughts are about periods of time, and less about moments. A scripted game, I think, is usually too short to produce nostalgia. The few scripted games that do, like the Myst series, took an unusually long time (for me) to play through, for example.


20 years from now, how will our current gaming scene be defined in the compressed, excerpted view of retrospective history? Sure, there are scripted megahits such as "The Last of Us" and the "Call of Duty" series. But the standouts in terms of pure revenue and popularity metrics would seemingly be the Grand Theft Auto series (particularly V) and, for indies, Minecraft (pre-Microsoft).


Call of Duty isn't really more scripted than GTA, and arguably it's less; both have a scripted singleplayer campaign, but Call of Duty has always been a multiplayer game first and foremost. I have clocked many, many hours in CoD4 since it came out, but only six of those were actually scripted.


Kerbal Space Program without a doubt.


I think of the current scene dominated by a lower barrier of entry to market and the current rise of indie and "early access" games so maybe it won't be defined by games at all but by the dominance of Steam.


The 80's had an even lower barrier to entry, but now we tend to only remember the best that was on offer.


Not really, if you count "access to a computer" and "access to distribution channels" in those barriers.


In comments under the article about civilization, definitely. It is going to be be different in comments under article about doom, wolfestein or some adventure game.


Anytime Civ 2 is mentioned, I just remember this. Just so amazing. https://news.ycombinator.com/item?id=4100032


My thoughts exactly; I can remember reading that the first time around on HN/Reddit.


CIV before pixel graphics, Empire: https://en.wikipedia.org/wiki/Classic_Empire

* VMS Empire: http://www.catb.org/esr/vms-empire/ (apt-get install empire, ascii graphics)

* http://www.classicempire.com

* XConq https://sourceware.org/xconq/

* A relatively more modern commercial variant (including mobile) http://killerbeesoftware.com/kbsgames/ednew/ http://killerbeesoftware.com/kbsgames/edme


Another fun fact, XConq has a lisp like game design language (GDL) built in. https://sourceware.org/xconq/manual/xcdesign_toc.html


Fun fact: Classic Empire was created by Walter Bright, of D fame.


Nice interview, much longer more in depth one on Soren Johnson's 'Designer Notes' podcast (the fourth part just landed the other day). https://www.idlethumbs.net/designernotes/


I never got to play Civ 1 but Civ 2 was one of the first games I ever owned and played endlessly. Its soundtrack still brings back those feelings.. https://www.youtube.com/watch?v=C55Ae3apn5s


I've been told by various people that Civilization was inspired by Empire (the first computer god game as far as I know), but I'd like to ask Sid if that is true!

(There's also the older Hammurabi game, but that one has no map.)


Civ 1 was the closest thing i've ever experienced to addiction, withdrawal and denial and it was frightening. Yet so memorable.


"The first Civilization was released more than a quarter-century ago in 1991" ...Thanks for making me feel old.


What are civ players thoughts on freeciv?


Compared to Civ1/2 it has nicer graphics tilesets. Sometimes I play (I am more into board games like Eclipse en Scythe) and then I use the 2D tileset and a tiny universe for quick play. For other open source civs - but much more retro - you might like VMS Empire, Classic Empire or XConq. Less is sometimes more.

http://www.catb.org/esr/vms-empire/

http://www.classicempire.com

http://sourceware.org/xconq/README


I suggest you give Freeciv-web a try: https://play.freeciv.org/


I played a ridiculous amount of Civ1 back in the day, and later Civ2 and Civ3. I deleted Civ3 around 2003 because I felt that it was sucking too much time away from my real life. Haven't played any of the more recent versions but I'm sure that they're great.

Instead, a few years back, I took up the occasional Civ1 again, downloaded as abandonware, and running in a DOSBox. A nice thing about Civ1 is that I can usually complete it in a few hours; as I recall, Civ3 could easily take a night to play through. I even played it with my son for a while. I think that he should know the classics.

As for strategy, I play with three civs, I always prioritize research over taxes, and I prioritize expansion over fortification early in the game (I quit when this backfires).

Once I get my civ rolling I have found the best strategy for getting a high score is to track down the other civs as fast as possible and to knock them out, starting at the capital. Once their capital has fallen and a few main cities, they will make peace. Then rest of the cities can be conquered at my leisure or they can be cheaply bought by diplomats and their building sold for scraps, decreasing the price further. An important prerequisite for the aggressive strategy to work is to get to them before they build city walls.

Often, however, I find myself gravitating towards a "tech hub" strategy instead. It rarely leads to scores above 100% on king level but if feels kind of satisfying. The point of this is to appoint some city with potential for trade and growth (ideally all of its accessible squares should be plains or convertible to plains). Then build, as soon as possible, the Colossus (+1 trade per trade square). Then, in some order, a library (+50% science), a granary (faster growth), Copernicus' Observatory (+100% science), university (+50% science), and Shakespeare's Theatre (everyone is happy). Bring in caravans from the rest of the continent to speed up production, and to set up trade routes from the tech hub. Elsewhere, build the Pyramids (switch to democracy) and J.S.Bach's Cathedral (two happy faces per city and only way to keep democracy while dispatching a sail or a frigate with caravans from the tech hub). Research railroad and, once achived, assign all settlers to put railroads on the area surrounding the tech hub, boosting growth (in preparation for railroad, clear any forests and swamps, and build roads and irrigation). Postpone researching electricity since it will cancel the Colossus.

With the tech hub strategy, the science light bulps of the tech hub will often overflow into the next panel on the city details screen and I'll usually have armors and planes and railroads before reaching 0 AD. After that, I guess the correct path is to switch to communism and eradicate all resistance. I tend to do this too late, though.


It's been 27 fucking years since 1990, and the civilization franchise is still a glorified boardgame, where the only agency is had by each player.

Totally unlike the real world, where ruling or governing is more about figuring out how to exercise power without getting overthrown.

Such a game would be way more interesting.


I think that's fine, if you want lots of intricacy in diplomacy Paradox have that covered. Civilization is very much a board game - in general Firaxis seem to have gone down the road of making incredibly polished boardgames that you can play on your own with a computer, and I think it's worked out really well.


>polished boardgames that you can play on your own with a computer, and I think it's worked out really well.

Paradox is also in the business of making boardgames, albeit with way more elaborate rules.

> your own with a computer, and I think it's worked out really well.

I don't think it has. The AI in Civ series is just bad, and only gets by through rampant cheating. Civ V was out for years, and they haven't managed to do anything decent with the combat AI. Nor prevent the AI from doing stupid things.


> It's been 27 fucking years since 1990, and the civilization franchise is still a glorified boardgame, where the only agency is had by each player.

Yes, that's​ pretty much what the game is designed to be; keeping the game what people have enjoyed it for is sensible.

> Totally unlike the real world, where ruling or governing is more about figuring out how to exercise power without getting overthrown.

And, sure, such a simulation would be interesting, but it's not what civilization has ever been, or advertised itself to be. It would certainly be good to have such a game, but why should it be in, or replace, the Civ line?


I'm not sure how you came to the conclusion that it would make an interesting game. Making something more like the real world is often exactly what you don't want to do in a game.


Because it'd be more complex and interesting and not just about optimizing to win.

No emergent gameplay at all in Civ.


Complex does not equal interesting. I suggest you take a look at Extra Credits' episode on Depth vs Complexity [0].

[0] https://youtu.be/jVL4st0blGU


By 'complex' I meant that it'd not be a boardgame anymore, but a boardgame in which the 'pieces' would have minds of their own and would do whatever they were could.

Have you read about the Russian frontier? As soon as it was at least somewhat secure so living on it wasn't a sure-fire way of getting enslaved by Muslim steppe raiders(Tatars), the empty land was filled up by peasants looking for land to farm.

An increase in population led to more slave raids, so Moscow was forced to push the border further south to protect the farmers. In this way the border kept moving south. In 250 years, this mechanism caused the Russian state to expand from Moscow to Crimea.


If you don't do it right, citizens rise up against you and pillage your empire from the inside. It's certainly frustrating, if not interesting.


You might like Crusader Kings II and Victoria II.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: