This one has a summary: http://gdc.gamespot.com/story/625325...e-video-insideOriginally Posted by ubik
My favorite line:
"...making the AI do surprising things will have bad effects either way. If it does something bad, the player will assume it's stupid. If it does something overly clever, the player will assume it cheated."
He gave us some great food for thought in that address; I do think the "Epic Journey" part was spot on. The AI part - less so.
I've finished now viewing it all. So my honest take on it, mentioning here that I am a CIV I/CIV III/Covert Action/Pirates!/Railroad Tycoon I mega fan:
At a global level, I think he started pretty well for the first two minutes and then, for the fist two thirds of the address I don't agree with him on most of the approach. The last third I mostly agree with SM.
Please bear in mind that Paradox games in general, games we enjoy most here are much more mature oriented than what SM usually produces. This has a serious impact in the approach considered.
Specifically addressing some keypoins:
Moral Clarity - Completely disagree with SM. Not desirable on anything barely mature. Moral choices ARE a very important part of emotionally engaging the player. Let him picture Gengis Khan struggling in his last city, licking the player's boots. Or, if the player has a particular issue with him, let him raze Gengis Khan last city, erasing him from the surface of earth once and for all... or perhaps show compassion, help Gengis Khan up again to some power and rejoice when he pays his debt to you by striking your foe unexpectedly or comming to your help at a later time.
Always Win in The End - Completely disagree with SM. If a player knows the odds are about his winning in the end, a huge part of the experience is forever lost. The risk of failure must be ever present throughout the whole exprience.
Battles - Completely disagree with SM. His problems stemm only fromt he fact he is reducing such an emotional moment of gameplay with a very logic and acessible math exercise, which ends up being poorly interpeted the player at a psychological level. There is too much quantitative information on what is first and foremost a highly qualitative experience.
Real Time - Completely disagree with SM. Johan proved him wrong with EU I. All other Paradox titles just underscore this. There are certainly more from other companies, but here SM is completely off the mark.
Rise and Fall - Completely disagree with SM. ...But it must be said the experience should never be canned into making the player fall. That must be part of the natural ever changing gameplay dynamic, not by a deterministic decision at some point. If hep layr reloads, that is his decision and his prerrogative. He may be saving his game (at the subtle cost of marring the experience a bit by having the knowledge of cheating) or he may be passing a great opportunity to live an experience of survival, come back and win.
Tech Tree - Mildly Agree with SM. Such defining elements in gameplay should be easy to plan ahead. however, a risk factor should always be present. So, I think if only 75% of the tech tree was ever present in a game but keeping the same path struture, that would be a great gameplay element.
Give Gold in Civ Network - Completely disagree with SM. If players were not taking advantage of this feature, the fault was in the game design.
Such a powerful concept will be included in anyone's strategy as long as the design makes it rewarding. This is one of his lowest points in the address.
AAA Games on a Shoestring - Agreed with him.
AI - Agreed with his talk, but not with his conclusions. AI should prove a fallible challenge, never something you can predict.
Protecting the Player - Mostly agreed with SM.
He started poorly as players should be fully accountable for their own decisions.
Load/Save issues Agreed with SM. Good tip for MM the Game.
Choices/Options/Settings - Agreed with SM as long as they interfere with the design. This is something that I cherish much as it was debated in the past to allow the players to select only parts of Magna Mundi to be active at a time. Obviously, that would lead to a mangled experience where the whole orchestra were missing some key instruments. The result would be crippling gameplay and wasting development resources. Fortunately was never implemented.
Cheatcodes - Agreed with SM. ALT-21 will do nothing in Magna Mundi
[BModding -[/B] Agreed with SM. When you see what we have in store for modding Mgna Mundi you'll be amazed
Listening to the Player - Partly agree. I agree with his approach and his assertion the player usually doesn't tell us the cause, but the consequence of problems. So, our work is to backtrack in thep layer mind to check the choke point that lead to his later conclusion.
He failed in my opinion when he stated that one should accomodate all players reports in a way or another. Of couse, a moment later he contradicts himself by rightfully telling us about player profiling. It's obviously that a powermonger will feel rustrated a subtle diplomacy, a diplomat willb e ever frustrated if he is permanently in a state of war and a trader if he can't make loads of money.
Epic Journey - Agreed. But the journey can only be epic in the player's mind if he knew he was really risking someting in the process. Progress should never be taken for granted. SM earlier comments contradicts this element that is for me central.
.
Last edited by ubik; 24-03-2010 at 04:50.
One thing that SM never actually spelled out, but seemed to be implicit in a lot of his comments was this principle:The player will accept/overlook bad event or limits more easily if they are constantly present than if they happen infrequently.This seemed very true to me-unexpected or unusual events stand out and take on a much greater weight in our minds than regular occurrence-so game designers can use that fact of psychology to help steer the player's emotions. For example, as SM pointed out, if the AI does something surprising, this is going to get the player's attention. In his view this is generally a bad thing, since the player will either assume the AI cheated or think it is stupid. So, to make the player focus less on whether the AI is making good or bad decisions, SM aims to have the AI act fairly predictably, thereby not drawing the player's attention to the choices the AI makes.
I think we can easily see how this applies to MMP/MMU. MMP had lots of events that put serious limits on the player ( Horse shortages, Cavalry tax, etc.) but the ones that got everyone's attention were the unusual ones like Framed, Noble opposition, or the Barbary Pirates. Perhaps the best example is the new naval system compared to the old Barbary pirates. The Barbary pirates events used to really grab your attention because they were so unlike everything else, and people HATED them. The new naval system on the other hand is a constant annoyance, so people get used to it, and now nobody complains about the Barbary pirates, because the whole naval system has made them less unusual, even thought the events are exactly the same. By changing the context, the Barbary pirates went from being the scourge of MMP to one more mildly annoying bad event.
Anyway, the lesson I got from this was that you can really control how the player feels about an event just by making the event appear more or less similar to other events. If you want something to be dramatic, it needs to stand out. Whether you loved them or hated them, the Barbary pirates used to be dramatic; now they're not. Whether that's a good thing or a bad thing is up to you. SM would probably say that the old pirates were too frustrating, since they were bad events that could not be avoided, but personally I kind of liked them, because I felt such strong hatred when they appeared. Similarly, I'll bet we could make massive noble oppositions far less frustrating, by having a whole series of events where nobles object etc. that always fire when you get a new monarch. But I think the point of those events is to create a memorable episode in the game, so it is important to make them stand out in order to grab the player emotionally.
In the end, I guess it's a matter of deciding which bad events add drama and should be emphasized and which just lead to frustration if they stand out. SM is surely right when he says that in general you want to minimize negative emotions, but I think sometimes you need to actually do the opposite if you want to create an exciting game.
ubik, I just watched the video with SM and was amazed to find myself to disagree with most of it (amazed because I like his games). Then I read your comments and felt relieved. Especially on 'Moral Clarity' and 'Rise and Fall', exactly my sentiments.
Same here! I think that when SM refers to "the player" he refers to the average or majority player, who of course is just plain dumb, just like the majority of people. This plain man always wants to feel more than he is (citing SM: "everybody is above average") and prefers routine over creative challange (think about SM's opinion on "significant randomness"). Paradox players on the other hand should share rather different psychological characteristics, especially the even more selected group that plays MM. Controversies like this always let me appreciate that MM is designed by ubik, who just won't be able to overcome his genuine evilness to make a dumbed down profit maximizing game for every other guy.
Last edited by Ignudo; 25-03-2010 at 11:35.
Actually, I think this is the key.
He doesn't want to design great games anymore. He wants to sell the biggest number of copies possible. What SM does is to find the greatest common divisor between the entire games' market and the game type he wants to deliver. To this end he always follow a risk averse strategy in terms of design to turn the masses happy, as long as a minimum quality of gameplay is assured.
Sometimes this works well (CIV III) other times it gives birth to the biggest turds ever (SM RailRoad Tycoon).
One interesting experience to check his approach was if he published a new game, with a new brand, with a hidden identity as the designer, using the same design approach and then check the sales. I doubt it would be any kind of even a minor success in terms of sales.
I think you are reading too much on him, to be honest.
He seems just to say:
Lets sweeten gameplay the most, as long as we still have a game in the end, in order to catter to the largest possible market. We trust in the most powerful computer gaming brand, my name established with all merit 20 years ago plus a millionare marketing campaign to take care of the sales.
With the corollary being:
The game that sells most has the best game design.
It's debatable, I do not agree with it, but it is indeed a very powerful argument!
Or the mod with most downloads? :P
Sorry for slightly checky comment, don't think it's been mentioned so far, I'm wondering if your going to have a similar general/leader system as in the Mod? As I'm a big fan of the GRIN mod which gives you loads of free (but usually crap generals) as RP wise I like having someone in charge, even if there an idiot.
Let's just say I interpreted his remarks in a generous fashion.
I am certainly not one who compromises with player complaints. I only compromise with designs that keep people playing.
Obviously some complaints are deserved and get properly addressed but most are not that relevant.
The fact that the mod is this successful is a testament to the validity of this design approach (and the other coders also share it, make no mistake) within Paradox set of players.
The biggest challenge for Magna Mundi (and Paradox) is to be "seen" by outsiders. I feel we have what it takes to catter to a significant segment of the strategy market not exposed to Paradox products but only the future will tell. The game's subject (detailed History) is too narrow to dream too high, I think.
Sorry for slightly checky comment, don't think it's been mentioned so far, I'm wondering if your going to have a similar general/leader system as in the Mod? As I'm a big fan of the GRIN mod which gives you loads of free (but usually crap generals) as RP wise I like having someone in charge, even if there an idiot.
General/leader system in the mod is the same as EU3 vanilla...
No, in Magna Mundi - The Game it will be FAR different.
Last edited by ubik; 25-03-2010 at 00:17.
Bah should have expected something so vague, but thanks for answer.
Don't suppose I could get an idea of what you current basic intention with it is?
My basic intention? Sure!
Make all the generals/admirals thing much more interesting than before.
Cunning ubik.. cunning.. but I'll get a non vague next time! Next time! *shake fist*
I disagree with some of the implied elitism here (not going to name names). Sid Meier's games are designed for a crowd that enjoys easier, less complex games, but that doesn't mean the people who enjoy that kind of thing any less intelligent. Let's face it: most people have no time nor interest in a game that takes hours and hours to learn and hours and hours to play and focuses more on small details of country development. The people who EU3 and other Paradox titles attract (avid strategy gamers and history buffs) obviously tend to be more knowledgeable because they have a great interest in learning historical facts and skilled at being successful at hard strategy games, but whether that translates into any kind of practical intelligence is debatable and highly variant from individual to individual.
Neither would I say that Sid Meier has definitely lost his touch. Firaxis has made some really crappy games (I agree 100% on Railroads!), but other recent titles, like Civ IV, are really very good. I thought IV suffered mainly from a crappy AI, tedious combat, and somewhat unsatisfying diplomacy, but the overall experience was great (much less tedious than I found Civ III), and you can't exactly say it was a cakewalk if you turned up the difficulty high enough (granted, it was because the player got handicapped, but MM the mod does that, too). Also, Civ V is shaping up to be a great game from what I've seen so far, with some really good radical changes (and not in the direction of dumbing down the strategy), so they seem to still have a few sparks in them. They are made to appeal to the general public, but so are games like Chess or Stratego.
Part of it, of course, is the different design philosophies that you mentioned in another thread. I think Sid still believes in giving the player a challenge adapted to what he wants, but where you seem to really diverge is on the AI and its behavior. Meier treats it more like the antagonist in a movie, who isn't expected to really stop the protagonist (unless you really mess up for your difficulty level), but to provide a thematic and emotional opposition. The difficulty levels in Civ are where much of the accomplishment comes from, as you become able to win consistently at Noble, then Prince, etc.; the accomplishment is spread out over multiple games, fairly certain of victory in the individual. A good analogy is to lifting weights: you don't pick up a massive weight and struggle with all your might to lift it, maybe failing; instead, you start progress from light to heavy over time, certain of "victory" at each. You treat the AI (and the awesome layers of events) more like the antagonist in a sports competition, who is expected to nail you to the wall if you aren't playing your best and not to forgive a few mistakes. The accomplishment there comes from, like you said, facing setbacks, losing a few times, but often being able to come back from failure to final victory, within one game. This has the effect of making each game a lot more dramatic, but also potentially frustrating.
I'm definitely not complaining here because I like both types. Variety is nice.
I also just want to say something about saving/reloading: I agree generally, but sometimes being able to do it is nice, and not just to blatantly cheat. For example, I'll admit to doing it often in MMU just to test possibilities, like "If I declare war on Lorraine, will France intervene or not?" and then I can see if I'm facing a winnable war or total annihilation. Now, maybe that takes away some of the suspense of not knowing or could be considered cheating, but sometimes I like to make sure a single decision won't destroy my empire. Besides, I'm sure this is one of the areas that MM the game is trying to make more predictable.
Nec audiendi qui solent dicere, vox populi, vox Dei, quum tumultuositas vulgi semper insaniae proxima sit.
—Alcuinus ad Carolum Magnum
They should not be listened to who keep saying that the voice of the people is the voice of God, for the riotousness of the crowd is always close to madness.
—Alcuin to Charlemagne
To make it clear, I do not subscribe to the notion where someone equated mainstream strategy gamrs as dumb.
Neither would I say that Sid Meier has definitely lost his touch.
Well, recently his only good titles seem to be carried on the shoulders of the original CIV gameplay ( a masterpiece on its own).
I don't know if he lost his touch or not, but in fact Sid seems very happy to just ride the brands he created 20 or so years ago. So, with the absence of touch, one cannot say one way or the other.
Firaxis has made some really crappy games (I agree 100% on Railroads!), but other recent titles, like Civ IV, are really very good.
I prefer CIV III. IV seemed more bland to me.
and you can't exactly say it was a cakewalk if you turned up the difficulty high enough (granted, it was because the player got handicapped, but MM the mod does that, too).
We need to buff the AI in first place because the whole game mechanics imposed to the player and the AI are stricter. The player can use his had and adapts... as for the AI, without access to the engine in the mod, we can't do the same.
So, to make it clear, we don't buff the AI to make it harder for the player, but instead to cope with the additional challenges of the mechanics introduced.
Well, he is certainly starting to sell the game. Int he presentation 80% of the screenshots were CIV...Also, Civ V is shaping up to be a great game from what I've seen so far, with some really good radical changes (and not in the direction of dumbing down the strategy), so they seem to still have a few sparks in them. They are made to appeal to the general public, but so are games like Chess or Stratego.
I think Sid still believes in giving the player a challenge adapted to what he wants, but where you seem to really diverge is on the AI and its behavior. Meier treats it more like the antagonist in a movie, who isn't expected to really stop the protagonist (unless you really mess up for your difficulty level), but to provide a thematic and emotional opposition.
Not at all. his approach to AI is not that different from MM. I only think if the AI is there like a static roadblock who always play he same to be overcome, the experience of th game is downplayed.
You treat the AI (and the awesome layers of events) more like the antagonist in a sports competition, who is expected to nail you to the wall if you aren't playing your best and not to forgive a few mistakes. The accomplishment there comes from, like you said, facing setbacks, losing a few times, but often being able to come back from failure to final victory, within one game. This has the effect of making each game a lot more dramatic, but also potentially frustrating.
Actually, we try to just allow the AI to survive.
And if possible, to make the AI seem to be playing like a country, with mitakes, but at least a semblance of humanity.
A good rule of thumb is to prepare for the wost if you think a critical decision may spell disaster.I also just want to say something about saving/reloading: I agree generally, but sometimes being able to do it is nice, and not just to blatantly cheat. For example, I'll admit to doing it often in MMU just to test possibilities, like "If I declare war on Lorraine, will France intervene or not?" and then I can see if I'm facing a winnable war or total annihilation. Now, maybe that takes away some of the suspense of not knowing or could be considered cheating, but sometimes I like to make sure a single decision won't destroy my empire.
Can you elaborate on this?Besides, I'm sure this is one of the areas that MM the game is trying to make more predictable.
One comment on Meier's take on AI, and a suggestion...
While I think he definitely has a point (ie, unpredictable AI seems like a bug if it fails, cheating if it succeeds) there's a way around this.
That is, to find some way to make the AI transparent, at least from time to time.
I would love to see a player be able to use certain assets (traders, diplomats, spies, etc) to gain some insight into what options an AI-controlled nation is considering. Historically, premodern espionage was as much collecting and processing public gossip (a job that would now be done by a foreign correspondent or embassy political office) as carrying out covert operations.
A minor point, but I completely agree that Civ IV was bland and boring... however, I thought FfH2 was a mod that made the game worth buying, and the best thing I have heard about Civ5 is that it has much greater Mod support.
But, frankly, I think Vanilla EU3 is rather bland and boring, and certainly way too easy, MM is what makes the game for me. Paradox' genius was in its modding system, not its game implementation, imo.
The real problem I think isn't one design philosophy vs. the other, or one developer vs. the other, it is the fact that historical simulation games don't make enough money to command huge developement teams/times, and it is left to deidicated players to improve the games from acceptable to great products. I usually like (and generally prefer) EU3/MM for its depth and difficulty, but sometimes I like some Civ4/FfH2 play for an easier, more fantastical approach... but in both cases, it took more development than the initial game could afford to make it fun imo.
Sad at the demise of MMtG; excited for EU4
Avid reader of Chris Taylor's England MMU AAR http://forum.paradoxplaza.com/forum/...ngland-MMU-AAR
Regarding Civ IV, I find it similar to EU3, a great game that fell to it's predecesors glory when released, later gets expansions that really do improve the game, and then comes "THE MOD", the one that does put things in the right place and order. the mod that makes the game completely different from all the rest.
Regarding saves, I like to have saves from times where I faced two or more paths to follow, very different from each other, not because of fear that I might fail, but because we play an alternate history game, and in a game you create a HISTORY, and I enjoy a "what if..." to that history; that makes a single game in MM last for months, I am faced with a revolution, OK: save up, defeat it, keep playing as monarchy roleplaying a weary empire holding to the throne and finish for glory or shame. Then I play losing agaist it, becoming the land of freedom and sharing liberty with those under weary regimes.
Well given the ipression,it seems the only common point between MM and EU III will be the game engine isn't it?
Posting Permissions
