Welcome, Guest. Please login or register.
March 28, 2024, 10:10:11 AM

Login with username, password and session length

Search:     Advanced search
we're back, baby
*
Home Help Search Login Register
f13.net  |  f13.net General Forums  |  The Gaming Graveyard  |  Game Design/Development  |  Topic: What if what we have now is all there is? (or, Dikus are the Steering Wheel) 0 Members and 1 Guest are viewing this topic.
Pages: [1] 2 Go Down Print
Author Topic: What if what we have now is all there is? (or, Dikus are the Steering Wheel)  (Read 15683 times)
Triforcer
Terracotta Army
Posts: 4663


on: July 21, 2009, 11:56:14 PM

Disclaimer:  Almost certainly none of the below is an original thought.


First, one fundamental assumption:  investors want big, continuous returns and smash hits.  There are investors, I suppose, who would be content with a 50-100k sub base (or the corresponding equivalent in a microtrans game) that made back production costs and a steady return.  "Niche" games could thus defy the ideas I have below.

But lets talk about the AAA, gold-plated MMOs, the WoWs and all the other current ( awesome, for real) and future ones that want to reach that pinnacle.  In most discussions of the future of the industry, there is an assumption- mostly implicit, sometimes explicit- that the genre will "evolve" past the diku standard.  Some technological breakthrough will allow meaningful story choice or lag-free mass PvP battle or deformable terrain or challenging PvE AI.  The specter of "grinding" for exp will go away and something that is constantly fresh and fun will take its place.  Things will change- this is very early days, the infancy of the genre, and by god things will CHANGE.  The day of the diku is numbered.  


But what will MMO developers do, when trying to develop the next smash hit?  They have to worry, above all us, about user sustainability (either in terms of subs, or buying microtrans stuff).  What will keep players playing, month after month?  Here is what they always discover:

(1) Improved monster AI- this makes players angry because dying is sad and they want to win most of the time and its not fair.  

Conclusion- subs/microtrans customers will leave, cannot be put in the game.  


(2) Meaningful story- this makes players angry because their choices mean they missed out on getting the best piece of healing gear for their 5-man raiding spec or they can't change their spec when BlizEA nerfs Witchaloks in PvP.  

Conclusion- subs/microtrans customers will leave, choices cannot affect gear/spec/faction in any way.


(3) Advancement methods other than grinding for exp/UO-style skill levels- players say they would love that, but no long-term sustainability as everyone is at full power in two weeks and then unsubs.

Conclusion- some months-long grind to max level/skill must be in the game.  


(4) Players actions in PvP change the PvP world- People are sad when their side loses and their city is burnt down.  

Conclusion- subs/microtrans customers will leave, cannot be put in game.


(5) Player actions in PvE change the PvE world, or PvE world has its own changing schedule- The vendor I need isn't available for business during the game "nighttime"?  The monster I want to kill doesn't drop the same loot and now I can't look at a guide to find my equipment?

Conclusion- subs/microtrans customers will leave, PvE must be Thottbot-able and unchangeable.  Same monster with same loot in same place, from now until the game shuts down.    



Through a painful 10 year process, we have discovered what the largest number of players want (and/or what the features are- like having a long grind- that are necessary for years-long sub/microtrans sustainability).  And its the above.  Its always the above.  There will always be PvP without lasting consequences, personal choice without lasting consequences, static and Thottbot-able PvE, and an exp grind.  Better technology or better ways of dealing with latency won't help us.  The possibilities that technological advancements open up can NEVER be used, because by definition they relegate the game to a niche game, incapable of even the possibility of smash-hit WoW-like success.

Dikus are the steering wheel.  We've had a steering wheel for close to 100 years now.  We have the technology to replace it with buttons or Okuda-grams or a thumbstick something else.  But we never do it.  Drivers would hate it, and therefore the tech advancements mean nothing.  The current shape is eternal- what we have now is all there is and all there ever will be.  It is elementally what people WANT, the lowest-common-denominator shape contained deep within a block of wood when everything that a majority "hates" is carved away.  Dikus aren't designed, they are what remains.    

Sure, there will be iterations.  Full voice-overs in SWTOR and LOTRO achievements and WoW's iteration of those achievements and more and more use of instancing and phasing.  Steering wheels have gotten covered in leather and powered and heated over the years too.  But in trying to develop a hit (not small-profit-stream niche, as least as a conscious design goal) MMO, the shape can never change.  Who will ever spend 50 or 100 million to prove the steering wheel is wrong?            
« Last Edit: July 22, 2009, 12:17:50 AM by Triforcer »

All life begins with Nu and ends with Nu.  This is the truth!  This is my belief! At least for now...
ezrast
Terracotta Army
Posts: 2125


WWW
Reply #1 on: July 22, 2009, 02:01:10 AM

I'll bite.

(1) Partially true, but "improved AI" does not equate to "greater difficulty" or "less fun." People are willing to die frequently in PvP when there is no death penalty and they don't get yanked out of the action for very long. People aren't willing to die in PvE because it slows their XP gain, which is where the actual enjoyment comes from. Make PvE fun in its own right rather than just a carrot on a stick, and people will be willing to stomach the occasional death. Of course, that runs into problems with (3)...

(2) Mostly agreed, although I think "meaningful" storytelling isn't done not so much because it would hurt retention, but because it simply isn't feasible; the player knows damn well that 10,000 other adventurers have saved Thardek Rumblebucket from the clutches of the Ice Fiend, and there's no way to maintain immersion. Anyway, who plays video games for story?  why so serious?

(3) Perhaps, but this doesn't rule out advancement based on skill diversity and epeen rather than strict power advancement.

(4) Yup. However, there is little reason why PvP - hardcore lose-all-your-shit burn-your-city teabag-your-corpse PvP, even - and PvE can't coexist in different parts of the same gameworld. Yeah, there's the argument that balance for one always screws up balance for the other, but that problem can be alleviated by better design (i.e. creating your PvE to work on the same principles as your PvP, see point (1) ).

(5) The weakest point out of all of them; I don't see why there should be a correlation between "Thottbott-ability" and player retention. The less predictable the loot tables are, the harder the "slot machine" effect kicks in - that armor you need could drop from the very next monster! Or the next one! Or the...

Most importantly, though, this:
Who will ever spend 50 or 100 million to prove the steering wheel is wrong?           
Lots of companies. They've been doing it time and time again for a decade. Just not in the way you mean.
CadetUmfer
Terracotta Army
Posts: 69


WWW
Reply #2 on: July 22, 2009, 11:31:53 AM

I think "MMO" will be a tag applied to games, like "3D", not a genre.  Originally 3D pretty much meant FPS, much as MMO means (Diku) RPG.

Soon all kinds of games will be able to take advantage of persistent multiplayer, just like they do 3D.

Anthony Umfer
Developer, LiftOff Studios
Lantyssa
Terracotta Army
Posts: 20848


Reply #3 on: July 22, 2009, 12:02:48 PM

The Steering Wheel has not changed much over time due to practical and safety reasons.

Entertainment on the other hand, has varied greatly over the same time span.  While subjective, there are good periods and bad periods.  Games are entertainment.  They'll change out of necessity because people will jump ship when something else comes around and new trends will form.

Hahahaha!  I'm really good at this!
DLRiley
Terracotta Army
Posts: 1982


Reply #4 on: July 22, 2009, 12:48:23 PM

I think "MMO" will be a tag applied to games, like "3D", not a genre.  Originally 3D pretty much meant FPS, much as MMO means (Diku) RPG.

Soon all kinds of games will be able to take advantage of persistent instance multiplayer, just like they do 3D.
tazelbain
Terracotta Army
Posts: 6603

tazelbain


Reply #5 on: July 22, 2009, 01:29:31 PM

I was thinking the same thing in the Kurt Schilling Thread.  To many people the idea MMO would deviate from the MMOG formula is unfathomable.

The research for current generation is all used up.  No one seems to be doing the research for the nextgen MMOGs.  An AAA MMOG costs are too high to do research themselves.  So new games don't have a blue print to go forward so we are stuck.

"Me am play gods"
DLRiley
Terracotta Army
Posts: 1982


Reply #6 on: July 22, 2009, 02:36:40 PM

Its not the lack of research, there is a proven formula that few devs know why it works. For example if you strip the hardcore roots from world pvp then we might actually have a non retarded "pvp" game that actually makes AAA money. Or another example, its 2009 xp curves should be nonexistent for any mmo charging $15 a month, but the because devs insist that their games don't compete with WoW, despite the fact of being an mmo and charging the same price, mmo's still continue crashing and burning before the alter of content/grind(because grind = content apparently) = fun != grind.

Or my personal favorite, the solo'ers vs group debate ended back in 2003, why do devs still revisit that dead and beaten horse only to relive the failures of mmo's made since 1999. What's even funnier is that devs still refuse to treat competitive pvp as a valuable alternative to the world butt fuck, despite having to expand less resources to make the former rather then the latter. Its like they close their eyes and shut their ears for 5 years, does no one other then blizzard and arena-net thinks that it is a good idea to have an semi-competitive arena the entire playerbase can participate in? I mean do developers really think that having only one preferred mode of play supported by your game actually keeps players playing your game longer vs blizzard other game which have several modes of play?
« Last Edit: July 22, 2009, 02:38:44 PM by DLRiley »
Kail
Terracotta Army
Posts: 2858


Reply #7 on: July 22, 2009, 05:59:47 PM

Conclusion- subs/microtrans customers will leave, cannot be put in game.

Changing MMOs will (probably) make current players leave the game, but so will cranking out World of Generic Fantasycraft XIV.  It's not a matter of "CUSTOMERS LEAVING = FAIL" so much as how many customers will leave as opposed to how many will it attract.

Features 3 and 4 are present in EVE, which has almost as many subs (over 300k) as Everquest did (over 400k), so it doesn't seem that unbelievable to me that someone will come along and give EVE the same treatment that Blizzard gave EQ.
DLRiley
Terracotta Army
Posts: 1982


Reply #8 on: July 22, 2009, 06:24:47 PM

Someone will come along, look at EvE and go...we actually want to make money. EvE will outlive LoTR, FF11, and lineage 2. If it grows to 500k subs that's when people go "hey this game is marketable", other wise its a niche game for several good reasons. One of them being its solidly hardcore simulator and wasn't even an attempt at UO2 ( i think EvE didn't want all the UO "PRE-TRAMMEL" boys migrating toward their game). Behind all the backwater bullshit in EQ, EQ was much closer to a game, though it reeks of SoE giving the LARP guys money to make an mmo. Blizzard used the EQ model vs the UO model knowing this.

And besides the only reason feature 4 doesn't make people in EvE rage quite in droves is because losing in pvp means going back to pve and considering that close to 90% of EvE sub-base are PvE'ers going back to pve isn't much a problem since they consider all the spreed sheets and mining fun anyway. As oppose to every "open world burn the city to the ground" mmo where losing in pvp means... having to lose again in pvp. The lack of something else to do in a "pvp" mmo's besides open world pvp often translates to win or quite and quite anyway because there is no one left to fight.  

And another and besides, World of Generic Fantasycraft XIV will continue to sell a million copies or so every time. Once these mmo's stop relying on big sub numbers the industry will crank out that shit every summer. Even now the rate that f2p generic fantasy games are being made is amazing. We are only a generation or two away from having a game with both a  subscription + heavy use of micro transactions. Of course that won't solve 2/3'rds of your playerbase leaving in 6 month, but the 1/3rd that stays can be milked for years.
« Last Edit: July 22, 2009, 06:34:46 PM by DLRiley »
Sogrinaugh
Terracotta Army
Posts: 176


Reply #9 on: July 13, 2010, 01:59:01 PM

(1) Improved monster AI- this makes players angry because dying is sad and they want to win most of the time and its not fair.  

Conclusion- subs/microtrans customers will leave, cannot be put in the game.  

(2) Meaningful story- this makes players angry because their choices mean they missed out on getting the best piece of healing gear for their 5-man raiding spec or they can't change their spec when BlizEA nerfs Witchaloks in PvP.  

Conclusion- subs/microtrans customers will leave, choices cannot affect gear/spec/faction in any way.

(3) Advancement methods other than grinding for exp/UO-style skill levels- players say they would love that, but no long-term sustainability as everyone is at full power in two weeks and then unsubs.

Conclusion- some months-long grind to max level/skill must be in the game.  

(4) Players actions in PvP change the PvP world- People are sad when their side loses and their city is burnt down.  

Conclusion- subs/microtrans customers will leave, cannot be put in game.

(5) Player actions in PvE change the PvE world, or PvE world has its own changing schedule- The vendor I need isn't available for business during the game "nighttime"?  The monster I want to kill doesn't drop the same loot and now I can't look at a guide to find my equipment?

Conclusion- subs/microtrans customers will leave, PvE must be Thottbot-able and unchangeable.  Same monster with same loot in same place, from now until the game shuts down.           
I agree with your premise that diku is a distillation, not a construct.

I'm not sure if Outback makes more or less then McDonalds, but i will eat their over Micky D's every single fucking time when i have the $.

MMOG's are some shitty shit.  But they are also the cheapest form of entertainment known to man (besides masterbation).  If someone made a game that was intrinsically fun (take the original Zelda or Metroid for example), mmog'd it, and not only added new spells/weapons/dungeons (i.e. "content") but continaully researched, designed, and implemented new ways to INTERACT WITH THE WORLD, and continually expanded on that world, you would now have something that wasn't deserving of a good flushing after the veneer wears off.

Most of the problems you are listing are the result of unengaginging, boring, repetative gameplay.  Think dungeons that are random (like Wizardry's from back in the day).  You can actually explore stuff, find secret passages, you won't have nerds flipping out on your stupid pointless forums when they ally with the Purple Eagle Orc tribe which prevents them from getting an Axe from the Hollow Rock Dwarves because they are busy enjoying the game rather then playing the game for a stupid "reward" - the reward should be the game itself.

I don't know about the rest of you but i hardly give a fuck about the Massive and Multi part of MMOG.  The worst thing about Zelda?  Or Metroid?  When the game ends.  Give me Zelda, slap a 50$/month tag on it, and keep making shit for me to do after i kill Ganon, and i'll suck your dick (AND give you some cheeseburgers).  Let me do it with a couple friends, dont make us play stupid "roles" that almost nobody wants to play (though I personally enjoy healing, i have come to realize most of the worlds population doesn't and it interferes with people just loggin in and playing, ditto tanking).  And don't be a pussy about putting Cool Stuff in the game that might induce teardrops from somebody somewhere.

The whole mentality of Players Can't Have Nice Stuff comes from companys not wanting to hiring adaquate staff to handle the maladjusted subhuman primates that infest their product.  Charge 50$/month, hire people, actual people with actual judgement, who ban the maggots who will cost you more subs.  If i'm pointlessly using my telekinesis spell to grab boulders and place them in front of a city gate, have a GM activate guards/archers to kill my ass (or simply ban me from the game).  Dont remove the fucking telekinesis spell because of assholes, remove the fucking assholes. 

The biggest reason these games suck so much cock is you can't do anything in them.  Nearly every single interaction is numerical, between your weapon/spell and a creatures life pool.  Why do none of these games have tombraider platform-like elements (but with dithering b/c camera angles suck)?  Have high-level NPC's run by GM's who create problems - increased Tarif's on your goods, or AH fee's.  Find ways (besides just collecting bear asses) to solve those problems, and that begins with finding out the why's.  This requires people (GM's again) doing stuff and making "content" - yes it costs money.  Make me use my water elemental to put out a forest fire for the elves.  Oh noes i guess thats a quest the rogue can't do, their might be QQ?  The rogue has quests others can't do, etc.  And only let me summon that elemental from a large body of water, and dont worry about the fucking teardrops related to BALANCE which is the bane of FUN.

Before people argue about the $ signs - consider:  If i go to a local bar, for 1 night, a shitty long island is 8$.  A shot of the good stuff is 11-14$.  Entry is usually 10$-15$.  Yes this does come with the (for me, remote) possibility of getting laid, but mostly people just go out to go out.  If you provide entertainment that is actually entertaining, and not just some doctor evil grab at the status climbing chimp in our subconscious, people will pay.
pxib
Terracotta Army
Posts: 4701


Reply #10 on: July 13, 2010, 03:04:38 PM

Thinking that they can produce quality, entertaining content that quickly (no matter how much money you throw at them) is Mythological Man Month pipe-dreaming. Content requires time more than it requires any other single resource. Deadlines do not improve it. The best art, the best craft, and the best science are ready when they're ready and not a moment before.

Worse, most great action RPGs are fun precisely because they end. When your character stops getting cool upgrades and learning new strategies, the story is over and you go play something else. That they leave you wanting more is what makes you remember them fondly.

If you played Zelda or Metroid for twelve hours a week every week for five years their gameplay dynamics would make you just as sick and tired as WoW's do today. A monthly fee for ten new maps and bosses wouldn't help (if, again, that were even possible). They wouldn't be nearly as interesting as the ten maps and bosses from some other new game that has had three to five years to innovate and perfect them.

Or the ones available in the occasional Zelda and Metroid games Nintendo is working on right now.

if at last you do succeed, never try again
vos
Terracotta Army
Posts: 60


Reply #11 on: July 13, 2010, 03:30:00 PM

Thinking that they can produce quality, entertaining content that quickly (no matter how much money you throw at them) is Mythological Man Month pipe-dreaming. Content requires time more than it requires any other single resource. Deadlines do not improve it. The best art, the best craft, and the best science are ready when they're ready and not a moment before.

To use WoW as an example, assuming they stayed within the current framework for content and didn't require additional programing, could Blizzard not hire additional artists/designers/testers to churn out more monsters/zones/dungeons/raids and therefore more content? If you are talking about the underlying engine and adding in things like phasing, or deformable terrain, then sure that applies. But couldn't they hire more people to say turn out 10 different versions of warsong gultch? Same underlying tech, just new art assets and new area design/balance....
pxib
Terracotta Army
Posts: 4701


Reply #12 on: July 13, 2010, 04:25:11 PM

To use WoW as an example, assuming they stayed within the current framework for content and didn't require additional programing, could Blizzard not hire additional artists/designers/testers to churn out more monsters/zones/dungeons/raids and therefore more content?
The idea works in theory, but in practice everything has to fit together. The artists have to work with the designers who have to work with the testers who will have to work with the programmers. Every new piece of gear introduced has to fit within a scheme that will balance with everything that's currently available. Every new instance must match the difficulty/reward and reward/time curves established elsewhere. Ditto every new outdoor zone and, to a lesser extent, new monster. Each piece of art, be it model or texture, has to match the tone of the entire rest of the game, and every individual creation must fit within the evolving storyline of what's come before, what's coming now, and what's yet to come.

On a more general management level, increasing the number of staff increases the inherent difficulty of distributing tasks to individual teams. Every group must be working at more or less the same speed lest they wind up either cramming for time or cooling their heels. The more groups there are, the more likely that at least one will run into a problem that slows them down... and slows down every other group that depends on their output. And every group that depends on their output.

It's a good old fashioned combinatorial explosion. Even if additional artists, designers, and testers manage to produce arithmetically more content, but they produce internal interactions and complications exponentially. No amount of new people or new money ever makes things less complex.

if at last you do succeed, never try again
Nebu
Terracotta Army
Posts: 17613


Reply #13 on: July 13, 2010, 04:37:46 PM

No amount of new people or new money ever makes things less complex.

I think this could be avoided by increasing the level of build redundancy with increases in the numbers of employees.  If you can create a means to add content in a modular form, then more builders would increase content without exponentially increasing problems.  The challenge comes in developing a product where each piece added would only add a limited set of potential problems.  Then your additions would add a linear degree of trouble rather than exponential?  Right? 

"Always do what is right. It will gratify half of mankind and astound the other."

-  Mark Twain
Malakili
Terracotta Army
Posts: 10596


Reply #14 on: July 13, 2010, 05:21:32 PM

No amount of new people or new money ever makes things less complex.

I think this could be avoided by increasing the level of build redundancy with increases in the numbers of employees.  If you can create a means to add content in a modular form, then more builders would increase content without exponentially increasing problems.  The challenge comes in developing a product where each piece added would only add a limited set of potential problems.  Then your additions would add a linear degree of trouble rather than exponential?  Right? 

This is more or less what they did with the Torchlight Editor.  Lots of up front work, but creating new content for it once that is done is a piece of cake (more or less).   
pxib
Terracotta Army
Posts: 4701


Reply #15 on: July 13, 2010, 05:48:46 PM

I think this could be avoided by increasing the level of build redundancy with increases in the numbers of employees.  If you can create a means to add content in a modular form, then more builders would increase content without exponentially increasing problems.  The challenge comes in developing a product where each piece added would only add a limited set of potential problems.
That's definitely the easiest way to flatten the log curve, and it's already standard practice. Good planning and smart, modular construction are (IMO) what's behind Turbine's fast and cheap development cycle on LotRO. Even WoW did it with their endless copies of the same buildings, caves, rocks, trees, and monsters. Complexity never quite goes linear, but it can remain manageable for a much longer period by following your advice.

Unfortunately, increasing staff increases complexity at a steep exponent even in non-game environments. Simply upping the number of teams involved in a project necessarily has diminishing returns in terms of development time saved. Past a certain point, adding staff can actually increase development time as they trip over each other. Unless instances and raids look a lot less like Zelda and more like Dungeon Runners, it's not going to drop down to a whole game's worth of fun every month no matter how much money you throw at it.

if at last you do succeed, never try again
Typhon
Terracotta Army
Posts: 2493


Reply #16 on: July 13, 2010, 09:09:37 PM

And before you're there (having hired way too many people all with their own personality problems and egos) it would be better to have the machine procedurally generate content (lets say zones) and hire... no word exists yet for this job so I'm going to call them "MMO critics", to say which zones are fun and which aren't.  If you're really good you have the machine try to learn what makes a zone "fun".  Machine cranks out a bunch, critics pick the top X, design guy takes a final pass through and smooths rough edges.

Yes, I know that SWG used procedurally generated content and it wasn't well received.  Doesn't mean that there will never be interesting-enough procedurally generated content.
Tarami
Terracotta Army
Posts: 1980


Reply #17 on: July 14, 2010, 12:00:50 AM

Statement: Procedual content the way it's usually meant is never going to work satisfactorily.

One reason is that art produced by humans is automatically infused with context and subtext: popular references; trends; past experiences; personal influences and collective knowledge all contribute to how we build things, and how we look at things others have built. We, as people contemporary to the designers, are going to pick up on this modern age context and form a subconscious continuity that the designer might not even have intended to include, but is there because we share a cultural heritage. I might even go so far as to say that this is where we find the true value in art: seeing who's and what's behind the creation of the piece, rather than in the piece itself.

When we're out questing and stumble across something that feels familiar - a landmark (a henge), a turn of phrase ("I have a lousy feeling about this") or a rare mob (a werewolf called Jakob) - we find satisfaction in knowing we are alike the designer and picked up on his or her subtext. The shared connection can be of varying tenuity and generally speaking, the harder the connection is to make, the more satisfying it's going to be to make. DeathSpank, for example, is completely dependent on people making these connections - people think feudal Japan is cool, so katanas exist as loot.

A machine, short of a fully-operational AI with 10+ years of personal experiences, is never going to give you real sub- or context. The chances of it happening randomly are just astronomical, even if you stick a human in there to touch things up, because subtext isn't an afterthought. It's a by-product of craft, something we inadvertantly add because we simply don't know better and it goes all the way down to how we choose to shape a cliff face or layout a village or itemize a mob.

Procedual tools have their uses, like filling in blanks that have natural regularity or as a starting point for something manually crafted. The visual pattern of lightning can be fairly easily described, for example, but a lightning bolt that was painted by hand is still going to look more interesting than a generated one, even if the latter one is technically more correct. The presence of artistry itself is vitally important for our appriciation.

The short of it is that procedual content that's able to pass as handcrafted content without significant altering is going to take longer to describe to a culturally agnostic machine than it would take to produce it by hand. That's without even going into the actual difficulty of formalizing the ideas.

- I'm giving you this one for free.
- Nothing's free in the waterworld.
Typhon
Terracotta Army
Posts: 2493


Reply #18 on: July 14, 2010, 07:30:22 AM

I'm talking about the model where artists create a bunch of reusable assets and then zone and quest designers use those assets to put together a total zone experience.  Really, I'm talking Diablo 1 taken to the next level.  Also, I just disagree that a routine can't create a map/zone that looks like it was formed via natural forces, this is actually done all the time.  I agree that a procedure hasn't been written that can create an interesting building, mob, weapon or armor.  I think you still need artists for all of that stuff.

Replace "zone designer" with "procedure" and "MMO Zone Critic".  Replace "quest designer" with "AI factions during real time".

Then, pick areas on a map that share traits that typically drive human settlement.  Use human generated art (read: buildings) to build up the settlements based upon how successful they are.  As you come into that zone, have the AI send ambassadors to the player to try to convince them to do tasks for that side.

The story will be non-existent.  The way the zone plays out over time will be interesting.

Currently we have: the story is mostly "meh", with a few bright points but if you are playing with someone else you don't really have time to read.  The zone is entirely static unless the developer has implemented phasing, in which case that will likely annoy anyone playing with someone else in a different way.  Even with phasing, you play through it once and your are done.  All the effort that went into that zone has now been "consumed".
Mrbloodworth
Terracotta Army
Posts: 15148


Reply #19 on: July 14, 2010, 08:03:05 AM

That's definitely the easiest way to flatten the log curve, and it's already standard practice. Good planning and smart, modular construction are (IMO) what's behind Turbine's fast and cheap development cycle on LotRO. Even WoW did it with their endless copies of the same buildings, caves, rocks, trees, and monsters.

Some of that is governed by pure technology limitations. Reuse of assets is key to ANY game development. So, if you can somehow make a number of objects able to be used in a variety of ways (See LOTRO Bree town, there are really only about 15 - 20 individual objects that make up the buildings, that are also used all over the world) with obscuring the repetition, you win!

Even "content" like quests are just a combinations of a handful of set functions mixed and matched in different ways. For any game.

Today's How-To: Scrambling a Thread to the Point of Incoherence in Only One Post with MrBloodworth . - schild
www.mrbloodworthproductions.com  www.amuletsbymerlin.com
Malakili
Terracotta Army
Posts: 10596


Reply #20 on: July 14, 2010, 09:36:44 AM

Statement: Procedual content the way it's usually meant is never going to work satisfactorily.

I'll mention Torchlight again, not to disprove or prove your point, just because I think its an interesting way of approaching it, and it may or may not be "procedural content the way it's usually meant."

So, they way Torchlight level randomization works is that there are level "chunks" that are hand crafted.  Any single one of them is relatively small.  Each chuck is a square, and it has a label based on where its "exits" are located (NE for a North and East exit, and so forth, with any combination of the cardinal directions).  So, when the game loads a level is says "Ok, this is supposed to be a "Lava Fortress" style level, so I'll pull chunks from that pool, and its supposed to be 10-15 chunks big, so I'll cobble together 10-15 handmade chunks into a new randomized Lava Fortress level.   (this cobbling together works something like a game of Carcassonne)

There are several advantages to this:

1) While any given level is likely to be fairly different from any other given random level, each chunk being handcrafted makes it feel much more cohesive and avoids the pitfalls of a totally random generator (pointless dead ends for example).

2) You can really easily create a new chunk for any given level type that can then be seemlessly added to any randomized dungeon of that type.  There are loads of Torchlight mods out like that in fact, that are simply "Adds 10 new chunks to the Lava Fortress tileset" or whatever.  In the case of developers instead of modders having control (as would be the case in an MMO), they could easily include new chunks that would increase the replayability and variety in level structure without having to make tons of new assets and such.

3) When you are creating entirely new level types  You have the advantage of being able to get something up and running with only a few chunks made, and you also have the ability to add on as you go without having to do a TON of work up front.


Now, this isn't your standard MMORPG (in fact its single player only at the moment), and it had lots of loading screens between levels, but it does lend itself nicely to replayability and variety, while still allowing things to feel cohesive, familiar, and hand crafted.  I full expect that Runic is using the same method to generate dungeons in the MMO version of Torchlight, so we'll see how it plays out, and I'm not sure how easily you could translate this idea into more open world MMOGs, but I think its a great way to go about it.  


Edited to add:  They also have rudimentary randomized quests that basically can be summed up as an NPC saying "Hey, go to level [A Level of the dungeon you'll be at pretty soon] and kill [A mini boss that having this quest will spawn there]" or collect an item that having this quest will spawn there.  Not the be all and end all of quest design to be sure, but you can see how that concept could be expanded upon.
« Last Edit: July 14, 2010, 09:42:38 AM by Malakili »
Mrbloodworth
Terracotta Army
Posts: 15148


Reply #21 on: July 14, 2010, 09:45:28 AM

There is a point though, where procedural content is outclassed by hand made experiences.

Today's How-To: Scrambling a Thread to the Point of Incoherence in Only One Post with MrBloodworth . - schild
www.mrbloodworthproductions.com  www.amuletsbymerlin.com
Nebu
Terracotta Army
Posts: 17613


Reply #22 on: July 14, 2010, 09:48:02 AM

There is a point though, where procedural content is outclassed by hand made experiences.

I think the days of finding hand crafted experiences in a mainstream product are over.  It's all about mass production to maximize profitability.  Can't really blame houses for this.  Making games in today's market is a big financial risk. 

"Always do what is right. It will gratify half of mankind and astound the other."

-  Mark Twain
Malakili
Terracotta Army
Posts: 10596


Reply #23 on: July 14, 2010, 09:51:00 AM

There is a point though, where procedural content is outclassed by hand made experiences.

Yes and No.  While a purely handcrafted experience is likely to be of a high quality, the long term experience of quality randomized content tends to be as good or better (Diablo 2 anyone)?
Tarami
Terracotta Army
Posts: 1980


Reply #24 on: July 14, 2010, 10:10:16 AM

I'm talking about the model where artists create a bunch of reusable assets and then zone and quest designers use those assets to put together a total zone experience.  Really, I'm talking Diablo 1 taken to the next level.  Also, I just disagree that a routine can't create a map/zone that looks like it was formed via natural forces, this is actually done all the time.  I agree that a procedure hasn't been written that can create an interesting building, mob, weapon or armor.  I think you still need artists for all of that stuff.

Replace "zone designer" with "procedure" and "MMO Zone Critic".  Replace "quest designer" with "AI factions during real time".

Then, pick areas on a map that share traits that typically drive human settlement.  Use human generated art (read: buildings) to build up the settlements based upon how successful they are.  As you come into that zone, have the AI send ambassadors to the player to try to convince them to do tasks for that side.

The story will be non-existent.  The way the zone plays out over time will be interesting.

Currently we have: the story is mostly "meh", with a few bright points but if you are playing with someone else you don't really have time to read.  The zone is entirely static unless the developer has implemented phasing, in which case that will likely annoy anyone playing with someone else in a different way.  Even with phasing, you play through it once and your are done.  All the effort that went into that zone has now been "consumed".
Bloodworth possibly already answered it - if you are referring to "building blocks", that's already what's being done. Each zone might get a couple of unique props to give it more flavour, but the vast majority is just preexisting props stamped in with a world building tool, just like how any RTS map editor works. It probably has a few simpler generators too, like one to produce a forest with a certain density that can then be modified to fit.

Diablo, NetHack et c. just clobber some prefabs together to produce quasi-random hallways (which is difficult enough,) it isn't like the dungeons actually make sense in any greater meaning (most people I know played D2 with maphack because the map generation was obnoxious) and it's that greater meaning that poses huge, huge problems for procedural content. Procedural content that can be recognized as such very quickly becomes as static as actually static content, but at a lesser build quality. There's no way you're going to write a generator that can produce Booty Bay, Stormwind and Orgrimmar dynamically in the time it takes to just build those three towns from scratch. What you get instead is SWG. SWG, however, didn't have a choice - they were populating entire planets, so they had to go with random blandness to get anything at all on the map.

The point the way I see it wouldn't be to use a generator to produce a couple of towns - that's just a waste of time, because it will take longer to write the generator than it would take to just build the towns by hand. Rather, a generator will have to be able to produce huge numbers of towns to motivate being written and that's where it all falls down - building the data model necessary to produce a huge number of convincing towns is more work than just building an even huger number of towns by hand to begin with. This is true even if you're going to have multiple stages of the towns.

This isn't to say there's no place for "dynamic" systems, but it's remains a sideshow to the handmade content if you want to build a world that actually makes sense and plays well.

Do you have any examples of games that do generate convincing procedual maps? By convincing I mean with a quality that makes it hard to discriminate between generated and handmade.

I'll mention Torchlight again, not to disprove or prove your point, just because I think its an interesting way of approaching it, and it may or may not be "procedural content the way it's usually meant."
Procedural tends to be used to describe systems which are given a small set of seed values that then give semi-deterministic results. Kind of the tiling approach you're describing, but usually much more complicated. The idea is that by building a data model of how the smallest common denominators (rooms, hallways, buildings...) interact with the seed values and eachother, you're to get "unlimited" variations that individually make sense. It tends to just result in randomness though, that ironically all looks the same.

Torchlight doesn't have to worry about making contextually weird maps. The player is there to whack foozles. If you're trying to sell the world as "real" though, it's going to hurt believability to see pattern repetition.

I think the days of finding hand crafted experiences in a mainstream product are over.  It's all about mass production to maximize profitability.  Can't really blame houses for this.  Making games in today's market is a big financial risk. 
What? I haven't seen non-handmade content in a game in ages. There's a difference between having procedural tools and relying on a generator to produce good content. The most basic map editor is procedural in some sense, it's just static once it's put into the game, which is what matters. The point is that regardless of what the original tools were, once the content is in-game there isn't anything that wasn't intended (well, hopefully. Grin) Mass Effect 2 looking samey is due to a number of technical reasons, not because someone wrote a map generator and pressed "make map."
Yes and No.  While a purely handcrafted experience is likely to be of a high quality, the long term experience of quality randomized content tends to be as good or better (Diablo 2 anyone)?
Slot machines? Meaning, you don't really have to have any variation at all for a game to be sticky if you have ding-grats. Like I said, most people I knew who played D2 disliked the random maps because it made it more difficult to get to the meat (i.e. farming Mephisto.)


LONG POST! Sorry.

- I'm giving you this one for free.
- Nothing's free in the waterworld.
Mrbloodworth
Terracotta Army
Posts: 15148


Reply #25 on: July 14, 2010, 10:24:12 AM

There is a point though, where procedural content is outclassed by hand made experiences.

I think the days of finding hand crafted experiences in a mainstream product are over.  It's all about mass production to maximize profitability.  Can't really blame houses for this.  Making games in today's market is a big financial risk.  

How many MMO currently feature procedural content? Also, just a note, but procedural content usually is constrained by a hand made template, as in a hand made example. There is also the conflict of static and dynamically placed assets.

Wurm is 100% dynamically placed server objects, even the terrain, it where most of our performance hit comes from, quite technically, before player hit the land masses, our entire world was procedurally created. I only bring it up because its somewhat relevant.
« Last Edit: July 14, 2010, 10:26:59 AM by Mrbloodworth »

Today's How-To: Scrambling a Thread to the Point of Incoherence in Only One Post with MrBloodworth . - schild
www.mrbloodworthproductions.com  www.amuletsbymerlin.com
Typhon
Terracotta Army
Posts: 2493


Reply #26 on: July 14, 2010, 11:04:51 AM

My post wasn't a derail (not that anyone is accusing me, I just want to tie what I threw out there back to the original post). 

I brought up procedurally generated content and smarter AI is because I think it has the potential to make points 1,2, 4, 5 mostly invalid.  It's very definitely pie-in-the-sky (but apparenlty there are already development shops doing some of this stuff already - thanks BloodWorth, what you mentioned Wurm doing was some of the stuff I was talking about - procedurally generated world and mob placement).

(1) Improved monster AI

Mobs (actually, faction leaders) acting with more intelligence makes for a more interesting game world.  It is necessary for all the rest of the stuff I'm babbling about.  Probably it's very hard to do and won't be done convincingly for awhile.

(2) Meaningful story. 

Players participating in a conflict between multple AI factions, and what happens to those factions based upon that participation seems meaningful.  Bad MMO fiction I can do without.  Maybe what the game would need to do is to pay a writer to play the game, then write stories about what happened in the game AFTER the fact for publication on the games website.  You could point a story and say, "heh, yeah, I participated in that".  Whether it was good or bad, the people reading it would actually be interested in reading it (as opposed to trying to play a game and being forced to read it in chunks).

(3) Advancement methods.

We've already seen leveling times come WAY, WAY down from our evil EQ overlords.  I think mentioning UO as a fast leveling game is odd.  I played as someone that thought using macros to level was... broken. I still think it's broken.  I tried to level in early UO and it was not quick in any way (ok, lag really, really hurt the game from my perspective).  EVEs leveling scheme is less broken that having to use macros to level your character. 

Also, I think the GuildWars experience model says, "yeah... if you don't want to level and just want to PvP you can do that".  I think this is your weakest point.  Yes, most games MMORPG games are diku and most are fairly tedious.  It doesn't mean that it's not changing.

(4)(5) Players actions in PvP change the PvP world/Player actions in PvE change the PvE world

The game I'm talking about has a changing world.  Player are the instruments of change (i.e. one faction should not be able to make significant gains against another faction unless players help).  It's all PvPvE, probably the hardcore shards allow you to attack eachother, where the "PvE" shards have you competing against each other via the mobs.

Mrbloodworth
Terracotta Army
Posts: 15148


Reply #27 on: July 14, 2010, 12:06:57 PM

what you mentioned Wurm doing was some of the stuff I was talking about - procedurally generated world and mob placement).

Yeah, but it brought a hole bunch of issues with it. Granted some of it could have been avoided by more rule checks, but its a once in a blue moon map generation. The largest issue is players found the first few iterations unrealistic, other iterations were not conducive to playable spaces, the finial version, depending on who you ask and what server you are talking about has to much vertical space. This created the issue of mob placement issues as well, as mobs can be placed in unreachable positions, throwing off our spawn count limit. There are 30k mobs in the world, however at least 5k are on mountain tops and other unreachable positions.

It also won't really work for other design types, ours is an open world with limited traditional "content". We have no regional habitats or historic areas in the common since (such as you would have with a world builder building the history of the area and its peoples). Our resources are regional by randomness alone.

Today's How-To: Scrambling a Thread to the Point of Incoherence in Only One Post with MrBloodworth . - schild
www.mrbloodworthproductions.com  www.amuletsbymerlin.com
Malakili
Terracotta Army
Posts: 10596


Reply #28 on: July 14, 2010, 12:27:46 PM

what you mentioned Wurm doing was some of the stuff I was talking about - procedurally generated world and mob placement).

Yeah, but it brought a hole bunch of issues with it. Granted some of it could have been avoided by more rule checks, but its a once in a blue moon map generation. The largest issue is players found the first few iterations unrealistic, other iterations were not conducive to playable spaces, the finial version, depending on who you ask and what server you are talking about has to much vertical space. This created the issue of mob placement issues as well, as mobs can be placed in unreachable positions, throwing off our spawn count limit. There are 30k mobs in the world, however at least 5k are on mountain tops and other unreachable positions.

It also won't really work for other design types, ours is an open world with limited traditional "content". We have no regional habitats or historic areas in the common since (such as you would have with a world builder building the history of the area and its peoples). Our resources are regional by randomness alone.

This sort of my makes me wonder if you you could take the system a step further and create some sort of method for change in the game world (I think Love claims to do this but I didn't play it long enough to see it in action, especially with the frequent server restarts in beta).   Mountains slowly erode, plants and animals spread through reproduction algorithms, etc.    It would be interesting to then, after the world has been created (procedurally or not), have it that simulation run for a long time with those given rules, THEN introduce players into the environment (that is ideally still as "alive", but of course not changing at a much slower pace).
Mrbloodworth
Terracotta Army
Posts: 15148


Reply #29 on: July 14, 2010, 12:35:18 PM

I take it you haven't played our game :)

Erosion (dirt flows down, rock does not). Check.
Plant spreading and reproduction. Check.
Mob/animal reproduction. Check.

We even have an entire skill line for breading traits in animals. Diseases even.

Today's How-To: Scrambling a Thread to the Point of Incoherence in Only One Post with MrBloodworth . - schild
www.mrbloodworthproductions.com  www.amuletsbymerlin.com
Malakili
Terracotta Army
Posts: 10596


Reply #30 on: July 14, 2010, 12:51:05 PM

I take it you haven't played our game :)

Erosion (dirt flows down, rock does not). Check.
Plant spreading and reproduction. Check.
Mob/animal reproduction. Check.

We even have an entire skill line for breading traits in animals. Diseases even.


Yeah, I knew you had the plants and animals thing, I have played Wurm a little, but I wasn't sure about erosion and other sorts of terraforming (not done by the players at least).  I know Love does stuff with this as well (wind that can be used to power generators, for instance).  I really should try that game again now that it is live and isn't going to be resetting every day to see how noticeable that stuff is.  I should also give Wurm another try too, though every time I do give it a shot never stick with it long enough to get past the initial "just survive" stuff.
ezrast
Terracotta Army
Posts: 2125


WWW
Reply #31 on: July 14, 2010, 02:27:00 PM

most people I knew who played D2 disliked the random maps because it made it more difficult to get to the meat
Those people are a menace to society and should be put down.

Unrelated: My take on procedural vs. handcrafted content is that, assuming a finite budget, any game has to choose where it falls on the spectrum of unique/artsy/handcrafted vs. different every time you play. A lot of people around here (or maybe the same few people over and over in multiple threads? I dunno) seem to have this pipe dream of magical content dev teams with infinite time that can have lots of both. But realistically, any game has to choose one or the other, or fall somewhere in between. For some games and genres being mostly or entirely handcrafted is appropriate, but in a game where my primary focus is to whack one thousand foozles so that I can whack my next thousand foozles 5% more quickly, any charm that comes from handcrafted areas is going to wear off somewhere around my 200th foozle anyway, and I just want something different.
pxib
Terracotta Army
Posts: 4701


Reply #32 on: July 14, 2010, 02:55:05 PM

For some games and genres being mostly or entirely handcrafted is appropriate, but in a game where my primary focus is to whack one thousand foozles so that I can whack my next thousand foozles 5% more quickly, any charm that comes from handcrafted areas is going to wear off somewhere around my 200th foozle anyway, and I just want something different.
This is an excellent point. Unfortunately, I'm unsure that random layouts can provide that difference alone. I've played a lot of Diablo 2 (random with occasional hand-crafted set pieces), and I've played a lot of Titan Quest (entirely hand-crafted), and while I got bored of the second a bit quicker than I got bored of the first, I blame that on their fundamental similarity rather than any benefit of the D2's random mazes. If anything it was kind of comforting to know where I needed to go and where I didn't the second and third and fourth time through TQ. In contrast, I didn't finish either Fate or Torchlight largely because their quests and setpieces were underwhelming and the character classes played so similarly. There just wasn't enough to look forward to in evolving gameplay, visuals, or narrative.

The same thirty chunks of maze arranged a gazillion ways doesn't feel like a gazillion different dungeons. Sometimes it doesn't even feel like thirty.

Then again, an alt-aholic like myself might get some additional enjoyment out of a bunch of random mix-and-match introductory areas punctuated by familiar, hand-crafted challenges. Nothing gets boring more quickly than endlessly replaying the introductory chunk of an RPG. YMMV

if at last you do succeed, never try again
Malakili
Terracotta Army
Posts: 10596


Reply #33 on: July 14, 2010, 03:46:46 PM


The same thirty chunks of maze arranged a gazillion ways doesn't feel like a gazillion different dungeons. Sometimes it doesn't even feel like thirty.


Its more about the fact that I never know precisely where to go.  I like the exploration side of things quite a lot and randomization means that even if things look a bit the same, the layout is always a bit different.  By comparison, something like a Dungeon in WoW is really neat that way the very first time, and then only if its a full group of people who have never done it.  Soon though, it becomes just going through the motions.

I know you could say the same thing about Mephisto runs or Pindle runs, and I've done my share of those in my day, but there is still something enticing to me about that fact that its a little different every time, even though its familiar. 

Maybe I'm the odd one out though.
pxib
Terracotta Army
Posts: 4701


Reply #34 on: July 14, 2010, 04:11:10 PM

Its more about the fact that I never know precisely where to go.  I like the exploration side of things quite a lot and randomization means that even if things look a bit the same, the layout is always a bit different.
I agree that our opinions differ, and I'm not sure yours is the minority. It's probably the case that we're each willing to tolerate quite a bit of the other's preference so long as the game keeps handing out rewards.

I don't need to know precisely where to go, and I'm not (generally) trying to maximize efficiency. Having choices is a lot of fun, but I can't stand feeling like I've wasted time. If, in D2, I wandered to the opposite corner of the map when the exit was just around the proverbial corner at the start... it got hard to justify the tedious trip with experience, gold, and loot. That wasn't enjoyable, it was frustrating. Torchlight seemed to have put some effort into making its random levels a bit more linear. More importantly, I enjoyed D2's set pieces and boss fights every time. It was the long meanderings between them that eventually drove me away.

This was also true in Titan Quest.

Like Triforcer says in the original post: grinding levels and loot isn't exactly fun... it's a skinner box. It's also the best way to keep people paying for a game for month after month. Disguising the skinner box won't change its fundamental nature, but improving the quality of the skinner box might change its profitability.

if at last you do succeed, never try again
Pages: [1] 2 Go Up Print 
f13.net  |  f13.net General Forums  |  The Gaming Graveyard  |  Game Design/Development  |  Topic: What if what we have now is all there is? (or, Dikus are the Steering Wheel)  
Jump to:  

Powered by SMF 1.1.10 | SMF © 2006-2009, Simple Machines LLC