Arcanaville

I code in my head, so when you'd start the clock is iffy on me. Keep in mind I've also written actual code for calculators that, while nowhere near the complexity or even direct intent of a game engine, gives me reference points I might not otherwise have.

I also have a nearly complete data dictionary which few people would be able to start with. I think the time spent figuring out what to do, separate from the mechanics of data dumping, really ought to count. Otherwise we'd be talking about a hypothetical person that knows exactly what to do and can just type out code like they are banging out the dictionary. I don't know about anyone else, but much of the time I spend "programming" is either time thinking about algorithms and implementation techniques while nowhere near a keyboard, and fast iteration of code that I write, and rewrite, and rewrite, and rewrite.

To be honest, though, counting hours is mostly quibbling. The question is whether the depth and complexity of the problem is such that it could be done in any small number of weekend efforts. The question is more about pace than anything else. Can what you do in any number of hours on a weekend even move the needle much at all.

And even an attribmod calculator, that only deals with what happens at a single moment in time for all possible sets of attribmods, requires a lot of complexity. Just getting resistance right is something I think most people are going to get wrong the first time, as trivial as that sounds on its face.

One other thing: my estimates were for C, where its harder to iterate code and you'd need to build up your own data structures (or use the appropriate libraries). My best guestimate for how long it would take me to code the engine in python, where I'm faster and can iterate code quicker, based on laying out the design last night, is about 900 hours.

I had the notion to use a project like this to teach me LUA. Its been on my mind even before Dr. Aeon's revelation LUA was spliced into the game engine. That would, of course, make it take a whole lot longer.

PS: writing the engine in Objective-C would also be a very interesting experiment, albeit mostly for the potential for meta architectural novelty than anything else.

Just like on every game. Here, you can do that on solo play, but shard play will be just as hard to do that as anywhere else.

Switching between trusted servers, just like here, will preserve everything. Trying to upload solo single player stuff into shards will be subject to caps if you aren't a trusted node. Will that "not fly?" I don't know, but I don't see why not, given the fact that you can't allow people to have unrestricted solo play and still import that uncontested into shared worlds. If this is something you're really concerned about, you just won't play in the pure stand alone mode. That won't preempt you from playing *solo* it will just be that while solo you'll be logged into an actual authenticated shard.


The same place it comes from for all other non-subscription MMOs. Selling the game client, and selling MTX and DLC updates. And, if you can make it work, the cut from the content store.

If you count all the time thinking about it, I'm probably close to the 3000 hour mark already.

Nobody said anything about a "weekend worth" of coding:

As far as I'm concerned, you may spend your weekends in any way you choose. However, most people consider the weekend to be Saturday and Sunday (unless you live in Israel) and its entirely your choice how many of the hours on those two days you choose to spend on this particular endeavor as opposed to all other activities including eating, sleeping, and interacting with other human beings.

What business speak?

That shouldn't take an economics degree to parse.

At least I got to see this "opinion" go from being wrong, to provably wrong, to completely moot.

When the datamining says the vast majority of players had far more difficulty playing blasters than all other archetypes, when the numbers objectively demonstrated the archetype was an immense outlier on the offense/survivability curve, and when the devs stated that the archetype needed help and made changes designed to help it, twice, saying yeah, but in spite of that I just know they were fine, is actually almost sad. Mostly goofy, but there's a little sadness in there.

The first, and sometimes last thing I notice about most of the other MMOs I've played since CoH is that I can't get over that hill over there.

No matter that I understand the design and gameplay issues, the fact that I can't fly over a hill I could climb in real life is usually the first two strikes against a new MMO for me.

When I first acquired fly, I just flew around. Literally, for a couple days I spent most of my playtime just flying around the zones and shooting things from the air.

Elsewhere, games are designed without damage types. Here, that statement is about as wrong as saying all attacks autohit, some less often than others.

I'm not spending very much time there while the official forums are still alive, but I do have accounts on unleashed and titan. I'll probably be around for at least a while there, beyond whatever happens here.

Umm, that article says Vivendi was looking to sell its stake in Activision, not that Activision was trying to sell WoW. And it was having trouble finding takers because Activision/Blizzard is worth too much money.

I haven't heard any news of Activision trying to unload Blizzard. Blizzard accounts for a third of Activision's revenue last I checked, and is still enormously cash-flow positive. That would be a brain dead move.

Vivendi is looking to unload because it has a lot of debt and wants the cash. Nobody is trying to dump WoW anywhere just yet.

Actually, I've been thinking about this in general for a long time: the Paragon shutdown simply brought the idea into focus.

But I believe a lesson to learn from looking at many different MMOs including CoH is that community building is not about shoving people into a crowded map. We have strong communities in CoH even though many of those people never see each other directly in-game. Even though we have fast travel and don't hang around the shared instances often, even though our content is highly instanced. Our communities live on the forums and in global chat and in coalitions and SGs. Our communities are meta-communities, not literal in-game clustetrs of avatars. Conversely, the meta tools for communities was much weaker on TOR, and all the shoving people into the same zone did nothing to build communities.

Its the meta tools for building communities that's important in my opinion. If you have strong community building tools and gameplay mechanics that let people easily play together with different thresholds of participation, it doesn't matter if they are not all running into each other in a shared (virtual) physical space. You can encourage people to play with each other without forcing them to, in my opinion. I don't consider the fact that not everyone will zone into the same Atlas Park after the tutorial to be a problem.

Consider how other games deal with the issue of multiplayer interaction and how we even started to deal with this: teaming queues. If you want to team, queue up for a team. Meanwhile, play solo until the queue opens. Why does that even need to exist if everyone is on a shared server anyway? Because that's not enough.

I think, to be frank, most people do not want to socialize in MMOs, at least initially. Most people are not extroverts, and its intimidating to jump into an established world. I think the reason WoW is so successful is in large part because its the Amway of MMOs: they not only reached a critical mass of players, they became a "thing" - something people played and felt comfortable sharing with their friends. People signed up because their friends were playing so they were not alone. No one wants to be alone while surrounded by a million other people. WoW solved the problem of how to attract casual MMO players: attract so much momentum players dragged their friends to play it.

There's not likely going to be another WoW, and even if there might be one day, you can't bet on being the one to make it. So I think this is an important issue to explore. Most MMOs don't: they can claim to be targeting the "casual player" but they still throw everyone into the deep end of the pool. Because they believe, as you apparently do, that that is important to powering the community.

I would want to try something different. I would want to allow people to play completely by themselves if they wished, but give them the tools to slowly opt into larger communities. As I said, I would start with using global chat. Solo or not, I would allow players to link into the global chat systems of the game. They could start off as lurkers, hearing but unable to speak, to see what the different subcommunities were like. The hope would be that they would find subcommunities they liked and were willing to participate in, by joining in the discussion. And that would lead to impromptu teaming, and eventually larger scale participation for many of those players.

It would be a grand social experiment, and it might fail. But in that respect, that would be no different than any other MMO released in the last eight years.

At some point, if and when the end seems certain, I'll have to say my adeius as well. When I do, I intend to point out that as visible and continuous as my presence as been, the community has always benefited from players willing to contribute to improving the gameplay of other players. There's a continuum from BuffyASummers tanker and scrapper guides to ZombieMan's guides of guides. Lots of people contributed to defense and tohit understanding from Amauros' SR guides to IronVixen (and many others) Invincibility testing to Stargazer and Pennelope's snowballs to my own guides. I'm embarrassed to say I probably don't remember everyone anymore although I'm trying to.

Even now we still have the legacy of RedTomax and Iakona and Mids. There are people who still use Brawl Indexes. Arcanatime and several other contributions of mine can trace their lineage back eight years to the Powers Quantification Project, which was my first encounter with TopDoc.

It would be false modesty to say my contributions haven't been large, but its not entirely clear what they would have been if there wasn't a community of generally cooperative and helpful players surrounding me. I don't do this for any other MMO I play.

The truth is if it wasn't for Snipefu's and Powerthrust's Energy and /Energy guides, I probably wouldn't be interested in the numbers at all. If it wasn't for the Brawl Index and the PQP, I probably wouldn't be interested in measuring the numbers at all. If it wasn't for the Tanker and especially Scrapper forums' unique character, I probably wouldn't have become a quant, much less the most visible one.

I only did what I did because I wanted to, and I wanted to because it seemed to be the thing to do around here. And that's what I will really miss most.

I think you lost everyone under 30 when you got to the curtains.

Should take about a weekend.

Did a lot of hiking at 17, did you?

He got temp banned for leaking I9 Hami data, but I think he got over it. He was still around in I10 but it may have contributed to his desire to move on.

And they aren't gonna. The fact of the matter is no one outside the parties involved is going to know anything until its all over. Anyone claiming to both know otherwise and be able say otherwise is almost certainly completely full of crap.

Its enough of a given that people who want to do things for the experience of doing things do not need to be addressed with game design mechanics explicitly, to make it generally unimportant to mention it as a disclaimer. The question is in what way will the game systems including the reward system support that game play priority. Given that normally moot disclaimer, I stand by my previous post as addressing that question by default.

Either we'll see, or we won't.

What was said at the time was that they had a massive spreadsheet that allowed them to adjust the balance for the game globally, by changing a few variables that would rebalance all the powers and gear everywhere, so that's why the launch nerf was so widespread: a global rebalancing decision made lots of changes all over the place to rebalance the game.

Not specifically to pick on whoever said that (I forget) because our devs have said some inexplicable things regarding the game's design as well, but there's no way that's true. Either its false and the developer who said it knew it was false, or its false and the developer who said it thinks its true. I'm not sure which possibility is worse.

This is easier to say than to do, but I believe the way out of that box is to make the primary reward for doing something being given the opportunity to do more of it in some way. No one who doesn't like to do it would farm for the opportunity to do more of it, and everyone who likes doing it would be getting a reward they were guaranteed to want.

That's separate from more generic progressional rewards which require multiple ways to earn them to avoid having to do only one specific set of things repeatedly.

When it comes to things like collectables, its important to make sure that if you do not intend for the players to collect all of them you make that impossible by structure. So for example suppose you don't want players to go nuts farming every single zone for reputation. You make a badge slot for something like "Home Zone" and a single badge slots into it based on which zone you spend the most time in. Token collectors can satisfy their completeness itch by making sure all such slots are filled, even though they do not, because they can not, possess every single possible option for that slot simultaneously.

This is a psychological, but important thing. If there are five, but the game restricts you to only one, the missing four will be an annoyance. But if there is a single slot, and multiple ways to fill it, there are no "missing" things the game is preventing you from having. Simple user interface design decisions can go a long way to presenting the gameplay in a much more enjoyable manner.

By mutual agreement, most of the people (I'm aware of) that were capable of seeing that data elected to show player data and not critter data, due to the potential for critter data to be used for exploitive purposes or to spoil the intent of certain content. For example, before I told the players what the LRSF towers did, I tested them myself to prove that knowledge was capable of being determined in-game by a player, and even so I left one thing out until the devs granted me permission to mention it - the momentary intangible buff that allowed Recluse to break toggle debuffs.

The difficult parts of making a real MMO engine are all the optimizations involved in making working map servers, real time network code and event queues, etc. Things I have no current interest in doing. Replicating the game engine would be, for me, an intellectual challenge to see how well I actually understand it. And part of the reason for the large time estimate is that I'm pretty certain there are interesting problems in the way the engine does things that I won't really understand fully into I try to implement that in code. In a real project, debugging and reimplementing code to account for problems represents a large fraction of the total time. The time to program something is not identical to the amount of time it takes to type the final source code.

I know I spent a couple of weeks, off and on, just musing to myself how the game engine must work for unresistable resistance debuffs to work they way they do and for Combat Training: Offensive's accuracy buff to work as it does. When I posted my results on ranged damage metrics in the beta fast snipe thread a legitimate question was whether I had properly accounted for sonic blast buffing its own DoT with -Res - which I had not, because my calculator (deliberately, due to time constraints) did not deal with effect over time specifically in that manner. But that has a very significant impact on Sonic's numbers.

Having written damage calculators, damage mitigation simulators, and other discrete calculators gives me a particular perspective on this topic. The sheer amount of stuff the game engine does is enormous compared to the amount of time it takes to account for each effect.


Also, it was only on average about 28% more damage except in some degenerate cases.

The intricacies of real-time system programming, mostly.

Dual box sometime. Then ask how its possible to see what you're seeing and still have a sane game.

Effects can have a Requirement in order to take effect, for example TargetType = player (for PvP effects) or source.kMeter > 0 (stalker hidden). These are entered into the power data as postfix expressions. It takes an integrated script processing engine to understand what "source.kMeter > 0" means, and it implies you either have to translate that into another concept on the fly programmatically, or you have to add entity tracking code that knows what kMeter's value is on the source of the power at the time its activated. Even the magnitude of the power is sometimes an expression: (source.TeamSize(30) * 0.03 + 0.07 >= rand()) * @StdResult -> the Assassin's strike non-hidden PvE critical.

If you don't know these things even exist, I'm not sure how you can be so certain about your estimates for how long it would take to reproduce all of the combat engine's behavior.

Technically, there's only one thing to do:

I never said anything about combinations of effects. If you think I mean combinations when I say the engine does hundreds of things, you don't understand what the game engine does. It does hundreds of discrete, singular, computationally different things. Magnitude resistance is different from Duration resistance: they require different code because the formulas are totally different. MaxMax buffs are totally different from Strength buffs. Zero duration effects are handled differently from non-zero duration effects. Extend stacking is different from conventional stacking. Cur is different from Abs. (Base) Defense is a special case of Attack type in the attack/defense type summation code. There are hundreds of different things in the engine, not dozens of things in hundreds of combinations. Hundreds of different things that require completely different code because they cannot be procedurally summarized.

This goes beyond the scope of what I was talking about in the OP, but if I was writing Super Powered MMO X that wasn't beholden to any of the City of Heroes mechanics or gameplay, it would almost certainly be levelless, but that would be because the rewards system I would use would make that practically a non-issue.

I too feel if you want to fight crime in Kings Row forever, you should be able to. But why would you want to? And the answer, for me, is related to what rewards combat grants. Right now, CoH and nearly all MMOs, even GW2, are wedded to the notion that progress is linear, and combat advances the needle. But if you're still going to have linear progress - XP - are you getting the full benefit of being levelless?

What if the reward for fighting crime in Kings Row was earning reputation as a crime-fighter in Kings Row? In a sense, Kings Row would be a meta-contact, and fighting crime in KR would increase your reputation with Kings Row itself. You'd get missions and other benefits other players didn't, for being a well-known KR crime fighter.

You could even expand this to be more finely grained. You could earn reputation for being a Lost-hunter in Kings Row, a Skulls smasher, a Hellion wrecker. It would be like the progress bars for defeat badges, but more open ended and with a gameplay purpose.

This wouldn't just be window dressing either: Kings Row progress would be fundamentally different from Skyway progress or Dark Astoria progress.

In other words, the way I would make the game levelless is not by eliminating combat modifiers and exemplar code. It would be by eliminating the notion of linear leveling. Without it, there's nothing to level. This would extend all the way to character progress. Fighting street thugs in Kings Row would be really good for improving martial fighting skills, shooting at things, etc. Not as good for learning other kinds of skill. But that part requires a lot more thought and discussion into the precise way combat and skills would work.