Thoughts on Hail Mary, one year later

[Drauger9]

Honestly I don't think NCSoft would ever reopen City of Heroes. They might eventually sale it to another company but never reopen it.

Even if they did, for me atleast. It'd have to be free to play with a sub option. As it was before it closed. I just don't trust them enough anymore to pay a sub up front and risk losing all that money. When the wind could blow the wrong way one day and have them decide to shut it down again. I think if they ever did win the players back, just to shut it down again a few years later. They'd have generation after generation of gamers, not willing to invest in them at all.

They burned a lot of people with this one. I'm sure old CoH players would go back if they re-opened but they'd do it hisatently and with leary eyes on NCSoft at all times for awhile.

I'd love to have City of Heroes back in any form but for now. I'm looking forward to the three projects it's closure has spawned. I don't expect them to be literal copies of City of Heroes. The only thing I hope for from them are their version of Master Minds, a modern setting, good community (which shouldn't be hard to accomplish with all the fine folks here) and the ability to farm (hopefully in instanced missions ).

[Arcana]

Matt's a funny guy.

Still, no game studio would want to inherit City of Heroes' codebase.  In fact, I think you're more likely to be able to make sense of City of Heroes design if you *aren't* a game developer than if you are one.  If you were already a game developer, it would probably be difficult to even see the code through all of the tears.

[Segev]

That's alright. I'll just program a neural network to learn the function represented by it. How hard could it be?





(*cough*  )

[Codewalker]

Still, no game studio would want to inherit City of Heroes' codebase. In fact, I think you're more likely to be able to make sense of City of Heroes design if you *aren't* a game developer than if you are one.

That's probably true, but mostly because game studios tend to suffer from worse "Not Invented Here" syndrome than any other software developers.

Even Oracle, who's one of the worst.

[blacksly]

Still, no game studio would want to inherit City of Heroes' codebase.  In fact, I think you're more likely to be able to make sense of City of Heroes design if you *aren't* a game developer than if you are one.  If you were already a game developer, it would probably be difficult to even see the code through all of the tears.

I COULD see one wanting the statistics tables, though, if you were trying to reproduce the game. Running them through a mapping function so they would be appropriate for your new code would be easy, and you'd end up with all of the CoH powers and mobs with statistics coded.

[LaughingAlex]

Honestly it's not as hard a game to replicate when it comes to the combat mechanics, the damage mechanics are actually simple, but yeah, CoX's code was very much a spheghetti mess.

I mean I ponder how much code was wasted when all we had were locational/damage type defense roles followed by resistance subtracting from damage on an actual hit.  It's simpler than CO's combat mechanics which go through about 20 layers that are all especially buggy.

[Arcana]

That's probably true, but mostly because game studios tend to suffer from worse "Not Invented Here" syndrome than any other software developers.

Even Oracle, who's one of the worst.

Perhaps, but CoH's design and implementation is still an outlier among outliers.  On a scale of one to ten, its implementation rates a 3+2i.

[LaughingAlex]

Honestly I don't think NCSoft would ever reopen City of Heroes. They might eventually sale it to another company but never reopen it.

Even if they did, for me atleast. It'd have to be free to play with a sub option. As it was before it closed. I just don't trust them enough anymore to pay a sub up front and risk losing all that money. When the wind could blow the wrong way one day and have them decide to shut it down again. I think if they ever did win the players back, just to shut it down again a few years later. They'd have generation after generation of gamers, not willing to invest in them at all.

They burned a lot of people with this one. I'm sure old CoH players would go back if they re-opened but they'd do it hisatently and with leary eyes on NCSoft at all times for awhile.

I'd love to have City of Heroes back in any form but for now. I'm looking forward to the three projects it's closure has spawned. I don't expect them to be literal copies of City of Heroes. The only thing I hope for from them are their version of Master Minds, a modern setting, good community (which shouldn't be hard to accomplish with all the fine folks here) and the ability to farm (hopefully in instanced missions ).

not a fan of mob farming much but I do want support in the games to not be all healing healing healing myself.  It's an easy trap to fall into even not intending to.  Which is why I often advocate making damage mitigating support very powerful, because if it's not, healing ends up becoming a be-all-end-all.

[Arcana]

Honestly it's not as hard a game to replicate when it comes to the combat mechanics, the damage mechanics are actually simple, but yeah, CoX's code was very much a spheghetti mess.

I mean I ponder how much code was wasted when all we had were locational/damage type defense roles followed by resistance subtracting from damage on an actual hit.  It's simpler than CO's combat mechanics which go through about 20 layers that are all especially buggy.

Its not quite that simple, but CoH combat mechanics are very well understood by the player community.  In and of themselves they would not be difficult to reimplement in another game.  The only real challenge is doing so in the most efficient manner possible.  CoH's implementation was not as efficient as possible.  I have often wondered, for example, if an implementation exists that would have allowed those mega two hundred player Hami raids to run at full speed.  I believe there is.

[thunderforce]

Honestly it's not as hard a game to replicate when it comes to the combat mechanics, the damage mechanics are actually simple, but yeah, CoX's code was very much a spheghetti mess.

I'm amazed at how well-informed people are on this subject given that none of them have ever seen a line of it.

[Arcana]

I'm amazed at how well-informed people are on this subject given that none of them have ever seen a line of it.

I suspect most people are just assuming from past dev statements.

[Ironwolf]

Because some players went on to work with the devs...........just saying.

[JaguarX]

I suspect most people are just assuming from past dev statements.

Yeah mostly from dev statements that I seen.

Hey, everything else they say or said was taken without question. I figured, they would know the code more than anyone and if they say it's nuts, I believe them.

Then again, a lot of stuff they said was impossible, no way, never,  suddenly became it's coming, new feature, it's here, normality.


Although still I wonder how did the code become or would have become a complete mess anyways? No one documented their work, basic filing, logical building of code and organization, or it once was there and something happened to it and eventually those that built it and or knew it end and out got out numbered by new people coming in as the old ones left? Or something happened back in the day that was unintended but was a PITA to fix and they kind of limped along with it as is and made it work as much as possible? 

I figured that for a product that is supposed to or meant to be upgraded and last for years would have an SOP so the new people can pick up on it as the old one leave and make upgrades as time go on without much hassle. At first I thought that maybe something happened or something and it got messed up but then I heard CO code is just as messy if not more so. So it sounds more like a certain group of people's modus operandi and ability with that engine.

[Arcana]

Although still I wonder how did the code become or would have become a complete mess anyways?

The same way most code does, I would imagine.  Although I personally wouldn't say the code is a mess, its more that it has (had) twelve years of design and implementation ideas accumulated within it, many if not most of which were formulated for a technological landscape that no longer existed.

For example there's a lot of weirdness in the code that only makes sense if you realize that RAM used to cost about $100 for a 64MB stick when City of Heroes was first being implemented, and today that same $100 buys you about 8GB of RAM - a factor of about a hundred times more memory.

Also, going Microsoft for everything is a more understandable decision in 2002 than it would be even a couple years later.  Although I wouldn't have.

The biggest weirdness is that the client and the server look like that scene in the Thing where the Thing is still in the middle of a transformation, half one thing and half the other.  The designer mindset appeared to be to think of an MMO as a single player game that has some of the engine ripped halfway out and made multiuser.  It is not as clearly separated into client and server as good design principles would otherwise recommend.

(On the other hand, Icon probably relies heavily on that original bad design decision, so there's that.)


True story about the codebase: the devs hired outside contractors to improve certain parts of the code around the time of and just before CoH:Freedom, and one of the things they broke was gravity.  For the last five-ish issues of City of Heroes, gravity did not work correctly.  Specifically in the physics engine that controlled things like knocked entities.  And no one had any idea how to fix it.  That's why knockback became insane damage mitigation: you'd knock something down, and it would often just keep bouncing on the ground over and over again until it died.

How you get paid for destroying a world's gravity I'll never understand.

[JaguarX]

The same way most code does, I would imagine.  Although I personally wouldn't say the code is a mess, its more that it has (had) twelve years of design and implementation ideas accumulated within it, many if not most of which were formulated for a technological landscape that no longer existed.

For example there's a lot of weirdness in the code that only makes sense if you realize that RAM used to cost about $100 for a 64MB stick when City of Heroes was first being implemented, and today that same $100 buys you about 8GB of RAM - a factor of about a hundred times more memory.

Also, going Microsoft for everything is a more understandable decision in 2002 than it would be even a couple years later.  Although I wouldn't have.

The biggest weirdness is that the client and the server look like that scene in the Thing where the Thing is still in the middle of a transformation, half one thing and half the other.  The designer mindset appeared to be to think of an MMO as a single player game that has some of the engine ripped halfway out and made multiuser.  It is not as clearly separated into client and server as good design principles would otherwise recommend.

(On the other hand, Icon probably relies heavily on that original bad design decision, so there's that.)


True story about the codebase: the devs hired outside contractors to improve certain parts of the code around the time of and just before CoH:Freedom, and one of the things they broke was gravity.  For the last five-ish issues of City of Heroes, gravity did not work correctly.  Specifically in the physics engine that controlled things like knocked entities.  And no one had any idea how to fix it.  That's why knockback became insane damage mitigation: you'd knock something down, and it would often just keep bouncing on the ground over and over again until it died.

How you get paid for destroying a world's gravity I'll never understand.

Makes sense.

And I just watched the Thing, the new and old version and got a clear picture of that analogy. Not pretty.

But seems something good came out that design, Icon.

But I do wonder this now, off subject, if mmos were are supposedly now being designed to be strictly multiplayer, then how hard is it now or in the future for games to say be released in single player mode instead of being shut down or else playing on potentially hack private server to find a good hearted private server of said game that comes after a few years that may or may not be around itself in another 1-3 years after it goes up due to various reason (i.e. private server owner doesn't pay bill or kick bucket)

And how the hell did they break gravity?

I always thought something was fishy about the knockback thing in that time but thought maybe I took too long of a break and don't remember.

[thunderforce]

I suspect most people are just assuming from past dev statements.

That's what I actually think is true, yes; I'm just - well, not surprised, but disappointed - by how much a few joking asides by devs have been turned into Gospel truth.

[Ironwolf]

Again actual PLAYERS were hired by the Devs at time - Arcana was one of them and there were several others.

I think it was Backasswards who was also hired. A couple others fessed up in the Forums after the closure announcements. So the code was a custom made engine that had a bunch of limitations they managed to kind of fudge around to make things happen.

I recall from the start them saying the engine did not allow custom power colors - then it DID allow them. It appears they made quite a few hack-arounds that went around the original engines perimeters.

[Arcana]

I recall from the start them saying the engine did not allow custom power colors - then it DID allow them. It appears they made quite a few hack-arounds that went around the original engines perimeters.

In a certain sense, nothing is genuinely impossible with software in the sense that you can always rewrite the software.  But if you didn't actually know the implementation details, it was basically impossible to simply guess at what the difficulty of a change was.  No player without first hand knowledge of some kind ever guessed anything right about the game implementation in all the time I was keeping score on the forums.  This was so absolute that whenever someone said something even close to true I knew they either had inside knowledge or was tinkering with the game on the side or both.

Custom power colors was one of those things that sounded simple but was a gigantic undertaking.  Here's what you needed to do to make that work, starting from the original game:

1.  Redo all the power FX art assests to allow tinting.
2.  Add tint information to the data channel sent to the game client
3.  Add power customization data to the character database
4.  Add tint settings to power customization database schema
5.  Add User interface to adjust color tint to powers

So, allowing players to change the color of their fire blasts involved changing the back end databases, the communications code, the server code, the game client, the character creation user interface, and redo a few hundred art assets.  If none of those things have been done yet and the studio had never before tackled a project involving even one quarter the scope, in the days of the Cryptic 15 you'd have rightly called that task - just to tint some blasts - basically impossible.

What was "hard" and what was "easy" sometimes had nothing to do with on-paper technical difficulty and more to do with what was simply things people said yes to and things people said no to.  For example, when I was asked to work on the Architect custom critter system, I designed a valuation system that had three tunable parameters, alpha, beta, and gamma.  However, by the time I was asked to work on it some preliminary work had already been done based on informal conversations I had with the devs way back in I14, and the fixed tables for those tunable parameters had already been written into the code for I18.  Alpha, and beta.  No gamma.  When I asked for that fixed data table to be extended by one more column I was told if I needed it the entire thing would be scrapped as undoable, because they didn't have the resources to make that change available at that time.  So I refactored my system to work with two variables instead.  I also had a proposed change rejected because it required adding a button to the AE custom critter screen.  A button.


I learned an awful lot about what the game could and could not do in the last thirty days, because I managed to acquire admin rights to the Beta server - that's how we were able to pull off that Immortal Game event there.  A *lot* of trial and error testing (and more than one multi-zone crash).  Hypothetically speaking, I may have granted that access to a few other players that made better use of that information than I was in a position to do so.