So what's Mass Market Again? Conan hits 1M (shipped).

[Falconeer]

It doesn't sound like a sign of bad health for AoC though. Which is kinda surprising, because due to incoming summer, bugs and bad performances I was under the impression that subscriptions were bleeding at an alarming rate. The f13 thread seems to confirm this quitting trend.

Now maybe this is just a "cunning" press stunt (or a symptom of more technical inadequacy), but took me by surprise.


EDIT: This video doesn't look like bad health too.

[Trippy]

I don't think this means just that, it could very well be that they looked at the total number of character's and then looked at the typical average for concurrent connections at peak time (estimated or potential) and said...no more.

Looking at the total number of characters is meaningless for estimating peak concurrency unless the game allows you to play multiple characters from the same account at the same time.

[Venkman]

It's probably just a database thing. Each character gets alloted a certain amount of unique items or whatever. Queue all of the early SWG armchair DB designer discussions. I don't know anything about this stuff, except to guess that in a sharded world, there's eventually a hard cap of some type.

[Mrbloodworth]

I don't think this means just that, it could very well be that they looked at the total number of character's and then looked at the typical average for concurrent connections at peak time (estimated or potential) and said...no more.

Looking at the total number of characters is meaningless for estimating peak concurrency unless the game allows you to play multiple characters from the same account at the same time.

You are assuming that the largest portion of characters are alts. We don't know this to be true, or false.

[Trippy]

It doesn't matter either way.

They already know their peak concurrency rate on that server. They already know the number of accounts on that server so they know the ratio (e.g. if accounts on that server is increasing by X a day then peak concurrency will probably go up by Y). Factoring in number of characters on those accounts is totally useless in estimating peak concurrency at some point in the future.

[Oban]

When was the last time you heard of this happening just for character creation? I.e. not talking about taking things down to upgrade hardware to support more simultaneous users or reduce the existing server lag.

SWG

[Trippy]

Was this to actually allow more characters into the database or to support more concurrent users? SWG was an exception because of their one character one server rule so in that situation characters == accounts.

[Miasma]

I assume it's either alts or some large PvP guild coming in all at once to give it a try.  As you get higher you need more bank mules if you are storing gems, alchemy supplies and other resources.

On a side note I actually took gemcutting yesterday and discovered that all the time I spent methodically banking gems, putting them in order by uncut/flawless, level and type was completely worthless.  To level gemcutting all you need is three of any gem for that tier.  Since virtually no one would wear crafted gear outside of level 70's and 80's there is no point to saving these gems as you level.  Just save three and then destroy the rest.  There is also a bug which means to get credit for each cut you have to zone or log out/in each time.

[Threash]

I assume it's either alts or some large PvP guild coming in all at once to give it a try.  As you get higher you need more bank mules if you are storing gems, alchemy supplies and other resources.

On a side note I actually took gemcutting yesterday and discovered that all the time I spent methodically banking gems, putting them in order by uncut/flawless, level and type was completely worthless.  To level gemcutting all you need is three of any gem for that tier.  Since virtually no one would wear crafted gear outside of level 70's and 80's there is no point to saving these gems as you level.  Just save three and then destroy the rest.  There is also a bug which means to get credit for each cut you have to zone or log out/in each time.

Actually thats very very wrong.  Some gems do not improve as they level so those low level ones are just as good as the level 80 gems.  Immunities, evasion, offhand chance, are all the same at every level,  i think the only thing that scales are the +dmg +magic and proc gems.  What is worthless though are the uncut gems, i doubt anyone actually bothers with green crafted gear.

[sam, an eggplant]

Female DPS bug not fixed after all, will take weeks to fix.

Ouch.

[Threash]

How hard would it be to analyse what the difference in dps is and then apply a temporary dps boost to females to balance things out?

[sam, an eggplant]

Well it's not just straight damage, many of the moves also have various status effects associated with them as well like attack debuffs, damage over time, stamina regen, etc. So even if you added a "if female damage = damage*1.25" the players would still be missing out. It's pretty bad.

[cevik]

I chose to think that the female dps bug is a grand social experiment by Funcom to show us how far Women's Liberation has come, and yet, how far it has left to go.

[Oban]

I chose to think that the female dps bug is a grand social experiment by Funcom to show us how far Women's Liberation has come, and yet, how far it has left to go.

Or, mangina deterrence.

 

[tazelbain]

Nice to know that you could hack the client and get infinite dps.

[Lantyssa]

How can every animation favor males?  Shouldn't there be some averaging going on?

[cevik]

How can every animation favor males?  Shouldn't there be some averaging going on?

Get back in the kitchen and bake us a pie, the men are talking.

[Lakov_Sanite]

just make it so only women could play female av's and then no one would notice since girls are such horrible gamers anyways.










 

[sam, an eggplant]

They spent too much CPU time modeling the movement of magnificent, soft, pillowy, pendulous, hypnotizing breastages which left little resources to handle secondary stuff like attack animations.

But really, my guess would be that nobody really thought to look at it, and figured that animation speeds would all even out in the end as they weren't trying to make females slower. But it didn't work out that way.

[Mrbloodworth]

How can every animation favor males?  Shouldn't there be some averaging going on?

Normally they would be the same rig, and share animations (or perpend the call to use a more female version, still the same rig though), this seems not to be the case based on the description of the problem. Its an animation issue, not a "scaling/playback speed" animation problem.

The only thing i don't get, is why the animation has anything to do with the damage taking effect. They should, in most cases be separate. Seems the length of the animation is used to determine when damage is done.

Perhaps its due to the more real time nature of the game. Basing it off the animation would at least ensure that on screen when something "hits" damage is done then. Not before, after, or in spite of.

Thing is (what confuses me), animation length time is normally just a variable in the code, not really triggered by the end of playback.

 

[Oban]

Maybe if they added larger cod pieces to the male models it would all balance out.

[tmp]

The only thing i don't get, is why the animation has anything to do with the damage taking effect. They should, in most cases be separate. Seems the length of the animation is used to determine when damage is done.

Wouldn't this be simply result of not being able to execute another weapon swing until the current one is finished (and that's determined by length of animation)? So the longer each swing takes, the less can be made in given slice of time... it seems like pretty normal approach, even for MMOs.

edit: though given that, i'm surprised the animators weren't given exact times each attack animation should take, to begin with...

[Tarami]

What I'm not getting is why they can't fix it, atleast temporarily, as it's interpolation animation via bones. They are just keyframes with interpolation that can be run at any speed, even if playing it too fast will make it look a little jerky. Every game ever tweaks the animation speeds in code, after the actual rigging has been done. Surely some visual anomalies are quite okay compared to a pretty huge bug?

Overall I find it really hard to see why they've used animation triggers rather than timers. If you want amazing animation you fix the animations to fit the game design, not the other way around.

[sam, an eggplant]

Surely some visual anomalies are quite okay compared to a pretty huge bug?

Assuming you're correct in your assumption as to how the client was coded, the answer is still no. Even the theorycrafters and powergamers found the female DPS disparity difficult to quantify. Only players reading the forums know about it. Everybody would notice stuttering animations.

[cevik]

Every time I run through all the possibilities in my head, I come to the conclusion that there is no logical explanation.

Thus I suspect that a design decision was made by someone that this is Conan, and it's a mature mmog and people can just handle the fact that women are weaker and fight worse than men. 

When someone found the "bug" it sparked a flurry of internal conversation where someone eventually said "Okay you fucking crack smoking assholes, this is going to not only cause a huge backlash with the player base, when people start saying 'LF2M for Cistern, must be male', but it's also going to cause a fucking explosion in the media.  Put down the fucking pipe and fix this shit now."

I'm willing to listen to other possible explanations, but I just can't come up with anything that makes sense.

[Lakov_Sanite]

Every time I run through all the possibilities in my head, I come to the conclusion that there is no logical explanation.

Thus I suspect that a design decision was made by someone that this is Conan, and it's a mature mmog and people can just handle the fact that women are weaker and fight worse than men. 

When someone found the "bug" it sparked a flurry of internal conversation where someone eventually said "Okay you fucking crack smoking assholes, this is going to not only cause a huge backlash with the player base, when people start saying 'LF2M for Cistern, must be male', but it's also going to cause a fucking explosion in the media.  Put down the fucking pipe and fix this shit now."

I'm willing to listen to other possible explanations, but I just can't come up with anything that makes sense.

If I am to believe any game designers to be his stupid, than all hope for humanity is truly lost and we should all be hoarding guns and canned food.

[cevik]

If I am to believe any game designers to be his stupid, than all hope for humanity is truly lost and we should all be hoarding guns and canned food.

I know, it frightens me too.

But any other explanation requires too much suspension of disbelief for my tastes.  I pray that someday I'll read a postmortem on this issue on Gamasutra.

[Tarami]

Surely some visual anomalies are quite okay compared to a pretty huge bug?

Assuming you're correct in your assumption as to how the client was coded, the answer is still no. Even the theorycrafters and powergamers found the female DPS disparity difficult to quantify. Only players reading the forums know about it. Everybody would notice stuttering animations.

Not jerkiness as such, but rather that some animations played a little faster than others. It would seem a little unnatural at most.

It definitiely qualifies as a "What the fuck were they thinking"-bug in either case.

[UnSub]

How can every animation favor males?  Shouldn't there be some averaging going on?

Animations that advantage females include bouncing, jiggling and same-sex kissing. These animations cause male players around the female avatars to drop fat loot straight into the females' inventories.

[Tarami]

http://www.tentonhammer.com/node/37257

Some details regarding the bug from one of the devs.

Edit:
That post makes it sound like there's some formula behind the combos that calculates the up-front damage based on multiple, intrinsic properties of the animation. As a developer myself, it sounds absolutely insane. They should learn about MVC or any decent 3-layer abstraction model. Data is not Logic is not Interface. Jesus.

[Lantyssa]

The only thing I can think of is when designing them the animator thought, "Hmmm.  Let's make the guys swing forcefully, so that is a quick blow.  I'll make women have a little more flourish and style to their swing," without realizing damage would be directly tied to animation length.  So the animations were consistent within each gender, however the guys programing damage somehow assumed all the animations were equivilant.

That someone couldn't catch this is mind-boggling, but I just don't see any other way there could be a consistant decrease in females' damage otherwise.

[Margalis]

Yeah it sounds like total amateur hour. How the hell can you program your own 3D engine and not get this sort of stuff right?

I might understand if animation length made it so that you could swing less often for lower total damage output, but making damage based off the length of the animation itself is crazy.

It reminds me of the days when 2D fighting game fans bitched that the hit detection was performed by boxes instead of pixel by pixel. Pixel by pixel sounds great until you realize that every time you want to change move properties you have to redraw the entire set of frames.

[Mrbloodworth]

To the animation conversation, i want to add something i didn't think about/forgot before.

All animations are mocaped. Meaning no shared rig.

Also:

The only thing I can think of is when designing them the animator thought, "Hmmm.  Let's make the guys swing forcefully, so that is a quick blow.  I'll make women have a little more flourish and style to their swing," without realizing damage would be directly tied to animation length.  So the animations were consistent within each gender, however the guys programing damage somehow assumed all the animations were equivilant.

Thats a very good, and logical assumption, i could see how this could have easily happened.

[cevik]

Damage is probably only indirectly tied to animation length, i.e. the client will not allow a new attack to occur until the previous attack completes it's animation cycle.  The server only checks for minimum time between attacks and doesn't care if the time between attacks is longer than that minimum.  Thus damage is tied to animation length, but only because 100 damage / 1.5 seconds vs 100 damage / 1.8 seconds decreases your dps.  If the damage was normalized over the animation length (if the algorithm took the length of the animation into account) then there wouldn't be as much of a problem.  For example, if the male did 100 damage on his combo and the animation was 1.5 seconds, the female should do 120 on her 1.8 second combo and they'd have equal dps.  This would still lead to problems though, as one would better at burst damage than the other (i.e. if all battles last 2 seconds, the female would out damage the male). 

And I suppose that animation length is encoded in the animation data itself, thus without outside verification (which was eventually performed by the theorycrafters) no one noticed the lengths of the animations being different.  One would have to assume that the designer told the animator the exact length needed for all the animations, if that's the case then the animator must have taken those specifications as "rough guidelines". ;)  Having worked with animators in the past I can almost see something like this happening.

The other option is that the designer didn't realize the importance of the length of the animation, thus only gave rough guidelines to the animator.  This would mean the designer is a total idiot, so I kinda don't like to go too far down this train of thought.

Or we come back around full circle to the fact that the game was intentionally sexist.  I like this theory best, but only in hopes that it will force Funcom to one day prove me wrong and tell us the full story!

EDIT:  One huge assumption I keep seeing people make is that Female animation times are wrong, Male animation times are right.  What if the opposite is true?  The "rebalancing" part of the fix could be to speed up all female attacks but then lower the damage of all combos to match the previous female dps.

[Mrbloodworth]

The other option is that the designer didn't realize the importance of the length of the animation, thus only gave rough guidelines to the animator.  This would mean the designer is a total idiot, so I kinda don't like to go too far down this train of thought.

HAY HAY HAY there pal!

The PROGRAMMER could have simply made the arbitrary choice to make his timing off the length of the animations as well.

Any one of the departments could be at fault here, animation, design, or programming. We all know when you design something, its not necessarily how the programmer implements it, and we all know that when you animate something, chances are, you got the wrong specs, OR the programmer could have just assumed all animations were the same length ... ETC...

I know when i design something, 9 times out of 10, its not working right because of the programmer and his great ideas and shortcuts, because they never care about look and FEEL of the system, just working, or not.