Arcanaville

Given enough time, anything is theoretically possible. For some value of "enough time."

Of course, players tend to mark time with stop watches, and the devs tend to mark time with the color of the leaves on the trees.

I think there's a fundamental difference between being picked by the devs explicitly for closed beta and being one of the randomly selected people in one of the random waves. I think if you are explicitly picked, the checkbox is not an impediment because technically an invite directed to you personally is not a mass mailing. Castle can invite me into anything he wants (or rather, put my name into the beta selection) and even if I've explicitly told NCSoft not to contact me, that invite is still going to come through**.

But the checkbox may impede your ability to be randomly selected into a closed beta wave.

Its also something they might only periodically honor: they may take extracts from the subscriber database periodically to use by the system that generates closed beta invites, and if the checkbox is unchecked you aren't in the extract. But if you've been put into the closed beta database at any time prior, it doesn't matter what the checkbox is set to, because you're already there. Unless they periodically purge that database, and I know if they do its not often. Either way, I think there's lots of possibilities that would make the statement true, albeit somewhat garbled through intermediaries.


** Although its also possible people don't get the email by malfunction. I've always had the checkbox checked, and yet you'd be surprised how often my beta invite gets screwed up. I've sometimes learned I was in a closed beta when a dev asked me a question about it, and I didn't even know it had started yet. Sure enough, account works, no email.

The devs can change their minds. However, if there are "no plans" to add enhancement slots now, even if someone changed their minds tomorrow there's essentially zero chance of them showing up in I19. Such fundamental changes to character powers would probably have to run a pretty significant internal gauntlet of discussion and testing.

These things happen. They happened before, and they'll happen again. If they don't make sense to you, its best to try to not overthink them.

"Balancing" not "rebalancing." In this context Tim is just saying that by adding inherent fitness, the devs have made us a bit more powerful relative to the PvE content. That's something that has to be watched carefully to make sure it doesn't change the reward-earning curves of the game too dramatically. I doubt it will, but its something I'm sure the devs will be watching very closely.

Next you're going to say that my Team: Paragon t-shirt is a fake. What about my autographed Matt Miller bobblehead, huh? Do *you* have one of those? I didn't think so.

You're just lucky I don't have Castle smite you. Don't think he won't do it, either. Two years ago I had a scuffle with a poster named PeterV and I asked Castle to smite him. If it wasn't for the fact his aim was really off that week... but I digress.

Technically, War Witch said she couldn't comment on personnel changes: she didn't confirm the Post's assertion that BaB was no longer there.

But its absolutely certain that BaB is no longer with Paragon Studios, and currently employed with another game developer. I can confirm that first-hand. There's no point speculating on that.

As to who the lead animator is now, it would be another animator. I know there were other animators brought on board a while ago. They probably haven't been formally introduced because that person does no work with customer-facing contact required unlike, say, Dr. Aeon. And Paragon Studios devs are not required to take forum red names or interact with the playerbase unless they voluntarily do so, except in the rare cases its required for their job (even then, they are allowed to communicate through the community reps, as has happened in the past).

The article didn't say where BaB went. Where BaB went has been old news on the forums for quite some time, and War Witch did not explicitly comment on that subject either, in accordance with corporate policy.

As to the rumor of Positron being shuffled around that the Post referred to, that's not a rumor I myself have heard up till now. But I have four separate reasons to doubt it, none of which I could articulate.

I am the BASF of City of Heroes. I don't invent a lot of the things in this game. I make a lot of the things in this game better.


Technically, I didn't invent Defense. I invented Elusivity.

International accounting practices basically mandate accruing subscription revenue when earned - i.e. when the subscription period starts or completes, depending on billing. Its extremely unlikely accounting practices for the game changed at any point in time because there would be charges for that, and an almost certain detectable change in the numbers due to that. Plus, no reason to do so. And NCSoft buying the IP didn't actually change their situation much from an accounting perspective: they were always the publisher and collected the subscription revenue. Buying the IP meant they were now footing the bill for development directly, and no longer paying Cryptic for anything, both of which would have affected costs, not revenue.

Things like add ons and perk packs are a bit trickier to account for, but its very likely they have no impact on a first order estimation for Going Rogue's contribution because we've had several consecutive quarters with add on packs which likely have statistically comparable revenue generation. In fact, the error bars due to attempting to reverse generate subscriber numbers from the published login numbers from the past (which are our primary means of extrapolation) are significantly higher than the reasonable estimate for add on pack variances, at least in my estimation.


One of the oddities that I'm trying to figure out is that the data suggests very few players prepurchased Going Rogue. I wouldn't have guessed that. Also, I'm almost hoping less than 100% of the existing playerbase bought Going Rogue, as weird as that sounds.

There's no denying it, but at the moment there's also no quantifying it. Such anecdotal observations have been used in the past to amake all sorts of positive and negative projections of subscriber numbers that subsequent data proved wildly incorrect.


My point was that I can foresee an accounting scenario whereby new customers, who might statistically be more likely to be retail customers, might actually be invisible to the Q3 numbers. Which means, barring a more sophisticated analysis, we could have anything from a hundred new subs to a hundred *thousand* new subs, and the Q3 numbers would be almost exactly the same. Which makes it dangerous to extrapolate Q3 without careful thought. I'm currently trying to think about whether there even exists an actual analysis methodology that might expose subscriber information better.

Its really, really hard to say. The range of possibilities is anything from Going Rogue doubling the subscriber base to growing it by exactly zero (and I mean *literally* exactly zero by calculations).

I've only just had time tonight to look at Q3 and think about the numbers. Its unfortunately something I don't want to post calculations on until I've had some time to think about and analyze further, because any calculations I release now would be alarmist. A trivial calculation I can imagine someone doing on the back of an envelope would actually show zero subscriber gain: just every existing subscriber buying GR, and no one else. That would be bad.

But this is much more complicated than it seems. For one thing, I'm thinking about realized revenue. Consider this: suppose someone bought the boxed edition from a retail outlet, and not the NCSoft store directly. Typical mark up for a game is like 50%, very roughly, so lets go with that. So when someone buys a $29.99 copy of GR from a store, we might expect NCSoft to get about $15 of revenue from that sale.

The catch: GR comes with a free month. Subscription revenue is, to the best of my knowledge, realized as it is earned, which means even if I pay for a year in advance, my revenue is prorated across the year I'm subscribed. So logically, if I were to buy the GR boxed set and apply the code, I would get a free month, which means the $12-$15 that NCSoft *would* have realized from me would be deferred to next month. So they realize $15 from my purchase of GR, but lose about $15 of revenue they would have realized from my subscriptions, which makes that sale revenue neutral on the books on an accrual basis. In other words, I actually added nothing to the bottom line. Or maybe only a buck or two. At least, that's what I think is likely to happen, given the few minutes of thought I've been able to put to it.

That makes the ratio of people who purchased from the NCSoft store verses people who purchased from retail outlets potentially critical to properly interpreting Q3. Or we can all wait for Q4, when this uncertainty will probably resolve itself.

It might not be until the weekend when I've had a chance to really think about what the numbers are saying. At the moment all we know is NCSoft made a bunch of money on Going Rogue. Whether they made it on existing subscribers or grew the subscriber base substantially the numbers currently aren't saying clearly yet.

Unless someone else beats me to it, I should be able to post that tomorrow night. I'm going to be a bit tied up until then.

Another combination that has some promise (although I haven't done all of the relevant calculations and testing yet) is KM/Fire, due to the unique mechanics of Fiery Embrace.

It is: critters typically burn 1.35x the endurance that players do when using comparable powers.

However, higher ranked critters also have larger endurance bars, so on a percentage basis critters of higher rank than minion burn less of their bar per power relative to players.

The circle is now complete.

It can't be adapted into an MMO (I never said it couldn't be made into a single player game) because:

a) The advantage/disadvantage system is logically incompatible with the free choice of combat situations intrinsic to virtually all combat-based MMOs. The same reason makes it difficult to design a reward system around custom AE missions when players can engineer critters with irrelevant strengths and targeted weaknesses. Good luck fixing that.

b) By its own admission its not balanced when the system is used in an unrestricted fashion, and human judgment isn't really achievable with current technology when it comes to build decisions as complex as that offered by the HERO system. Just the mitigation powers alone highlight the problem: most versions of the HERO system I've seen going all the way back to the original Champions rulebooks warn GMs to be very careful about allowing players to stack defenses; its considered an optional place for the GM to prohibit player activity that the rules otherwise say is legal (and Champions Online amazingly fell into the exact same trap when they designed their own totally different system and originally allowed players to stack every single defensive power). So the system itself says to use judgment that a computer won't have in an MMO or you'll have game-breaking situations arising. QED.

c) The HERO system has known one-dimensional combat optimizers, particularly surrounding the mechanics of killing attacks. Difficult to get around, but not impossible in a game managed by a human being that can explicitly craft encounters that sidestep those known overoptimizations. Impossible for a computer to do so in a multiplayer game where the computer would have to account for multi-dimensional optimization among many interacting players.

d) The HERO system fails to account for non-linear stacking, even separate from the defensive powers. This is a deal breaker.


This is not to say that a points-based combat system is impossible. Actually, I've thought about how to construct one of those that would work within the confines of an MMO for quite some time. In fact, even within the confines of *this* MMO. However, starting with the HERO system is a non-starter. You'd be better off starting from scratch.

You could say that's just an opinion, and everyone is free to disagree. However, it comes on the heels of a lot of actual work analyzing points-based powers frameworks. Its not just a guess. Its an extremely informed judgment I'd put up against any actual game designer that wants to debate the point.


For the record, I'm not just convinced about what a points-based HERO-like system *cannot* be, I've also looked at it enough to have a pretty good idea of what it *has to be*. A points-based powers system for any combat-based MMO with foreseeable technology would have to incorporate four features that are, I believe, non-negotiable:

1. Non-situationally manipulatable disadvantages. The only such system I'm aware of are primarily opportunity-cost disdvantage systems. Which means, in effect: exclusive skill trees. Actual disadvantages are extremely difficult to make work.

2. Limited proportional returns. Aka the misnamed "diminishing returns" effect. Points must be calibrated to deliver the same proportional value or less across the entire system. This means no accelerating mechanics, and extremely strong controls on multiplicative effects on critical balance-significant metrics.

3. Provable offensive/defensive balance. Alternatively, multiple points currency for defense vs offense. Bonus trick issue: powers other than direct damage dealers and passive defenses which offer both mitigation and offense would need to be handled under such a system.

4. Randomized combat. Predictable combat such as that offered by City of Heroes, where every attack does the exact same damage every time its used in the same circumstances (as opposed to PnP systems where attacks do Nd6 damage or other randomized effects) is too gameable in a pure points-based power generation system. It allows for discrete optimizations that are different from statistical average ones that hide within the combat system. An example that actually occurs in CoX: farming builds designed with *just enough* AoE damage to defeat minions and/or LTs of a particular critter type in farming missions, so that they maximize return on damage: they don't require more shots than necessary, and don't do overkill damage that doesn't generate a return.


If someone that demonstrate that *any* of these four is not necessary for a points-based power generation combat system, I would love to hear it. I doubt anyone is going to be able to do so, however.

I did a bunch of analysis on Power Siphon during the I18 beta, but oopsie I forgot to save the threads.

The gist of it was that without too much difficulty you could average about +55% to +65% damage buff from Power Siphon, and that included the downtime of the power, when slotted with SOs (recharge) and using any reasonable attack chain (this was for Scrappers). This wasn't a computed estimate, but an actual measured estimate based on combat logging.

That's actually fairly consistent with assuming that you'll have about the peak buff for 20 seconds, and be up about a third of the time. Why this tends to be true is that so long as you have enough attacks to be attacking fairly consistently, you will stack the buff enough to get at or near the buff cap, and the only difference between a slow attack chain and a fast one is the ramp up time to get to the peak buff. Almost no one has an attack chain that is so bad that (if they have enough attacks at all) it takes longer than ten seconds to emit five attacks, so the ramp up should reach the five buff limit within the duration of a single buff stack.

Ramping up slowly doesn't hurt because if the buffs ramp upward slowly, they will also tend to decay downward slowly with about the same stagger. Something many people forget to consider is that while the Power Siphon power itself has a 20 second duration, it enables 10 second duration buffs. The last buff you get will linger for up to 10 seconds, so you'll actually have close to 30 seconds of at least some buffing. The ramp up to full roughly cancels the ramp downward to zero, and you end up with something close to 20 seconds of 5-stacked buffs (excluding missing).

So you'd guess that the buff would be something close to 5 stacks (+156.25% for scrappers, +125% for brutes and tankers) for 20 seconds, with approximately 60 seconds of cycle time (slotted with recharge SOs). And that is about +52% average buff. That agrees well with the measurements I took that were between +55% and +65% (the higher numbers included concentrated strike, which can insta-recharge Power Siphon and give it a better availability).

The biggest thing that will reduce the effectiveness of Power Siphon, besides being inactive, is being tohit debuffed: the buff only happens if your attacks hit. In that respect, its vulnerable in the same way that Follow Up and Soul Drain are.

For comparison, Soul Drain's peak peformance is to saturate the buff at +150% (again, for scrappers) for 30 seconds, with 60 seconds of up time. That is +75% average buff. Conversely, to beat Power Siphon with Soul Drain you have to average hitting more than 6 targets every time SD is recharged, like clockwork. *Waiting* for six to be available after Soul Drain recharges would reduce the efficiency of SD. You don't have that problem with Power Siphon, because you only need a stream of single targets to keep it saturated.

Stacked Follow Up has the same peak performance as Soul Drain (+75% damage when stacked, and it can be perma-stacked with enough recharge), and can be generated against single targets like Power Siphon.

Still, I think for Scrappers Power Siphon has the potential to reach or possibly exceed them both due to the potential benefit of Concentrated Strike. But only at higher levels of recharge than I tested with.

My alts make me. Although for a multitude of reasons, I'm probably the worst possible person to address this question.

This is an odd persistent myth. Critters use endurance in exactly, precisely the same way we do. Powers that have endurance cost burn that endurance upon use, and if the critter lacks that level of endurance the power cannot be used.

In particular, critters recover endurance with the same basic mechanism, which means they recover endurance in ticks of 6.67%. So a minion with a 100 point endurance bar will only recover endurance in 6.67 point jumps. If you are trying to drain them with drain but without -recovery, you won't be able to prevent them from getting at least an occasional tick of endurance. If they have attacks that need less than 6.67 endurance, they will be able to use them and will probably fire them off right away. But they will be stopped from using powers with higher costs if you can keep them to only one tick. Even with the 200 point bar of Bosses, holding them to 13.33 points or less does mean you'll tend to negate their heavy hitting powers and AoEs. AVs, on the other hand, are virtually impossible to inactivate without heavy -recovery, because a single recovery tick on an AV is enough endurance to use just about any power in existence (~53 points of endurance).

No, that would be the "regeneration should have more regen and less heals, because it's called Re-gen-er-a-shun, not Clik-Heal-ing" posts.


He collects those. He prints them out and stuffs them into a pillow and then every quarter he gets a stick and turns it into a pinata. The author of the first PM to fall out after Castle's sustained blindfolded swinging gets a free perk pack code. The rest are incinerated. The PMs, I mean, not the players. Although I've never actually clarified that, and subscribership does tend to dip a bit in September.

Actually, there's a discernible reason. Jack is a conceptual designer. Whether you like the concepts or not, his strength is conceptual design. He's not an implementer, or a refiner, and has, honestly, almost no real mechanical feel for the game systems implementing his concepts.

So his strengths are launching games, not managing them. It was my theory very early on that prior to the Atari buyout Cryptic's strategy was to become an assembly line MMO developer, which plays directly into both Jack's strengths and his interests. He thinks he can be, or rather is trying to be, the Johnny Appleseed of MMOs. And actually, with three titles under his belt successful *enough* that all three are atill in operation, and at least one more on the way, it's not like he's failed in that endeavor.

Personally, I wouldn't let Jack balance a checkbook much less design the details on an MMO engine, but he has a proven track record in being able to get games out the door in good enough shape to keep the lights on for the long haul. It's not what most MMO studios aim for, but it's more success than most people can point to in this industry.


By the way: would we have a City of Heroes without Jack? Based on what I know, no. What COH looked like in the Rick Dakan era would have been, i believe, frankly beyond the ability of the developers to launch. It was out of the *current* ability of our development team to get right, and I don't say that lightly (it's also out of the range of what I believe to be the ability of the current Cryptic team to be able to get right also). The quantitative design tools to pull it off did not, and still do not exist in either team, and the resources required to build a suitable game around it were probably too high. It was too "open" of a design framework for a dev team that to put it bluntly took months to figure out smoke grenade was broken by a power of ten. It would have required more numerical skill than that possessed by the team that didn't know instantly that allowing players to stack every defensive power simultaneously might be a bad idea.**

By focusing the design on deliverable targets, COH became realizable. I honestly think that simplification was as much done to make the game simpler for the players as to make it simpler for the developers, but the one unanswered question I never really had the chance to ask Jack was whether he realized that the COH alpha design was just beyond the ability of the team to pull together. I have a sneaking suspicion that he did, and archetypes were as much for them as for us.

For all the mistakes Jack made, i believe he made the most critical decisions correctly, given the resources available to them, and that's why we even have a game to complain about. In that sense, Jack and the launch team deserve a lot of credit, especially since they had no real prior experience to guide them in what to focus on.


** beta iteration of that other superhero genre MMO that only testers would have seen

I should point out that among the melee defense sets, there is no true "resistance focused powerset." At one time, there was a genuinely Defense-focused powerset: SR. Its still primarily only defense, but the scaling resistances blur the point. At one time, Ice Armor was almost all Defense, but not quite. And until not long ago, Electric Armor was the "all resistance" set, but it now has a significant heal (significant enough to blur the point the greater degree than the scaling resistances do for SR). Separate from that, however, *none* of the powersets traditionally called "resistance sets" are resistance only: all of them have either sizable defense, healing, or regeneration in sufficient magnitude to make them only partially focused on resistance. That includes Dark Armor, Invulnerability, Fiery Aura, and Granite (stone) Armor.

And I've found that there are many situations where Dark Armor beats soft-capped SR unless SR also has either aid self or highly invention-boosted regeneration. Soft-capping alone is great, especially in high-order debuffing situations (barring cascade failure) but resistance + healing tends to beat soft-capping alone so long as its not recharge-debuffed to the point of nullifying the heal (on the other hand, soft-capping plus healing tends to be better most of the time, outside of cascade failure and tohit buffs, or ultra-spike damage situations where you can be essentially one-shotted without resistances).

My experience tends to be that Dark Armor isn't often killed directly: it either runs out of endurance, gets recharge-debuffed to the point Dark Regeneration is taken away, or gets tohit debuffed to the point that DR can't hit anything. Then its killed.