Arcanaville

Actually, we have many versions of Super Strength in the game. Fifty levels of tanker super strength, fifty levels of brute super strength, and a lot of versions the critters have. Statesman has a level of Super Strength that the players really don't possess.

There are two definitions of "Tanker" in the actual game. There is the conceptual one, which is vague, and there is the game mechanical one which is not, and which ultimately dictates the strength of powers. The unambiguous one that actually affects our gaming lives most of the time is that a CoX tanker is something that trades offense for defense. Scrappers, meanwhile, are things that have balanced offense and defense.

You can't compare Spiderman to the Hulk and ask which one is the tanker, because you're comparing things of such massively different scale of power that its a worthless comparison. Its no different than comparing a level one tanker to a level 50 scrapper and asking which is the better tanker. The real question in this case would be, if Spiderman was scaled up to be as tough as the Hulk, which would then be offensively stronger? My guess is that they would be equally offensively strong. And that means - in CoX terms - they are either both scrappers or both tankers in archetypal design.

In fact, given that the Hulk is much more often portrayed as the irresistible force rather than the immovable object, I think the Hulk's concept skews much more towards high offense than high defense, even though his defense is obviously formidable. Which means even though he is harder to kill than almost anything else, he's still a scrapper. Just like a level 50 scrapper with five billion worth of inventions that outperforms 95% of all the tankers in the game is also just a very strong scrapper.

In the actual genre, there is a separate opinion skew that scrappers are simply weaker overall than "tanker-ish" things, but that belief doesn't translate to CoX: CoX doesn't honor that belief because it doesn't define scrappers as a second-class powerlevel type.

Its actually very difficult to find comic book representations of CoX tankers as the game actually implements the strength of tankers. So much so that I really consider CoX tankers to be an invention of the game, and not an homage to a comic book genre element. Some of the conceptual connections to tankers that the devs try to honor are comic book staples, like the Super Strength concept itself. But not its actual in-game strength.

I am not an employee of NCSoft, and therefore not specifically subject to that rule.


Big or small, its a request you will not get a red name to answer. And if you aren't interested in being informed, and choosing to question whether I'm even qualified to inform you, then that's cool. I cannot, and will not, persuade you otherwise.

The middle ground is where gameplay enters into the equation. Suppose there turns out to be a way to create an AE mission where, just by entering it and standing still, you could automatically earn the exact average XP/min for your level, continuously, for ever minute you remained in the mission. Your cap wouldn't affect this exploit, and you're also suggesting that the reason *why* your cap wouldn't affect this exploit is because it really doesn't matter so long as the earning rate is not above average.

Exploits aren't necessarily about a specific level of earning ability. Its a relative comparison between reward rate and activity - what is sometimes colloquially and misleadingly referred to as "risk/reward."

Put it this way: the devs would probably not care if they datamined that a particular player was earning rewards twice as fast as the average player. They *would* care if it turned out that the average of all Fire Blasters was earning 50% more than the average of all players, even though the difference was much lower. The reason is that in the first case, there is a presumption that the higher earning rate was related to a higher skill level in playing the game. That might be true, it might be false, but they are willing to presume that unless they have evidence to the contrary. But there is an equal presumption that not all players that play fire blasters could be significantly more skilled than all other players in general, so that would represent a clear imbalance in the powerset, the main difference between the two groups. And averaged across a large number of players, fifty percent is a very large gap.

Balancing rewards with the activities that produce them is a highly imperfect science, and one the devs do not have precision tools to craft. But they still have a general sense of what should earn more, and what should earn less rewards. Exploitive imbalances aren't just a question of people exceeding a singular limit. More likely, its equally a question of breaking the grey, imperfect, but still important activity/reward balances.

Making a mission that can level you twenty times faster than average is clearly over the top. But a mission that can level you twice as fast as normal while doing nothing is much more exploitive than a mission that requires a significant amount of activity that can level you *three* times as fast as normal.

In effect, saying there's only a singular maximum reward limit, and all levels of reward below that must be fine, is erasing all gameplay distinctions in earning rewards. Its essentially taking the "game" out of the game.


There's one more practical problem with caps. If exploitable content still exists and the caps are actually doing something, there's always the alt-switching workaround. If we set the maximum limit of the AE to, say, 80% of the average XP earning rate per hour, but there's a mission that can earn that 80% in 10 minutes, all we have to do is run it, switch to an alt, run it again, switch to a third alt, rotating between six alts, and we can exceed the AE cap by a factor of six, earning (plus or minus overhead) a net 4.8 times the average per hour, collectively across multiple alts.

The bigger the window, the worse this problem gets. But set the time window too short, and you run into burst XP problems. Just killing a boss or AV could get capped. And there's no sweet spot that's obvious to me, rather there's actual overlap between too short and too long, such that in between the two are windows that suffer *both* problems rather than neither.

The original version of MoG had a -90% max health and +75% res (technically, slightly lower, but slottable to the cap if you wanted) which did not balance out. That was eventually corrected to -75% health and +75% res which did basically balance out, except for psi. Separate from that, it had a heal to full and negated regeneration. The regen you lost was stronger than the heal to full was eventually, but there was a break even point. On top of all of this, MoG had +Def. The net result of all of this was that someone using MoG was stronger than someone that wasn't, in most non-psi situations, for the first 60 seconds after the activation of MoG. Then somewhere between 60s and 90s (there were some ambiguities in the calculation) the player that didn't use MoG - assuming they were still alive - would have a net survivability that was higher.

In other words, MoG gave you better burst survivability at the expense of lower sustained survivability. You could say it gave you the option of frontloading protection (something Regen normally cannot do very well) at a cost of some regeneration. That's not intrinsicly bad, but in my opinion the cost was a bit too high for this option and a large percentage of players didn't view this option as a net positive. Basically, MoG was a little bit broken and a lot image-challenged.


On the subject of SoW, the reason why Castle hasn't applied the SoW design to other godmode powers is actually pretty simple: he doesn't want to. SoW had a very specific design challenge. Castle wanted the strength of the power to be intermediate, not massive, because he wanted Willpower's performance over time to be more steady than other sets - that's one of its intended fortes (remember: the powerset was originally envisioned to be an all-passive powerset). That means SoW's intrinsic strength needed to be lower than other tier9s like unstoppable. But that meant it needed to have better uptime to make its overall value similar to powers like Elude and Unstoppable.

But *that* meant it was in danger of being slottable to be perma, and Castle did not want Willpower to be perma-SoW all the time. He wanted SoW to have *good* uptime without being perma. There's no way to do that with the current tier9 design rules. So Castle engineered the power so it would have good uptime but not be slottable for recharge. That way it did not need to obey the tier9 "recharge > 6x duration" rule.

None of this applies to Elude or Unstoppable (or other similar tier9s) which is why Castle is so confident in saying that the SoW design wouldn't likely be ported to other tier9s. It won't, because those powers simply don't need it. They are *intended* to be "ultra high, but infrequent." SoW isn't intended to be: its intended to be "around a lot, just not all the time."

The only reason Castle hedges that statement is the simple fact is MMOs evolve. One day, someone might decide that Elude *should* be an "intermediate strength, up a lot, non-perma" power. If that happens, logically it would almost certainly adopt the SoW treatment. But until that happens, there's no mandate for Elude and Unstoppable to conform to SoW. SoW isn't the "new way" tier9s should be designed. Its just *another* way tier9s can be designed, if that's what the devs intend for the powerset.

While it directly addresses the issue of exploitable levels of reward, in my opinion its the least elegant solution, for two reasons. First, its a very crude sledgehammer. It offers nothing in between "exploitable" and "not exploitable." It throws up its hands and says that some critters' XP will suck, some will be ridiculously high, but we're not going to bother to try to expand the range of the moderate middle. Its specifically, on design-elegance that I would oppose such a solution in the general case.

Second, it encourages people to make farms that fly just under the cap. So "Exploit" is bad, but "Exploit-1" is fine. And it suggests the devs implicitly approve. In this sense, it can actually make the overall problem worse, and force the implementation of a cap that isn't "definitely exploitable" but "so low it cannot be exploitable." And that level might be so low it creates the problem of AE missions that simply run out of XP long before they are over. The complaint level for that is likely to be at least as loud, if not louder, as it is now over low per-kill rewards. The psychological implications are not trivial to predict here, and if I had to roll the dice, I would say it wasn't worth the risk.

The devs are specifically barred from posting anonymously under another account. Even when his redname was malfunctioning, Castle identified himself when posting under a different account (which was also named obviously). That policy dates back to the Cryptic days, and to the best of my knowledge that policy is still in effect. And I last checked just a few months ago.


Although this is certainly true, its also true that perhaps counterintuitively because the devs have their own work to do, they are not always very aware or try to be aware of what everyone else is doing. There's no "thrill" to finding out before we do that Castle is making a banana-grenade powerset in Going Rogue, because they work there. Sometimes *I* find out what Castle is working on before BaB, and vice versa, when its not a big secret, and the one doesn't directly affect the other. BaB may know, he may not know, he may be so busy with Going Rogue and other projects he doesn't currently care to know. Its impossible to know without asking him.


On these forums, oh yes it is.


That's hyperbolic. Under no set of circumstances would someone be modded for saying what BaB said, first because its relevant, second because its true, and third because it doesn't violate any of the forum rules. Just because you don't like it, doesn't make it generally offensive.

Also, BaB *is* one of us, when it comes to forum posts. BaB has actually been modded in the past. The community managers don't answer to the designers; BaB is not higher in the food chain than Mod8, he's in a completely different part of the company.

Actually, in many ways the devs are held to a higher standard than the average poster, specifically because they represent the company when they post, even when they are not specifically representing their official position in the company. I can say things they can't get away with saying.


If BaB didn't say it, I would have eventually. Its completely relevant to the suggestion that the AE *must* provide a levelling rate comparable to the rest of the game. Even when its "fixed" there's no guarantee that will be true. If it was up to me, that almost certainly wouldn't be true on average, for a number of reasons.

And if you think its not constructive to be told that one of your assumptions about the AE might be wrong, and therefore all of your conclusions about the AE that derive from that assumption are suspect, and any argument you make to the devs that include that assumption have a significant probability of being dismissed, we have a different definition of "constructive."


And by the way, I actually don't know if BaB had first-hand knowledge, or was just guessing when he said what he said. So I'll reiterate: the AE's "alternate levelling path" makes no guarantee that it will provide an equivalently fast levelling path. And I am not guessing.** I think its fair to suggest its too low because its drastically too low. But its not fair to suggest its too low because its lower at all. That's a target you have no right to expect, in the sense that the devs have not stated or implied you'd ever get, all assertions to the contrary notwithstanding.

Comparing this to "Jack-speak" and saying that "alternate means equal" is ironic, because that sort of redefinition of English is exactly what Jack was often accused of. Alternate means equal in exactly the same way that Enhancements aren't powers: only from a very specific and narrow point of view.



** I've had discussions with many devs about the AE, from before I14 was in beta, up to the present day. Their collective position about rewards in the AE during those discussions has been consistent ever since Positron laid down the law on it in the open forums: the AE is intended to have rewards (it was originally conceived to be reward-less) but those rewards may not be equivalent to the rewards achievable in other parts of the game. They may be qualitatively different, and they may be lower. That's been true ever since the AE's design was changed to allow rewards at all.

Actually, BaB is not a reward designer or an AE system designer, so BaB's comment is not an official dev response. BaB was commenting colloquially. BaB's statements about the design of the AE have no more official weight than mine do, unless BaB specifically indicates he's speaking in an official capacity. BaB has said this more than once himself. BaB is the animation-lead. He's the reward-nothing.

Also, if the devs were working on a solution to custom critter XP, it would be hard to say that they were still working on it because they would probably be barred from saying so. There are rules about speaking about future game developments that are not locked-in certainties, especially about controversial matters like AE rewards. All they could say is probably what has already been said, which is that the current state is likely not the final state. I doubt any of the red names would be allowed to say anything further.

Actually, I tend to use IH as the Regen "tier9" rather than MoG, and I consider it very valuable, if not essential. In fact, I find IH and Elude have more in common than Elude does with Unstoppable. With Unstoppable (especially given Dull Pain and Invuln's resistances to start) you can wait until you get into trouble (to a point) and then activate Unstop. Its no big deal to get down to half health or a little less and then go unstoppable. But with Elude, you're much better off using it somewhat proactively to protect your full health bar. If you wait until you get low, Elude won't protect you from one unlucky hit, and won't reduce the level of damage per attack. It will greatly reduce the likelihood of two unlucky hits close together, and reduce the chance of three in a row even more, which is why it works best when you still have full health. IH works in a not terribly dissimilar manner. You don't want to wait until your health gets too low, or its recovery won't be fast enough to bail you out of one last big hit (or burst of hits). But if you use it while you still have lots of health, it will basically erase a high level of incoming damage and keep you at or near full health. I found the transition from using Elude to using IH to be smoother in terms of gameplay than the transition from Elude to Unstoppable for that reason.

So for me, IH is optional in the same sense that Elude would be optional to SR, if Elude didn't crash and you were not invention-buffed to the soft cap to begin with.

Actually, animation costs increase linearly. If you presume that clicks "pay" for their benefit in terms of endurance and cast time, then the ratio of cost to benefit remains constant right up to the recharge limit.

You're probably thinking that as mitigation root costs increase linearly, offensive opportunity as a percentage of the total decreases inversely. That's a completely different thing.


No, actually I forgot the movement penalty was removed. Slip of the keyboard on that one, my mistake.


You're stating that Regeneration (the set) has differences in mitigation strength as justification for looking at the cast time penalty without first demonstrating those differences are a net minus. In fact, the way Regeneration was implemented, those things are a net plus. Because the devs have no good model for comparing Regen to sets like Invuln and SR, Regen gets set much higher than it really ought to be (the same thing happened with Willpower, although the error there was slightly different).

It doesn't help your case to mention all the ways Regen gets to break the rules in ways that *benefit* the set. Trust me: if the devs had a proper model for Regeneration, Regeneration would be lower. Instead, it was originally set high, then reduced in strength until the datamining couldn't tell it was too high.


Oh, one more thing:

1. The contribution of recharge to sets like Invuln, Dark Armor, Regen, and Fiery Aura is not a minor contribution.

2. When SR was designed, Willpower and Shield Defense didn't exist. Its design rules predate those sets, in terms of the unique aspects of the set. I was assuming you were considering the original four scrapper sets, because:

3. Accepting the current state of mitigation sets available to scrappers, its not true that Regen has especially high rooted time costs. Regen has reconstruction, which has a base 0.73 cast time with 60s recharge, and Dull Pain which has 0.73 cast time with 360s recharge. Assume we slot both to 1.95 recharge, and we get a net rooted efficiency cost of 0.73/(60/1.95) + 0.73/(360/1.95) = 0.024 + 0.004 = 0.028, or 2.8%. Fiery Aura has Healing Flames, a 1.5s cast time 40s recharge power. Equally slotted, its rooted cost is 1.5/(40/1.95) = 0.073 or 7.3%. Fiery Aura has nearly three times the cast time cost of Regen. Adding in IH (0.004) and MoG (0.011) doesn't help matters. Regen increases to about 4.3%, which is still only about half the cost of Fiery Aura.

4. The specific ways SR would still be unique, even among all the currently available scrapper mitigation sets is:

a. It has no way to improve on the intrinsic endurance limit for offense, which you can do in one of three ways: improve DPE, increase endurance recovery, or provide an endurance-efficient damage aura. Regen and Willpower have quick recovery. Shield Defense has a +DMG power. Invuln has a tohit buff. Dark Armor has a damage aura. Fiery Aura has an endurance recovery power *and* a damage aura *and* a damage buff. Quickness does *not* do any of those three things, and does not, as a consequence, actually increase the damage per unit time you're capable of generating while obeying the endurance recovery limit. The endurance recovery limit is actually one of the foundational balance pillars of the entire game, which makes this not a trivial difference. In fact, its the difference that comes closest to being game-breaking, because it tampers with a balance-significant parameter (as the devs define balance).

b. It has a large class of attacks which exist at essentially all levels and all ranks for which it has no damage mitigation of any kind, proactive or reactive. There are no such attacks for anything with health recovery, which includes Regen, Willpower, and Fiery Aura. It also includes Invuln on a technicality because DP has a heal, but more importantly and not a technicality Invuln has +health, which acts as resistance against all forms of attack. Even against psionics, Invuln has the equivalent of about 37% resistance with fully slotted DP. Dark Armor has resistance to all damage types. Shield Defense has +health. That leaves SR as the only scrapper mtigation set that has literally *zero* protection against a large class of attacks: non-positional psionics. This is significant because SR's original design didn't intend that: SR was designed to offer protection to all attacks. Non-positional psionics was essentially an override of SR's original design intent.

c. Its the only scrapper mitigation set for which essentially all the damage mitigation powers in the set synergize sufficiently strongly with at least one other power that the set loses more than proportional strength when any one power is eliminated. In other words, SR-1 is significantly weaker than any other set minus one damage mitigation power. That was due to a laudable but faulty design decision. To make sure the SR passives were not seen as "worthless" they added scaling passive resistances to each of them. This meant every passive synergized with each other strongly (in fact, having two provides only about half the resistive mitigation of having three, not 2/3rds - the math is complex to prove it though), and synergized with its corresponding toggle. Removing, say, slotted dodge cuts the passive scaling protection in half, and also increases net incoming melee damage (assuming SO slotting) by 45%. In normalized terms, assuming each positional vector blocks roughly 1/3rd of all damage, removing dodge reduces net strength of the set by over 22%. Without this strong double-synergy, removing any one power should reduce the set by something closer to 1/6th, at least on average: instead the ratio is closer to 1/4.5.

What's specifically unique about SR is that *every* damage mitigation power is like that: all three toggles and all three passives have an incremental reduction factor higher than 1/6 (the passives are about 1/4.5, the toggles about 1/4.2). This interlocking is the basis for the colloquial comment that with SR, you need to take everything. And that runs counter to the design intent for the damage mitigation sets. It should be optional to take any one power, for most powers, with the only cost being a loss of effectiveness proportional to the strength of the power. Some powers will of course be more valuable than others, but overall the average should hold.


Those are the top three on my list. There are a few more that are weaker distinctions, so I'm not including them here. Each of the above is balance-significant working from first principles of the devs' own statements about game balance, or things I happen to know about the design of the game. Collectively, its enough for me to prove beyond any shadow of a doubt that *something* is wrong with SR.

The only reason why none of those problems has been addressed is because Castle holds a trump card: SR isn't underperforming as a result of those problems. That makes it a low priority item to even bother looking at them. My guess is that the same holds true for Regen. Things like MoG were a thorn in Castle's side, but something like reconstruction taking too much time (when its one of the fastest clicks around) is probably too far down the priority list to devote significant time to at the moment. And for *both* SR and Regen, I would tend to agree.


By the way, a significant fraction of Regen's damage mitigation comes from the +health of Dull Pain, which is not a health recovery effect.

I have found that while the visible population of the servers changes from day to day, my mental association with "number of people" has tended to be the number of people chatting on the global channels I'm subscribed to, not how many people I literally run across in the shared zones. Its been that way since before there were global channels and all I had was the supergroup channel and the friends channel.

It takes a certain mindset, especially if you come from a game that doesn't use a lot of instancing, not to think you're the only hero patrolling the entire server sometimes. But I would recommend new players concerned about the server populations to spend a little time lurking the various server forums, find out what the communities are like, and perhaps picking up on that server's global channels.

I'd also recommend having a slightly thick skin initially, until they get into the groove of socializing within a long-established community.

The original game didn't balance cast times at all. Attack cast times were only examined because the wide variations in cast times on attacks had a provably unintentional effect: many powers were much less effective than intended overall.

That same argument doesn't translate to Regen directly, because the specific argument relates not just to full attack chains, but the average damage output of attack sets when slotted with SOs. Attack cast times have *both* an initial effect (or rather a mid-game effect) and place a constraint on the amount of benefit an attack set can get with recharge bonuses.

Comparing two defensive sets with comparable numbers of clicks, one with higher cast times, would be at least an analogous situation. But when comparing two mitigation sets, one of which has more clicks and one of which has less, two counterbalancing effects of clicks come into play. Clicks cost more than toggles and passives in terms of activity time. But clicks benefit from recharge more (because in general recharge can't benefit toggles and passives). That makes this a far more complex argument to make, and the burden is on the advocate in this case (well, in all cases really).

Its not a new assertion: Castle wasn't a red name when it first came up on the forums back in I4 (I don't think), but I'm assuming he probably saw discussion of it. A lot of regens suggested when the focus shifted from toggle IH to reconstruction that reconstruction's activation cost was in effect applying an offensive penalty on regens. However, there was no way to demonstrate numerically that this offensive penalty wasn't actually part and parcel of the qualitative mechanical difference between clicks and toggles, specifically that clicks could benefit from recharge. In other words, no one has suggested yet why the difference isn't (or couldn't be) an intended difference.

On the subject of uniqueness: Regen is unique - among scrapper secondaries - in terms of its rooted cost, but its not, for lack of a better way of putting it, unique in a unique way. Dark Armor is unique in its endurance penalty. SR is unique in its inability to increase damage mitigation strength with recharge bonuses (ignoring the tier 9 emergency power). Invuln is unique in its movement penalty. Uniqueness alone cannot justify taking action against an intentional qualitative difference. If it did, Regen would lose Quick Recovery: its the qualitative distinguisher among scrapper secondaries with the absolute highest game-balance impact. To be valid, a uniqueness argument must be able to state that, all other things being equal, the thing in question must be specifically equalized. But there is no rule I'm aware of that says, all things being equal, all mitigation sets must have the same number of clicks and incur the same amount of rooted penalty. If a set had to incur the rooted penalty for a large number of clicks but somehow couldn't benefit from recharge or other click mechanical advantages, then that would be a case where a balance argument could be made. That makes the uniqueness observation intrinsically weak as a balance argument. If you had an argument that could justify *why* the imbalance you've calculated shouldn't exist, relative to all of the other qualitative differences that exist, you'd probably have a much more persuasive argument.


(By the way, on the subject of Clicks, its been suggested that there is a rule that says Clicks > Toggles > Passives. To the best of my knowledge, there is no such rule. There is a rule that says in general Toggles are stronger than Passives, but no rule that says Clicks must be stronger than Toggles. If there is such a rule, the original devs didn't follow it, and the current devs consistently ignore it. Clicks tend to be stronger per activation than Toggles, but that's mainly because Toggles activate more often. In *that* sense there is such a rule, but no rule that says the *overall* net effect of a Click must be superior to comparable Toggles). So while there is a qualitative mechanical difference between Clicks and Toggles, there seems to be no requirement that Clicks explicitly end up significantly stronger than Toggles either ignoring those qualitative differences or when completely factoring them in).

Castle listens to a lot more players than just me. And regen *is* numerically stronger than most of its peer sets**. Not specifically because of animation time, but because the mechanics of regeneration dictate that as a necessity when balancing against a range of performance.

On the other hand, I do know that Castle doesn't tend to respond to most PMs that tends to start with "I think X should get a buff because." When I first started PMing him about game mechanics, he didn't tend to respond to any of those from me either. Historically speaking, these do not go well or end well for the devs.




** The arguable exception is Willpower, and Castle knows I think Willpower was set too strong in Issue 11. If Castle listened to everything I said, Willpower would have launched a lot weaker than it actually did. But apparently Castle has a lot more spine or I have a lot less influence than most people give credit for.

You know, its not really my math skills that are valuable to the devs for closed beta testing. Primarily, its because I always try to pick things no one else is looking at, and test that to death. That makes me a consistently unique tester in that regard. Now, some of the things I look at require significant math skills to analyze, or require perspectives and knowledge fairly unique to me, but in every closed beta there is a giant laundry list of things that I think should be tested to a higher degree than they ultimately are, because not many people try to test something other than what everyone else is. And many of those things don't require specialized knowledge or computational skills. They do require methodical testing ability.

Most people test what they are familiar with, what they think will be the most interesting thing to test, what everyone else is testing, or what they think are the "most important" parts of the issue. I always spend most of my time testing something else entirely.

I cannot say how to get into the closed betas on any consistent basis (or at all), but I suspect that doing that makes me more valuable to the testing process, and by extension anyone else looking to make themselves valuable to the testing process might want to consider a similar testing strategy (even if its conducted during open beta, it will still be noticed by the devs if done well). No guarantees, and it will probably seem like work - but that's really what closed beta is: work.

Separate from the semantic point that Sam pointed out, your statement above is not true either. They attempt to "support" the minimum system requirement, but "support" only means the game will still run on that system in theory. Even without Ultra Mode, the current game engine will not run on the minimum system specification with all features turned to maximum. The minimum system is too slow for that.

What the programming team does, and is supposed to do, is target the *best* system that nevertheless won't completely orphan the stated minimum supported system. In other words, if they can add a feature that degrades reasonably on the minimum supported system, that doesn't break the minimum system support.

Which is why I even pointed out my little semantic pet peeve in the first place. I know the reason why "lowest common denominator" has semantically drifted to colloquially mean nearly the opposite of what it is actually defined to be. The concept of "low" is getting stuck on "lowest." The system the devs have to target is lower than the average system out there, so its probably the "lowest" something or other.

But their target should not be the lowest anything. It should be the highest target they can get away with.

They should be targeting "greatest common" - the best that we all share. Not the worst or the lowest possible target. So even colloquially, the term "lowest common denominator" has a poor connotation. It suggests the devs should aim low. They should aim high. Just not so high that we orphan too many customers. But as high as possible. The Greatest target that we all can Commonly play.

They didn't have enough time to implement a system that was close to par, or whatever target they wanted instead of par. There is no compromise between removing exploits and making balanced rewards: they hit the system hard enough to remove all possibility of exploiting it. I'm sure if they had unlimited time and resources they would have implemented a more nuanced system, but they did not have either at that time.

The lowest common denominator would be the cheapest card that nevertheless had every single feature or capability of all video cards. The greatest common factor video card would be the best card that had nothing but the features everyone else had. You target the greatest common factor, not the lowest common denominator.


Sorry, pet peeve. Carry on. Gravity, light, swirl thingy.

There are lots of things the devs aren't crazy about, but not enough to take draconian measures to prevent. Just because the devs haven't implemented ways to prevent an activity that you believe would stop it, doesn't necessarily mean they are giving tacit approval. There's a lot of ground in between "devs deliberately intend" and "devs will stop at nothing to prevent." Actually, most of the game probably exists between those two extremes.

The dev statement relevant to this topic is: the devs are generally against players performing repetitive actions focused on the primary purpose of, and with the net result being, optimizing the rate of earning rewards significantly above and outside the range of most players. Such activity may be curtailed at any time at the devs' sole discretion. If your version of "farming" doesn't match this definition, you're safe. If it does, you're not.


On the general subject of fixing the AE rewards, lets just say there is no simple solution that will accomplish that, unfortunately. There are solutions that I think would work (all significantly more complex than any suggestion I've read so far), but none that could be deployed immediately.

I don't consider swearing to be a "playstyle" within the context of this discussion. But even if it were, the closest analogy that can be made, given the differences between chat and power effects, is that the only thing you can do in chat is opt-out: you can filter what you hear. You have no ability to stop what others say. Short of harassment violations or game-wide disruptions, you have no recourse on stopping someone from swearing.

There is no way for a player to make themselves immune to the collateral effects of knockback without suppressing the player's ability to generate it, so there's no way for a player to "opt-out" of knockback in that sense. But that's as far as I would allow it to go, possible or not.

Its more correct to state that I believe as an axiom, visibly unbalanced is less fun. More to the point, though, I believe that there are other considerations besides "fun" in a well-designed game. It sounds good on paper, but its far too nebulous a thought to attempt to justify every single game design or implementation decision on whether its "more fun." Its almost a nonsensical question to ask whether 0.84 or 0.94 is more fun in most cases. And in fact, even if dual blades stabbing you in your own head when you superjump turns out to be "more fun" I doubt that would affect the animator's decision making process, nor should it.

I don't always know if a suggestion of mine will end up generating a more or less fun game overall. Sometimes, its unclear whether its even possible to judge. Usually, in the absence of knowing with certainty, I assume its best if the game at least follows the devs' own design rules with consistency, and then only attempt to argue against those rules when they very obviously create problems in overall entertainment value.


This is actually the specific reason why I generally oppose KB-suppression suggestions. It adds a weapon to the arsenal of the playstyle-intolerant.

Its ultimately purely a matter of subjective choice which way the devs decide to go here. They can decide that the kind of game they want is one where some intolerant players avoid playing with other players altogether, but the rest learn to deal with each other because there is no choice. Or they can decide to make a game where everyone has lots of choices besides teaming or avoidance, but those choices can allow some players to discriminate in more subtle, but probably much more common ways. I happen to fall on the side of the former rather than the latter.

I can't stop people from exercising playstyle discrimination. But I will not support any attempt to provide them with cover.

Cringley tends to think lots of things. Most of the time, his predictions tend to be right only when he predicts the continuation of a trend. When he goes out on a limb, he's worse than random chance.

I think anything is possible in the chip industry right now, but Intel acquiring NVIDIA seems highly unlikely to me. It all hinges on his theory that the ATI acquisition by AMD gives AMD a strategic advantage over Intel that Intel has to respond to. Larrabee suggests that Intel wasn't thinking that.

Intel doesn't care how many graphics cards AMD and NVIDIA sell, anymore than they care how many hard drives Seagate sells. What Intel wants control of is the motherboard, and everything on it. The motherboard is the defensive perimeter around the CPU. Intel *does* care about embedded graphics, but embedded graphics did not appear to be directly on Intel's mind when they designed Larrabee. A multicore superscaler x86 architecture is not the sort of thing you're going to then add to a motherboard or a SoC as built-in graphics most of the time.

Instead, Intel was most likely thinking two things. First, the real threat was NVIDIA (and to a lesser extent AMD) pushing the graphics cores into the computing space, ala HPC via GPU. If people start deciding that encoding video or other computationally high tasks can just run on their graphics card, they may decide they don't need the next big Intel CPU. That's bad. They probably felt they needed to have a foothold not in the GPU-space, but in the *ultra-highend* GPU space to make certain they couldn't be outflanked in the computational high ground (for those that don't follow graphics card technology, look up the specs on the top of the line NVIDIA and AMD GPUs sometime).

They were probably also, in a related line of thought, wondering if super-core x86 could eventually scale up to compete with the superscaler GPUs. Since Intel's current strategy is to keep increasing cores until people can't make use of them any more, *one* way to make sure people are buying those Core i-eleventybillions is if they push the software industry to start writing code designed for multiple x86 cores. One way to do that is to release a GPU designed around multicore x86. If graphics driver writers could be convinced to start writing code for Larrabee, those same skills would be useful for eventually writing for Core iX. Eventually, Core iX would implicitly *have* a built-in GPU without even needing to integrate a special graphics core (they'd basically just need to extend the cores with Larrabee vector extensions). You'd just have an Intel CPU with twenty cores, and four would run your graphics. This is Intel's best-case wet-dream scenario.


There are a number of reasons why Intel might pull Larrabee. If you're Cringley, Intel is lying. But if they are telling the truth, then Intel has said two contradictory things about Larrabee. Prior to the announcement, Intel was all roses about Larrabee's performance. In fact it achieved a teraflop on a real benchmark not that long ago. But they are now saying there are performance issues with it. Two big possibilities loom for me:

1. In spite of prior information to the contrary, on real workloads the system was unable to achieve its benchmarked performance levels, or unable to achieve *stable* performance levels. My guess: internal memory bandwidth or architecture issues. This has bit the big two in the past as well.

2. When actual silicon was tested, it turns out the design has fabrication issues that would make it difficult to ship in volume. It may even be analogous to the problems AMD and NVIDIA are having now with their 40 nm process.

Its also possible that the skills required to make an actual video driver for the thing that will work under all conditions at high performance simply don't exist yet. In that case, Larrabee might be a "software development platform" (what Intel is now saying) specifically to build the skills internally at Intel just to program the thing. A GPU without video or HPC drivers is just a hunk of sand.

I will concede that nameless minions possess KB in CoX significantly more often than they do in genre on average. But that's an issue with the PvE critters' use of KB, not the players. The players do not, most of the time, numerically outnumber the critters in a fashion where we become the nameless minions.

On the other hand, the likelihood of seeing KB seems to increase in the genre the higher the numerical superiority of the opposing forces, which does tend to match the general situation of players relative to critters.

I have my doubts that Larrabee was ever going to apply downward pressure on the high end. I think Larrabee was designed at a time when Intel did not think AMD and NVIDIA would get to the stream/vector processing levels that they did. Although they are blaming general problems, I think Intel realized Larrabee wasn't even going to be competitive with the current gen of AMD and NVIDIA silicon, and decided to go back to the drawing board. If their architecture doesn't leapfrog AMD and NVIDIA completely, they are toast (because their novel architecture is almost certainly going to require significant time to software-optimize, so they need to be way faster in raw performance just to compete with the current GPUs in the short term).

If the cards are still hard to get by xmas, I'm probably going to just scoop up a 5850 (or possibly a 5870) at whatever price they are at, and call it a day.

What I said was: "immunity to knockback is far rarer in comic books than in the game."

Immunity from knockback in comic books generally falls into three categories:

a. some specific power that grants general immobility or rooting to the ground
b. a level of power so high that impact generally has no effect
c. A massive size differential

In CoH, immunity from knockback is granted to:

a. Nearly all melee
b. Anyone that takes acrobatics
c. Anyone buffed with one of the KB protection powers

I'd say its objective fact that KB protection is far more common in CoH than in the comic book genre overall.


I'll give you Claws. But force blasts are much more common (Iron Man's repulsor rays, for example) and there are other forms of knockback that parallel the game, such as Storm's control of winds or Magneto's control of magnetic forces. And Assault Rifles in the game have knockback primarily in the grenade, and comic books are often depicting explosions as knocking things around (probably far more than they would in actual fact).

Well, when you put it that way, go climb a tree.

I don't think you can argue successfully that Quickness might lag Hasten, but it has other buffs and doesn't cost endurance, so its fine. That's not an argument that its balanced, that's an argument that the situation is too confused to make a judgement. Its essentially obfuscating the situation and saying obfuscated situations are probably ok.

Unslotted, Hasten's benefit is slightly higher than Quickness' recharge benefit for a relatively negligable endurance cost. The question is whether Quickness' other benefits are sufficiently high to *both* overcome the fact that Hasten's recharge benefit can be (essentially) enhanced *and* that Quickness is supposed to be intrinsicly stronger by a significant margin.

Although this isn't strong enough to prove the point, it is a significant set of data points that Quick Recovery is 20% stronger than Stamina *and* Stamina has a two power selection prerequisite, which makes QR's true value significantly more than 20% higher. Fast Healing is nearly twice as strong as Health, which itself has a one-power prerequisite.

*At Best* Quickness' value is a little more than 50% stronger than the strength of its power pool equivalent, given that Quickness' buffs are about equal to Swift and half of Hasten with a plus/minus hedge for the slotting opportunity (essentially, 1.5 power pool choices). And that presumes swift has the same value as Hasten does.


The power pool argument is not the sole argument suggesting that Quickness is underpowered. Its just one supporting line of thought. There are several more that converge on that conclusion, including the issue of what effect +Recharge has on SR. Practiced Brawler has a 120s duration and 200s recharge. With its default slot slotted with an even SO its not perma: it takes +66.7% recharge to make PB perma. PB also has a 0.087 eps cost associated with it if its slotted to perma without endurance reduction (10.4/120). That's actually pretty high relative to defensive toggles which average about 0.13 eps (its 67% of a defensive toggle) considering it only offers status protection *and* SR has three defensive toggles, so PB isn't especially powerful given its inability to be made perma out of the box. If Quickness was +0.33 recharge rather than +0.2 recharge, the combination of quickness + one SO would make PB perma. At its current strength, Quickness can't close the gap on PB alone, forcing an additional slot (or Hasten) whether you take it or not.

Another line of thought considers the question of endurance. Endurance is designed to be a bottleneck on performance, and it is at the slotting levels the game (and the powersets) are balanced for. Most QoL (meaning non-defensive, not non-functional) buffs improve effective DPE, or increase total recovery which increases the bottleneck. Quick Recovery does this, and the tohit buff in Invincibility does this. Its pretty clear of the original four scrapper secondaries, Dark Armor was intended to be limited by endurance (the design of Dark Regeneration demonstrates this), which is why it doesn't also have an endurance cost mitigator. But SR doesn't follow this design theory, and yet Quickness' recharge benefit does *not* improve the DPE bottleneck. Recharge actually doesn't improve your net kill/sec at all; it only allows you to burn through your endurance quicker. For advanced builds with very low or non-existent endurance issues (especially with inventions) this is not significant, but for the balance point target for the powersets it is.

In total there were about six different lines of thought that suggested that either:

1. Quickness be increased to 0.33
2. Quickness have an eps benefit to balance the +recharge benefit
3. Quickness have alternate mechanical advantages.

At least, so I thought when I9 came out, Quickness could be an invention mule. That turned out to be false, reinstating the third possibility.

I'll dig up a copy of the original analysis (if I have it) and update it with current numbers, and see if the argument ends up stronger or weaker. I'm not 100% certain which it will be, but I'm pretty sure it won't be invalidated by anything that happened since that time.