Arcanaville

Yes and no. Yes, its apparently possible. No, you're not going to actually last long enough to fold it properly before setting it and yourself on fire.

I would assume the latter, but I'm honestly not sure. I'm actually less worried about my molecule-thin skirt, and still coming to grips with the fact that a surprising number of my characters are actually apparently walking around in little more than body paint, and not actual clothes.

There are some ramifications to 3d printing when your top is only airbrushed on. I wonder if there are any clothes that are just bump-mapped?


You get two separate capes that float about ten feet away, unless you use that other trick that of course I'm not using.


I would try to continue to hypothetically pursue this project that I'm only theoretically working on, to try to resolve the hypothetical question of what barriers stand in the way of making this work all the way down to a hypothetical printed model, if I was working on this, which of course I'm not. And hypothetically speaking if I did do that I would hypothetically keep the devs in the loop on my highly inappropriate and unauthorized actions. That would hypothetically allow everyone to potentially benefit from this hypothetical activity, which I'm of course unable to actually do in reality.

Hypothetically speaking, if someone did figure out a way to make this work at least in theory, I wonder if under those hypothetical conditions Marketing would consider investing a small amount of hypothetical resources to create an avenue for hypothetical customers to purchase these hypothetical models, and if they would consider this a hypothetically valid use of marketing resources. It would be a shame if hypothetical customers willing to buy 3d printings of their characters were hypothetically forced to use hypothetically questionable means to accomplish that task.

Hypothetically speaking, it could be inferred that while the technical skill of the person that made those models is practically infinite, they might possess the 3d modeling experience and talent of a three year old chimpanzee and can't figure out how to actually map the textures to the model correctly in any software package available, although the textures do in fact exist to do so.

Hypothetically speaking, of course.


Hmm...

Which is why I'm specifically interested in the tier 1/2 powers on primaries and the tier 1 power on secondaries (not coincidentally, the three powers that are first available at character creation).


I don't think that is true. What I know for sure, with absolute certainty, is that every low level player has at least one of the first two primary powers and definitely has the first secondary power. These are the powers that, if they have long recharge or high endurance costs (relatively speaking) will have the greatest impact on a player's ability to actually do anything while playing the game.


The reason why this isn't better is two-fold. First, it would affect all powers and not just a focused few. That *would* start to create problems for powersets that are designed in different ways or are used in different ways. Every powerset is designed on the assumption that those three powers are special: every player is going to get at least two of them, if not all three. A player has to be able to solo with just two of them, because at one point two is all we have. There's a certain dependability to targeting those powers. To the extent that they are different from powerset to powerset, archetype to archetype, those differences tend to be deliberate, and not accidental, and generally in keeping with the archetype's priorities.

Blanket recharge and endurance buffs would affect things like the early Build Up that Energy Manipulation gets.

Second: blanket buffs would be extremely difficult to balance. The side effects and unintended consequences of doing something as drastic as cutting the recharge of a tier 1 attack in half is relatively minor, and the benefit descales exactly the way you want it to: as the player takes more and better powers, the benefit of that one power's discounts gets diluted. This actually makes the buff do something closer to what I want than a blanket buff across all archetypes. In particular, it dissipates when the player has a lot of attacks - whenever that is, which is different for every player and archetype - and not by an arbitrary linear scale.


Those are actually much more drastic changes than the one you are suggesting is drastic ("changing the formula"). And I should point out that the formula is a balance formula, which doesn't preclude buffs beyond the formula when they are explicitly called for: Claws was designed with specific discounts relative to the formula (prior to its revamp) specifically to implement a conceptual goal of speed and "lightness" to the set (which is debatable if they accomplished correctly, but still). I could argue that the early powers are specifically intended to be, conceptually, the abilities we are most familiar with and thus most competent in exercising, and should have a discount that represents that fact.

This happens in reverse: powers like the tier-9 Nova-class powers are actually *more* expensive than the formulas would suggest, because they are intended to reflect the fact that at those levels of power, the formula doesn't really apply because the effort required is more than normal (conceptually: game-design-wise the reason is that by formula those powers would be ridiculously too strong).

The tier 1/2 attacks, particularly at lower levels, are actually the higher endurance drain on a EPS basis.

Just for illustration purposes, consider Power Bolt: 1s act, 4s recharge, 5.2 endurance cost; and Power Blast: 1.67 act, 8s recharge, 8.684 endurance cost. If you fire them both as fast as possible, and you don't have recharge slotted, you'll be burning 5.2/5 = 1.04 eps with Power Bolt and 8.684/9.67 = 0.898 eps with Power Blast (Arcanatime not factored in). Power Bolt is actually being used often enough to overtake the endurance burn rate of power blast which costs over 60% more per use. When Power Burst becomes available (2s act, 10s recharge, 10.4 end) it will only burn 10.4/12 = 0.867 eps if its fired as often as possible (technically, if Power Burst followed the rules it would burn 0.919 eps, still under the burn rate of Power Bolt). Firing all three as often as possible, and not factoring in collisions (when two powers recharge at the same time) Power Bolt (the lowest cost power per use) will be responsible for about 37% of your endurance drain due to attacks, while Power Burst (the highest cost power per use) will be responsible for only 31%. While Power Bolt is only costing a little more than Burst, it is definitely costing more. Reducing the endurance cost of Bolt would improve endurance burn rates more than reducing the endurance cost of Burst.

Once you have the ability to slot or acquire significant recharge, these numbers change and the high damage/high endurance cost powers begin to have a greater impact on endurance burn. But until then, the lowest tier powers tend to have a greater impact on your endurance simply because they are used far more often.


I should point out that EPS is not my primary motivator. In fact, if you decrease both recharge and endurance as I suggest you'll end up running out of endurance in a similar amount of time. But your activity rate will go up at lower levels without commensurate cost, which is the actual intent of my suggestion.

Running out of endurance is not the core problem in my opinion. Being idle is the problem, and whether that is due to having no endurance to attack, or no attacks recharged to use, the problem is the same root issue I'm looking to tackle.

Not necessarily, and probably not in my case. Since none of the cores on my system were maxed out during that load, its likely there were other bottlenecks constraining performance (disk, for example, or less likely memory IO).

On an i7-860, its also likely I would have to disable hyperthreading to get the maximum possible overclock, and that might also be a less than optimal change given my workloads.

For the most part, overclocking only helps if you have a saturated or nearly saturated processor core (even if the others are idle and your net CPU utilization is low). If you don't, overclocking usually can't help.

Your best bet for gaining ground with overclocking on a multicore processor like an i5 or i7 is if you don't do a lot of stuff simultaneously, but you do one or two extremely CPU-intensive tasks for which the CPU speed itself is the critical bottleneck. For people trying to maximize City of Heroes performance, the CPU is probably not going to be the problem if you're starting from an i7-860 or 920, say. There are other games, on the other hand, that could tap out those CPUs, because they would saturate out a single (or pair of) cores. I just don't own any at the moment.

I'll be honest though: I haven't bothered to even check to see if any overclocking settings exist in my Dell, because I haven't pegged the CPU in my i7-860 yet. If I really wanted to, I'm sure I could, but when I dump normal workloads (normal for me) onto it, it really doesn't seem to notice. A week ago I had two instances of CoH loaded, Vmware running a copy of my old XP workstation, a python simulation running, Real converting movies to iTunes in the background, and a bunch of miscellaneous foreground apps (like browsers), and I think I was at 78% CPU utilization. Wouldn't want to do that with less than 8 gigs of RAM, though.

Actually, I've been thinking for a long time about taking the first two primary powers, and the first secondary power, and applying a recharge and endurance discount to all three (if it happens to be a passive, the discount just has no effect). Something high: 30% - 50%, but not cut to zero.

I believe I can make the case that its actually better for game balance overall, but I haven't had the time to do all the leg work, and the devs are too busy with I17/GR to entertain the idea anyway. Once things settle down a bit, though, its actually the next big thing on my todo list.

Hypothetically speaking, the devs could make something, oh, that has a software renderer that doesn't require vertex programs to perform mesh translation. Theoretically, such a thing would be forced to spit out opengl rendering primitives that could be intercepted and converted into a complete mesh model almost completely automatically in many cases. Such a thing could, if it existed, be converted into a model suitable for 3D printing.

If only there was a piece of circa 2005 technology out there that was capable of rendering City of Heroes player models, but wasn't optimized for OpenGL 1.3 or higher. And if only there was some magical software capable of intercepting OpenGL calls from OpenGL rendering software packages. And if only there was a plug-in for said software capable of converting OpenGL primitives into something like maybe wavefront format.

Of course, this is all completely hypothetical, because who would be crazy enough to try all of those things and discover that it generates perfectly formed player mesh models. I mean, that would be silly, Rube Goldbergian, and possibly violate the EULA. Hypothetically, however, if a red name happened to be curious about how such a thing could be accomplished, I might hypothetically be able to sketch out a hypothetical way to do that which would hypothetically work, at least hypothetically. Maybe with some hypothetically active emotes.

Whatever he did to draw that assignment, he's probably posting from above the arctic circle.

Actually, I just thought of a possibly insane way to do this. Might give it a try tomorrow if I have the time.

Its an unsafe presumption that doing it with the raw data is simpler than doing it outside of the raw data. In fact the safe presumption, historically speaking, is that doing anything outside of the raw data would be simpler. It almost always is.

The rule of thumb in situations like this is that its never as hard as the devs say it is, its never as easy as the players think it is, and the best way to do it is most likely to be a way both groups would initially think is insane.

Usually, when I use the phrase "non-trivial" I'm usually usually referring to the fact that either the problem is such that a cursory examination doesn't confidently uncover the entire scope of the problem so it has an ill-defined scope, or it does uncover the scope of the problem, but that scope contains elements that themselves do not appear to suggest straight-forward solutions. There's no other term that tends to convey the same sense of unknown complexity. Saying something is "difficult" or "time-consuming" suggests that you know enough of the problem to guestimate the amount of time and resources it will take to resolve, since you know its difficult or time-consuming. Saying its non-trivial is saying you don't know enough about the problem to be able to estimate that correctly, *except* that you know that whatever the problem is, it doesn't seem to suggest a straight-forward solution.

Which traps powers specifically seem to take longer to execute than their cast times would suggest?

I'm pretty sure the last time he started a 15 page thread, and he wasn't actually being burned in effigy within it, he had to steal a thread title, topic, thesis, and possibly a few words I just made up, from a thread I started back in 2005. The man has no shame.


That said, my question is: did you draw the short straw, or did you just lose a bet?

That is certainly significantly harder than the Impossible Cypher from a few years back. Might take a couple hours at least.

Since your actual on-topic contributions to this thread so far is zero, you'll excuse me if I consider your moderator credentials even more suspect than you consider my dev credentials.


Also I think I'm shorter than Castle. But a little less crazy than pohsyb. I'm probably tied with BaB on the forum trigger finger. Look at my ego go. My ego has a message for you: it asks "guess which one is my forum trigger finger."

Now *that's* a pot shot. My response to A_F is a rebuttal. I can provide you with more examples if you wish, just ask, and ye shall receive (unless I'm on a plane).

What does ego have to do with anything. I don't have to guess not because I'm so amazingly smart that I can deduce everything like Sherlock Holmes: I don't have to guess because I discuss these matters with the devs often, and have for years, and because in many cases I've actually seen, first hand, the specific things I'm talking about. When I say "Castle can't just trivially sort the powers system into damaging and non-damaging powers" that's not a guess, and not an attempt to impress you with my powers of observation. Its a simple statement of fact.

Its no more egotistical of me to state that the powers system doesn't work that way than it would be for Castle or one of the powers designers to state that, or for another player to state that Power Blast does 1.64 scale damage. For me, its just a simple, completely non-controversial fact.

If you think I'm being egotistical because someone is calling me essentially either a liar or a really bad guesser, and further implying either malice or intellectual dishonesty by claiming I'm deliberately constructing strawmen to obfuscate the problem and impede a solution, when what I am stating are the actual facts of the situation, and I've decided to pack the dirt onto the hole after pushing him into it, well, *that* sort of ego I have no problem with. If you're going to try to capture the intellectual high ground with me and you think you can get away with doing that balancing on one leg while standing on beachball, I might decide to tip you over face first and then laugh. I make no apologies for that. Heck: link to the thread and proclaim that facet of my egotisticalness anywhere you want, because I'm not ashamed of that.


(Getting better at iPad typing)

It *can* but apparently, given what the devs have stated, it apparently isn't on average, because the devs actually gave a formula for determining the number of merits that a task gives, which is *also* at the link I provided and I guess you didn't bother to read:

Lets shuffle that up a bit. The TimeModifier was defined to be just MedianTime/20, and the TimesRunModifier was just an uncertainty factor for any task which had not been run often enough for their datamining to generate good averages at the time the tables were computed. We'll assume TimesRunModifier was 1.0 for all the tasks we're concerned with at the moment the latest merit values were computed. And the ArtificialModifiers factor was acknowledged as a fudge factor mainly used to round up certain merit values that were just short of a random roll. Dropping that also, that gives us:

MedianTime * MedianTime * TaskModifier / MPM / 20

That's the actual reward computation. But we were talking about reward rates and how they related to difficulty. The reward rate is the reward divided by the time, and for our purposes, we'll call the intrinsic reward rate the reward divided by the median time. MPM is the baseline merit earning rate of 3.7, and that gives us:

Intrinsic reward rate = Median Time * TaskModifier / 74

I already listed the TaskModifier values. According to the devs, then, the merit value of any particular task is going to be proportional to a factor that increases *precisely* in the same way you ranked the difficulty of the tasks above, *and* also to its median time to completion. So basically, the reward rate of a task is proportional to its difficulty and the amount of time it takes to complete it.

The best explanation for the discrepancy between your estimates for the time it takes to complete those tasks and the dev computed numbers is that your estimates for what the median player does, as opposed to just what you've personally seen, is highly inaccurate.


No. What I said was that there were about twelve thousand powers in the system, of which probably a few hundred are the problematic ones. If you're not going to spend time actually reading my posts, don't expect me to take you seriously for much longer.


Yeah, lets. Lets not assume I've actually already broken the problem down as if I was interested in solving the problem, because I'm actually attempting to solve the problem. Because that would be crazy talk. I mean, who here thinks that's how I spend my time.


Stop. Problem #1. Are all buffs ally targeted? Accelerate Metabolism is not. Healing Aura is not. Dispersion Bubble is not.

To buff an ally, a power must at least be able to affect allies, but it doesn't have to target them. But even that is a little hazy. Some really important powers don't actually do *anything* at all. Fulcrum shift doesn't actually *do* anything. It casts two pseudo-pets that do all the work.

Problem #2. "...but do no damage." You'd think that was an easy one. But again, many powers work through pseudo-pets which you would have to trace manually. And damage itself? There's no flag called "damage" that is *guaranteed* to flag a power that does damage. There's a flag that usually indicates that a specific attribmod was defined to do damage, but there's no 100% guarantee its accurate at the moment, because until recently that flag wasn't used for anything by the game engine: if it was set wrong, no one would ever know (its now specifically being used by Dual Pistols, so it has to be correct for at least that set).

By coincidence, I actually had this conversation with a couple of the devs not that long ago. The logical filter for "power that does damage" was basically "power that affects Foes with attribmod that affects Target with an Abs for any of the attributes Smashing/Lethal/Fire/Cold/Energy/Negative/Toxic/Psionic/Special/Unique." Except it turns out to skip quite a few powers, like Rains.

An interesting odd exception to "power that targets enemies but does no damage" that I came up with: Gang War.


As everyone who studies this game and wants to know probably already knows, because its been common knowledge since at least 2006 if not earlier, the entire powers system is designed in Excel sheets.

I posted this "elsewhere" and rather than rewrite it, I might as well quote it:


There is no such database. There is no such entity of any kind query-capable that exists anywhere within Paragon Studios itself.


Your actually trying to argue with *me* that I'm creating improbable strawmen when discussing the actual internals of the game and its implementation? I sort of assumed that since you seem to think its open season on me, you might actually have some idea what the hell you are talking about. Since you don't, I'll spell it out for you:

Unless I specifically use the words "I'm just guessing" then when I say anything about the powers system, the powers implementation, the game code, the developers design tools, the developers design methodology, or the general game mechanics I am not guessing.

I almost never *have to* guess. In the specific case about how Castle manages and implements the powers system, the only person on Earth that knows more about that is Castle. After his morning coffee.


Custom allies are not necessarily exclusively the problem.


Accurately? As a matter of fact, I have a database of the custom critter powersets and reviewed it when this issue first arose. There are currently eight custom critter control primary sets that exist:

Demon Summoning
Earth Control
Fire Control
Gravity Control
Ice Control
Illusion Control
Mind Control
Plant Control

67 powers between them. But in terms of ally buffs or nondamaging debuffs, those are not exclusive to control primaries. In my first pass through the system, I found 27:

Dark Pit
Soul Drain
Quicksand
Earthquake
Volcanic Gasses
Lightning Clap
Smoke
Cinders
Wormhole
Shiver
Ice Slick
Glacier
Ice Patch
Chilling Embrace
Deceive
Flash
Spectral Terror
Confuse
Mass Hypnosis
Total Domination
Mass Confusion
Spore Burst
Seeds of Confusion
Spirit Tree
Drain Psyche
Fault
Hand Clap

There's more on the secondary side, because that's where so many of the buffing powers are, 107 in fact (I'm not going to list all of those). Definitely "more than a hundred" though. And that doesn't count iffy powers like Acid Arrow, which do minor damage but significant debuff.

It took me probably about a half hour to go through all of my own internal data to generate the two lists above, but those were very rough-cut passes through the system, and only for custom critters which I don't think you can say arbitrarily are the sole source of the issue they need to resolve (and actually, I wouldn't). To do a professional sweep, I would estimate an hour to double check those lists, and another two hours to scrutinize a set of borderline powers. And that's for me, and that's factoring in the fact that I just coincidentally have just the right tools and data in just the right form to do that analysis, purely by coincidence. I would need a day to generate such data for the rest of the powers system, including a cross reference to the critters allowed in the AE, and a couple days at least to analyze those data tables to come up with a list of all the potentially problematic problems within this context. And I would probably tell the devs that since they keep adding new critters to the AE, they should really do the ones that are in right now first, but immediately sweep the rest of the critters to prevent them from reopening this issue in a couple months.

And I really don't think there's anyone alive that can realistically do it accurately much faster than that, given the form the data exists and the amount of time it would take to become familiar with all of the corner cases that exist in the powers system, which I've spent years becoming proficient at recognizing at a glance.

If you have any other questions for me, they may have to wait because I'm packing up for some business travel. Whatever odd retort you compose I might not see until tomorrow. Although I'm crossing the dateline so I'm not sure if that is tomorrow tomorrow, or the next day tomorrow, or later today tomorrow. I hate long range flying.


* Note: Demon Summoning is not a control set, its a mastermind set. Rather than stealth edit that mistake out, I'll just mention that there's a reason why Demon Summoning was originally classified in my data as a control set and I never bothered to go back and reclassify it, and the list is a copy and paste from that data, but that's as far as I can explain that error.

CoH2

Your list conveniently ignores the parts of the game that the players probably spend 90% of their time playing. And yet:


Fact:

Unless the devs changed the merit formula and I didn't hear about it, this clearly demonstrates the intention is for raids to be worth the most, trials next, task forces next, and story arcs last. See: Merit Rewards.

So perhaps the devs made a computational error, which is possible, but would not change their intent**. Or maybe the error is with the person making up out of whole cloth the relative rankings of the rewards without actual hard data which is reasonably extrapolatable to the average player population, which is what the rewards are based around, and not, say, the average of the people in your household.


Yeah, reviewing twelve thousand powers should take about an afternoon. Why not help the devs out and gen up that list and PM them, it would save them at least half a day and they could get a jump on writing the suppression code Monday morning, probably be done by lunch.



** Their formula actually factors in both difficulty factors and duration factors in a way that is slightly odd but not inconsistent with the concept of acknowledging difficulty thresholds in different classes of activity.

But most of that is not relevant to what I said. In fact, I specifically said that its invalid to specifically compare two different things of different difficulty and expect the rewards to be predictably or proportionately different. What I said was that the game was designed more around thresholds of difficulty. So task forces, for example, which have a specific set of rewards, tend to have a specific threshold of difficulty. Superimposed on that is usually that things at higher security/combat levels are intended to get more difficult. But its clear the difficulty level for task forces tends to be higher than any normal story arc. The fact that Task Force A is more difficult than Task Force B but generates the same reward is actually expected: in a threshold model everything only has to achieve a particular level of difficulty, but anything higher isn't necessarily rewarded more.

An exception to the rule would be a case where a story arc with normal story arc rewards was determined to be objectively higher in difficulty than a task force with normal task force rewards. I don't think that is true for any task forces.

Also, optional difficulty settings have no specific requirement to generate special rewards by any game design rule I'm aware of. Just because you care about reward balance, doesn't mean if the players want a way to crank up the difficulty you are required to reward such behavior. The degenerate silly case is asking for enhanced rewards for jumping off a building into a spawn. Saying if the devs aren't going to give extra rewards they need to make sure you can't jump off of buildings would be bordering on psychotic.

In a threshold system, you can't go from difficulty to reward. You can only go from reward to difficulty. If A has higher difficulty than B, that doesn't mean A must have a higher reward. But if X has a significantly higher reward rate than Y, X should have a higher difficulty threshold.


You could start by making a list of all of the "problematic powers" and then extrapolate from there how much effort it would take to figure out a way to disable those powers on the fly from buffbots only. Or at least attempt to estimate the amount of time it would take just to list them much less fix them.

My back of the envelope estimate is that there are about six or seven hundred of them, out of about twelve thousand different powers in the game. Of course, only a fraction of those powers (probably about half) can ever show up in an AE mission, but I have no real way to automatically eliminate the impossibles from contention.

Risk vs Reward was always a phrase the devs used too colloquially, and the players abused into meaninglessness. But it has always had the same meaning, from the origin of this game to the present day. The game design has always reflected a design principle - even when the devs didn't describe it as such - that "rewards" are actually reward rates, because rewards measured outside of the time taken to earn them are meaningless metrics. And "risk" refers to specific thresholds of difficulty. Different reward rates have always had (sometimes subjectively) differing difficulty thresholds. Its the threshold concept that most often trips people up who think that "risk" is supposed to be *proportional* to reward, and that therefore increased difficulty *must* always be rewarded, and lowered difficulty should have only proportionately lowered rewards.

Because this isn't quantitatively enforced in both the reward systems and the critter design methodologies, there are lots of exceptions to the rule. But its not difficult to see the trend overall, if you are actually paying attention, and its always been consistent with how the devs have made balancing changes to the game.