Arcanaville

One reason why I think there's no gate on the crafting at the moment is because the Notice recipe is doing double duty as a component sink. I don't think the devs want to limit it: they want it to be expensive and they want people to use it to suck out excess components that are otherwise just sitting around. Doing that creates an incentive for the fastest earners to keep coming back for more.

Another reason why I'm usually very cautious about functional overload in any system design.

I don't know about the latter, but I am usually exhibit A on the former. Arguing against extremes usually means everyone forgets you're arguing for some position in the middle, and only sees you're apparently arguing against almost everyone else. I honestly don't see (m)any people survive there long besides me and still come back for seconds, and even that is for some questionable definition of "survives."

The problem is that its actually easy to avoid being influenced to change your opinion to be what someone else's is. Its not so easy to avoid being influenced to change your opinion to be the exact opposite of what someone else's is when you're arguing against them.


After looking at the numbers carefully, and given all the parameters I believe the costs should be balanced around, I believe a strong case can now be made that Notices are too expensive by about half. First, it was always illogical that Notices are made of the other components and shards because anyone that can earn all those components cannot *possibly* be unable to get the Notice directly. It was always about the base shard cost to craft everything. Those extra components appear to be filler, possibly added after the initial shard costs. Second, if we assume that crafting costs are intended to approximately double with each tier (separate from consumption of the previous tier), you'd expect a notice to cost somewhere around 40 shards, not 88. Third, if we assume the average soloing earning rate is about 1.5 shards per hour, twice the floor, then the time to acquire rare is about 75 hours, which is somewhat higher than my benchmark time of acquiring three combat levels in the upper forties: about 30 hours. You'd have to be able to reach nearly four shards per hour to approximate the benchmark time, which is unlikely for a soloer.

Three separate lines of thought are all independently converging on the same number, which tells me there's a specific reason why that number is significant and probably the correct one.

My guestimate is that somewhere between 40 shards and 88 shards is the correct crafting cost for a Notice, and I believe its closer to 40 than 88, if we're balancing solely around solo earning capabilities. I am still very concerned that lowering the costs that low will make it easy for people who team to exploit the low crafting costs, and no matter how much that gets hand-waved, I'm not going to dismiss that concern as being trivial or unimportant. I still believe that making the solo path to crafting use things teamed people will always, always earn faster is a fundamental mistake in design. But if the devs refuse to remedy that situation, they will be left with no choice but to create an exploitable teamed path to place the solo path within reasonable reach, or add the code to add some form of crafting gate to the Notice crafting. And even the gating is liable to cause some issues.

Difficulty is relative. We know the higher level of difficulty is difficult relative to our current playerbase. That makes it difficult by definition.

Some MMOs are more difficult than ours are, but I would argue that Going Rogue was and is not the opportunity to fiddle with the standard difficulty model, particularly at such low levels. There's nothing to be gained by that. There's only two possibilities at the moment: it scares off new players that don't want that level of difficulty, or it attracts people that want that new level of difficulty that will then leave when they discover the rest of the game doesn't have it.

Even if you think raising the bar across the entire game is a good idea, I see no logic to raising the bar an order of magnitude in the first twenty levels of an introduction section of the game only. You don't just mess with global difficulty standards haphazardly, because they are one of the foundational elements of the entire game. You can't just phase in a ton of difficulty at the beginning. You can theoretically ramp it up at the end, because there's nothing past the end. But you can't do it at 20 without doing it at 30 and 40, and making an entire leveling path that looks like it.

That's assuming you want such a path, and choose to introduce it without warning anyone, and have it ramp up as fast as it does, all three of which are questionable.


Very very far back. I don't think you want a piece of this one.

It makes it easier for AoEs.

As to my lower end test numbers, I tested my Energy Blaster at 0x2 with bosses off, and my Illusion/Radiation controller at 0x2 with bosses on, and in both cases I completely deslotted my Alpha. I then normalized the kills to one hour and carefully counted them up in herostats.

Blaster kills per hour at 0x2 with bosses off (rounded to nearest whole number)

Minions: 214
Lts: 86
Bosses: 2

Estimated shards per hour given Paragonwiki drop rates: 0.83 shards/hour

Controller kills per hour at 0x2 with bosses (rounded to nearest whole number)

Minions: 232
Lts: 63
Bosses: 12

Estimated shards per hour given Paragonwiki drop rates: 0.89 shards/hour


A couple of notes.

1. That's not a misprint on the Blaster boss numbers. Even with bosses off, there are lots of odd ways to actually get a boss in a mission anyway. In fact, it would be a good trivia question and test of game mechanics to figure out how many there are (I can think of four). In this case, the boss came from a merging Herc

2. In both cases, the number of shards per hour is less than one. That would be pretty slow to get 480 of them, assuming you had to literally get everything with shards, and could not solo above 0x2.

3. 0x2 is not a good setting to get bosses. The boss ratio is 1:5 Bosses to Lts. At 0x8 its much better. Somewhere between the two bosses start to gain in ratio, which improves the shard rate substantially. If you can defeat them without dying. My guess is that you have to solo at or above 0x4 to have a decent shard drop rate.

4. The blaster number actually includes two deaths. Factoring that in, the Blaster and Controller had almost exactly the same kill rate. Which is what you'd expect, actually, at such low numbers. Illusion is not a good AoE damage controller, its not high damage in general, and 0x2 is too low a target count to efficiently deploy radiation debuffs. And when you are averaging only three or four targets, deceive has nothing to work with.

5. The difference between my best measured rate (MA/SR in 0x8) and my worst (En/En in 0x2 no bosses) is a factor of over three to one, and that's not the widest possible range in performance for solo players by far. My guess is that the best solo performance is at least twice as fast as my MA. Balancing just the solo earning rate around a performance spread of six to one or better is non-trivial.

6. Assuming the critical target is Rare for a casual player that doesn't team and only solos at low difficulty, 128 shards at these rates comes out to about 150 hours. That's actually probably within a factor of two of the correct number in my opinion, considering we're talking about the absolute worst case scenario of someone that doesn't convert Vanguard merits, doesn't run the WST, doesn't run task forces that drop components, doesn't team, solos exclusively, and while solo cannot run at higher difficulty. This is the limiting case of someone that has by circumstance or design eliminated every possible way to gain heightened rewards and can do nothing except essentially kill targets one at a time.


I'm guessing this will provide some ammo for both sides of this debate, but its a data point regardless. I think its fair to say that the range of solo shard earning (on average) is likely to be somewhere between 0.75 shards per hour at the low end and 8 shards per hour at the high end, discounting people that are unusually slow in combat.

No, Praetoria is difficult. Its actually outside the standard difficulty model. It keeps you and me on our toes. But the original stated purpose of the Going Rogue expansion was to attract new customers. People who wouldn't have the benefit of prior experience. The tutorial was lauded as a better introduction into the game than Outbreak or Breakout. Which it is. Then at level 15 the game beheads new players with a very rapidly escalating difficulty level.

I have veteran attacks, seven years experience, and so much knowledge about the game I know what the AI is going to do two seconds before the server does. I think Praetoria is significantly more difficult. If it wasn't for inherent fitness, I'd consider only playing it by emailing resources to the lowbees to outfit them with maximum enhancements and inspirations constantly. And I never, ever do that now. That, or only playing masterminds there.

They really should consider having a lower level difficulty story path for new players that buy Going Rogue. I don't mind the existence of the newer more challenging path: for people like me its just that: a challenge. For most people, I suspect its an exercise in frustration.

Like those damn DE in the tip missions. Seriously, once Black Scorpion gets out from under I20, he and I are going to have a long talk about those. In fact, Zwillinger, you can tell Tim that I'm a comin', and hell's a comin' with me.

The point was to make the system offer something for the fast players to pursue, while balancing that concern with allowing the slower players to gain the majority of the benefit of the system relatively quickly. It takes only one week to get the one notice it takes to craft a rare alpha, which gets you probably 80-90% of the full benefit of any alpha slot. It then takes more than twice the total resources plus three more notices which takes three more weeks of pursuit to get the very rare alpha, which only adds about 10-20% more strength.

You're looking at it from the perspective of, since its not that much more power, why not let everyone have it quickly, since its not game breaking? You're not seeing it from the flip side, which is that the devs have carved out a deliberately small benefit from the total and allowing the fast players to go chase after it, and that benefit is so small it minimally affects everyone else. To me, the Very Rare Alpha is actually representative of the extreme care the devs have taken not to exclude most players from the vast majority of the system given nominal effort.

Seriously: games are supposed to be fun, but different people have different ideas of fun. Some people like pursuing long term goals. Some people like making model ships in bottles. Its not work: its fun. This game has to address both kinds of fun, and a lot of variants in between. This much is true: the more kinds of things you find fun, the more of the game will be accessible to you. That's not just the way it is, that's the way I want it to be. Something for each player not everything for every player. The latter sounds laudable, but its extremely limiting. What I want is not what you want. Making the game be only the intersection between the two, plus everyone else, is not in the best interests of the game.

Team of three = 2.66666666666666666666666666666666 per person.

I believe its reasonable to have a balance between random rewards and predictable ones. I believe the original suggestion preserves that balance. You're free to disagree. But tell me this: is it ok if I refer to your preference as "this predictability bull-****?"

South Korea follows international GAAP, which requires, I believe, such income to be amortized.

As to the curve, given the release of Going Rogue and the requirement to account for the free month, the most logical explanation for the sales numbers is that the majority of GR sales were to existing customers, and not to new ones. New customers were added at only approximately the normal rate for the game. Meanwhile, the game retained customers at approximately the same rate as before, with only small changes in total subscription base over the year.

When the Q3 numbers came out, I mentioned that this was my guess, but I needed Q4 to refine that guess. Q4 seems to confirm that guess. The drop in 2009 didn't continue in 2010 like some people predicted, but it wasn't reversed by GR either. We seem to have very roughly stabilized at a new subscriber base level in 2010.

Optimistically, you could say that at least the 2009 trend wasn't a continuing trend, which would have been problematic. But contrawise, I think its important to start to consider that the devs need to find a strategy that eventually gains net subscribers. At best, GR didn't cost us any. And I have to admit that while adding a new starting game like Praetoria is a strategy that can theoretically gain customers, end game is mostly a strategy aimed at retaining customers.

You have to wonder if the problem with Praetoria is that only experienced players can tend to survive there. That's a very problematic logical contradiction.

Until someone explains why their prediction of the game shutting down in two years is correct, while every other person who has made that prediction has been wrong, I don't find the prediction to be particularly interesting. Back in '08 people were predicting that within a year we'd be below 50k subs and non-viable. People were saying if the trend in '09 continued we had two years at best, one at worst. Now the accounting wizards are saying we have until 2013. Eventually someone is going to be right, but I don't see why they should get credit for that prediction when its made consistently, and entirely randomly.

The problem with these predictions is that they cost nothing to make. When someone comes back and admits they were completely wrong and as a result have revisited their understanding of MMOs as a result to attempt to prevent such an incorrect line of thinking from happening again, I'll be impressed. Predicting the demise of the game is not interesting. It just means its a day that ends in Y.

Praetorians are, what's the word I'm looking for, somewhat overdesigned.

When blasters were being reviewed in I13 and defiance changes were being tested, the subject of "stances" came up. The notion was that when the blaster was standing still, for example, his or her damage would go up, and it would return to normal when they moved, like when kiting or moving around in melee range. This was supposed to be a form of ranged buff: if you're shooting from range you have less need to move.

The reason why this was not pursued was because, as Positron stated through Castle, any benefit while doing something will be seen as a penalty when not doing that thing. They didn't feel that fight was worth fighting. The only action for which that was a difficult assertion was actually attacking. Because obviously no one needs a damage buff when not actually attacking.

This psychological trap that associates any circumstantial benefit with a converse penalty is a pox on all game development. Its irrational, and as a result it has only one remedy: psychological trickery. Personally, I'm not fond of that, because it doesn't treat players like adults. But I find its also absolutely necessary because, to use the logic against itself, the devs are always penalized for not practicing it.

On average, I think I die once or twice out of every five 0x8 tip missions I run, not counting the times I say to myself "I think I can beat this Nemesis spawn of 8 LTs clustered together this time."

You can in theory, but you're never going to convince me its a good idea to adjust the drop rate to place this within striking distance of all but the faster and higher activity players. For the same reason we are unlikely to balance drop rates around people acquiring Very Rare Alphas for a hundred alts. You get to choose which you want to do: progress deeply or widely. Giving people a drop rate that allows them to basically get everything for everyone with a low amount of activity defeats the purpose of having a progressional system with multiple paths, tracks, and tiers. Choices are worthless if you can just choose to have everything.

Deep progress is orthogonal to alts. For each player there is only so far their efforts will be able to go upward and outward. That's deliberate: each player will have to choose the appropriate balance point for themselves.

A full team running the same tip missions I'm running should clear them much faster than I can solo, and as a result each member on the team should earn shards significantly faster than I do. Each member of that team will probably, on average, earn the same *number* of shards that I do per run, but my guess is that each run will be approximately three times faster. I would not be surprised to see a clear-all team averaging closer to six to eight shards an hour.

A full team steamrolling 0x8 would probably also be bored out of their minds. Solo, I could (and occasionally do) die, so there's real risk and tactics involved. I don't fall asleep in those missions.

(Also, I don't care how strongly you build: no solo build can *completely* ignore *all* the threats in random tip missions. Hand-picked opponents, yes. But I don't think the solo build exists that can *completely* ignore the threats from sappers, quartz eminators, stacked vengeance, master illusionists, DRMs, arachnos assassin's strikes, and the fruit cocktail of powers from Longbow bosses equally well. *Something* will eventually get you, even if you are the world's best stone tank. Maybe only once in a hundred missions, but they will get you.)

I have two tests I'm planning on running tonight if I have the time. Blaster at 0x2 with bosses off and Ill/Rad at 0x2 with bosses on. Ill/rad is actually an interesting test case because its almost impossible to die in a 0x2 anything, but its actually not a very quick killer, especially in a 0x2 mission. It will deal similar damage to my blaster actually, but with absolutely no need to pause at all, for endurance, health, or anything else. Neither my energy blaster nor my Ill controller are speed killers (point me in the direction of an AV and I'll dispatch it without breathing hard, but killing masses of minions, LTs, and Bosses is not what Illusion/Radiation does best; killing them one at a time is actually almost painful).

I was going to do some badge catchup tonight before the next WST, but I think I can squeeze these tests in. I was on a team this weekend where I had the *lowest* badge count on the team. That's never happened to me before while on my badge hunting main. At the time, my badge count was 796. With the new explore accolades and side switching, apparently eight hundred badges is like having two veteran badges. Its cute, in a minimalistic sort of way.

Unfortunately, I'm currently farming alignment merits to try to complete my I19 build updates, so I can't side switch and pick up the bulk of the red side badges on that character yet. Oh well.

Anyway, why 0x2 instead of 0x1? Because I want to see the difference bosses make, and I'm afraid if I go all the way down to 0x1 I'll stop seeing bosses altogether (except at mission end). I'll be satisfied extrapolating downward from 0x2 to 0x1 for now.

I have a separate suggestion for the difficulty issue**, but the splinter suggestion does offer a difficulty escape valve that does not exist for a *locked* solo mission arc, and that is you could ask for temporary help. That may not be a general purpose solution for pure soloers, but it is an option that exists. Furthermore, it generalizes the solution from just solo vs teamed, to teams of all sizes. Players who don't pug but do duo or organize small teams with friends would benefit from a scalable solution better than one that actually forced them to solo.

Forcing people to solo is no different from forcing them to team. And the important target from a size of demographic perspective is not pure soloers but rather casual players. Casual players often, but not exclusively solo. We would, to be frank, never notice the loss of pure exclusive soloers. We would not survive the complete loss of the casual players. So any solution that purports to help soloers has to be able to work with more casual (but not necessarily arbitrarily time-limited) players as well. It cannot be targeted at pure soloers and no one else in my opinion. We need to address soloers as a non-exclusive subset of casual players, so that the overlapping class of mostly but not exclusively solo players of average performance work reasonably well, for some definition of reasonably.


** Two, in fact. The first one is mostly mine: it involves creating a structured mission arc that allows for randomizing the mission maps and foes but with a consistent theme, with the in-game story justification that the Well was testing the player in a variety of situations of increasing difficulty. The randomization would be done in a very specific and structured way to prevent mission burn-out, and would have side-tasks designed to assist the player with the escalating difficulty within the mission. The second was inspired by a counter-suggestion by EvilGeko where we use carefully selected Ouroboros arcs to complement the WSTs, but with heightened "challenge settings" and a similar way to earn temporary boosts to counter the increased challenge.

Fun fact: during I14 I abused the heck out of the ambush mechanic while testing the architect, and I think in no small part because of me they reduced the number of ambushes you could add to a mission to 3 (its now 6 I believe). I was told that adding that many ambushes to a mission was simultaneously abusing, and potentially exploitive.

The number I was using at the time? Twelve.

I had come up with an alternative reward mechanism to do that:

In 2005?

If I remember correctly, I ran from Ms Liberty to the Skyway tunnel, from Skyway to Talos, from Talos to PI, and from PI to Portal Corp. The last "shortcut" I took to win the race was to get immediately killed walking out of the Ferry, rez at the PI hospital, then go out to sea and swim to the north end, zig zagging over and down to the courtyard.

That prompted a second race with a rules change: get there the quickest without dying. I lost that one by about ten seconds. Without teleporters, this can really test your knowledge not just of the zones but also of the critter distributions in the zones.

At one time in this game, before Pocket D and other teleporters, it was a small mini-game to figure out ways to get from point A to point B with the minimum travel time. If you fly a lot like I do, this was an important game to get good at if you expected to team at all.

I think Snow Globe started getting much better drop rates ever since he started monitoring his drops with herostats. Which supports a long-standing theory of mine that herostats is one of the better buffs you can have running. In fact, I think while it was monitoring his shard rate his purple drop rate increased by a factor of a hundred as a side effect.

Actually, what I've been noticing is that when people get far lower than expected shards, or anything higher they tend to speak up. When someone gets slightly lower they almost never say anything. That tends to skew the psychology of reported numbers.

I was on a task force once where someone said "I got six." Three other players said they got six or seven. One said they got two. I said I got five. No one else said anything. That was not an isolated incident. I presume the other two likely had some number between five and three, but didn't say anything. Thus, the numbers people will tend to remember are six, and two. If I hadn't said anything, there would have been no reports of drops between two and six.

You on the other hand always know how much you yourself got, and it will statistically often be near the average. So you will tend to get the average, and hear about everyone that gets more than you or a lot less. You'll tend to think most people either somewhat luckier than you or very unlucky.

My blaster solos at 0x2. I'm trying to get the stats for her earning and kill rates, but that's a bit trickier because she teams more than she solos (lately) so I don't have a lot of whole herostats files with pure solo stats. I'll have to do some surgery on them to get clean numbers without any teaming at all.

I might just set up a test run and run a bunch of missions at 0x1 and 0x2, just to get a rough idea of kill rates separate from looking for average shard drop rates.

I'm not sure a blaster (datamined to be one of the lowest performing, primarily due to dropping dead a lot) running at 0x1, especially with bosses off, will be helpful in judging the drop rate, since that might be, essentially, "off the map" in terms of balancing performance. Technically, it will also be repudiatable given that I'm not the typical soloer. But it will be a datapoint, even if its an extreme floor in terms of character, and a high datapoint in terms of player.

One more thing: we are extrapolating player performance in earning shards given their current performance. We need to factor in that with every slot they open and craft, their performance goes up. My scrapper might only notice a small difference in net kill speed with level shift, but I do notice it get substantially easier to run x8. My blaster notices a *huge* difference using level shift. I'm actually thinking about cranking up to x3. And that is just Alpha. With each slot, the next one should be easier to get (while soloing), because the player will be all the more powerful. There's no reason not to assume that a player that just has commons for all four Incarnate slots (so far) and manages to snag just one rare in Alpha (which gets them level shift) won't be able to dramatically increase their shard earning rates if they weren't already high to begin with. I can't go any higher than 0x8 in a way that materially increases shard drop rates. But someone that is only able to play 0x1 or 0x2 has a lot of room to increase drop rates significantly.

I don't know. El/Shield should smoke me left and right. I run tip missions specifically. I notice tip missions often generate a lot of bosses, perhaps more than newspaper/scanner missions.

I also run heroside, so my tip missions are a mix of freakshow, carnies, arachnos, longbow, nemesis, Rikti, Malta, and DE for the most part. I often drop the DE on my SR scrapper (cheating bastards) but I usually try to plow through the Nemesis (cheating bastards). Malta, Carnies, Rikti, and Freakshow might as well be barrels of gasoline to my MA/SR: except for the MIs in Carnie missions, which I specifically save and burn insps for, they go down very fast. I haven't tried this redside yet: perhaps the mix of enemies red side might be harder on average: Arachnos and Longbow are a bit harder (at least for me) than average, and maybe they are overrepresented red side (just as they are in all red side content). There are whole days when I don't see either in blue side tip missions.

One other thing: my older stats are from having Alpha, but no level shift. My newer stats have level shift within them. I'm trying to see if I can tell what that did to my kill speed and earning rate. They fluctuate from day to day so its important to try to average that out. In *very* broad terms level shift didn't increase my kill speed dramatically enough to notice. But even if it increased my kill speed by 20%, I wouldn't necessarily notice that yet in the rough numbers I am looking at (it should increase damage on target by 11%).

I'm not saying the level shift is worthless, just that given the really big numbers we're talking about and the large error bars in my measurements, its probably not large enough to swing the numbers from "ok" to "unacceptable."

It does not. There's only one way in which the random number generator generates behavior that I can determine to be non-random, and its a way no one could possibly ever notice.

It would be difficult to explain, but let me try this way: suppose we were to make a game in which we would bet on what the *last* digit of the next tohit roll would be. As in: "you rolled 36.57%" - we'd be betting on the "7." And suppose we were picking numbers like on a bingo card, where I could bet on different numbers depending on the previous two, so that if the previous two were "65" I would bet "7". But if they were "55" I would get "3" for example. In that case, I would win if the roll was X6.57 or X5.53. There are one thousand possible combinations of bets, and they should be all random. However, if we both pick 500 of those possibilities and I pick first you'll lose to me about 55% of the time.

This isn't due to the RNG being bad, I don't think, but actually the round-off algorithm they use to round the rolls off to the nearest hundredth of a percent. Its slightly alignment-biased.

On the other hand, when that Oberst boss crits you in the face, that's basically random.

There's one other oddity. Technically speaking, there should be ten thousand possible rolls: 0.00 all the way up to 99.99. But there are in fact ten thousand and one: 0.00 all the way up to 100.00 - that is actually ten thousand and one possibililities. That is also due to round off. And 0.00 and 100.00 are both only about half as likely as 0.01 or 99.99 or all the other rolls in the middle, just because they are in effect "sharing" one ten thousandth of a chance. Also not a problem in actual play because there's a 5% miss floor and 95% hit ceiling anyway.