Arcanaville

When an engineer says something is colloquially more or less complex, the meaning of that statement is usually pretty unambiguous.


I can actually feel the pain all the way down to my engineering degree. I would say these statements are impossible, unworkable, and inexplicable, in that order.

A 100x100x100 building would be about 30 times larger than one of the two WTC main towers, by the way, which were considered one of the most complex super skyscrapers ever built at the time of its construction. Some design elements, like the express elevator system, rewrote the textbooks on skyscraper design.

Please tell me you are not pursuing a career in civil or structural engineering.

And for the record, ten 10x10x10 buildings are only 1 percent the volume of a 100x100x100 building. But correcting that numerical error doesn't save the assertion. Not even a thousand 10x10x10 buildings would approach the complexity or maintenance level of a single 100x100x100 building.

No "free to play" game is genuinely free. Someone is paying to keep the lights on. All free to play models I'm aware of are predicated on two assumptions. One: that some percentage of people will pay to have things others will be willing to work for in-game, and the amount of money that percentage of people are willing to pay is high enough to pay for the rest that won't pay. Two: by not charging a subscription, the number of people that will try your game will increase dramatically, and you'll be able to retain a large enough fraction of them for long enough to make money off of them.

To make "free to play" work, you actually have to figure out how to make a lot of money selling things to your players other than monthly access rights. This is not a trivial decision, because it requires a fundamental shift in how you develop the game. Even the way the game functions has to change in some critical ways because concepts such as "balance" and "progress" tend to mean completely different things in a free to play game where people can generally buy what others have to earn through time. It isn't a question of whether you can make money any other way, but rather if you can make money this way, because that's not a guaranteed thing.

You also generally have to assume that you're writing off the current player base. Some will stay, if you're lucky even most will, but the values of the game will inevitably change, and many of them will not appreciate that, and there will be nothing you can say to change their minds.


If it happens here, it happens. I won't be predicting it, but I also won't be shocked if it happens. But if it happens, I can say this: I will be rooting for the devs, I will even be doing my part to help the devs if I can. But I won't be betting big money on them for the simple honest reason that its just not an easy thing to pull off, no matter how many success stories you point to in other games.

I'm not sure Rags to Riches was really all that interesting.


As to what Castle is up to, I hear Floyd is going to be working on the next generation in electronic entertainment. It involves latex and high voltage, but that's all I could get out of him: its all hush-hush.

As you point out, the physics of the situation is often a wash. But what tends not to be is the simple fact that at some point, the systems you use to solve a problem become so big they generate problems of their own, and that tends to happen as things scale upward. Eventually your cooling systems need cooling systems, your backup hydraulics require backup hydraulics, your engines need engines to start them. The space shuttle main engines use high pressure turbopumps that are themselves basically very powerful jet engines just to pump fuel into the *real* engines.

That's a potential danger, but there are lots of ways to safeguard against it. Although to some extent its not a big deal: its only a big deal in a few critical areas such as mitigation or powers so strong we don't need to encourage people to buy them already, like the emergency radio.

I don't care nearly as much, on the other hand, if blasters can buy 500 charges of the slugger, or scrappers can have 300 charges of the revolver for runners, or if people can have an unlimited amount of power analyzers. Powers with long cool down times are also good candidates for this kind of thing, and to the extent that there aren't that many, we could make some.

The goal is to make the powers good enough that they are worth using, but not so good that they become essential gear. Right now, many temp powers are so weak that you're almost self-nerfing yourself by using them because they aren't worth the cast time, and certainly not worth the crafting time. They probably drop at one hundred times the rate they are actually used.

Examples of powers that could be useful, yet not overpowering to have in high charge quantities (in PvE: PvP would likely have to exclude or modify these):

Vitality booster: +20% health for one minute, five minute cool down.
Teleportation Snare: long range teleport foe (for runners), seven second cast time.
Accelerated Healing: +300% regen for 20 seconds, +100% regen for 40 seconds, +50% regen for one minutes, six minute cooldown.
Emergency Force Field: 10 second emergency PFF, 0.5 second cast, ten minute cooldown
Critical boost: next five attacks have 15% chance to crit, five minute cooldown
Temporal Accelerator: +500% recharge for five seconds, ten minute cooldown (useful to break out of super-slowed situations).

Stuff like this, if not exactly this, would likely fall under the category of "nice to have, not mandatory to have, but probably at least worth crafting and having around, or using while I have it." The idea is to use either cooldowns to make them limited, high cast times to make them not as good as alternate analogs such as power pools, or situationally powerful without being globally powerful (numbers negotiable).

Why its a bit hazy is that *some* of the added complexity of the cruise ship is inevitable, but some of it is voluntary. The propulsion system of the cruise liner is more complex mostly by necessity, as it has to drive a much larger craft. But things like the computerized bridge is partially voluntary: obviously a hundred years ago large cruise liners didn't have computerized bridges. The cruise liner has a theater not because all big ships have to have theaters, but because the cruise liner's requirements include entertainment facilities for paying customers, something the houseboat doesn't have the same requirements for.

The real question is: if you wanted to make a seven hundred meter houseboat, could you make it just as simple as the small one if you wanted to. And the answer is: probably not, not even if you tried. But you could make a seven hundred meter houseboat that was far simpler than the cruise liner. And that's what makes the cruise liner a blurry example of the complexity scaling.

On the other hand, while there's some of that fluff in a commercial skyscraper, they are usually so focused on efficient usable rentable commercial space that the vast majority of complexity in the structures is mandatory. 70 story elevator systems are not just taller 7 story elevator systems: it just doesn't work that way. 70 story plumbing is not just 7 story plumbing times ten. My guess is that a starship would have similar scaling issues: you couldn't just glue a hundred little ships together and make a functioning ship a hundred times bigger. You'd have to design for that size, and that design would be unavoidably more complex. Power distribution becomes more complex. Thermal regulation becomes more complex. Internal transportation, fire suppression, communications, emergency escape facilities all do not scale linearly and would need to be designed for the target size explicitly.

That's actually why I mentioned warships separate from commercial and pleasure craft. Warships generally have very targeted design requirements, and all of the complexity in them is mandatory, not optional, relative to task, and the tasks for comparable ships is (or can be) similar. An aircraft carrier is far more complex than a battleship, but that's not just due to size: its due to the aircraft carrier having a totally different design direction. But the complexity differences between battleships, cruisers, and destroyers, say, are much more comparable due to their much more similar (but still not identical) design objectives.

Both of these reasons are explicitly things I was thinking about when I formulated the PB change. And its a variant of a change I first suggested for scrappers back in I5 (to compensate for GDN), reiterated in I6 (to compensate for ED), again in I7 (when defense was normalized against critter accuracy), and especially in I9 (with the introduction of the invention system).

I don't even know why I have to do it.


Actually, I kinda know, but am not currently employed by NCSoft.


Which is why, out of respect for them I make sure any rumor I invent about their departure is suitably epic.


For example, I heard the other day that this whole thing is Castle's fault. One day Castle decided to redesign Fireball so that every time someone used it, the game servers deposited a penny into his retirement account (ever wondered why fireball has such a short cast time? now you know). Castle figured in a few years he could have retired sitting pretty. But then datamining showed that blasters were dying too quickly and not using fireball enough. The ones that were not dying had such high defiance buffs that they were one-shotting everything and not using fireball enough. So he reduced the damage buff of defiance, and made blasters partially mez-immune. This helped greatly, but then blasters starting doing so well that blasters were using hundreds of thousands of fireballs. Pretty soon NCSoft didn't even have enough money to pay BaB and had to let him go, much to his surprise. At that point Castle knew the game was up and by year's end the auditors would catch up with him. Knowing that his finances were being monitored he transferred his ill-gotten gains to an offshore account under the name of an accomplice. Castle would then leave Paragon Studios and his accomplice, citing new opportunities out of state, would leave California and meet up with him in Aruba a few weeks later where they would split all that perk pack money.

And that, people, is why blasters were buffed.

I'm not sure if those are the best analogs because those craft are designed with completely different requirements. However, it is a general rule of engineering that complexity is not a scale-invariant. A better example is skyscrapers. You can find skyscrapers with comparable design targets and intents in all respects but scale, and you find invariably that separate from all the complexity issues of every five story building, fifty story buildings have to solve problems unique to their scale. They never gain the ability to ignore the smaller problems in the general case, though. The same thing holds true for things like naval warships, computer data centers, cities, even integrated circuits. Complexity tends to increase usually by some small factor relative to scale.

Its a truism of systems engineering that no one (knowledgeable) disputes.

There is a heavily underutilized opportunity to do that: temp powers. Most are totally worthless (slugger). Some are actually worth a lot (the recovery aura one, whose name escapes me at the moment). If the crafting costs of those were made level-dependent so it could be set to reasonable costs for any character to make them, and such things had other properties (like being able to store extra "charges" for them up to a limit that was higher than "you can only craft one") I think you could make a bunch of those that people would be willing to craft, and thus spend inf on. Basically, disposable powers.

Temp powers can't be traded (the recipes can, but not the powers) so these things could be cost-normalized to level and draw inf from all levels.

Of all those changes, the one that I think could most be backported reasonably to non-tankers is the practiced brawler ones. The defense and scaling resist changes are specifically intended to meet the intent of archetype scaling and thus explicitly intended for tankers. Normal archetype scaling doesn't work real well for SR: the passive resists wouldn't get any stronger due to their implementation, which is wrong, and SR defenses would go up to 40% defense on tankers, which is probably way too high. They'd be basically perma-eluded to everything by the twenties with slotted combat jump.

Keep in mind Black Scorpion runs the show now, and he'd have something to say about all of this. And I'm guessing its something like "oh hell no."

That is what I was thinking.

If you mean "wouldn't deal" in the sense of not giving the player tons of influence to buy purples, it doesn't do so deliberately. What it does is attack the problem on the supply side, by adding supply that the reward tables already say should go into the markets and eliminating the barrier from them actually getting to them.


I thought of that, but it has some problems. First, it requires players to actively check overflow. Second, it creates the problem that if the system forces the player to sell at a particular price they can feel cheated if they see what's being sold. It would create a source of complaints I would not want to deal with. Conversely, if it allows the player to sell at their own chosen price, overflow is all but functioning as additional slots and it provides the opportunity for players to manipulate the system that are in this case avoidable.

The concept is not that players shouldn't have recipe caps and this is a clever way to outsmart the devs. The concept is that when a player is supposed to get a drop but their own packrat behavior or lack of selling prevents them from getting that drop, the entire economy shouldn't suffer because of that. So in fact, conceptually I see this as a phantom market bot getting all the drops that players are prevented from getting and listing them on the markets to increase supply. The fixed price return to the player is a token gratuity that is really there less to give them more influence (I'm trying to suck influence out of the system, remember) and more to give a constant reminder that they are losing drops by not selling, and tease them by never specifically telling them what they could have been.

Some say he abused Excel so much he made Clippy cry and run away. And he considers decimal points completely optional punctuation in powers definitions. All we know is they call him The Castle.

Some say game design, it is a hunger; an aching endless need.

Some believe that this game here, began out there. With tribes of developers who may have been the forefathers of the Islandians, or the Ultimans, or the Evercrackians. That they may have been the architects of the great RPGs, or the lost MUDs of the Internet. Some believe that there may yet be brother game designers who even now fight to survive, somewhere amongst the development community.

If I was tasked with porting SR to Tankers, I would start with the Brute version, add +Health and toxic resists to Practiced Brawler, and increase the scaling resistances from their current 20% max per passive to the original 25% max per passive from CoV beta.

*However* I would also rebalance the SR defenses so that they did not just scale upward with archetype. Toggles would increase from 13.875% to 15%. Passives would increase to 7.5%. Fully slotted Tanker SR goes up from 30.42% (scrapper) to 35.1%.

SR for Tankers:

Defensive toggles: 15% to melee, ranged, AoE
Defensive passives: 7.5% to melee, ranged, AoE, plus scaling resists zero to 25% all but toxic and psi per passive
Practiced Brawler: Mez protection; +10% MaxHealth 120s dur, enhanceable; 15% resistance to toxic, enhanceable.

Something like that would be a good starting point, tweakable in beta testing. Not saying I would necessarily recommend this, just that this is what I would do if I was tasked with doing it.

That wouldn't work. The sale would execute normally, which means it would execute against the lowest offer, not yours. Only if you found an item that had no sellers would this work, and the system would need safeguards against that. That's one of the reasons I suggested compressing recipes to 5 level increments rather than individual levels.

In a sense, it does inject influence into the system, but that's a compromise attempt to strike a balance between liquidity and inflationary pressure. I do not believe focusing on inflation at the expense of liquidity or trading volume is appropriate at the moment. Liquidity is the more serious problem. With proper inf sinks, inflation eventually solves itself.

This is an incredibly dangerous thing to do. You're assuming the markets behave like economic textbooks, but even the textbooks will tell you that those rules only operate under a certain set of conditions, and illiquid ones violate those rules. When you suck demand out of the system, you don't end up with lower prices, you end up with less sellers who can't sell anymore, thinly traded items that wildly swing in price, and obstinate buyers and sellers that are more willing to retain huge bid/ask gaps because they know they can't expect quick execution anyway.

This is really another example of trying to solve a problem of the markets by destroying the markets themselves. Any solution to market issues that has as a component removing participants from the markets is a market-destroying suggestion. Its illogical to enact because it presupposes that the problem with the markets is the markets. Again: if you don't want the markets to set prices, you eliminate the markets. You don't make the markets a marginal sideshow to the stores.


A lot of the talk surrounding the markets reminds me of a guy I know. Once upon a time he wanted to sell his house and buy a better one. The problem was that it was always the wrong time. When the housing markets went up, the houses he wanted were unaffordable. When the housing markets went down, his own house became difficult to sell to get out of. So basically, high prices hurt him and low prices hurt him. What he wanted was a magical moment when his house went up and the kind of house he wanted went down simultaneously.

I think that's the problem a lot of people have with the markets. High prices shouldn't hurt players. They make it more expensive to buy, but it should also make it easier to earn influence by selling. The problem is that they can't sell what they have and buy what they want because there's still a gap between the two. What these players don't realize, I believe, is that this is not caused by inflation pressure. Its caused by scarcity. Some things will be worth more to players than others, and you cannot convince the players that have them to give them up for what you have when its not worth as much. It doesn't matter how many zeros are in the numbers: it will still be the case that the LotG I have I will never give to you for the two sniper recipes and stack of scientific proofs you have. And the game cannot really make me. Of all the problems you can attempt to solve, this fundamental one is not solvable, because its not really an economic problem. Its an attempt to force players to value things in a manner other than what they really value them as, and that's basically impossible. If the game said I could only sell LotG's for 50k, I would sooner give them away: the person I give it to would appreciate it more than I would the 50k. You cannot make me value it less than what its worth, no matter what sort of economic engineering you attempt.

I was sort of giving you the benefit of the doubt that you believed the assumption was both cynical and reasonable. Because if not:

Then you've just said that there is no possible way for something to happen, and anyone that thinks otherwise is being ludicrous, based on an assumption you admit is not explicitly reasonable.

And people call me the numbers-crazy one.

I never said to neglect anything. I was replying to this:

Because anyone who thinks an objective look at Dark Melee is likely to conclude it needs significant improvement is brain damaged, or statistically challenged.

The devs don't really nerf things unless they think they have to, but I would place the odds against an honest balance pass on Dark Melee improving the set, verses reducing the capability of the set, at about 75 to 1 against. That is not an exaggeration for effect: those are betting odds.

In todays balance terms, Soul Drain is too strong, Touch of Fear is too strong, Shadow Maul is too cheap in recharge and endurance, Siphon Life pays nothing for its heal, Smite's DPA is too high, and the entire set's stackable -tohit offers too much mitigation. If the powerset was made today, the odds of it surviving closed beta in its current state are exactly zero.

None of these are guesses on my part either. They are based on the balance discussions and tweaks that have occurred for other powers, including very recent ones such as Power Siphon and Cobra Strike. Basically, Dark Melee, and for that matter Claws, are sets with design rules that will never be replicated or even approached again, because their very existence breaks rules that are otherwise so strong today that when they are broken in other sets that is usually considered a bug that must be fixed.

Wow, that's pretty good. That's seven of the eight I was thinking of.

The eighth one was one I thought would be difficult. Blind. Few people probably even know or remember that Blind has a sleep component, much less that it uses a pseudo-pet to generate it.

The engineer in me really has only one thing to say: don't put all your eggs in one basket. What matters is not how easy it is to fix one ship verses hundreds. What matters is what do you do when your one ship breaks. And no matter how much redundancy and fault tolerance you put in it, it will eventually break with your entire species in it. And I'm not thinking break as in light bulb. I'm thinking break as in genocide.

Even if you had two or three, that's still way more concentrated value than I would ever recommend in a system intended to keep an entire species going. When you're talking about the survival of your race, six-sigma doesn't seem all that impressive anymore.

Suppose the Vulcans were actually as on the ball as you expect them to be. We have an unknown vessel that is firing an energy weapon at the planet that is disrupting communications and transporter technology. It has destroyed a fleet of starships. At this moment, the logical thing to do would be to order the population in and around the area of the ship into civil defense shelters probably underground. The rest of the planet would be similarly warned to keep away from potential military targets such as spaceports. It would be illogical to head into space at this point in the face of an unknown enemy that is primarily shooting up starships and generating electromagnetic disruptions. No one either on Vulcan nor on Enterprise had any idea that the safest thing to do was *not* to head underground shelters until Nero shot the red matter into the borehole. At that moment, most of the population was not primed to evacuate the planet.

How exactly do you prepare to evacuate an entire planet without actually doing it? Perhaps to their detriment, the Vulcans are unlikely to go on TV and say every man for themselves: make a break for it. They would have proceeded orderly and logically and my guess is many shuttle pilots died on the ground waiting for orders to depart or their passengers to arrive. They would not leave with empty craft when there were people heading to them for evacuation, and few if any knew exactly how much time they actually had. It might have been as little as two minutes. And panic is an emotional response.


Edit: I'm reminded of the Decker error. You're in command of a starship that is investigating a string of destroyed solar systems. You come across an incredibly powerful weapon that you cannot dent and is blowing your ship apart. So you beam your crew down to the safety of a nearby planet, whereupon the incredibly powerful weapon decides you're not interesting anymore and proceeds to destroy the planet you transported your crew to. In retrospect, perhaps not the best move. You are, after all, investigating destroyed solar systems. But you believe at the time that you're facing a weapon that will eventually kill you all and can do so easily, so you make the call, and it ends up being the wrong one. The Vulcans were in a similar situation. Unknown powerful enemy vessel capable of destroying an entire fleet. Logic suggests not sending up any more vulnerable starships, and protecting your people on the ground. Because logic doesn't warn you that the vessel is about to eliminate the ground.

Except even the Romulans don't dispute the basics of the Vulcan/Romulan split. Maybe it was because Surak had bad breath and that fact is now lost to history. But its still obviously the case that the militaristic and empire-building values left with the Romulans, because there's no evidence of a Vulcan Star Empire and there's ample evidence of a Romulan one. If Vulcan had even a tiny fraction of the imperial aspirations of the Romulans, Earth would have been conquered by Vulcan before Cochrane's flight.

Moreover, Surak's teachings were distorted at the time of Enterprise but its clear that this was for the most part corrected with the discovery of the Kir Shara. By the time of either TOS or the Abrams Trek, the teachings of Surak would have been properly known. I'm pretty sure they included stuff like not conquering neighboring worlds somewhere in the fine print. Its also clear that Romulans rejected Surak's teachings of using logic to control emotions and passions.

Why is that a reasonable assumption? If the objection is that Vulcan is relatively insular and isolationist, then no percentage is necessarily reasonable. To take it to an extreme, if I were to say that millions of Vulcans were cannibals, based on the assumption that only one tenth of one percent of them eat other Vulcans, its clear the percentage is meaningless. There's no basis for assuming that percentage without evidence. Similarly, without evidence, there's no reason to assume some percentage of Vulcans must be off-world.