Arcanaville

Outside of PvP, very little. Worse, its almost impossible to know that it ever had an effect, because there's no way to know if your target would have moved if you hadn't immobilized it, and the immobilize doesn't have an obvious visible sign that it even took effect.

Ranged immobilize makes sense for runners. Melee immobilize has some benefit, but as I said a single target melee immobilize that is only mag 2 by default is probably the lowest possible foe effect that can still be called a foe effect. It would be interesting if it was in a PBAoE - it would then paralyze potentially a whole bunch of minions. If it was ranged it would be interesting: it would then be Impale. If it affected mag 3 or 4 it would be, well not exactly interesting but not exactly trivial either (bosses are less likely to run usually anyway). Single target melee ranged mag 2 immobilize is trying to make the worlds weakest mez and succeeding in spades.

I suppose 10% chance for single target melee ranged mag 2 immobilize would be weaker, but I think even Geko had to draw the line somewhere.

If the damage extended outward, we'd have the AoE factor issue again. If it was just the secondary effects - immobilize and slow - then AoE factor would not likely be a problem, but honestly that would require a design exception and I personally would not want to waste a lifeline on that request.


I'd certainly take it, but combined with Storm Kick MA's post 32 single target damage would start to rise to very high levels under significant recharge, and as I mentioned previously, I haven't ever successfully made the case to the devs that MA's raison d'etre should be best single target damage set.

There was actually a brief period of time when it was, but that was when TK was broken with too high a damage number, and every other scrapper primary had much less single target damage than they do now. The speed ups that BaB put into the melee sets, while welcome, made it much more difficult to make MA the undisputed single target king without getting too far out of hand.


That's ... unlikely.

It was discussed long ago that its an urban legend that MA ever had a 10% accuracy buff. It always had a 5% accuracy buff going back as far as Castle checked. The 10% number was a player guess that took on a life of its own but never confirmed through testing or dev confirmation.

Its my idea, and I think it represents the current state of the powerset as the devs see it, but its not everyone's idea.


Unlikely. The devs took away mag 4 from total focus because they didn't feel it appropriate to possess a mez strong enough to take out a boss in one hit when you were not even a controller or dominator.


Suggested that one a couple of times. Has some balance issues that would need to be addressed.


Maybe. It would depend on whether or not CAK could get an exemption from the AoE factor equation.


Adding critical chance gives a way to increase the damage of EC without tampering with its recharge. But I think in general I like the idea (not originally mine, but I've repeated it a lot) of EC having a chance to do knockdown to complement its stun. Knockdown would work against bosses better than the mag 3 stun, MA's stuns are almost *designed* not to stack with each other anyway which makes stacking to affect bosses for any significant length of time usually impractical for MA. The animation also seems to suggest the effect so the effect isn't inconsistent with the visual appearance of the power (at least the standard animation of it).


Overall, though, I'm thinking a lot less "minor tweak" and a lot less conventional. I think in keeping with MA always being the unofficial "don't hate me because I'm beautiful" powerset, any change to the set now should follow MA's *unofficial* design concept: flair.

Just to be clear, my focus on CAK has to do with what it says about how the secondary effects are constructed in MA. Its a good starting point because separate from the fact that MA is clearly lacking overall, its a good question to ask whether CAK represents what MA should be delivering with all of its attacks, or if its too low, or if (although I think this is basically impossible) it is actually too high. Its hard to have that conversation objectively about, say, Crane Kick. CAK is interesting enough to be non-trivial (TK is trivial to discuss) without being so complicated that the discussion is guaranteed to get bogged down in details (CK).

Its simultaneously also one of the better candidates for buffing. Its DPA is low enough that there are no balance problems with considering it a strong candidate for high order foe effects (unlike Storm Kick which already has very high DPA) and its in the right spot in the powerset progression (unlocked at 18) to have a meaningful effect on the powerset relatively early in its career (unlike buffing Eagle's Claw, which would only have an effect on the set post 32).

But that doesn't mean CAK is turning into Power Build Up or anything. That would be silly. Being a good starting point doesn't mean its a good candidate for becoming the most powerful power in the set or anything. There's just a lot of latitude in the power to get creative, especially if its immobilize/slow isn't tampered with as part of the change.

For MA, this is actually important. Its why I do not advocate radically speeding up Eagle's Claw (BaB actually surprised me by finding frames to shave off that didn't significantly alter the visual look of the power). Doing so would significantly improve the damage of the set, but at the cost of altering one of the signature powers of the set: a lot of people probably take and use the power just because they like the way it looks (honestly, that's why I have it).

But contrawise, when you are trying to make improvements to a set where you *want* to honor all of these other constraints, it makes it all the more critical that the numbers you *do* tweak are very carefully chosen and altered to generate the results you want. So there is still a very good reason to understand how the set performs numerically, separate from how it performs mechanically.

Its tempting to do it just to see you proven wrong again.

Yes.

Just because they didn't always know the consequences of what they did, doesn't mean they acted completely randomly. You just have to put the pieces together correctly. And my sounding board over the years have been the devs, both the current dev team and the devs that were here originally and are not here now. So yeah, I think the picture I've pieced together is reasonably accurate.


I mention the past not because I'm bound to it, but because the devs usually don't just wake up one day and do things completely differently. They have their own "cottage rules" regarding maintaining continuity of design. While they can refine their methods, they still have conceptual rules about things they generally don't discuss often except when they mention them as the justification for certain kinds of changes - or lack of changes.

In any case, the current picture of MA is still the same: it doesn't match its current conceptual design, and no one has yet created an alternate one for it to meet. And my original statement is thus still true: MA either needs a new conceptual framework to meet - and must then be redesigned to meet it - or it needs to be reengineered to meet its current one. I don't mind if the devs change their minds. In fact, for a long time I tried to get them to do just that: change the conceptual target for MA, to allow it to get features its currently barred from having. They've actually been resistant to that, which is why spend more time trying to adjust MA to fit its current conceptual model rather than proposing alternates.

Although I've been talking to the devs about MA for four years, this is not just some distant memory. The last time I (briefly) talked to the devs about MA was December. And I prod Castle about it pretty much constantly. So much so that (jokingly) Castle struck back with glee (although I can't explain how quite yet, that quote is an MA-related quote - but before anyone gets any ideas, it has nothing to do with an MA powerset change: no MA change is hinted at in that quote. Its about my *thoughts* about MA, not about MA itself).


Well, I don't know that. As the set intended to have the *best* overall secondary effects (which within this context means foe-affecting secondary effects by the way) it has a tremendous amount of leeway given the strength of secondary effects in other sets, particularly DM. Unless someone declares DM's secondary effects to be broken, MA is automatically granted enough latitute to at least reach DM's net overall effectiveness, which is the combined effect of all the tohit debuff in the set plus Touch of Fear. Right now, by any reasonable measure, that already exceeds the damage mitigative effect of all the effects in MA combined.

Lets line them up:

Immobilize: DM's has longer duration and higher mag (reliably).
Single target mez power: MA's is harder (stun vs fear) but DM's has over four times the uptime.

Other:
DM has a pervasive -5.625% tohit debuff (-11.25% in ToF) stackable (usually 10s duration)
MA has a 4.8s mag 3 stun, 10% 7.2s mag 2 stun, 60% knockback, 75% PBAoE knockdown.

Debatable, but all of the DM tohit debuffs are perma or better relative to the attack they exist in. MAs effects are stronger but not perma relative to their power's cycle time and usually less than 100%. No clear winner in the remaining basket.


Keeping in mind that as of right this second DM is supposed to be the utility specialist, its actually powers like Siphon Life and Dark Consumption that are directly part of its concept. Its getting damage mitigating foe effects as basically an extra over and above its set concept and beating the tar out of MA with them. And this is not just a paper exercise: its really obvious in-game that DM's secondary foe effects are very strong, and combined with Siphon Life DM has almost as much damage mitigation as some actual scrapper secondaries.


So actually, I'm really not all that afraid of adding significant secondary effects to MA. The set is so distant from any balance rails that its a rare moment when I really don't have to be as restrained on powerset suggestions as I normally am. By my estimation, its not just DM that trumps MA: arguably MA is numerically near last in effective secondary effects, and not even unambiguously second in foe secondary effects (separate from DM which is unambiguously better, its arguably no better than Claws in foe secondary effects).

I should point out that I already won this argument with Castle years ago. And it shouldn't be "unbelievable" - its obvious that the devs are concerned about the amount of mez and other secondary effects that are placed individually into specific powers. Contrawise, and in defiance of reasonable balance, they are not - or rather historically have not - been concerned about pervasive secondary effects. The net effect of Dark Melee's tohit debuffs is *huge* but the thought never crossed anyone's mind to eliminate it from some of the powers for balance purposes. A similar situation occurs with the swords and defense debuff for the most part.

But when the devs put secondary effects individually into individual powers, they have historically been, and continue to be, much more cautious. Dark Melee's secondary effects besides tohit debuff are obviously hand-crafted. MA's original design also reflects this: TK had a mag 1 stun, cobra had the mag 3, and eagle's claw had its own chance for mag 3, CAK had the immobilize/slow, and dragon's tail and crane kick had the knockback. Storm Kick was a cone and didn't have a secondary effect.

CAK was a special power: it was intended to be a "tween" power that was in between a true mez-carrier (i.e. Cobra) and a genuine mid-tier attack. Its original damage was 1.0 scale, and it had a 2.0 cast time. Its DPA was one of the worst in the entire game's existence, but it wasn't intended to be an attack, it was intended to be a mez that also did damage. The devs were valuing that immobilize/slow combo *very* highly.

The players weren't, though, and when CAK was revamped its damage was bumped to its current value in I6. That was basically an acknowledgement from the devs that the immob/slow in CAK *wasn't* all that fantastic: certainly not enough to make the attack a weak attack (even though the devs historically did not factor in cast time or DPA, by their own reckoning the power was weak due to its low damage for its tier).

But MA was and is conceptually intended to be the best secondary effect scrapper primary (Dark Melee, the *true* secondary effect king, was always intended to be the utility king). I'm tired of explaining that one, so just ask Castle yourself if you don't believe me. The devs clearly considered CAK to be a valuable secondary effect power at the beginning of time, and a requirement to fulfill MA's secondary effect prowess which was also necessary at the time. When the devs acknowledged that CAK's secondary effect was not, in fact, very powerful, they in effect proved the point that CAK's secondary effect was underpowered. That could be corrected by leaving CAK alone and buffing all of the other powers to compensate, but CAK is not, and has never fulfilled its obvious original intent to be a strong secondary effect bolster for the powerset.

I can't quite prove that the devs overvalued the combination of immob/slow relative to each separately *because* of a belief in synergy that didn't exist, but I can prove all of the above, and within that context this is a reasonable assumption to make, and almost certainly true.


Also: a lot of people including some of the devs originally thought CAK debuffed recharge. Why? Because its combat spam said it did, so most players assumed it did. There weren't many ways for a player to verify that one way or the other for a long time after release, and even the devs wouldn't know unless they looked specifically at the power spreadsheets. And honestly there wasn't a lot of checking back then relative to now**. I was actually *told* by a red name back then that one of CAK's strengths that made it worth taking was its recharge debuff. Proving that CAK didn't debuff recharge was one of the very first non-defense mechanical tests I engineered in-game actually. The fact that anyone had to prove it at all should say something.



** The very first "numbers" debate I ever got into on the forums was with a poster that insisted that all blaster snipes took the same amount of time to fire. Proving that they didn't is what caused me to invent the demorecord timing system I still use today. But what's worth noting here is that I actually communicated to various red names my belief that this poster's numbers were wrong, and had two separate red names *confirm* his numbers, before I unambiguously disproved them. Prior to about I4, getting accurate information about the game even from the devs themselves was, sad to say, comparable to asking a magic 8 ball.

The problem is not the immobilize per se. The problem is putting the immobilize and the movement slow in the same power. If I were to take the stun from TK and the stun from Eagle's Claw and put them both into Cobra strike, so Cobra strike had three chances to stun the same target and those stuns didn't stack the "set" would have the same number of secondary effects, but they would be less effective. The same thing is happening to CAK.

Now, if every MA attack slowed, and CAK was the special attack that also immobilized, that would be different. Dark Melee has this property where Touch of Fear has both a very strong -tohit and a fear. Stacking -tohit on a target that is feared is less beneficial than the sum of the two effects, because a target that is feared is not attacking (or at least not attacking as much). But its clear Touch of Fear didn't pay extra for that effect: Dark Melee wasn't given a limited amount of Tohit debuff and "wasted" some of it on ToF: it was basically given an unlimited amount of tohit debuff to put into all the attacks in any way the designer saw fit.

But MA's secondary effects are very clearly limited. It was granted only a limited amount of individual effects, including slow and immobilize. Putting them both into the same power only makes sense if the two effects were meant to be packaged as a singular "super effect" - the effect of "immobilize, or at least slow if not immobilized." But that means in the grand scheme of things it should really count as one effect simulated by two game mechanical effects.

Either way, either its one atomic effect which means the rest of the set is free to have additional immobilize and/or slow, or its two effects being counted too highly, in which case MA is still weak on secondary effects. Either way, CAK is one of the best, if not provably the best target for set tampering. The cottage rule would prevent me from removing the immobilize anyway. It would not prevent me from making the immobilize a guaranteed mag 3, or adding other effects to the power.


Also, in PvP its like a 50% chance for a two and a half second mag 2 immobilize. Its better than nothing, but its also not an especially interesting PvP effect.


For a variety of reasons I can't discuss, I've basically taken every idea I've ever had for buffing MA and set them on fire. But I've started rethinking the entire process from scratch based on the current state of the game and the latest greatest game engine mechanics. After Going Rogue launches, I'm going to take a brand new approach to reconceptualizing the design of MA.

CAK is the power I have the least reservation about changing in the MA set. Its DPA is only average, it has a history of having the least support for its animation of any MA attack, and its secondary effects are at best quixotic. A single target minion immobilize (with a chance for LT immobilize) is relatively low compared to almost every other secondary effect in existence, and CAK creates the additional illusion that MA has two additional secondary effects where it only has maybe 1.5: by packaging movement slow in a power with immobilize it makes the whole less than the sum of its parts - because slowing something you've immobilized is less than interesting.

When I finally get MA looked at, you'll know it when CAK is significantly altered.

Although Scrappers didn't crit, some scrapper powers did: Eagle's Claw and Headsplitter in particular used to crit. I also believe Full Auto used to "crit" from the beginning too. In giving "Scrappers" the ability to crit, you could say that took away some of the unique nature of those powersets. So Scrappers can't really complain if other things got criticals later. They really stole crits first.

Aren't all forms of critical unique? Scrapper crits don't scale with team or work from hide.

Or, to put it another way, suppose Corruptor scourge didn't range from 0% to 100% with target health, but instead from 5% to 100% with target health, with the absolute floor for scourge chance at 5% instead of zero. Would that imply that Corruptors just got the Scrapper crit, plus a bonus on top?

If you found out you were going to die in six months, would you break up with the person you were with and look for a terminal person to hook up with?


As to the cancer thing, immortals have to have a way to deal with genetic damage over time, or they'd just plain die.

Well, technically maybe. But Statesman, immortal or not, hasn't lived all that much longer than actual people have lived yet. He's just physically better off being immortal. He hasn't even caught up with this woman yet.

The only problem Statesman has had to face that real people haven't already faced is the realization that his situation could last indefinitely longer. But he hasn't experienced anything other real human beings have. Real people have lived longer. Real people of his generation saw just as much death and loss in World War I (as he did).

Ignoring anything that immortality would do to his brain, being physically 30 forever means being physically in your sexual prime forever. He's going to be a testosterone-fueled male forever, which means while that doesn't dictate how he (or any other male) will necessarily behave, he'll always have the same physical attractions he had when he was 30. Somehow, I think even Statesman isn't so depressed that he's going to be a monk for eternity.


You know, now that I think about it, immortality for women is a much more interesting situation. So far as I know, there's nothing about male biology that couldn't be sustained indefinitely. But that's not true for women. Women are born with all the eggs they'll ever have, and hypothesizing them just growing back isn't a trivial exercise. We don't assume that immortal people can necessarily just grow back arms and legs: growing back eggs and follicles is kind of a similar thing.

The problem with that is that eggs themselves are an integral part of female biology. They are a critical component of the hormonal cycle. You pop an egg out, and that sets up the cascade of events that eventually lead to hormonal resets, periods, and triple fudge chocolate ice cream. Usually, most women have about a half million of them when they reach puberty, and lose about a thousand of them every period. So you have about 40 years of sexual maturity, maximum. If that process could be made biologically more efficient in theory, you could get to as much as 50,000 years of sexual prime, but really you're more likely to get a few hundred years at best.

Then what? I can only think of two possibilities here. One: immortal or not, youthful physically or not, you go into menopause. Two: the rest of your body somehow picks up the slack or doesn't need the estrogen from those eggs in the first place, and you keep humming along. In either case, we're speculating about women with significantly different hormonal biologies: women who go into menopause but don't physically suffer from that, or women who somehow dodge menopause by having hormones that override anything their ovaries might be doing.

Either way, the physical requirements of immortality might have a much different effect on women than men (unless its all just magic). Heck, if it turns out that immortal women are just hopped up on powerful levels of estrogen and testosterone all the time, then the question of who Statesman dates could be moot. It could be whichever immortal woman gets to him first.


What's my name, Statesman? Say my name, b*tch!

This could make for some interesting fanfic, anyway.


(Then again: the amount of estrogen and testosterone clearly flying around Paragon City even among the mortals sometimes makes me wonder how they even *notice* the effects of Superadyne.)

This was changed in the patch on the 18th: the Strength buffs were converted into Cur buffs for Mutation: Heightened Speed. They now buff movement speed directly rather than movement buffing powers. I don't think this was noted in the patch notes.

Apparently War Witch is the one that was prophesized to bring balance to the force.





Concept art for War Witch's update for Going Rogue compared to Issue 17. Note the reflections added for Ultra Mode.

I don't think so: I just think that thinking about how those things work can expand your horizons into thinking about how neural networks work, which is in a much more holistic fashion. A lot of times when you describe how neural networks work, people can get the idea that its all weird mumbo jumbo and maybe its all just made up. Then you look at holograms (which store images as interference patterns contained within every piece of the material they are stored in) and its a physical object you can actually see (in theory) and its a much stronger way of illustrating that this holistic form of information storage is not just hypothetical.

But I don't think brains are holograms mechanically. I think they do with electrical patterns what holograms do with optical interference patterns, but only at a high level.

I also don't currently buy the notion that human brains are quantum computers, except insofar as a potato chip is a quantum computer. I think that is just a defense mechanism for people who need to believe there's something particularly special about human brains that complex order itself cannot explain. Which is weird to me because there's nothing a quantum computer can do that a classical one can't simulate, albeit slowly. Computation is computation. Declaring the brain a quantum computer is just declaring it to be a slightly more sophisticated pinball machine.


My opinion, and its just an opinion, is that attaching music, rhythm, and harmony to a sequence of words to remember engages both the right and left brain to create keys for the memory, and this radically improves the strength of the memory.


Interesting thing, that. There's actually a lot of interesting studies that have been done on phone numbers, which are one of the most common non-trivial things people in the twentieth century and beyond are asked to remember. For one thing, phone numbers haven't always been quoted that way. And interestingly both 3/4 and 2/5 numbers were both very easy to remember. I think at least some of the structure of phone numbers is learned: you're exposed to them in a particular way, and you quickly build a rhythm for learning them that way, which then conflicts with any other way to learn them.

Also some of the structure is really mandated by the structure of telephone numbers: country code, area code, number. I met someone from South America not long ago that quoted his phone number XX-XXXX-XXXX. Same number of digits as mine, but rhythmically different because of the way their numbers were structured (the first two were the local area code). Once he said that, I actually found the number easier to remember that way. Even though I still cannot think of my own number that way without difficulty.

I am never going to let the devs forget that I told them they would rue the day they created the concept of archetypal inherents just to put a little icon on people's buff bars and avoid the need to update the damn manual. They managed to turn a game balance patch into a reward.

I wonder if they called the IH-toggle change the "Regeneration inherent" if everyone would have demanded an equal one of those for themselves also.

Besides, crits aren't just possessed by Scrappers and Stalkers. Containment is a crit. Scourge is a crit. The notion of "double damage under certain conditions" is all over the place.

You don't know that. Experiments show that dreams are one way for the brain to commit short term memory into long term memory, and reorganize and review long term memory. Interfere with dreaming, and you can impair the ability to form new long term memories. Those memories could have been sifted through unconsciously in many ways.

Memories can also be used implicitly as part of learning. The brain can review memories as part of learning other activities entirely outside the perception of the conscious mind.

Even if the memories are not touched at all, how fast a memory degrades and how it degrades is based on a lot of other factors, including how closely related it is to other information being processed.

This is *exactly* how neural networks "forget" things. The first thing to go are the fuzzy connections to the memory. In other words, no input into your brain can recall that memory. You've "forgotten you know it." But its still there and can still be retrieved if inputs sufficiently close to the original are presented. No thought running around in your head has the "key" to unlock that memory, but sitting in front of the game suddenly presents a set of inputs that retrieves that long term memory because its a close enough match. And once its retrieved, simply thinking about it causes other long term memories to be retrieved, and soon you can recall large portions of the memory.

Nothing of what you describe contradicts memory as neural network learning.


My memory works better that way also, but the question is whether you're really remembering, and really forgetting. What's more likely is that the "understanding" structure you've constructed in your head acts like a very specific "key" to the memory of the subject. Just like you couldn't remember Lost Vikings until you sat in front of it, you can't remember those lessons until something very specific triggers those memories. In this case, you've created a key you can construct in your head on demand which unlocks it, rather than have to reexperience something to refresh your memory. The memory trick of associating difficult to remember things with familiar and easy to recall things is very old. In the past, when human memory was the *only* form of recording things for many people (before the invention of computers, paper, and even widespread literacy) the trick was to take things that had to be memorized, and connect them to a process that was familiar enough it was unlikely to be forgotten. Sometimes it was connecting facts to be memorized to the things you'd see around your house, or in a walk around town. The facts were reinforced every time this process was repeated, reinforcing the memory of the facts and also connecting them increasingly stronger to the other, easier to retrieve memory.

You've replaced a walk through town with a logical process you do in your head. But that process is familiar and repeatable to you, and every time you walked through it and reconnected it to the class lessons you strengthened the memory of the lessons and connected it to the logical thought process. The act of rewalking the logical process creates the right signals to retrieve that memory, so you don't have to rely on the memory being strong enough to be "stumbled upon" by less specific thought processes.

City of Heroes really hasn't resembled that model for a long time: the devs certainly haven't designed under its premise since maybe a couple issues after launch. Granted: at launch archetypes like Blasters were very likely designed with weaknesses intended to emphasize their team role. Things had strengths and weaknesses in a way that would complement each other in teams in a manner similar to the classic MMO triads.

But now, things are different less to make them fill a specific MMO-tropical need on a team, but more to make each archetype offer a unique experience, and their strengths and weaknesses intermesh on teams mainly to make teaming reasonable, not necessary. That's why controllers and tankers overlap a lot: they are allowed to, becaues there is no requirement that each fill a unique role on a team. Furthermore, City of Villains archetypes all but threw the standard MMO model out completely.

The archetype balancing requirement really comes from two CoX design requirements: a) every archetype choice should have a reasonably unique and valuable gameplay offering (factoring out player preference) and b) combat is the primary mode of advancement in the game. The limitations of (b) combine with the requirements of (a) to make overall combat effectiveness, averaged across all aspects of the game from soloing to teaming and all situations therein, have to be roughly equal.

Actually, no. It only assumes brains remember things in the same way that neural networks do. Neural networks don't remember things in any way you can possibly analogize to "files" and "formats."

Suppose you make a small neural network of about ten thousand nodes. And then you start training it to respond to a particular set of inputs with a particular set of outputs. Once its trained, in a sense it has "remembered" something. But *what* its remembered is something that is a very complex thing to describe and explain. First of all, its not a local phenomenon usually. You can knock out any ten, fifty, a hundred, sometimes even a thousand of those ten thousand nodes and the neural network will still "remember" what you taught it. They can be any set of a hundred or a thousand. The memory is not stored in any one place, but as a holistic pattern throughout the net.

Furthermore, something interesting happens when you knock out nodes, and its related to the analogy to holograms I mentioned earlier. As you knock out nodes, the neural network continues to "remember" what you trained it to remember, but it starts to make mistakes. Its memory becomes "fuzzy." Normally, values close to the original input still generate the same output - the neural network can "remember" what it was taught with only vague "hints." But as you knock out more and more nodes, it takes greater and greater fidelity to recover the same memory. Eventually, even perfect inputs don't always generate the right output: it starts to forget. And eventually, there's no correlation between the inputs and the outputs: its "forgotten."

Holograms work in an analogous way. Take a piece of holographic material and store a hologram into it. What happens when you break it in half? Do you get the left side of the hologram in the left piece, and the right half in the right piece? Actually: no. You get two copies of the *entire* hologram in *both* pieces. However, each half gets slightly fuzzier: the image loses resolution. That's because the material has a maximum recording density, and the way you recorded the hologram also has a recording density. As the object storing the hologram gets smaller, the amount of information available to recover the hologram drops and the hologram is generated at lower and lower levels of information density - it gets fuzzier.

Neural networks and holograms are, in my opinion, important keystones to understanding human memory (and memory in brains in general). Its a metaphor that defies most people's common sense notions of how memory ought to work mechanically, especially people whose common sense is informed by computer technology.


As a totally separate matter, human brains clearly "compress" memories. In fact, experiments show we compress *experience*. We virtually *never* experience the actual world our senses detects. We experience a "model" of the world our brain constructs for which our senses are only one input.

Memories have to be constantly reinforced, or they fade. Much of that reinforcement happens subconciously, but it happens. When we remember something, we send inputs into the parts of our brain dedicated to memory (still not fully understood, but this is regulated by the hippocampus). Outputs come back which are the stored memory. We "feedback" those outputs back into the memory system to reinforce the memory. But if we focus on some parts of the memory and not others, we tend to reinforce some parts and let others weaken. Eventually they fade. This is also why memory is mutable, and can be greatly in error. Keep telling yourself something happened, and eventually your memory system will say it happened, like a bad screen burn-in from a old arcade game that persists even if the screen itself no longer shows that image.


Memory and skills are not exactly the same thing. You can see this in people with amnesia how can still execute an algorithm, but cannot remember it or how they learned it. Its actually possible to be unable to answer the question "how do you do this?" and yet be able to do it, seemingly by magic, when actually put in the situation. Somehow, data is turned into code in the brain when functional learning occurs. However, this too is subject to the same neural network rules of learning, non-locality, and fidelity.

Its *extremely* dangerous to think that you're fully capable of introspection about things like this. Studies have shown that when people try to think about how their own brains work, their own brains will lead them astray because so much of conscious introspective perception is an illusion. Probably the most interesting experiments related to this are ones where people have had their corpus collosum severed: the part of the brain that connects the left and right hemispheres of the brain. What generally happens is that both halves seem to have their own separate conscious agenda, even though the "person" doesn't perceive two separate halves. When they cannot communicate internally, they automatically try to communicate externally. Classic experiment: have the person draw something with their left hand - which is controlled by the right side of the brain. Now ask them what they drew and why. They will correctly describe what they drew, and give a complete explanation for why they drew it. Now block their line of sight to their left hand, and repeat. This time, they will either not be able to state what they drew, or give an incorrect answer. That's because language and speech are controlled by the left hemisphere. The part that can talk cannot know what the part that can draw with the left hand could possibly have chosen. Which means when they answered correctly the first time, and gave a reason the left hemisphere made one up and convinced itself that was correct. There was no attempt to lie: the brain simply did what it always does, but usually invisibly: it fused together what the right and left sides of the brain were thinking, and spackled over the discontinuity so our conscious mind didn't perceive the seam. This happens in normal people all the time, but we just cannot detect it when it happens internally. Only when confronted by evidence, like in experiments like this, do we get a hint. And by the way, when confronted by these experiments, most people with this condition still refuse to believe they made up a story to describe what they drew. To them, there was no process of deception going on in their heads.

There was a deception going on, but the deception was in the brain itself. The left side of the brain and the right side of the brain decided to collude together and try to create a unified view of the world in the background, and lied to their own conscious mind about what was happening. There's actually evidence to suggest that consciousness itself is One Big Lie: the lie that the brain tells itself to allow people to function as singular beings rather than cachophanous scatterlings.

Wonderful. I made a pistol blaster named "double action" which I can now no longer play.

The main issue with that suggestion is that its ultimately no better than the much simpler, much easier to implement solution of just sell everything in fixed price stores. Inflation gone by definition. Influence farming neutralized by definition, because without inflation there's nothing to spend it on. It actually eliminates beating around the bush and tackles the core fundamental premise of the suggestion: make the auction house redundant.

Its a very predictable solution, because we've seen it before: its exactly what the game was like before the auction houses and the invention system were created in I9.

All you have to sacrifice is the player run economy. And if you think that the dev-run economy will be a place of plenty, you're ignoring that pesky game-balance thing everyone thinks can be hand-waved away. In spite of the *best* items costing sometimes astronomical amounts of inf, the one thing that the I9 system did accomplish was to provide an avenue for influence to flow rapidly from high level characters to low level ones in a manner that made influence constraints on low level outfitting neutralizable via gameplay options.

The phrase "via gameplay options" is key. Provide a way to gameplay your way out of constraints, and you're still making a game. Give it away automatically for zero effort, and you're a monkey with a keyboard banging randomly and hoping the players like the pretty serifs.


I'm actually quite amazed, given how much I'm accused of unnecessary verbosity, the massive number of words expended on extremely complex ways of saying "the way to eliminate the problem of people complaining about not having enough stuff is to just give them more stuff."


In any case, marginalizing the auction houses seems to me to be an act of game design cowardice. If the devs were to do that, I would basically tell them that if they don't feel the auction houses are a valid economic system, they should just eliminate it. It only works if the game design encourages participation. Explicitly putting in encouraging ways to bypass it is saying you don't want players to use it, but you don't have the guts to pull the plug on it. And when I say "I would basically tell them" what I mean is that *is* what I'm telling them. You work to fix it, or you dump it. Period.