Arcanaville

Marketing is like the Terminator. It can't be bargained with. It can't be reasoned with. It doesn't feel pity, or remorse, or fear.

I've never thought marketing people hate us.

Often I wonder what the hell they are thinking though.

I'll try to get as many of my dead computers to connect to the site as possible then.

Bah. I'm just volunteering to put wheels on the camera tripod. Well, wheels and a jetpack, but still, its what you do with the camera that matters.

In general that will work but there is a catch. When you select the splines for each dimension, if you aren't careful you could pick odd combinations of parametrization for the different splines and end up with "odd" motion. In particular, you could exceed the nominal absolute curvature limits you want to adhere to in order to prevent the motion from becoming too jumpy. For relatively "gentle" curves, this isn't usually a problem.

Tell you what, next time you're experimenting with camera work for a video, or just screwing around in general, let me know what you want the camera to do and I'll try to set up a spreadsheet for you to play around with. I have no talent for making videos, but I'm pretty certain I can make the camera do whatever you want it to do.

Back in the day, we used to write those for POVRAY, before they added splines to camera movement for animation and made the effort moot. There used to be programs that you could enter camera positions into and they would split out interpolated camera coordinates, but those might be hard to find these days.

Its not a difficult thing to write a script to generate these values for an ad hoc instance, but I'm not aware of a specific tool you could use that would automate this completely, especially because there are different techniques that would accomplish different things.

For example, based on watching the video and reading Aral's description quoted here, I don't think he's using splines for his camera movement. I think what he is doing for some of his smooth camera motion is creating target-centered spirals by placing the target of the camera at the center of a sphere - or rather an inclined circle - and the rotating the camera around the target while simultaneously decreasing the radius by a generally accelerating curve.

For simplicity's sake, lets say you wanted to do this in two dimensions, and the target is at 0,0. I place the camera at 10,0, which is (for my example purposes) at 3 o-clock. I want to move the camera in a smooth arc around and approaching the target. The x,y equations for a circle, if I wanted to move the camera from 3 o-clock to 12 o-clock would be:

x = cos(theta) * 10
y = sin(theta) * 10

as theta went from zero degrees to 90 degrees (or pi/2 radians, if you prefer). I could make a spreadsheet that calculated that easily, for any interval of theta. Using Aral's 40ms rule, if I wanted the motion to take 2 seconds, I would want 50 data points, and I would calculate the motion at 90/50 = 1.8 degree intervals.

To make the camera also approach the target, I would reduce the radius. I could make the camera approach the target evenly, in which case I would reduce the radius evenly from 10 to some small value like 2 (because zero would place the camera up the target's nose). So I would make a spreadsheet with 50 (theta,radius) values where theta incremented by 1.8 degrees and radius decreased by 0.16 (8/50), and calculate x,y values based on the equations:

x=cos(theta)*r
y=sin(theta)*r

If I wanted a more dramatic approach, I would change the radius non-linearly. If the target is not at 0,0 I would only have to add the target's coordinates to the calculated ones to translate the motion to the target's correct location. And then I would just need to calculate the camera's viewing direction for each data point. And in this case, that's actually easy because we know what the camera angle is from the target to the camera, because that is how we calculated the camera location. Theta gives the angle to the camera, so the angle from the camera to the target is the inverse of that. When the camera is at theta=10deg, the target is at view=190deg. I'm thinking this might even be why Aral chose this method of camera movement: you kill two birds with one stone if you move the camera as if the camera was connected to the target's eyes with a stick, and the target swept her eyes around in a smooth motion watching the camera. The camera does the inverse of that motion so you do not need to use heavy math to calculate PYR for the camera.

Doing this in three dimensions is slightly more complex, but the math is basically the same.


Splines make more sense when you are moving *through* an environment rather than around one target. You can use elliptical splines and such, but its often overkill.


Of course, this just simplifies the math. You still need to use the math in aesthetically interesting ways, which is of course the hard part.

Clearly, marketing doesn't market itself very well.

Positron would like to clarify the matter with the following statement:


(Click to view Dev-o-meter translation)


I've always wondered why Matt calls you "Donny" behind your back. "Donny" is not usually short for "Christopher."

BaB and Positron would like to issue the following statement to any of their loyal fans that wish to offer game implementation suggestions in person at their Paragon Studios offices:


(Click to view official Dev-o-meter translation)


Thank you for your attention.

I'll take your word for it that is what you meant. But its not what you said:

That says the exact opposite of what you meant.


I do agree that Going Rogue has set very high expectations, specifically because of the implied amount of time the devs have worked on it. I do not believe any one area needs a minimum amount of attention, but I believe overall GR needs to have a level of content that, if not equal to, is at least in the general ballpark of the content in CoV. Not because they are both boxed expansions, but just because the level of content in CoV just coincidentally happens to be in the general vicinity of what I believe is reasonable for Going Rogue at the present time. If GR ends up having even half or two-thirds the content of CoV, by my subjective standards, I will consider it to have at least nominally achieved my expectations.

I have eight, and no one thing I ever do requires all eight.

However, I run virtual machines, including a complete copy of my older system for reference, which I almost never shut down (unless I'm shutting down my entire computer). If you run virtual machines, more is better.

Slight sidetrack: if you have nine, and its a relatively recent computer, you probably have a core i7-9xx in a LGA 1366 motherboard (triple channel memory architecture, so memory is usually slotted in triplicate thus the odd amount of RAM). So you certainly have a powerful enough system to be doing all sorts of unusual things if you wanted to. But without knowing what you actually do, its hard to say whether you would benefit from more memory. You could bring up the task manager and look at free physical memory. If that never seems to drop below gigabytes of ram (thousands of megabytes) then I'd say more RAM would have very limited usefulness to you.

Well, its nice to know the Radeon 5850 I have on order is a 1,044 Ghz part.

Well, regardless I will at my next opportunity to study the situation. I'm not sure what the loophole is in this situation. The fire weakness alone can't explain it (they are only -5% resistance to fire). AoEs are certainly more efficient, but ordinarily you'd expect that efficiency to be muted by travel time between spawns.

Although those XP numbers seem wonky to me. What do you mean by exemping to 24 to get those numbers?

(The number I have for a level 36 even con minion is 484 XP).

Only by default.

Fixed.

You have a thing for conjoined twins?

I consider GR to be the *first* boxed expansion. City of Villains can't really be considered an expansion of CoH because it was intended to be a stand alone game - and *was* a stand alone game when it was first released. An expansion can have a new AT, or two, or three. City of Villains absolutely positively had to have five archetypes and a lot of content, because it was possible to buy only CoV, and even if you had both games your villains were going to be locked into CoV content only regardless.

As an expansion, I believe GR should devote about half its effort to adding new content for both sides, and half its effort enhancing the existing content in some way. That means I don't consider the number of new archetypes in GR to be significant. In fact if they had the resources to make five all new archetypes I would rather they spend them making just a couple new archetypes and add more powerset choices to the existing archetypes.

One more thing: it was a lot easier to make the CoV archetypes than it would be today to make all new archetypes. And that's because CoV archetypes reused a lot of powerset types and powers from CoH. Corruptors took less time than a brand new archetype would take because the devs could start by flipping Defenders. Stalkers are mutated Scrappers, they aren't a totally new archetype, at least in terms of the raw work necessary to construct the data for them (so are Brutes). They could get away with that because none of those concepts existed anyway on the red side, because blue-side characters couldn't cross over. They could generate a lot of the CoV archetypal diversity by shuffling, and didn't need to invent as much from whole cloth.

Today, especially within the context of soon-to-be side-switching, a new archetype actually has to be new: it can't be too trivial of a rehash of existing archetypal concepts, because otherwise it will collide with the other archetypes (which will now potentially exist on both sides).

I think GR should be judged at least in part on how it enhances the existing components with the game, and not exclusively on how many new things it adds.

I'm reasonably sure its not. I wouldn't expect GR beta to begin for "a while."

That's "a guess."

In terms of the law, fundamentally speaking, there really isn't any difference between the two situations. In *both* cases his creation is protected by copyright.

Where the practical difference comes up is that in both cases, to use the character outside of the game you'd have to recreate elements of the character which use NCSoft IP you have no right to use outside of the game. So for example, in both cases you'd (probably) need a visual representation of the character that was different than the game's visual representation of the character.

However, in the case where the character already existed prior to the author playing the game and recreating the character, the author would almost certainly have a visual representation that predated the game, and would not need to invent one. In the case where the character is created in the game initially, one would have to be invented. In the latter case, its potentially tricky to prove that your new visual representation is not too derivative of the game's graphics to be a unique creation of yours. In the former case where the character predates the game, its a lot easier to prove that.

But that's a question of proving the elements are non-derivative creations. In both cases the author's rights are identical if he or she can prove it.


In terms of the game rules its probable that an author could overturn a genericed character if he or she could prove they were the actual copyright holder. But that's not trivially easy to prove to a game publisher, and they could still bar someone from playing such a character if they felt it would cause too much disruption in the game to do so: players could keep petitioning the character because they did not know the player was the copyright owner, and NCSoft could simply decide it wasn't worth it. Since they can pretty much prevent you from playing anything for any reason, they do not have to produce a legal justification for genericing a character. Convenience alone could be sufficient justification to act.

This is a frequently debated question on the forums. However, there are a few things that are essentially fact, and not subject to interpretation (although people argue them anyway):

1. Copyright protection is granted to the author of a creation, at the moment of creation. This fact cannot be altered by contract or other legal agreement, except in the specific case of work for hire. Therefore, if player characters are artistic creations at all, the player owns the copyright to them, at least initially.

2. The artistic components that are in the game itself and the character creator specifically, such as the costume designs and body elements, are still owned by NCSoft regardless of whether players incorporate them into their creations, and ownership of the whole does not mean ownership of the parts under copyright law. So even if you own the artistic creation of your character, you don't own the individual visual elements of the character as they exist in the character creator. So you can't arbitrarily use the visual appearance of your character as it exists in the game. You could use a drawing of your character based solely on your own artwork.

3. NCSoft requires you to grant them (at the very least) a non-exclusive license to use anything you make in the game. This is entirely legal for them to ask for, and it grants them the right to use your creations in virtually any way they see fit, even if you are the owner of them. However, they do not grant you a similar license, so you do not have the same right to use their stuff. Its highly problematic to use a character in the game when you are not allowed to use its visual appearance (without recreation from scratch) or any of the backstory elements of the game (no mentioning the Crey or Lord Nemesis in your backstory, for example).


What's at least somewhat debatable are the following:

1. Is the EULA sufficiently specific that it meets the requirements under copyright law for a copyright transfer? In my opinion, it does not. But if a court were to rule that it did, then theoretically speaking we could be the authors of our characters under copyright law, and then we transfered those copyrights to NCSoft as part of agreeing to the EULA.

2. Are player characters even "creations" under copyright law? I think this is bogus myself, but there are some people who contend that players do not put enough work into their characters to constitute an artistic contribution, and therefore characters are really just trivial assemblages of NCSoft's intellectual property. The amicus briefs that any court hearing this argument would be flooded with staggers the mind, but its at least a theoretically non-frivolous legal challenge.

3. Can players be considered "employees" for the purposes of work-for-hire statutory requirements, and therefore NCSoft considered the author of anything the players create? I mention it only because some people have in the past, but I think this is *so* out there that it falls just outside the boundaries of being frivolous. I think it wouldn't be considered literally frivolous only because while its probably stupid, its legally interesting to rule on this theory and set a precedent, so I think a judge would consider it sufficiently intellectually interesting to maybe hear the case out.


Bottom line: in a legal sense you almost certainly own your characters. However, "your character" may not mean what you think it does, and in particular if you want to use your character in another setting, you'll have to figure out a way to do so without using the direct visual elements of the game, without using the protected elements of the game's story, and without using any other element of the game that is protected by copyright that you didn't create yourself without permission (this includes other players' characters).

Conversely, whether NCSoft owns your characters or not, they do have the legal right to use them in just about any way they want.


And of course, as some people do rightly point out:

1. Tort is expensive. It is very likely to cost you more to defend against a suit filed by NCSoft than the character is worth, unless your name happens to be Stan Lee.

2. Since you can't use any of the CoX elements in your character if you attempt to use it elsewhere, its far simpler to make a similar character that lacks those entanglements than it is to try to "purify" an existing CoX character. This dodges the whole issue of character ownership to the best extent possible.

Allow me to reverse the inputs into the Dev-o-meter:


(Click to see Player-o-meter translation)

I think that encapsulates the current playerbase evaluation of NCSoft marketing to within 1.2% margin for error.