Slashdot Mirror
Article about The Lord of the Rings MASSIVE Crowd
TheOneRing.net has posted an article going indepth about LotR CGI, and specifically the rendering of extremely large crowds being done byWETA Digital. With the special edition due out soon, and TTT coming out in december, well let's just leave it at "Yay".
The Two Towers in the book are Orthanc and Cirith Ungol, not Orthanc and Barad dur.
Fuzzy logic doesn't use probabilities--just weights. Probabilities add up to 1. Fuzzy values can add up to more than 1. For instance, If I say a cup of tea is hot, and on a scale of 0 to 1 weight the "hotness" of the tea, I may rank it at .7 . Under a fuzzy system, that doesn't mean I rank the "coldness" of the tea at .3 . It could be more or less or equal to .3.
He painted a unicorn in outer space. I'm askin' ya, what's it breathin'?
For those interested, you can purchase Massive. Stephen Regelous, the brains behind this app, showed it last SIGGRAPH. You can check their website here:
Massive Software
Softimage also just announced their own system:
SOFTIMAGE ANNOUNCES SOFTIMAGE®|BEHAVIOR
Actually, you're half right; Tolkien's British publisher insisted on breaking up his 6 "book", 1 volume work into 3 smaller volumes of 2 books each (due to a paper shortage), which compelled Tolkien to come up with new titles (as the original 6 "book" titles didn't apply broadly enough to the new groupings - TTT was originally "The Treason of Isengard" and "The Ring Goes South", IIRC).
;)
In his letters Tolkien discusses his dissatisfaction with the title (though he came up with it), and vacillates as to exactly which two towers are referred to (!). Other towers he referrs to include Minas Morgul, and even Minas Tirith (although IMHO the last would only really be approp. for ROTK). I believe that Minas Morgul itself is described as having two towers on either side of its gates as well...
Anyway, I'm happy with it being ambiguous - but as far as the promos go, it makes sense to "nail this down" to shut down the ridiculous WTC connections.
And yes, you are correct, I don't have a girlfriend.
Yes, the software used in "The Lord of the Rings : The Two Towers" to produce the massive Orc armies at Helm's Deep is based on that used for the Jurassic Park movies. The cool thing about the Weta software, is that each 'Orc' has a small about of A.I. behind it, which will supposedly give the effects a more life-like appearance. DAPence Webmaster, THELORDOFTHERINGS.com
Here's the history behind the Four Towers. Any two will do: http://members.cts.com/king/e/erikt/tolkien/2tower s.htm
This link explains that Tolkien changed his mind about that a few times. Really, the name was rather forced on him, for a volume that he didn't want released in that manner anyway. Book 3 and Book 4 of Tolkien's six-book epic became "The Two Towers." Tolkien himself drew three different covers for the book, one showing Minas Tirith and Barad-dur, and the second and third showing Minas Morgul and Orthanc.
So you might as well call it The Four Towers, as Tolkien changed his mind about which two the title refers to.
Here's a Tolkien quote that shows that for awhile at least, the Two Towers the movie refers to were possiblities left deliberatly vague.
"The Two Towers gets as near as possible to finding a title to cover the widely divergent Books 3 & 4; and can be left ambiguous- it might refer to Isengard and Barad-dur, or to Minas Tirith and B; or Isengard and Cirith Ungol (1)." [Letter #140] -J.R.R. Tolkien
Later Tolkien did settle on Cirith Ungol and Orthanc, because of the text of books 3 and 4, but conceded that Barad-dur and Minas Tirith, seemed less confusing.
I think, as long as it's any two of the four, you can't go far wrong. I can see easily how Jackson's choice of Barad-dur and Orthanc makes a great match. "There is now an alliance between the Tower of Orthanc and the Tower of Barad-dur." It spells out, in a simple sentence, the power of the threat to Middle-Earth.
Wait a second. I just posted a Lord of the Rings factoid on Slashdot! Coal to Newcastle! You guys probably knew this when you were in Kindergarden!
I am not a Tolkien fanatic and not "wise in the lore" but didn't Tolkien write "The adventures of Tom Bombadil"?
this what you want?
It was for sale on ebay. That's all i could find, though. Google didn't have much, sorry.
Who is this Anonymous Coward character, how does he post so much, and why is he always such a whore?
I remember seeing a television show about this kind of logic used to animate some scenes from "The lion king". Apparently no one was able to animate a stampede of several thousand zebras by hand.
Fuzzy logic is not the same thing as probabilities. You and the original poster are both confused.
Fuzzy logic assigns a proportion of truth (between 0 and 1) to each fuzzy logic value. A proportion of truth is not the same thing as a probability. A probability assigns a value to the chance of a particular outcome for an as-yet-unknown event. Fuzzy logic assigns a value - the proportion of truth - to already known data. The difference is not even subtle or semantic: it's a huge difference.
I'll try to explain this better. In boolean logic a statement is either false (0) or true (1). In fuzzy logic the truth of a statement is any value between 0 and 1 inclusive. So for example you might have a value OLD with a baby being 0 and a pensioner being 1 and a linear slope between those two extremes. You don't have to choose a linear slope, of course.
Once you have the truth of a statement as a number you can use logic tables. So NOT(x) equals 1-x, OR(x,y) = maximum(x, y), and AND(x,y) = minimum(x,y). So the statement
Is equivalent to
You then convert raw data for a given car into truth values and pass it through the fuzzy logic statement to determine the truth value of a complicated concept like "cool car". A pink station wagon would have (sports,red,suv) values of (0.0, 0.5, 0.0). A green Ford Explorer with an Type-R sticker would have values of (0.2, 0.0, 0.9).
So to take this example further, a Ford Explorer would rate highly on the suv value, so 1 - suv would be low and the minimum(..., 1-suv) statement would drive cool car towards zero. Imagine that cool car was 0.1 after doing the calculations. What that value means is that the truth value of the statement "The Ford Explorer Is a Cool Car" is 0.1... which is not very true.
Notice how this isn't a probability! If that value of 0.1 was a probability then it would mean that every 10th Ford Explorer is cool while the other 9 Ford Explorers are not cool. Clearly that's nonsense. Ford Explorers are all equally uncool. I hope this example makes it clear that Fuzzy Logic is not the same thing as probabilities.
like 30 years ago
Actually it was written in 1954, 48 years ago. Close enough though...
No, the "proportion of truth" isn't a subjective thing. You don't get to pick another person and get a different answer because the person's opinion plays no part in it. We've already explicitly defined the fuzzy logic statement:
And if my example was more complete I would have explicitly defined the mapping between the car and the tuple (sports, red, suv). I didn't explicitly list the mapping but that's just my laziness. Create whatever mapping you like and we'll go with that.
There's no room for subjective interpretation. That's why it's called fuzzy logic. Two people can start with the same statement and the same tuples and get the same answer. That's the whole point. The fuzzy logic statement isn't asking whether you think the Ford Explorer is cool, or whether somebody thinks the Ford Explorer is cool, but rather it's saying "this is how we are defining coolness for this particular problem".
Now to be perfectly clear, the statement and the method of getting the tuples is the subjective part. You might think "red" has nothing to do with cool. You might think "suv" is very cool. You might disagree that a Type-R sticker turns a car into a sports car. That's OK. Fuzzy Logic isn't about defining an Absolute Truth. It's about defining a convenient definition of truth for a particular problem. You get to choose what suits your particular problem.
So an air-conditioner might say that 18C and 35C with a linear scale are a good definition for "hot". A refrigerator might choose -4C to 2C with an exponential scale. A quantum physicist might pick 0K to 10^27K with a logarithmic scale. But once you choose the mapping and you have the statement there's no room for argument. Fuzzy Logic is not about probabilities!
The other contender is Walmart.com. The price is the same, and you have to pay $3 shipping, but if you preorder, you get a free t-shirt.
I'm not affiliated with either, just sharing my findings (and hoping if someone else knows of better deals, they'll post back!)
Cool funny t-shirts for geeks, gamers and everyone else
No, the people having sex were real. What happened for ther US release is that they composited this people in the way so that it wouldn't be so graphic. You can spot like a few couples that don't move stratigically placed here and there.
The worst part was that they claimed Kubrick wanted it that way, but if that was the case then why did the European version didn't have that "censoring"?
Fuzzy logic assigns a proportion of truth (between 0 and 1) to each fuzzy logic value. A proportion of truth is not the same thing as a probability.
Fuzzy logic theory is just that, a theory, and in its pure form has nothing whatsoever to do with probability, but applying it to make classical logic decisions requires unfuzzing the truth values, and this is generally done by interpreting the values as probabilities.
For example, the software package that is the subject of this article does exactly this, I'm sure. Each character has a variety of fuzzy characteristics, like anger, but when it's time to figure out what the character is going to actually *do* the software derives a fuzzy value for the character's state of mind and then interprets that value as a probability, rolls a virtual die and converts the probability to a decision.
Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
the big deal is that until recently, those characters would have needed to be animated by hand, or by using motion capture. with a crowd of that size, animating each character individually would have been prohibitively difficult, requiring the animators to fall back on instancing, or even simple cloning of animation curves from character to character, with slight modifications for behavior, direction, and the appearance of randomness.
but this way, you can create a mess of characters, tell them which side they're on, and set them loose (essentially), without having to worry that groups will look the same, or worse, having to choreograph their movements individually.