PDA

View Full Version : WETA to release Massive Freeware


Grey
07-12-2002, 08:16 PM
Computer Graphics World reports that Massive, the now famous Crowd scene software, will be available commercially AND as Freeware from the soon to be opened Massive Software website.

The freeware version will reportedly lack features sported by the commercial version, however the move appears, on the surface, to motivate 3D Artists to familiarize themselves with the software on a grand scale.

Massive was used by WETA to create the battle scenes in Lord of the Rings in what can only be comparered to a very high end Artificial Intelligence that some games sport.

Reportedly, Charcters, or "Agents" are assigned several motion sets and turned loose with a tag similar to a military I.F.F. wargame tag so only the "enemy" is attacked.

Press release and a full story on http://www.cgw.com

Massive Software site is not yet operational, but will reportedly be http://www.massivesoftware.com

JamesDeschenes
07-12-2002, 08:27 PM
Cool :drool:

I can't wait to read up more stuff on this. I will probably never use it but its cool just to read about.

Grey
07-12-2002, 08:33 PM
My guess it that WETA's thinking is that the more people using it, the more experince out there, the more likely Studios will take advantage of that experience and use their software in production work...

I think it's a good supposition :D

So they're probably hoping you'll play with the freeware version a bit....

MacRonin
07-14-2002, 03:27 AM
Sounds interesting, wonder how it stacks up against the AI.implant from BioGraphic Technologies...?

www.biographictech.com

Would be interested to see someone review the two against each other someday...

#3

Grey
07-14-2002, 06:02 PM
Took a cursory look at the site and the content, I couldn't get any details as to how it works save for that it's a plugin for Maya and LW.

Fortunately, with Massive, we've been given some of the details about what's involved in construction...

There was apparently an article about it in 3DWorld, but no text that I could find...

l_farley13_l
07-14-2002, 06:33 PM
Sounds great and definitley somting to work in if your trying to get into that specialty (AI/ group simulation) Looking foward to it.

Farley13

The Magic Pen
07-16-2002, 12:45 AM
I wish ILM would release Deep Canvase ... :hmm: ...I have heard that program is everything Deep Paint should be .. :shrug:

Grey
07-16-2002, 12:53 AM
I'm no fan of Deep Paint.

BodyPaint gives you plenty of capabilities at a fraction of the price. You have to buy Texture Weapons just to get decent UV editing functions.

Overpriced and a resource hog, not to mention that damn dongle...

ambient-whisper
07-16-2002, 01:39 AM
ya...but only if bodypaint had a better interface ;) ( more photoshop-ish )

it takes about 2 minutes to get familliar with deep paint. mmm.
( btw. deep paint 2.0 isnt nowhere near as much of a resource hog as it used to be.

Grey
07-16-2002, 03:16 AM
<---actually thinks UVMapper is a better value than any one of them.... http://www.uvmapper.com

The neat thing about it mainly is that it forces the user to learn the ins and outs of UVMapping and why it works the way it does...

The only real problem with it's function is that you have to use a 2D paint program, such as photoshop, with it, and that won't paint on multiple channels.

Other than that, it's just fine for most texture work.

beaker
07-16-2002, 09:51 AM
>>I wish ILM would release Deep Canvase ... ...I have heard that program is everything Deep Paint should be ..

That is Disney that made Deep Canvas, and not ILM. The big thing it was popularized on was for Tarzan doing all the tree's.

Grey
07-16-2002, 04:43 PM
<---just read an article about that last week :D

didn't put two and two together.

Now one thing I don't understand. Deep Canvas, wasn't that also an old disney technique for set design back in the days of yore when everything was hand drawn?

MDC
07-16-2002, 07:05 PM
Sounds interesting, wonder how it stacks up against the AI.implant from BioGraphic Technologies...?

I'll go out on a limb and say Massive would hand AI Implant it's ass. Not because AI Implant is bad, but because Massive has proven itself in one of the largest FX productions ever undertaken by Hollywood. If AI Implant were up to the task, WETA would probably have used it. They seem to try to use offf the shelf solutions wherever possible to cut costs.

I have no use for this plugin, but I can't wait to check it out.

Originally posted by Grey
Now one thing I don't understand. Deep Canvas, wasn't that also an old disney technique for set design back in the days of yore when everything was hand drawn?

Yeah, I don't think Deep Canvas is a particular piece of software, but a process. From what I gather they are using PaintFX or a custom software very similar in methodology.

Mightymaggot
08-14-2002, 10:51 AM
Originally posted by Grey
[B]Computer Graphics World reports that Massive, the now famous Crowd scene software, will be available commercially AND as Freeware from the soon to be opened Massive Software website....

that is very true, but according to the massive software website they are not releasing a freeware version... check out http://www.massivesoftware.com/faq.html

and contory to popular opinion massive is not a plug in for maya or anyother program but a stand alone programme!!!

Grey
08-14-2002, 07:58 PM
:(

Oh well...

I still think we're looking at the future of 3D Character Animation...

Everybody and his brother will be trying to replicate that technology.

Crazy Max
08-15-2002, 02:12 AM
"I'm no fan of Deep Paint. BodyPaint gives you plenty of capabilities at a fraction of the price. You have to buy Texture Weapons just to get decent UV editing functions. Overpriced and a resource hog, not to mention that damn dongle... "

Grey You obviously have no idea what your talking about when it comes to Deep Paint 3D. It does not have a dongle and never has. Find out the facts before putting you foot in you mouth.


:shame:

Mightymaggot
08-16-2002, 01:08 PM
Originally posted by Grey
:(

Oh well...

I still think we're looking at the future of 3D Character Animation...

Everybody and his brother will be trying to replicate that technology.

how very VERY True....

Time to port openAI to Maya i think! could be a fun project!!, any body out there had any experience porting software to/ or making plugins for maya?

I am quite serious about this, it is about time the community stop asking for one and start making one... the technology is all ready there!!!!!

lets rock on :buttrock: :buttrock: :buttrock: :buttrock:

arvid
08-16-2002, 06:50 PM
What do you mean Freeware???

http://www.massivesoftware.com/faq.html

Q: Is there a free version of Massive?
A: There are no plans to make a free version of Massive.

Grey
08-17-2002, 02:12 AM
Before the website became active there was a report in CGW magazine that there would be a freeware version.

Either CGW got it wrong, or Massive Software changed their minds.

castelis
08-17-2002, 04:04 PM
You are looking forward to release? Its too expensive even for big studios.

dantea
08-17-2002, 05:08 PM
I tried looking for revenues of these big studios on the web and couldn't find anything. It seems that these guys are mostly private companies. Does anyone here know how much these bug studios make?

Grey
08-17-2002, 06:54 PM
The Wall Street Journal might have archives on that.

I know that Media outlets are able to get those figures, but I've never figured out exactly how they get them.

And BTW, $40,000 is a drop in the bucket when the system is designed to replace huge numbers of extras which would easily cost that much for two minutes of film... that may be the rational for it. Massive's connection to Weta would give them a great deal of experience on how cost effetive that would be for the major studios.

That said, if there's that much money involved, I'll bet a bunch of little garage programers are out there as we speak looking to get a piece of that pie. The market may become saturated with clones and wannabes very quickly.

Massive's being first, however, says quite a bit. It will be a very very interesting and exciting thing to watch unfold IMHO.

jtk77
08-17-2002, 09:43 PM
I agree Grey 100%..
40 gs should NOT be a big deal for a compnay who needs a tool like this.

And MASSICE is VERY much intended as a studio tool.
have no mistake on that.. it is NOT intended for a hobbists or an indie artist ..or even a small group of 2 or 3 people.


Thats why the freeware thing was so odd...its not he kinda of tool to want to target the masses.

Grey
08-17-2002, 10:32 PM
in retrospect it may have been a mistake for me to parrot it...

However: let us assume for a moment that Massive does indeed become THE studio tool for high end 3D animation, and there's every reason to believe this a possibility...

That cuts EVERYONE else out of the picture.

jtk77
08-17-2002, 10:41 PM
wait wait...i think you either misread what MASSIVE is or mis typed here..
it IS NOT an ANIAMTION tool in the sense you described it....nor will it ever be. it is what it is ...
a massively intense crowd simulator.
It does not take over animation as far as I can see.
It is NOT in competion with any of the highend aniamtion tools...
it is a very specialized tool that does what it does well.
Main reason its a studio tool. When somthing is that specialzed it is more than a few bucks , and it is geared toward a studio.

AS far as i know, only inhouse developed for a single purpose crowd sims are out there right now..so what compedition is it killing?
Forth coming compeition?
Well honestly that will be few and far between I think, there isn;t enough money out there for a very nich oriented tool set I think to be feasable to spend the amount of R and D to compete with MASSIVE except perhaps for a different group than what MASSIVE is seemly targeting...
Well just a thought. we shall see..
but, it WILL not become the aniamtion tool for stuidos , becouse that is not its intent nor what it will be feasably used for as near as I can see.

Grey
08-17-2002, 10:53 PM
I"m not sure exactly what lead me to think these things, I need to go back throught that site...

I hope I'm wrong and you're right, however :D

jtk77
08-17-2002, 10:55 PM
massive is a crowd simulation that runs on rules and mathmatics . its fully automated like a particle sim..not a "aniamtion" program like MAX/MAYA/or XSI
in fact , its more of a plugin than anything else..works WITH your highend aniamtion piplines.
its very much a specialized peice of software.

Grey
08-17-2002, 11:07 PM
uh, wrong.

It's been well known for some time that it uses mocap, says so right on their site that it imports and exports the Acclaim Mocap Format.

You can also hand keyframe for it getting animation in using acclaim.

The "agents" it would appear from everything I've read about it, and knowing something of how Mocap Works, would need to be boned like a standard 3D figure.

jtk77
08-17-2002, 11:09 PM
well I'm sure it can.but as far as the actions and how it picks it , its definitly a sim.

beaker
08-18-2002, 12:03 AM
>>That said, if there's that much money involved, I'll bet a bunch of little garage programers are out there as we speak looking to get a piece of that pie. The market may become saturated with clones and wannabes very quickly.

There are allready a whole bunch of crowd simulation plugins for every single 3d software package out there. Atleast 5 for max and lw, and 3 for maya that I can think of. Also LW and XSI come built in with crowd simulation options. Also there is atleast 2-3 free plugins and scripts for maya and max that do this. There is no lack of options for the people who can't afford 40k.

The big thing that sets massive appart from everything else is that it is rule based and can handle millions of characters, not just thousands like all the others. Not many people who can't afford Massive will even have a project that has a crowd simulation of 100k-1million characters all fighting in a field, or if they do its going to look like ass anyways.

Grey
08-18-2002, 01:35 AM
Comparing the crowd simulaters available for Max or Maya to Massive is like comparing a PONG to the War Craft.

They're not just rules based, with an AI for each agent, not one AI managing all at once, all it requires is settup.

I remember CGW reporting on this great new plugin for Maya a few months back showing how it made figures avoid a boulder coming down a hill, and was considered a big deal...

beaker
08-18-2002, 05:57 AM
>>Comparing the crowd simulaters available for Max or Maya to Massive is like comparing a PONG to the War Craft.

For any moderatly sized crowd these plugins work just fine. Just look at the end of the movie "The One". They used one for xsi just fine and it did it's job.

You made the statement earlier:

>>However: let us assume for a moment that Massive does indeed become THE studio tool for high end 3D animation, and there's every reason to believe this a possibility...

>>That cuts EVERYONE else out of the picture.

All I was saying is that its not really that big of a deal cutting everyone else out of the picture because Massive isn't really made for the needs of the person who can't afford it right this sec. Everyone else will be perfectly fine with any other crowd simulation plugin.

Though it will be nice if they lower the price and I believe they will. Since most of the big studios allready have their own in house crowd simulation scripts and plugins or they can have a programmer/TD there wip one out to suit their needs. This leaves a very small crowd of people to purchase the software at this price range.

Grey
08-18-2002, 06:05 AM
Beaker, if most large studios had code that did what Massive does, don't you think they'd be using it?

Stephen Regelous started programming it in 1996 by order of Peter Jackson.

The code reportedly can MODIFY motion, it doesn't just use the presets.

Here's a quote from the LOTR article at CGW:
The brainchild of Stephen Regelous, Massive generates artificial intelligence agents that respond to their environment. "They select which motion to execute, modify that motion, and blend the motion on the fly as they run around in response to the environment," Regelous says. The software was primarily used to create battle scenes with as many as 70,000 fighting warriors, each a Massive agent with its own "brain," but it also helped animate digital doubles and a flock of crows.

Regelous began working on Massive in 1996 at Jackson's request. Once the software was ready, Weta Digital's crowd department began creating brains and bodies, generating libraries of motions, and designing variations. It took two years of this pre-production work before the Massive agents could be used in shots.

The agents were built with primitives that have physical properties. To allow for physical simulations, the program has rigid body dynamics built into it so that, for example, warriors fall believably onto rough ground. Because the goal for this movie was to have each agent (warrior, goblin, digital double) look and act uniquely, the Massive crew designed a variety of tools, pieces of clothing, skin colors, and so forth for each type of agent. Included in Massive are methods for easily generating variations to change geometry or characteristics. "We have dozens and dozens of variables for each agent, which can be anything from how muddy his boots are to how aggressive he is," says Regelous. "We can change proportions of the skeleton, shader parameters, and brain variables." The agents' bodies are assembled as they are rendered by "Grunt," an A-buffer renderer developed specifically for this purpose by Weta's Jon Allitt, who leads development for Massive agents.

At the same time the agents' bodies were being created, the crew also created their movements. "First, we list what each type of agent has to do and break that down into specific actions such as, for the warriors, strike, side step, pull the weapon back, and block," Regelous says. "And each of these will have variations because the agent will be in different contexts."

Of course, this meant that hundreds of motions needed to be captured. "We generally have between 150 and 350 moves for each type of agent," Regelous says. An animation-blending engine built into Massive modified the moves on the fly to let the agents aim a weapon, for example, or grab another agent. "The agents are able to control their limbs with inverse kinematics," he says. Massive's Tree Planner program helped create this complex network of motions, and the result became part of the brain.

The brains were built with modules, which are networks of input and output nodes, rule nodes, and fuzzy logic nodes; the brains typically have 6000 to 8000 of these nodes. The agents can "see" scanline rendered images of their surroundings, "hear" frequencies of sound, and determine where the ground is underfoot and then respond based on rules that use fuzzy values to approximate the way people make decisions. Each type of agent has a particular brain and each agent in an animation has its own brain and therefore, its own, unique responses. The agents make decisions at a rate of 24 frames per second; choosing to, for example, strike an Orc in the middrift when its weapon is at a certain height and it's most vulnerable.

Massive was used in The Fellowship of the Ring for thousands of warriors that, once set loose on a battlefield, would find an enemy, pick a fight, and fight to the death. It helped digital doubles of the Fellowship navigate a steep staircase in Loth Lorien and the Orcs (aka goblins) taunt the Fellowship and climb pillars in the Mines of Moria. The agents can be placed in scenes within big circles drawn over a terrain, into rows and columns, or in particular places. "Once they're in place, we just let them go," says Regelous.

Everything else is PONG... :D

Grey
08-18-2002, 06:15 AM
Originally posted by beaker

>>(Grey Said) That cuts EVERYONE else out of the picture.

All I was saying is that its not really that big of a deal cutting everyone else out of the picture because Massive isn't really made for the needs of the person who can't afford it right this sec. Everyone else will be perfectly fine with any other crowd simulation plugin.

Though it will be nice if they lower the price and I believe they will. Since most of the big studios allready have their own in house crowd simulation scripts and plugins or they can have a programmer/TD there wip one out to suit their needs. This leaves a very small crowd of people to purchase the software at this price range.

The part your missing is that they used MASSIVE for INDIVIDUAL CA for Digital Doubles too in LOTR... (it's right up there in the article :D)

arvid
08-19-2002, 09:05 AM
Originally posted by beaker


The big thing that sets massive appart from everything else is that it is rule based and can handle millions of characters, not just thousands like all the others. Not many people who can't afford Massive will even have a project that has a crowd simulation of 100k-1million characters all fighting in a field, or if they do its going to look like ass anyways.

1 million characters? They might as well apply a noisefilter ;)

Grey
08-19-2002, 07:15 PM
I don't think Beaker had read up on Massive when he said that...

The huge number of characters is the least of it's abilities...

googlo
08-20-2002, 12:28 AM
You can already do all of that with Reactor controllers in Max..

Grey
08-20-2002, 01:29 AM
Tell us about "reactor controls" then...

googlo
08-20-2002, 03:17 AM
I was joking but if you don't know what they are in Max here's some info about them from the user reference manual.

*****
The Reactor controller is a procedural controller that reacts to changes in any other controller within the software. Reactor comes in five different forms: Position Reactor, Rotation Reactor, Point3 Reactor, Scale Reactor, and Float Reactor. Any animatable parameter in the software can react to changes in any other animatable parameter. Reactor is not based on time, but is based on other variables in your scene.

You can use a reactor controller to turn on a light as an object nears a given point. Muscles can bulge as an arm bone rotates. A ball can automatically squash as the ball's Z position nears the ground plane. Feet can rotate as their heels are lifted from the floor.

A particle system can be triggered by any given event. Morph target percentages can be controlled by events
******
Control Info:

******
React To—Click, then select an object in the scene or use Select by Name on the main toolbar.

Choose a position, rotation, scale, or float track. Changes in the selected track will effect the parameter you are controlling.

A default reaction is created after selecting a track. The Reaction Value is set to the current value at the time the track was picked. The Reaction State is set to its current state. Both of these can be changed at any time.

React To Name field—Displays the name of the track you are reacting to.

(Current React-To Object value)—This unlabeled field displays the current value of the parameter being reacted to.

Reaction List—Displays all the reactions.

Select a reaction from the list and edit its value, state, influence, strength, and falloff.

Name Edit field—Changes the name of the selected reaction.

Create Reaction—Creates a new Reaction.

The Reaction Value is set to the current value at the time you picked the track. The Reaction State is set to the current state. Both of these can be changed at any time.

Delete Reaction—Deletes the selected reaction.

Set Reaction Value—Changes the selected Reaction Value to equal the current value.

Setting this value causes the reaction's state to be reached as the current value reaches the newly set value.

Edit Reaction State—Adjusts the reaction's stored state.

Move, scale or change the object (with the Reactor controller) in the viewports to edit its state for the current reaction.

The Reaction State, is the position, rotation, scale, or floating point value the controlled track will reach as the current value reaches the reaction value.

Float controller—When assigned, the state parameter becomes a spinner instead of a button.

Adjusting the spinner at any time will adjust the selected reaction's state.

Influence—Set a range of influence around the reaction value. Position=world units, Rotation=degrees, Float=number range.

When the current value is within the given influence range surrounding the reaction value, the controlled parameter will approach the reaction state. If two reactions have overlapping influences, the parameter is blended between the two reaction states.

Strength—Sets a strength for each reaction.

When multiple reactions have overlapping influence, the strength value will weight the reactions in comparison to one another. This can make a reaction stronger or weaker within the same range of influence.

Falloff group
Falloff—When Use Curve is off, the numeric Falloff setting specifies the ease-in and ease-out speed as the current value travels from the reaction value to the outer limits of the influence range.

The Falloff setting uses an exponential function (1/(x^falloff)). Falloff values greater than 1.0 will cause the controlled parameter to start off slowly and speed up. Values less than 1.0 will cause values to start off fast and then slow down.

Use Curve—When Use Curve is on, click the Curve button to open the Reactor Falloff Curve dialog, to create a custom curve graph for the falloff. See the dialog description at the end of this topic.

Reaction Value field—Displays the value of the highlighted reaction.

As the Current React-To Object value reaches the Reaction Value, the controlled value approaches the Reaction State.

This value can either be a float, a Point3, or a quaternion, depending on the track selected to react to.

Current Output field—Displays the current value of the client track.

This value is updated as the client track value changes.

Falloff and Reaction Curve Dialogs
You can use curves to control reactions. The type of control the curve dialog gives you depends on they type of controller being reacted to.

When reacting to a position, rotation, scale, or point3 controller, the curve editor contains multiple falloff curves. There are as many curves as there are reactions in the reaction list. Each curve represents the falloff rate for its corresponding reaction.

When reacting to a float controller, the curves you see don't represent falloff. Instead, they are reaction curves that control the combined motion of all reactions.

Falloff Curve Dialog
The falloff curve dialog appears for reactions to a position, rotation, scale, or point3 controller. The curves it displays are ease curves that control the falloff for each reaction. Three-dimensional reactions use a volumetric influence approach, where the falloff is the speed at which the reaction's influence over the controller's value diminishes as it gets farther away from the reaction value and approaches the outer influence range. The falloff curves give you more control then you get using a falloff spinner with exponential falloff.

Reaction Curve Dialog
The reaction curve dialog appears for reactions to a float controller. The curve or curves represent the reactor controller's output for different reaction values. The number of curves depends on the current type of reactor controller. If it is a float reactor, there is only one curve. If it is a position, rotation, scale, or point3 controller, there are three curves that correspond to the X, Y, and Z axes. They are shown in red, green, and blue, respectively.

Then number of keys on the curve corresponds to the number of reactions. For example, if you have three reactions, your curve contains three keys. The horizontal position of the keys represents the value of the float controller the controller is reacting to. The vertical position of the keys represents the state the controller is in when the given value is reached. You can move keys horizontally and vertically to change the reaction's value and state. Inserting keys on the curve creates new reactions. Deleting keys deletes reactions.

When reacting to a float controller, the curve entirely describes the motion of the reactor-controlled object. For each reaction in the list, there is a corresponding key on the curve. The vertical value of the key represents the reaction's state (the reactor's output value). The horizontal value of the key represents the reaction value (the value the object you are reacting to). A vertical gray line represents the current value of the object you are reacting to. The reactor current value is the value of the curve at that line. With this model, there is no influence or falloff, just the slope and value of the curve, and the distance between keys.
*******

urgaffel
08-21-2002, 02:03 AM
I met Stephen Regelous at Siggraph, and he described much of what was in that article. The way Massive works is way more advanced than your regular crowd simulation tools in for example max and xsi. Since all the agents have their own brain, they can react a lot more competently than a regular system.

One of the examples he used at Siggraph was when an orc raises it's sword to bash a gondorians head in. The gondorian sees the orc raise his sword, estimates how much time he has before the blow lands and decides wether he should block with his shield or stab at the orcs unguarded side. Or if he should just chance it and stab anyways :)

Seeing the FOTR opening scene in realtime was pretty cool too :)

Grey
08-21-2002, 02:20 AM
I wonder if I'm the only one who's reminded of a high end Game AI when reading about it...

urgaffel
08-21-2002, 04:17 AM
In my opinion it's a lot more advanced than game AI...

Grey
08-21-2002, 04:21 AM
I'm sure you're right... but even so...

googlo
08-21-2002, 05:15 AM
It makes me think of object oriented programming being applied to 3d objects, like assigning behaviours and everything through hierarchies, classes, methods, etc.

Grey
08-21-2002, 05:22 AM
objectoriented programming with fuzzylogic?

urgaffel
08-21-2002, 11:10 AM
Om the other hand, if game AI started to use Massive... Woohoo! Intelligent behaviour from enemies! ;)

Grey
08-22-2002, 02:50 AM
hehe... and impossible to beat :D

Could you imagein a version of Doom with such an AI?

DotPainter
08-22-2002, 12:21 PM
What is really new is not the AI so much, since like someone else said, it has been used in games for years. The real achievement is using it for 3d animation. Games are built around AI and some type of real time physics engine, but most 3d for film has been done by hand. What would really be neat is to see the technology developed to the point where you could develop a model a character and all of its expressions and movements, then use some kind of AI system like massive to allow the AI system to animate the character interacting within an environment .... all in real time. The more sophisticated the model, expressions and morphs, the more like virtual actor or hands off animiation.
Add speech recognition and the ability to define objectives and ...
sounds like something for the lab.....

dukeofhasard
01-15-2005, 10:02 AM
i cant wait to try this demo !

whats really cool though is the 3dmax comunity have a nice opportunity to ally a rule based particle system with a great "procedural" animation system to create a "mini-Massive" that kick a***.. I m doing some deep test with Thinking particle 2 and Character animation tool and its really promising..

anyway , great news , hopefully we could know soon when its gonna be out !
Cheers,

onlooker
01-15-2005, 10:09 AM
i cant wait to try this demo !

whats really cool though is the 3dmax comunity have a nice opportunity to ally a rule based particle system with a great "procedural" animation system to create a "mini-Massive" that kick a***.. I m doing some deep test with Thinking particle 2 and Character animation tool and its really promising..

anyway , great news , hopefully we could know soon when its gonna be out !
Cheers,

in case you didn't notice the last post before yours was 08-22-2002. I don't even remember what happened to this demo, or if it was ever released.

Wizdoc
01-16-2005, 03:03 PM
I just checked the Massive website. The "no plans for freeware version" answer is there, so I'd guess they ditched the idea.

While we're on the subject, I'm pretty interested what this 18,000$ license includes. If it comes with a full ready-to-use behaviour library, then I'd say it's very affordable for its type. But if the prices is only for the "core", and you need to get those 100-150 mocap animations per unique agent yourself and set all environmental properties from ground-up, then it's another thing entirely.

shingo
01-16-2005, 06:44 PM
How Advanced one crowd simulation application is versus another is a matter of debate. The fact that Stephen Regelous considers Max's and XSI crowd solution to be the same thing suggests a limited undestanding of both.

Max's crowd system is a rudimentary AI tool whereas Behavior (XSI's crowd system) is a Finite State system - virtually an entire game engine in it's own right.

Massive was originally priced at US $40K and then reduced to $18k over a year ago. IMO, this move from Massive Software has been hastened by the fact that Behavior (used to be US%15k) is now included with XSI Advanced. Also, because Massive is a standalone tool, implementing it involves trasnferring data from a 3D animation app to the crowd simulation engine, which often requires some technical expertise and maybe even the development of cutom tools. Because the data sets that come out of a crowd system like Massive are potentialyl HUGE, wrangling the renders is often unavoidable.

Furthermore, I read somewhere that the version of Massive that ships is not the complete system utilized by WETA. I believe that the rendering component (GRUNT) is not included, which is the reason that Framestore decided to write their own tool for Troy rather than purchase Massive.

Like Massive, Behavior is a stand alone application. V2 introduced a direct rendering pipeline to Mental Ray, which enables it to render crowd scenes in the hundreds of thousands.


I met Stephen Regelous at Siggraph, and he described much of what was in that article. The way Massive works is way more advanced than your regular crowd simulation tools in for example max and xsi. Since all the agents have their own brain, they can react a lot more competently than a regular system.

One of the examples he used at Siggraph was when an orc raises it's sword to bash a gondorians head in. The gondorian sees the orc raise his sword, estimates how much time he has before the blow lands and decides wether he should block with his shield or stab at the orcs unguarded side. Or if he should just chance it and stab anyways :)

Seeing the FOTR opening scene in realtime was pretty cool too :)

motionstorm
01-16-2005, 07:59 PM
I've worked with Massive since since the beginning of 2003 when stephen was still working on getting a version out to studios interested that could implement it into their pipeline. The major problem was that so much architecture that made it work well was assisted by the proprietary tools that Weta had written (Camera Tracker, OrcBuilder, Grunt and other tools). Now with a lot of those features built in that needed to be added the software stands on its own fairly well in my opinion. Pay close attention though, the article that was posted from CGW is BRAND NEW and FYI...

THE NOTE on the website which says there will be "no freeware" version was from the original site, far before the time they ever had an executive staff and is VERY dated, before I was usig the software. I'm sure that page will come down soon because its out of date... so don't fret all you wanna be massive users... the powers-that-be at Massive are smart and have been wanting a broader exposure to the software for some time.

I think it would be the best move possible for them and the CG industry at large. I'll look forward to a freeware version being available to the masses for their work... now we'll see realistic crowds all over the place. Honestly, the software rocks if your just doing something like 3-6 people interacting.

nullUser
01-16-2005, 09:20 PM
My guess it that WETA's thinking is that the more people using it, the more experince out there, the more likely Studios will take advantage of that experience and use their software in production work...

I think it's a good supposition :D

So they're probably hoping you'll play with the freeware version a bit....


weta is a customer of massive. they don't own it.

Ezekiel19
01-16-2005, 11:36 PM
this is interesting and cool.


Thanks

zappenduster
01-19-2005, 03:52 AM
what i would like to know is how much processing power is needed for this kind of software like someone stated before it was shown on a expo in realtime was that only the final thing or did they show also the simulation process ?

nullUser
01-19-2005, 05:45 AM
what i would like to know is how much processing power is needed for this kind of software like someone stated before it was shown on a expo in realtime was that only the final thing or did they show also the simulation process ?

I would imagine the simulations like the ones in LOTR with agents fighting each other and interacting and stuff would not be able to be run in realtime, you would have to calculate exactly what was happening and if the scene was complex I imagine you would have alot of tweeking to do after that.

If they said it was going in realtime I imagine it would have to be a very simple simulation. But I don't know anything about Massive, so I'm just guessing here...

jeffthomann
01-19-2005, 02:57 PM
Anybody tried Crowd Simulation Framework? http://sourceforge.net/projects/crowd/

zappenduster
01-19-2005, 04:18 PM
I would imagine the simulations like the ones in LOTR with agents fighting each other and interacting and stuff would not be able to be run in realtime, you would have to calculate exactly what was happening and if the scene was complex I imagine you would have alot of tweeking to do after that.

If they said it was going in realtime I imagine it would have to be a very simple simulation. But I don't know anything about Massive, so I'm just guessing here...

yeah thats what i thougt also but question is would a nice quad server do the job already or do they have to crank out the half renderfarm for 3 nights to do a full simulation

i see there are 2 ways for such problems you can just break it with bruteforce and slash as much processingpower in as goes or you can walk the intelligent way and maybe save time and money

CGTalk Moderation
01-19-2006, 05:00 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.