Well, this is certainly a new feature. Thx

Thanks for the preview Frank. Can you tell us if the models are interchangeable or at least editable? Are different or custom action sets supported? Or is the purpose of the tool only to generate quick people for ArchViz?

I played with this last year and at that time thought it was a really neat idea.

I'd like to see some more information on customizing the actual models and/or modifying behaviors. If my memory is accurate, there was a way to edit some XRefed models... but I was not really clear on the intended workflow for any kind of customization.

Another useless thing which cant be used in real production, i realy dont understand why max developers team are doing this toy features it is for kids?? Why you cant do something which can be comparable with massive or goalemcrowd or even softimage crowd? Why you didnot put more money to develop the old max crowd system but you start with this limited solution, we dont want this 0.01 beta version of 6 year developement. And also please open your eyes and start to look on production standarts like alembic, bullet etc... And if you want to make archviz guys happy, please put this feature to max designt and for regular max start proper developement

I use 3ds Max (not design) and I would find a great use for Populate. Thank you! I wonder if there is anything new in the drop down menu labeled Hair. Couldn't tell it was closed.

gheppetto is a good tool for arch users but seems can't high editable and not usable for non-arch users

Thanks Frank! Hope development team make it more useful for production teams by adding more features and customizations like custom rigs and animations, custom characters and materials, more character<>character character<>environment interaction options, etc.

Custom behavior and AI maybe in the future?

Thanks for the insight.

But first of all - AFAIK it's Project Geppetto, not Gheppetto...

@larex339

USELESS? Oh boy. I made so much money with Geppetto already, my clients get horny about it.

@c0cllc
so u are archviz user...
but there are many people which ear money from more complex things like vfx and no placing random people into scene with one click

@larex339:

I think you a right that this seems to be more fit for the archviz user.
But don't forget that you are requesting an unrealistic feature here. AFAIK Goalem Crowd is a plugin for Maya coming with an ~$6.5k pricetag and of course 3rd party, not Autodesk. Massive is even more. You did not really expect such a product/feature would slip into 3ds Max all of the sudden ?

re larex339: I agree with you. Why does AD create these island solutions and doesn't create plugins that works in max, maya and softimage at once? Or why not migrate existing solution from one package to the other...

I'm an ArchViz user. Some says many people spend on VFX. How about Arch people? They don't complain much and just need some amount of tools such for modelling, rendering etc. They don't care much about Particle, Rigging, Dynamic etc like in my zone. Max is for all. But hope it can be used in anyway you want.

What happened to the second video (protestors?)?

I didnt get a chance to see it? Was it about waiting for XBR or something?

@spacefrog

If somebody is developing something for 6 years, are you expecting something like this or something which can be comparable with commercial crowd systems? For my point of view i think is super cool to be programmer in autodesk and in project like this. If your end resulf from 6 years developement is this feature, i think i want to work in autodesk :-) And if you know softimage you also know that they have the technology so they can share it with max developers team, I i realy dont know why they didnot do that but they start their own way and this is the result..

@jedie exactly that should be the point, where the softwares can share technology and everybody is happy.

@Jedie
"Or why not migrate existing solution from one package to the other... "

Sorry but that "migrate" thing is not always as easy as that. Sometimes it takes much more effort than just writing your own solution from the scratch.

@timd1971
the link is here but sorry it says it's private!
http://www.youtube.com/watch?v=6AcnJ8FQL-g

@timd1971 "What happened to the second video (protestors?)?"

It was someone who used geppetto to create a crowd scene for an album cover, it was a neat creative approach from a talented artist, but showed nothing of the capabilities of geppetto, probably because they are not that great.
What this guy did could have been done in a fraction of the time without geppetto.
You missed nothing that isn't shown in the first video.

Looks like a decision was made from above, and it was pulled, Frank has been back here, removed the link, fixed the "gheppetto" typo, and said nothing about it.

Way to go Frank, keep up the new communicative approach, can't wait for the next layer of marketing hype to appear.

Give my love to the shareholders.

What I find the difference btw Frank and Ken is communication. Frank seem to be quiet and did not reply or any interaction to user's comment. It seems "You can say whatever you want, I don't care. I'm silent." At least some responses, but not all.

Hi RioMiv, of course I care. Its a sneak peak, not a deep dive into how it all works unfortunetly. But your right, I wont be able to respond to every comment.

As for Geppetto, it is designed specifically for design viz and bringing worlds to life. Its not meant to replace high end crowd systems, this is an easy to use system that used a predefined set of characters and animations that takes only minutes to setup.

That said, it is exposed to maxscript to do some amazing things with it and you can override materials \ textures. On the roadmap we are looking to give you further access to modify your own characters.

As for the video I removed, unfortunately it was released too early, they are waiting for the album to be released first. (It was a customer example)

Really nice to have your response. Some users will provide negative and positive response. It's normal. We know that you're limited to reply, But we need your response to clarify bit more. Then we will feel that "Autodesk is communicating with user". By the way, I miss Ken. But we're living with you now in 3dsMax world, frank.

a simple-to-use crowd system is also useful for vfx, commercial and game cinematics. Sometime you need to make some background, a city, a shopping mall, an enemy base, and fill it with moving people to give it some life.


>And if you know softimage you also know that they have the
>technology so they can share it with max developers team
>I realy dont know why they didnot do that but they start their own way and this is the result..

not sure where that comes from .. The ICE crowd in Softimage from 2010 is a set of TD tools to help you program your own crowd tool. It's just the basics, based on particles, and simple motion blending. It was developed two years ago and there is nothing that the max team doesn't already have.

Sofitmage had another crowd system called "Behavior" many year ago, based on an game middleware acquisition. Development was abandonned around 2004, years before autodesk acquired Softimage. The guys who worked on that have left softimage almost 10 years ago. It was not actually ever meant to be a crowd system, it was a game AI middleware system that didn't succeed in the market, which Avid repurposed for crowds.

You had to program your own crowd system and behavior in Javascript. It was a programming environment and not an interactive crowd layout tool. It was a separate application and not integrated in the Softimage UI. you had to export to .mi2 files and render at the command line.

"a simple-to-use crowd system is also useful for vfx, commercial and game cinematics. Sometime you need to make some background, a city, a shopping mall, an enemy base, and fill it with moving people to give it some life."

And totally useless if you need custom meshes.

Hi Mr. DeLise!

From a developer POV how this crowds work?
They are seen as triobj from the sdk?
All them are a single triobj or there are one per ragdoll?

Probably the most common engines will support all this now but the poor-man ones could be really interested on this questions.

(Note my intention is to support it HERE)

I shared this idea with the Geppetto team last year... but I would like to see integration with CAT (which of course means more stability in CAT). It would be really cool if we could use arbitrary CAT rigs.

Ideas for integration:

* Set the models used along any specific path/area to come from a probability pool (like Itoo's Forest does for trees and plants--one area can have an arbitrary probability set of arbitrary trees).

* Allow the controls for the paths/splines to allow "tagging" that can be linked to motion layers in CAT. For example, I might tag an area with "silly" and then any character that goes into that area will now have a chance of acting silly if it also has any motion layers that are tagged silly. You could tie into each side a probability factor to help control it.

I had some more ideas but don't recall them all. But those were ones that I think would give a lot of leeway... and could open the door for non-human environments.

Thanks for reply Frank, and thanks for the info about maxscript and roadmap

will this work for steps, stairs and uneven ground?

"will this work for steps, stairs...."

...and cricket noises.

(Note: by 'cricket noises' I mean absolutely no response/silence, and not the sounds of someone painfully going leg before wicket.)

It works with ramps, but we still need to add animations for steps.
Its in our plan to support.

This is an updated version of the 2nd 'populate user example' video that was referenced in this blog post around a month ago.

3ds Max - Populate user example | Youtube.com

It was removed early at request of the Band's record label, who delayed the release of the album. Now it can be shown, I've made an updated version of the video.

As well as showing the Populate feature, The video includes an example of the Nitrous Viewport improvements and use of the 2D Pan and Zoom tool.