3D Studio Max instead of Poser?

Can I use it instead? Has anyone tried? Thanks

Hello

I too was looking for a BVH exporter for Max. Nothing found as of yet. Pitty since CAT and Character Studio (in my opinion) are superior in that it allows you more freedom to pose your character and can do so without the aid of cluncky, mechanical dials.

I am new to SL and was wondering if I could use 3D Studio Max to create objects for use in SL. I played with the SL tools.. and they aren't bad.. but since I'm already fluent in max.. it would just go faster and the result would turn out better for me.

Anybody have any input for me? I know Max can import Poser stuff.. but could it re export it back to poser to then use in SL? just a thought.

thanks,

Dirk Gorham

I arrived in SL yesterday and asked one of the greetings/helpstaff about importing from max. I was told that while you cant right now, their planing on bringing tools to import from max and maya in around two months. Take that with as much salt as you wish.

Funnily enough it kinda took the wind out my sails - no point struggling with the supplied tools if ill be able to import from max come summer.

Mungus,

I would suggest you reblow your sails since usually these kind of big implementations are delayed for production. Importing tools next summer? or should it be the summer aaaafter that?

Mangus, was the person who told you this an employee of Linden Labs, or were they just a resident hanging out in the welcome area? In other words, was their last name "Linden" (and if so, do you remember which Linden it was)?

The reason I ask is we've all been told since the dawn of time that this kind of thing was likely never to happen. Now all of a sudden it's 2 months off? I can't help but seriously question the validity of whomever gave you that information. Please elaborate if you can.

I am SO hoping this is in the works!

Yes, it was definately a Linden guy.. he gave me a cocktail and a treasure hunt and all the things I assume their supposed to do. I can't recall his name but I do remember he said his job title was lead evangelist (I remember that because I made a joke about him asking me for money).

I may ask a question again in the Hotline to Linden forum - if im the only one to hear about this then i definately want confirmation.

*Update*
Looks like someone beat me to it.
/invalid_link.html

LMAO

Lindens are supposed to do that? Where's my cocktail? Did the treasure hunt lead to mountains of L$????? I must have missed out.

It led to free objects, animations and some money trees, which yeilded a total of 2L$

I just found out he left me a calling card too, his name was Ruben Linden.
He was a really nice guy, I hope I haven't gotten him in trouble.

I read that 3DS-Max has a scripting language. I just developed a BVH exporter for Maya as an open-source. If you wanted to, you could probably take my script and translate it into Max's script language. My code is (I think) reasonably well commented. See the post on 'Free BVH exporter for Maya'

Only Max and Maya? *sob* I'm gonna miss you my dear Lightwave... *sob*

The reason they currently use primitives is to save bandwidth and storage space. Meshes allow for more detailed models using fewer polygons, but each vertex must be stored in the file along with the texture map for each face. Using primitives you just store the center point of the prim, it's orientation, dimensions, texture, etc and you're done, it's smaller and can be sent faster. The client does all the work figuring out all the vertices. Also it allows a quantized method for counting render engine use (prim limits). If they do allow mesh importing which I would actually like, how will they be keeping track of prim limits? Two months? well, it was apparently a linden that told you this, but they don't actually get their estimates too close usually. I wouldn't place any significant bet on 2 months... maybe 2L.

Hmm, all this thread sounds suspiciously like "wishful thinking" on a wrong assumption...



Most likely that means: they're giving a skeleton for importing animations created with either 3D Studio Max and/or Maya into SL - not objects.

There are so many reasons that speak against "importing 3D meshed objects" in SL from external editors...

1. Bye-bye collaborative work
   The SL platform is announced as being the best example (well, at least, the best among the ones that are beyond the alpha stage...) of collaborative work online, since you can "glue prims" together with your friends. Even amateurs can build together, since the tools are so simple (ok, you may qualify them as "primitive" if you wish) to use. So now you exchange all that for individual work by professsional designers - the ones who are able to master Maya, Blender, Max, Lightwave, whatever - who do all their work off-world, alone, and just import it. Hmm. There goes one of the key features of the SL platform: collaborative work. I guess LL's marketing won't like that.
   
2. Bye-bye physics engine
   Meshed objects are simply not easy enough to work with with the current and the upcoming physics engine (which has been "under development" for, uh, 2 years...?). So this means trashing all the existing code on the prim-based physics engine and start from scratch. And they have quoted you on "two months" for that? Hmm. I'm very, very skeptic on that:
   - either they will allow prim-based objects only to be uploaded
   - or they have figured out a way to have the physics engine working with both meshed objects and prim-based objects
   - or they're dropping the whole physics system altogether. Weird. If you read Philip's blog, he's also very keen to point out that one of the key features of SL is its physics engine. And if you search around in the forums, you'll see for how long people have been waiting for Havok 2. I don't think that would be a sound move.
   
3. Bye-bye prim-based allocation system which translates cooly into CPU resources
   So, if you have meshes, you need to have a completely new system which also translates easily into CPU resources being consumed. A polygon-based system comes to mind. Sure, it's tougher for a newcomer to understand why prim-based hair consumes up to 380,000 polygons and "normal hair" just a few hundreds, but we residents are able to adapt to a new model. However, again, this would do dramatic changes to SL. Most (if not all...) really dramatic changes in SL have been discussed with the residents - things like the upcoming permission system, for instance - and I don't really believe that they would develop something in the next 2 months that would change radically our economy. And in the middle of releases? Nothing of this was ever "planned" for 1.7, which is not planned to be released in "the next two months" anyway...
   
4. What about compatibility with existing objects?
   Yes, what about our lovely prim-based objects? Will they be silently converted into meshes overnight? I would imagine that a "conversion system" will take much longer than 2 months to develop. And I still don't think that they would "allow" a double-system to work - prims for "old objects" and in-world rendering, and meshes for designer masters using an external 3D tool. While the SL engine is able to use both (our avatars *are* meshes, after all...), I'm not sure if they could have two types of objects working together.
   Notice that the "new renderer" (2.0), as shown by Ben Linden on this thread, seems prim-based to me... and the "new renderer" is due to come out only with 1.7.
   
5. Lindens rarely disclose upcoming features with people they don't really know (when they do it, it's mostly through an "official announcement", either in a Town Hall meeting or through the forums)
   Well, this may sound snobbish, but most Lindens are very careful not to trust any residents with their "secret plans" and "hidden agendas" or "timelines". Yes, it's true, there have been people disregarding their trust and posting "confidential" information on the forums. And I believe that some Lindens are more "careless" than others, although I have yet to find one (and no, I haven't ever talked with Ruben) Let's just say, for the sake of the argument, that such a "bombshell" wouldn't probably be "common knowledge" among the Liaisons - most of the "future development" of the SL platform is revealed to the Liaisons on a "need to know basis" - and I would expect that this kind of "information" would really be highly confidential and not so lightly "given away" to any new resident simply passing by...
   

So, my guess is...

1) Ruben was talking about importing animation files from Max/Maya instead of limiting the animations to BVH files created by Poser. This makes a lot of sense to me, and I also think that 2 months is a relatively conservative timeframe for developing those. Also, that is not sensitive of confidential information. It will not impact any major change in SL, except for allowing animation artists a bit more choice of platform, which would overall be better (Poser is easy to learn, but if you have been doing anims in Maya or 3D Studio Max for ages, you'd prefer to do them in the tool you know best).
2) LL is planning to allow people to create prim-based objects (and only those!) inside Max or Maya and allowing them to be uploaded, thus feeling that this will not threaten their "collaborative work" platform much. While I disagree a lot with that view, I have very limited resources to do some serious market studies - perhaps LL has done them thoroughly and reached a different conclusion.
3) SL's change from a prim-based environment towards a mesh-based environment will take much, much longer than "just two months". If that is the way SL is going to "grow", I'd point it out for SL 2.0 - in about 2 years or so, not 2 months. We only have a few references on what we can expect for 1.7 and 1.8, and a tiny bit beyond that, which will keep LL busy for most of 2005 and part of 2006 at least. In case you haven't noticed, a mesh-based SL would probably mean "no backwards compatibility" with existing content, or at least "limited compatibility". Now, I really can't believe that this sort of thing would go "unofficially announced" by a Liaison to a new resident as an off-the-hand remark.

I usually trust LL's common sense a lot Yes, I know, we have been surprised very often...

I wish id known what I know now when I talked to the guy, id of asked a whole heap of questions. Its nowhere near as simple as it seemed at the time. I also hope I can find him again and tell him how I put myself on the spot

I feel better building stuff out of lots of spheres knowing that the engine is optimized to instance them. Sure, id love to get my max meshes and anims into SL. But this aint work, its pleasure, im not too proud to get primitive =P

I know from experience that theres probably a very good reason why things are the way they are. In the end, everyone wants the same thing, we want SL to be the best it can be. We want fast, seamless, rich worlds populated by vivid living avatars.

I'm happy enough waiting around for whatever LL think is the best way to achieve this. Meantime, pass the primitives. =)

Okay, here's the official Linden response to this question (thanks Robin):



That's pretty much what I thought she'd say. It would be really exciting if this kind of thing were possible, but all signs are that it isn't. SL has to work with only a common set of objects due to its streaming nature. Every machine already knows how to make cubes, spheres, cylinders, etc. because that functionality is built into the client. That way no actual geometric data need to be sent over the iternet. The only information that needs to stream back and forth is where to put each prim, what direction to rotate them in, and what size to make them, which is a very small amount of data. Were custom meshes to be imported, the entire kit & kaboodle of information about exactly how to build each one would need to be streamed to every user machine, which would take forever and a day.

It would be great if we could do this, but I doubt it will happen any time in the next few years.

Well said, Chosen. Actually, I've talked to several die-hard RL 3D modellers and they all get tremendously disappointed because they have to use a "new" tool for doing 3D modelling, instead of relying upon their own, favourite tools.

Interestingly enough, while I would think that there would be an immediate appeal to SL for 3D modellers, I guess that they are rather distrustful (or even scared!) of the way LL has dealt with things. After all, you don't have many 3D modelling tools designed for collaborative, real-time use under a $5000 price tag. Someone who is proposing a system which does exactly that and only charges $9,95 for the "tool" is viewed with serious suspicion.

Of course, many of those very talented 3D creators are much more open-minded. Sure, SL is not 3D Studio Max or Maya or Lightwave, and probably never will be, but it certainly does lots of interesting things that none of those tools are able to do. One of those things is putting advanced 3D modelling technology in the hands of amateurs. As basic human nature goes, this scares 3D modellers off - and naturally enough, they shun SL, or disregard it as a "children's toy", not to be taken seriously, and voice their comments loudly among friends, colleagues, and, most importantly, customers.

As things go, failure to adapt to change leads to stagnation. Remember those days where you had to be an expert in software programming to put up a Web site and type HTML with a text editor? Well, these times are over - you just need some sort of blogging software to put up your own "amateur" content up, and it still looks good. I think the same will happen with technology like SL. It's for the "normal human beings" like myself. I'm not a programmer nor a 3D modeller or professional animation designer. But in SL I can create dance bracelets which work well In RL, I wouldn't even dream to start working with professional tools to achieve those very same results - it would be pointless, even as a hobby, since I would need years and years of practice and patience just to achieve mediocre results. The cool thing of SL is that everything is pretty easy to do and use, and after a few weeks (or a few months, in my case, since I'm a slow learner... hehe) you can get high levels of quality and maturity on your work. It's very, very rewarding. But, of course, a serious threat to professional 3D modellers.

Although importing meshes from Max is not in the cards, I think it would be possible to create a plug-in for Max that would allow it to be used as a powerful offline SL editor. Meaning an emulator of the existing prim-based editor in SL, but with many more powerful features. The resulting work could be imported through SL scripts using llSetPrimitiveParams() etc. You could export the SL terrain into Max so a build was in the correct context. You could combine SL snapshots and camera tracking to composite a work in progress into an SL scene. (I've done the last two actually.) Not that I expect LL to do this. But there is a difference between limitations of the rendering and streaming technology and limitations of the editor.