Questions about modelling

Hi, I just got the client today. I haven't logged in yet but I was browsing the documentation and I just had a couple questions about creating objects and the like:

1. Is it possible/going to be possible to use an external 3d program such as Lightwave in order to create custom objects for the game?

2. Is it possible/going to be possible to create custom avatars?

that I'll leave up to someone at linden, but as far as I've seen, neither of the things you're suggesting are currently possible.

If you think about it, that would be nearly impossible. The increased polycount alone would be massive. Once you get in game look at the avatars, you'd have to torch all the current customize if you were uploading arbitrary avatars.

Thanks Celerity,

In short:
Q1, no, though we haven't ruled out an offline tool sometime WAAAAAY in the future. Please note. An offline tool is much more difficult than you might expect for this world, please don't email me asking when it will be done, it's only a pipe dream at the moment.

Q2, Customization? Oh Yeah! No uploading of "other" avatars, but unlimited options on customizing your own. None of the avatars that you see on the screen shots are "pre defined", they are all created in world by users.

We honestly are quite proud of our avatar customization, please, knock yourself out. And while you are at it, think about this question: How would you dress if you had NO social limitations, and knew that EVERYONE dressed, er, a bit oddly? Would you wear pink bunny slippers all the time? Would you wear a zorro costume? Do you want wings? Scales? Tendrils and tentacles? Wings? Funny hats? Would you be a block of wood with a target on it?

The answer to that question is easily given with your avatar.

Have fun,

-Peter

P.S. Sorry for the soapbox diatribe, but after watching so many people play with avatars yesterday, I can't help myself.

and if you don't like something as basic as say.. legs.. you can use attachments sometimes to overcome these difficulties. Hey peter, do you know if you can use transparent skin? then you really could remove limbs and replace them =)

Seriously, there are some neat avatars in game. Nada's is fantastic, lotsa work.

and yeah, I tweak something on my avatar almost every time I'm in game, haven't got it perfect yet =)

I was visting BP at his undergound hidaway a couple nights ago and he commented on my heinous flesh colored pants. He proceded to take a snapshot and attach it to a billboard sized primitive right before my eyes. I was a bit suprised at what I saw. Although I saw my av wearing a red furry jacket and blue bellbottoms, the image showed my av without jacket and transparent bellbottom pants.. with my whitey tighteys showing thru!!! What's up with that?
I tried changing my appearance a few times and some of the changes BP saw and Zoli (who was there as well) did not. In fact zoli saw me wearing only a white undershirt at one point.
My av is afraid to go out in public now unless I attach a barrel to him.

Hey thanks Celerity! It is amazing what you can do with a couple primitives, i wanted wings, so i made some... its all about the go getter mentality, oh and a willing to fiddle with all the little numbers that make up the object panel. Also for all the new beta-types, if you have questions or are trying to figure something out, find another white dot and ask! There are a lot of us on who will stop whatever we are doing to help you guys out .

Oh yeah, i agree with you celerity, work/school = necessary evil.

ok, how does *that* work?
Is there a way to tunnel or is there just a hollow/valley roofed over?

It's just roofed over... but it feels like you're underground.

In fact... I'm planning on ditching my lame tower for something similar. Especially with the anti-tower posts circulating lately.

Interestingly enough in the old grid (old avs) we could make av parts go transparent with an alpha channel texture applied. Had fun floating around as a head. =0) Now the hair (really no use now) is the only 'part' that will work with the transparent texture.

I had spoken to some Lindenite and our conversation was regarding an alt idea that perhaps we could just unattatch a limb or part.

I would like to hear if they are still considering this as it would really, as you guys suggest too, help the flexability of our customizations.

Towers are a more effecient use of space. Build more, cover less ground, what's so bad about that?

Hi Celerity et al,

Its not that towers are bad, they are an efficent use of space. It's that they are generally ugly.

Assuming that the land is somewhat scenic, rolling hills, lakes rivers, and then someone builds a large cooling tower, the aesthetic is just offensive to our sensibilities (well, some of our sensibilities). Frank Lloyd Wright had lots to say about making structures part of the land, and not on the land, but architectural philosophy was not my area of study.

On the other hand, if someone were to build a good looking tower, we could all easily point out how well they can be made, but so far, that hasn't happened.

Begin form/function debate here.

My 2 1/4 cents...

-P

ok, so I admit it, I have the aesthetic design sense of a myopic wombat (which is to say, not a lot).

But I'm trying to make my place pretty. If anyone has suggestions I'd be more than happy to listen. I think the largest part of my problem isn't with the actual building (though that bites too) but is more centered on my texturing.

My tower has a lot of wall. it gets pretty monotonous pretty quickly. On the inside I'll be doing all kinds of crazy things.

on the outside I'd rather be a little more sedate so as not to offend the neighbors =)

And since I can see my tower from half a world away..... Everyone's my neighbor.

Though aside from people dropping in to give me new textures I've gotten more approving feedback than negative.

I would be very interested in this as well. For example, not everyone wants that many fingers, legs, arms, etc. in their design. Yet right now there's no easily apparent way to reduce the "basic number of appendages" short of an attachment to cover them up.

Obviously using an attachment is less than optimal given the attachments don't "move" like the avatar itself does for gestures, etc.

As an example, I want my avatar to have a more "simplified" fingers/hand/arm design that existing, but still retain the associated motions intrinsic to the avatar's model.

In pgms like Poser this is easy to handle, where just the "remaining" pieces continue to operate as designed in the motion.

In SL, though, since I cannot actually remove any parts, I seem to get this mess of trying to "reduce them to invisibility" which doesn't seem to work well with the existing gestures, etc.

Having the ability to remove parts, and have the remaining parts continue to follow the translations/rotations designated for them in the motion/gesture would be very useful.

Oh yes, and being able to use "custom gestures" by importing standard BVH files with identically named parts following the designated motions would be very very nice. It'd allow folks with Poser, Character Designer, and high-end MoCap tools, and even those who just find nice BVH files on the web (there are many out there) to use them as gestures.

not all towers are ugly There are some truly beautiful lighthouses in SL... Val's on Elysian Island comes to mind.

First... wow, this is an OLD thread.

Second -- yeah, another one that comes to mind is the spirally turret with all the steps on the outside. I think we can blame that one on Alek Wu

It may be old, but it still has some excellent (and as of yet unimplemented features) worth exploration. I only wish it were staying a bit more focused on feature requests, as the topic drift makes it unlikely it'll garner "official" attention.

One way to shorten the development time of the "offline" design tools would be to simply extricate the existing tools and package them with a kind of "local runtime", thus allowing designs using precisely the same data formats and restrictions of the actual system. All that would be needed then is support for serializing the "local system" output and importing it into the SL environment.

At least for modelling, I suspect doing such repackaging of the existing tools would be much, much quicker than writing new offline tools.

For generating custom gestures, I firmly believe that allowing import of standard BVH files, and publishing the existing "group names" for the avatar parts, would provide a far better solution than reinventing the wheel by writing offline tools. Then it would be simple to use Poser to generate the gestures, and simply translating the Poser group names to SL group names after the fact (a simple task working with BVH files).

If Poser's constraints weren't quite the same as SL's avatar joint constraints, etc. then LL could simply provide the constraints they're using (as it seems likely they are using such tools for gesture work). All they'd need to do is provide a Poser-importable constraints set.

The nice thing about using BVH is that there are so very many tools designed to work with them for animation, from Poser (towards the low-end) to real-time MoCap tools at the high-end. For those who aren't inclined to buy or use such tools, it would also provide another marketable in-game item -- new gesture anims.

I'd love to see this stuff, but who knows what will happen. Given the age of this thread, and the stuff still not being available, my hopes are pretty constrained on this.

If LL was really evil, they'd open the formats of certain files (like gesture files), explain how they work a bit, and let us upload them.

At that point we'd have everything we need to start hacking up/working on our own little 3rd party utilities or whatever to create these animations...thus passing on all that work they'd have to do to us. It's obvious the community has the talent base for it, and I'm sure some of us are just stupid enough to do some of LL's work for them without pay. lol

As far as 3rd party modelling programs, I initially thought it would be nice too. Well, it still would be. But without loosening the contraints that we have with current modeling tools...what good would it do? Currently you'd have to make your models with so many rules in mind that it wouldn't be worth it unless the way objects are uploaded to clients changes.

Okay that's a stupidly verbose thing that probably makes no sense but hopefully you get what I'm trying to say. Basically I think it's more complicated than just letting us upload models...they still have to send them the other way, too. Also I'm not sure how feasible it would be to do something like say, twisting an object in a 3d program and have it work exactly the same way that it does in SL.

So unless the object transport protocol is rewritten totally, I guess there just isn't any immediate benefit other than a better interface. But with all the problems that could arise with model format conversions and crap, heh...bleh.

I actually expect that's precisely what they'll eventually do for gestures. Then folks can write Poser->SL exporters (or BVH->SL converters, there's loads of really cool MoCap BVH files out there for use).That's why I suspect the best approach is to simply give us a version of the existing tools, but one which can run standalone at home. It would provide the (quite useful) ability to design objects offline, thus allowing unconnected work. It would also provide a workspace that allows for "persistant" experimentation without cost, which is not possible now (since the sandboxes are cleared nightly, and work elsewhere costs). I think there'd be much value in just being able to roughly prototype things, over multiple work sessions, (with cost estimates but no actual cost).

While it wouldn't add all the spiffy features of other modeller tools, as you noted most of those features wouldn't be compatible in most cases anyway.

Long-term, opening the object format (complete with descriptions of all the constraints) would make more sense, thus allowing folks to design their own modellers or add exporters/converters which could meet the constraints to existing programs.

Either that or give the "offline SL object editor" a 3rd-party-accessible plug-in interface to allow adding new tools that still fall within the existing constraints. The benefit of this approach is that it would also provide third-party tools that could be incorporated into the online SL editor if LL wished to pursue that functionality but didn't want to do it themselves.

I just hope some lawyer doesn't come along and scare them off when the time (hopefully) finally comes that 3rd party software is possible.

If this is the case and anyone from LL reads this--look into Ultima Online's system of player created support software. I remember awhile back they started officially "certifying" 3rd party programs that were created to be used with their game...

Thank you so much Panda love.