Standard Sizing Package For mesh Clothing Design; Good or Bad Idea, or a Bit of Both?

Still need to do some edits to this… There’s a lot of argument (hardly debate, more rantish most of the time) against the ideas of standard scales and proportions for avatars. Folks often pick a horridly distorted shape, or make one for sale; and then become attached to defending that choice when made aware of its problems. Others choose a ‘fun-house mirror look’ on purpose, for effect – but these folks tend to actually know what they’re looking at, and have purposeful distortions designed for a specific intent.

Various arguments have formed over what can be done, or even if anything should be done, to address this.

Mesh now complicates the whole feud – due to the lack of its ability to ‘auto fit’ some shape dials, while perfectly matching others… Or in short; mesh will auto size itself to any height, limb length, or body width. But if your ‘bust-waist-hips’ fail to match the model used to make the mesh clothing, you’re out of luck.

Go shopping for mesh clothes and you’ll discover that this is rapidly dissolving into exactly the chaos people predicted when they warned of the problems of Linden Lab not including a deformer… There’s a motley variety of dimensions to clothes on the market, and the labels are near meaningless. Small from brand A is ‘the fat lady sings’ in the eyes of brand B… And even if two brands match for waist and hips, they might have very different ideas of bust…

I’ve read a little bit on the idea of starting a standard, and have been somewhat critical of it in past topics. Still not convinced, for some reasons noted below. But it also may be the only real solution we have unless and until Linden Lab decides to grace us with the “gift” of including the code community members had to fund a third party contractor former Lab employee to make without any guarantees it would even be adopted… That deformer… the sort of thing that really should have been a ‘cannot ship mesh without’ line in the sand element. This is the kind of lack of core feature that in most development communities, management should have laid down the law and put people’s jobs on the line to get in there before Mesh was released.

I’m ranting…

Ok, so, the above in mind, I stumbled across something interesting last night while looking for a completely different thing on Marketplace (I was looking for an SL version of a Bobo Shanti turban hat for a female Rastafarian – which is NOT those touristy hippie tam things – and that somehow landed me here after giving up on search and just hitting a few random browse buttons.)

Someone on marketplace is attempting to develop a standard sizes fashion system. 5 sizes based on these dimensions:

Body Fat: 11
Torso Muscle: 38
Breast Size: 58
Love Handles: 31
Belly Size: 6
Leg Muscle: 56
Butt Size: 44
Saddle Bags: 36

Body Fat: 7
Torso Muscle: 35
Breast Size: 52
Love Handles: 28
Belly Size: 3
Leg Muscle: 50
Butt Size: 38
Saddle Bags: 33

The ‘package’ is a freebie with 5 shapes, that I assume have those numbers alongside the other values – the notion being that you just need to stay close to the dimensions here and any mesh clothes made for it will fit you.

It also comes with a long notecard explaining the objectives of the project, which seems a very noble goal. Some anticipated concerns are answered, such as ‘what if my shape doesn’t match these numbers’.

Finally, it has a download link to a zip file of some mesh objects – bodies to load into a 3D application to use for building clothes to match.
– I need to learn how one takes the SL avatar mesh out of SL and into a 3D app, to see how this part was done.

– The problem is they have 3 sizes under M, and only 1 above it; which is way above it. Some of the reviews on Marketplace have slammed the project for this. I would say it needs at least 1 more size in between M and L.

Note how the dimensions are basically bust-waist-hips related. Not height. Also not ‘Thickness’. That’s because mesh doesn’t care about height or limb length or avatar thickness; it rescales to those automatically, but not to these other things.

But it is missing cleavage which is extremely important. That said, its something that in RL adjusts depending on what you wear anyway, so no one has any business complaining a clothing item fails to match their cleavage setting, unless they’ve never seen a real woman’s body. 😉
– Honestly no ‘breast’ setting should seen as something to get attached to. Its the nature of wearing or not wearing clothes that it ‘mashes up’ the soft tissue there. Folks just need to get used to changing the dials for that part of the body to fit their outfit, because anytime you put something on in RL, that’s exactly what happens to you.

The standard sizes ‘team’ claims to have surveyed 500 female avatars, but I see no link to a blog or to the survey for it to be checked for accuracy. While 500 is a valid sample size (anyone who has actually studied polling at a university level knows this) – the question that is more important is how were they selected… Again, never saw any forum post or group notice or blogs announcing it. But maybe I just missed them. So the product should link to them.

Curiously M and S are very similar to my own dimensions for those parts.


Body Fat: 3
Torso Muscle: 37
Breast SIze: 50
Love Handles: 29
Belly Size: 0
Leg Muscles: 50
Butt Size: 45
Saddle Bags: 30
– So I’m going to test some items made with this system and see how well it works for me.

I’m also wondering now if my avatar has a bit of a ‘bubble butt’ on other people’s screens… 🙂

Almost every shape setting not listed above (again except for cleavage) DOES NOT MATTER for mesh. This means two people can be the same “size” on this system but actually have completely different ‘shapes’.

I’m also of the opinion that for standard sizes they should have used dimensions from the real life fashion industry for some top brands, and not SL. Second Life avatars are notorious for being deformed monstrosities… 🙂 Something to help people look better would be good.

Something like perhaps, a Proportion guide, if not fashions with sizes based on the variety real life fashion designers contend with.

Addendum from a comment I left on the nwn blog:
SL of a year from now will likely be something very alien to the SL of today. Mesh frees up a lot of past limits.

Right now folks are paying most of their attention to clothes and avatars, because that’s “sexy”.

Things like tears, ‘male bits’, hair, feet, tails, ears, and so on will bring dramatic changes even for ‘non-furries’.

But the big shock is going to come when things that you rez on the land get a serious looking at – since those need to consider prim cost, and the prim cost of a mesh is set by its -SIZE- when rezzed.

The smaller you make it, the less prims it uses.

And that will cycle us right back into discussions about things for avatars: since mesh items worn on avatars -AUTO- scale up or down regardless of avatar scale – the need to be the size the item was made for now no longer exists (you just need to have the bust-waist-hips it was made for).

– And that too will go away if we either adopt some standards like this project, or Qarl’s code makes it in. Hopefully Qarl’s code will make it in.

But do get ready to make your avatar smaller… because pretty soon that’s going to be coming up as a hot topic again when this discovery of how size affects prim cost becomes more widespread. I predict lots of flame wars, a few accusations of ‘techno-communist-child-avatar-agendas’ and other irrational insanity… but once people realize they can wear all their mesh clothes on any size, many will likely just go with the size that lets them have the most prims on their land – not for the reasons of scale so often advocated in past, but simply so they can more fully fill up things like the space of a linden home.

4 Comments (+add yours?)

  1. Shug Maitland
    Feb 23, 2012 @ 10:11:33

    With a little luck the whole thing will become moot when the Parametric Deformer is fully functional and implemented. Mesh clothing will just fit if the creator built to the default avatar. For now I am not buying mesh clothing, I can wait for things to settle out.


    • Pussycat Catnap
      Feb 23, 2012 @ 17:54:26

      Yes greatly hoping that deformer hits.

      We all need to remember though that the SL fans hired Qarl, not LLs. He might make the best thing since sliced bread, and LLs could turn around and just ignore the entire project like it never happened. They’ve certainly made dumber moves…

      As for current mesh – for me it is already so much more vastly improved over non-mesh clothing that I’m not waiting, and have been shopping like crazy since I caught on to the potential and found lower cost suppliers.


  2. Bubblesort Triskaidekaphobia
    Mar 22, 2012 @ 04:51:35

    It also bothers me that they did such a large survey so quietly. I don’t buy it. Also, the default avatar shapes are all incredibly ugly. Uglier than most avatars I meet in SL. I don’t think that they were created off of any kind of survey at all.

    In my experience, mesh is invisible to a lot of people. I wore a mesh suit quite a bit for a week or two up until last week when I got tired of troubleshooting people’s viewers when they could not see me. Every time I am in a crowd of more than 5 or 6 people with my mesh suit on, at least one of them sees me as a floating head, and that is just the ones who speak up about it. I assume more people can’t see me and don’t care enough to say anything, so I assume about half of SL can not see mesh at all.

    As far as how to export avatar shapes from SL in to other programs, though, that is done all the time. Mesh clothing would be impossible without being able to do this. Learning how to do this should be a first step for anybody who wants to make mesh clothing in SL.

    There are a few ways to pull your avatar shape from SL and put it into a 3D modelling program. I don’t know how the people who made the mesh standards did it, but these are the ways I know how to do it:

    The first one is free, but it is time consuming and requires Phoenix build 1185. On the upside, you get the .obj file directly from SL and you can just work with the simple avatar mesh with this method. You don’t have to deal with any rigging or shape keys if you don’t want to (which is very useful for techniques that involve breaking the avatar apart). You can find instructions for this method here:

    The second method involves using a plugin suite for blender called Avastar (it costs L$6000 right now and the price will be going up to L$7000 soon). It involves exporting your mesh shape settings to an .xml file and importing them into blender, letting the Avastar plugins interpret the data and create the model for you to work on. It is quick and easy. I make custom mesh things for my friends by having them export their shape data to .xml then copy paste that data into a notecard and pass it to me in world, then I copy paste it to a text file and bring it into blender like that.

    Here is a tutorial on how to import with Avastar:

    The downside to Avastar import is that you don’t always want to work with a fully rigged and weighted model like what Avastar creates. I don’t mean that the rigging gets in the way visually, because that can be hidden easily enough. The big problem I run into is that the shape is created from shape keys, which I find limiting as far as what I can do with the shape, because it is not ‘real’. For some things I love Avastar, but for some things I just want the simple ‘real’ mesh, so I usually try to get the .obj files using Phoenix 1185 as well as the .xml file to feed into Avastar. Sometimes that means I have to recreate the shape with Avastar off of the .xml file then read the numbers Avastar displays in Blender to recreate a shape by hand in SL to export the shape using Pheonix 1185 (the first method). It’s time consuming but it gives me more flexibility when I have everything set up like this.

    There is a third method that I have not tried yet and it is called Wiz importer. It is L$3000, which is half the price of Avastar, and it looks interesting, but I don’t know anybody who uses it so I never tried it. Network effect is king in the whole ‘learning new tools with steep learning curves’ game:


    • Pussycat Catnap
      Mar 22, 2012 @ 07:57:31

      Now that Direct Delivery just hit, anyone -not- on a mesh viewer is going to have trouble using Marketplace… So the non-mesh crowd is going to fade very fast.

      But for the survey… yeah I’d really like to know who was surveyed where and when and how.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: