Maeve Balfour

Pamela: You'd be surprised at the potential in land impact savings, regarding separating interior meshes from exteriors. It's something I've been experimenting with a lot in my personal projects. I treat the exterior of a big structure as a totally separate "shell", and model it accordingly. Generally I opt for relatively clean, low triangle geometry (but still visually pleasing), and try to design for modular texturing (ie: have multiple usage of the same textures via efficient UV mapping / stacking etc). The exterior tends to eat most of the land impact cost, due to its long view distance. However, for the interior rooms, I deliberately design them so they are hidden from long range view, either by windows or completely enclosed by the exterior "shell mesh" walls. This is so that their LOD changes will be totally unseen (preferably) - in these cases, I simply create a very basic set of triangles for their LOD2/3/4 mesh (same mesh for each LOD) - one triangle for each material used in the interior room mesh - so if the room uses six materials, I simply create an extremely basic six-triangle flat mesh plane, with each triangle being allocated to each material. If you design your interiors carefully, chances are these LOD swaps will not be seen at all, even within the structure itself in most cases (especially if you mask out LOD swaps with other room module's walls. IF the interior room has a window to outside, I simply create opaque "glass" faces to cover the window on the inside (in the LOD2/3/4 mesh), utilising a separate material and texture for this effect - still an extremely low triangle count. Generally, this method allows me to create nicely detailed rooms, with pretty low land impact costs. A room of about 15m x 15m with a 7.5m high ceiling is generally under 1.5 land impact, depending on complexity. When linked together, you gain additional land impact savings. (I've found from my experiments that it's better to keep room modules as individual meshes and linking them together inworld. If you try uploading a whole set of rooms as one object, you will lose much of these land impact savings. It's associated with how soon the LOD swaps occur - hence a single room object will have a much lower land impact than a group of rooms in the same mesh object, due to the smaller volume and quicker LOD change). Definitely something to experiment with. Treat the exterior shell as the most expensive part in regards to land impact, and design accordingly to keep it reasonable. And then be savage with the hidden interior rooms' LODs. You should get a surprisingly low land impact like this. :matte-motes-smile:

Another thing which can really cause your land impact to soar are the LOD meshes (level of detail). If you are only just starting out with mesh building, it can be a little daunting for sure, but this is something to definitely keep in mind for future projects. Getting a little technical here, but for longer distances, a simple LOD mesh with only the exterior modeled would still give a good representation of your house, and a simple texture representing the outside details and windows (with opaque glass) would remove the need for geometry for the insides. Keep in mind that your ORIGINAL house will still have the interior etc - it's only when the camera is viewing your house at longer range that the simplified exterior-only mesh would kick in. The reduced triangle count would have to help reduce the land impact cost. (The simplified exterior texture would probably need to be used on a separate material etc, and would be revealed via material swapping on the lower LOD mesh - this is something you could explore once you have a solid grasp of mesh) - Sorry for the overly technical description there! Also a good practice for largish mesh builds is to try and keep the triangle count low - the higher the geometry detail, the greater the penalty for large meshes in regards to land impact. Looking at your building, I agree with Drongle... I would assume that your round windows and subsequent holes needed in the walls would require a lot of geometry - that is probably a major factor in your high land impact cost. The rest of your main structure appears relatively "geometry friendly" - meaning that you could create it with relatively low amounts of triangles. Your railing outside could also be expensive in triangles (hard to tell from your image). And yah, as Drongle said... a simplified mesh to describe the physics hull is highly recommended. All you would need are simplistic box shapes to describe where the solid sections are, leaving gaps for doorways. Large, simple shapes for physics are actually CHEAPER than lots of small ones - this is because large simple flat areas are easier for the physics engine to calculate. I would guess that your railings, if used as they are for a physics representation, would be severely costly. For those, a simple cubic mesh to represent their width and height would be all that you would need. My guess is that your physics for this whole structure shouldn't be more than a few PE at most, if done in this manner, and definitely NOT be the most expensive aspect of your land impact costings. (This custom physics mesh is uploaded with your main mesh via the physics tab - this physics hull is invisible to the eye, so ignore it's "ugliness" - it's purely for physical representation). NOTE: The size of your physics mesh needs to be the same bounding box size as your original mesh (the outermost extremities need to match) - this bounding box needs to match so that your original mesh and the custom physics mesh align properly. Another idea for future projects - Consider modeling the INTERIORS as separate mesh objects to the exteriors. In SL, you would simply combine the interior and exterior meshes (each having separate physics meshes of their own). This independence can potentially result in serious land impact savings if planned for correctly (hidden LOD switches for interiors etc). Sorry if I overwhelmed you with information Keli - I know mesh can be pretty daunting at first, with so much to learn. But definitely DON'T give up on it - stick with it, learn as you go, and mesh can be seriously FUN and open up a vast new world of creative potential! Just a matter of knowing how to take advantage of land impact in given situations. :matte-motes-smile: EDITED TO ADD: You posted just before me, Keli - You mentioned about your railings being SCULPTIES. Did you attempt to LINK them to your mesh? If you did, that is a BIG NO-NO! If you linked the sculpties to your mesh, they are instantly converted to the mesh measurement system, and their land impact cost will SKYROCKET. Try UNLINKING them from the mesh, if this is the case. Also, the other prims you linked into the mesh (for the windows especially - the cut cylinders etc) will also be adding to the mesh's high land impact cost... as an experiment, try unlinking the mesh from everything else, and see what it's land impact ALONE is... I would reckon it would be substantially lower than what it is currently with everything linked TO it.

Have you checked how many material zones you are using, and whether your LOD meshes have the same number of materials? The fact you can upload each separately hints that there might be issues there (just a thought). Dependent on the viewer version, mismatched materials can cause crashes when using the uploader. From my own experience, when I selected LOD meshes that didn't match the main mesh in regards to the number of materials used (due to oversights in my modeling), it would bomb me out to the desktop every time. Possibly when you combined your two meshes (shirt and skirt) into one, it's done something with your materials. I'd suggest checking in that direction - see if the total number of materials is eight or below, and that your LOD meshes have the same number of materials. If you have MORE than eight materials in the combined mesh, that could be your problem as well. To try and narrow down the error, I would also suggest uploading the mesh as a pure, unrigged version on the Aditi test grid - just in case it's the rigging causing the problem. Don't give up though - mesh can be a pain when it doesn't want to work, but it's well worth persevering with! :matte-motes-smile:

Drongle: Yah, I would use proper baked ambient occlusion for complex geometry, definitely. However, in my current project, where I am creating simplistic mesh modules to combine into a much larger, complex structure, I personally find it quicker to fake the shadowing inside of Photoshop (just my own preference). The concept I am working on (which I mentioned in a different thread a while back) involves creating an inventory of simplistic, modular mesh shapes (low triangle count), designed to be pre-UV mapped and utilising combinations of texture segments to mix up their appearance. The texture sets are heavily packed with a variety of UV-texture pieces (with texture variations for each piece), all with this Photoshop shadow effect, to try and give a nice consistent ambient occlusion effect throughout. Thus far, it's working nicely - with a total texture palette of about a dozen 1024x1024 images (slightly more than desired, but still acceptable for performance), I'm managing to create a large, sprawling structure via repeated usage of mesh modules, without the dreaded obvious texture tiling. The repeated textures are still there, of course, but the structure overall retains a nice aesthetic, giving the illusion of more detail than is really there. Will post some screenshots when I have something ready. :matte-motes-smile:

Pamela: In regards to wooden flooring, the method I described should (in theory) be able to handle it, with a little bit of creative compromise. I would suggest say, one texture (and material) for parquetry tiling around sections of floor where ambient occlusion shadows would be needed (utilising a faked shadow effect on the relevant edges). This parquetry texture could be repeated without it looking too obvious, since by nature it is designed to tile repetitively in RL. For the main flooring areas of the mesh, I would separate that into a different material space - which can hold an entirely different wooden flooring texture. This would preferably be a seamless texture (both horizontally and vertically). In this material zone of the mesh, inworld via the SL texture tools (build menu) you could repeat the wooden floor texture as required to fill the space and keep the image nice and sharp. Due to this being a separate material, the repeats won't affect the parquetry tiling around the perimeters, thus preserving the ambient occlusion shadow. So BOTH the parquetry tiling (with shadows) AND the main wooden sections of floor will be sharply defined. For your proposed light effects... I think a separate mesh overlay would do the trick (possibly in the same mesh object, separated on another material). You could create a series of simple mesh rectangular shapes, with a simple bright colour (or texture with alpha) combined with glow and transparency to get your desired lighting effects. Not the only way necessarily to achieve the result - just the method I would utilise. I hope this helps a little for you. :matte-motes-smile:

Pamela Galli said: Soooo -- no one has any ideas about how to texture a large floor to incorporate baked shadows? ---------------------------- Pamela: I've been experimenting lately with "faked" AO (ambient occlusion) shadow effects, via usage of the same texture space as the original base image. My reasoning is that I prefer to avoid shadow overlays as much as possible with main surface areas - to avoid that flickering effect that often happens, plus it will allow smaller objects WITH shadow planes to sit atop of these surfaces without "alpha fighting". I've included an example image below, explaining my method. This technique, when used over large surface areas, tends to work best with a tile pattern - it just makes it easier to hide the UV joins. However, with a bit of imaginative work, it would be possible to create non-tile patterns, although it would obviously require a lot more work to get seamless results. In Photoshop (or whichever program you prefer), I create a grid pattern - in this case, a 10x10 grid. I use this to create the "grout" setup. In a separate, underlying layer, I create the base texture (stone in this case), using the grid to select random tile segments to rotate, adjust tint etc - to randomise it a bit. Using a multiply layer on top, I simply used a few gradient fills to create the faked AO shadows in all the combinations I would need (wall sides, inner corner and outer corner). For UV mapping it takes a bit of work, but it's fairly simple (being a flat plane mesh). The mesh floor in this example was designed to MATCH the pre-made gridded floor texture. During the mesh construction, I divided the floor up to match the grout lines - the outer edges where the faked AO is, the mesh has seams running down the equivalent grout lines. Each corner piece is isolated, so I can use the single relevant texture AO corner tile (inner or outer). The main floor spaces (and sections of the AO edges) I divided into segments - generally in multiples of 6x6 tiles for the main floor, and between 5x1 and 7x1 tiles for AO edges (depending on wall length) - so I could randomise the tiling and avoid obvious texture repetition. The UV seams were adjusted to suit the texture, and the UV islands were mixed (randomised locations in the UV space, rotated, flipped etc) to achieve the final result. The end result is fairly effective - not true ambient occlusion, of course, but a reasonable representation of it. The texture (1000 pixels x 1000 pixels) is on a ratio of 100 pixels per 1 metre inworld (each tile is 1m x 1m), so it remains nice and sharp in comparison to AVs standing on the floor mesh. The extra geometry needed to allow the effect isn't overly costly - this mesh's floor area inworld is about 22.5m x 30m (bounding box area), yet its Land Impact was barely 1 from memory (physics plane included, and using aggressive LODs (since the room where this is used will be hidden from outside view). For the visual result and efficient texture usage, I think the Land Impact cost is quite acceptable in most cases, dependent on LOD techniques used (I have reused this texture in many rooms in the same build, all with different shapes and sizes of floor area). Hopefully this example might give you ideas with your workflow, Pamela. :matte-motes-smile:

Faces which cross THROUGH each other (which is what I assume you mean here) isn't a problem at all - I do it all the time with my SL meshes. It definitely helps reduce triangle counts, and I use this method to eliminate duplicate faces wherever possible (texturing intentions allowing). As a simple example, if I have a single mesh object consisting of a row of windows in a long wall (with the windows separated by pieces of wall between each instance), I will stretch a single long pane of glass between all of them as a single shape (two triangles per side of the glass - four triangles in total). If I plan it carefully, and space the windows evenly, I can make use of a repeating window texture (with blank texture space where the non-visible pieces of glass will be). The glass section of the mesh itself will be narrower than the wall sections, so it will be hidden within the wall cavities where applicable. This example (wall section with windows and glass) I would upload as a single mesh object (with suitable LODs), hence the importance of reducing the vertice count. Compared to having individual glass pieces PER WINDOW, there can be a considerable saving - of course this depends on the layout etc, since nicely spaced windows might not be what is intended. But yah, definitely utilise this to the maximum - you'd be surprised how many vertices you can reduce by eliminating duplicates. The only potential issue might be Z-fighting in rare circumstances (flickering faces, but this would only be for faces occupying the exact same space where they cross over, and is easily worked around with vertice shifts). :matte-motes-smile:

Alisha: Yah, this is something I am quietly watching with anticipation too. So many, many ideas come to mind at its possibilities - and judging from the video on the user dashboard page, it should be relatively easy to script for too (even for script luddites like myself!). Just things like swimming underwater, and having little fish schooling around, darting away from you as you get close to them etc, will make for awesome new levels of immersion (assuming they can swim up and down in the water - I'm not overly sure if that aspect is feasible; although if limited to ground/seabed level, I can envisage using inelegant "invisible prim" tactics to get around that). Definitely looking forward to playing with this, though! :matte-motes-smile:

Qarl began working on the mesh deformer project following a SL community fundraiser to help cover his RL costs while developing it. (The funds were "crowd sourced" towards the latter part of last year). This Parametric Deformer project that Qarl is working on should allow for mesh clothing to fit most avatar shapes, as the video you mention shows. As far as I know, once the project is completed Qarl will release the source code, which hopefully will be adopted by all viewers (official and third-party). If this is the case, mesh clothing will become usable on a far higher level - pretty much any shaped avatar with a human form should be able to wear any item of mesh clothing (within reasonable constraints I would assume). Still, very exciting stuff - fingers crossed the project is seen through to completion. Qarl is well regarded in the development field, so if anyone can get this to work, he can! :matte-motes-smile:

My idea... is having a mesh shape... lots of "pointy hills and valleys" vertically, facing outward... LOL, kind of hard to describe in pure text.... imagine it like a big row of: VVVVVVVVVVVVVVV (this would be vertical, the lower part of the VVVVV would be furthest away from your view, and the tops would be the pointy bits facing you... like a piece of paper folded and facing you (LOL, this is hard to describe!). This mesh shape would then be sandwiched inside a transparent flat panel, to give it a flat glass effect (either a mesh or a prim). Add some glow to the VVVVVVVV mesh and transparency, and I would guess it would achieve the effect. Not necessarily the same method used at the club though, but an idea for people to play with if they want to try it out. Still.. it's always fun dissecting things to figure out how to be creative. :matte-motes-smile:

Just wondering... Is it possible it was done with mesh? I reckon (but haven't tried it yet), that this effect could be achieved by taking advantage of the way that mesh faces work. (Individual faces on a mesh (and prim for that matter) can only be visible on one side). Anyways... in theory... if the "prism" effect inside the glass has various faces in a mesh set at different angles, ideally with transparency and a little bit of glow, I would assume that as your camera moves past / around it... individual faces would move in and out of camera line-of-sight, and pop in and out of visibility (I hope I worded that so it makes sense!). The same effect could be achieved with prims, but would be very expensive. With mesh, since you can create a shape like this in a single object, it would be quite feasible I think. Not entirely certain, but I reckon that's how it could be done. Would make for a fun experiment anyway. (Sidenote: In a mesh, you can allocate up to eight different texture faces via materials... and unlike faces on a prim, individual faces in a mesh don't have to be in a single continuous space - they can be scattered across the object and still have the same properties of a single face that a prim face would have. This could also be used to enhance the prism effect). :matte-motes-smile:

Chosen: I've been quietly following this thread after my initial couple of posts. I'd like to thank you for your valuable and extensive knowledge in regards to copyright laws / protections etc. Much of it I already knew and understood (mostly good old commonsense, really), but after reading your replies throughout this thread, I have also learned quite a bit more, especially the deeper, more detailed aspects of it all. I sincerely appreciate your input. Thank you for your insights. :matte-motes-smile:

If you can see a recognisable texture, I would assume the UV map you created is functional. Have you tried rotating the texture inworld, via the build menu (texture section)? I know for my own work, depending on my method of UV mapping (in Hexagon), the texture is sometimes rotated by 90-degrees in SL compared to what I had on my screen in Hexagon. A long shot, but worth trying. :matte-motes-smile:

A statement along those lines would definitely be of help... but I assume it would only be reliable IF such statements were verified somehow. If it was just based on an honesty system (such as a tickbox to state that you have permission etc), I imagine it would be abused by those who simply don't care - just like uploaders of stolen mesh blatantly choose to ignore their obligations despite having done the mesh upload quiz. I guess from a legal standpoint, it would be a bit of extra ammo usable against thieves, since they would have knowingly stated that they have "permission" to upload when they do not.... so if their actions were legally acted upon, I guess they wouldn't have a leg to stand on. It's a pity that often it's the customer who gets burned in the scheme of things, when they buy in good faith. Buyer beware, of course, but still... it's such a minefield out there. :matte-motes-sour:

Chosen: Yah, that's probably the best approach (notifying the rightful owner), and I have done that on a couple of occasions when I have spotted stolen DAZ content. And I can understand the potential quagmire with LL trying to police things (owner permissions etc). Still, it's a really irksome topic. The sheer brazen stupidity of examples I am seeing lately is gobsmacking. I'm not wishing to be a vigilante, ever. Just trying to protect LL, SL, and ourselves from the idiocy of the few. You're right though - that the owners of the IP are the best to deal with it. I guess going to the relevant websites of the IP owners, and contacting them from there, would be the best approach? A bit of legwork, but worth it in the long run.

This is a topic I hesitate to bring up, but I feel it's something that needs discussing.... that of ripped / stolen game meshes. I browse through the Marketplace several times daily, searching via the keyword "Mesh" and sorting by "Age: Newest First". Mostly because I just love looking at what people create in general, and also as an occasional buyer myself. However, every now and then, up pops something (usually an avatar replacement), which is either a mesh ripped directly from a game, or a very blatant intellectual property breach (or both, if the merchant is really stupid - labelling their avatar mesh WITH the name of the actual game character, for example). I know this thread is technically suited to the merchant forum or other.... but... for us as mesh creators, who are often familiar with content in general (from outside of SL - games related, Turbosquid, DAZ, Poser etc), we can often spot a mesh rip from a mile off.... BUT we have no clear avenue to report these issues. I know that copyright breaches can only be reported by the actual IP owners themselves... which I guess is there to prevent nasty actions from rival merchants in general.... but, at the same time, it prevents us from responsibly reporting blatant breaches. So my question is... For us as responsible mesh creators, who can often easily spot blatant (or suspected) mesh rip-offs on the Marketplace... what is our best course of action? Just ignore it? Or is there an avenue we can use to alert LL directly? (For the record - My only motive in this issue is to give LL a heads-up when I spot things like this, as they cannot possibly monitor everything with their limited resources - our eyes can help with keeping a lid on this kind of illegal activity). Thoughts?

Margaret: I'm not a Blender user myself, but there are plenty of expert users who frequent the forums here - so a wealth of knowledge is on tap relating to specific SL questions you may have. If you are just starting out with Blender as a complete newbie to mesh in general, I would strongly suggest learning general mesh creation before trying to create specifically for SL itself - that way, you will get a proper understanding of the concepts involved with mesh creation overall. It's a little bit of an initial learning curve, but worth persevering with as it will avoid a lot of headaches long-term (trust me, when I first started making meshes several years ago for generic personal usage (non-SL), I didn't learn the basics properly, and it caused me a world of headaches for a long while after LOL!). THIS could be a possibly useful starting point for you to learn Blender 2.5x onward (I think it's the version with the more up-to-date user interface (more user friendly)). From appearances, it will take complete newbies to 3D through all the aspects of Blender (reading through the Syllabus dropdown menu). I personally have it bookmarked for when I will eventually need to pick up and start using Blender myself (I am primarily a Hexagon user at the moment). In regards to your question with UVMapper - I am pretty certain it will be fully compatible with any meshes you create for SL. HOWEVER, be sure to give the demo a proper testing prior to purchasing the full version of the software (just in case). Also, be sure to explore UV-mapping options in Blender as well - since it's free. Being a stand-alone, UV-mapping ONLY program (from what I can tell), you would need to export your pre-made mesh as an OBJ file (or compatible format) FROM your mesh modeling program TO UVMapper, where you would then UV-map your mesh to accept a texture. A simple re-export as a new OBJ file would allow you to work further with your mesh in your preferred mesh editing program (material mapping, rigging etc), prior to final export as a SL compatible DAE file. In general, it is quite often handy to utilise more than one program in your overall mesh project workflow - Often, specific programs will have more powerful or easier to use functions than others, so often it can be logical to jump from one to another. So if, in your case, you prefer to UV-map with UVMapper, then go for it! I hope this helps - have fun! Mesh is a great addiction! :matte-motes-smile:

(Just a quick reply (typing this on my work break!)) With physics for mesh, especially for a complex organic shape such as this example, it is highly recommended you create a separate, low triangle count mesh to use as physics hull. This mesh is selected in the physics tab of the uploader menu. All that this mesh does is simulate interactive physics - it isn't visible at all. A physics mesh (often referred to as a physics hull) really only needs to describe basic interactivity for AVs... such as boxes or similar primitive shapes to represent the legs of your mesh, for example. Ideally, if you keep your physics mesh very simple in triangle count, with LARGE triangles instead of lots of tiny ones, the physics cost will actually be very low - probably less impact than the overall land impact of your original mesh. Basically - The more complicated a PHYSICS MESH is, the higher its associated Land Impact cost - this is purely due to how difficult or easy the shape is for the SL physics engine to work with. Large, flat surfaces with few triangles are actually MORE EFFICIENT than complicated surfaces with lots of small triangles. Also, the BOUNDING BOX (the total volume) of your physics mesh should match the bounding box of your original mesh (so that it aligns correctly, when the uploader combines the main mesh and the physics mesh). I am pretty sure that with a relatively simple, low triangle count PHYSICS mesh, you should retain your original Land Impact cost of your main mesh object. :matte-motes-smile:

I fully agree with your statement, Jennifer. Alpha clothing layers SHOULD ALWAYS be fully mod/copy in my opinion. This is especially the case where multiple alphas are required if you wish to wear more than one mesh item of clothing (this applies with sculptie clothing as well). The ability to actually combine several alpha masks together via Photoshop would be a huge help when creating customised outfits for one's self, and tweaking the alpha edges if necessary. If the day comes that I actually decide to become a merchant with my mesh clothing (I only create things for myself for now), ALL my alpha clothing layers will be full perms - it will allow full customisation by customers. :matte-motes-smile:

Wow Kiyo, that's turned out really nice! :matte-motes-smile: Getting a bit off topic: Looking at its geometry, I'm wondering what kind of Land Impact cost it has. Is it pretty high? Reason I ask is due to the combination of its size AND the amount of complicated geometry (triangle count) I think I can see there. (Mostly the rounded columns on the bottom floor, the curvey / rounded shapes in the decorative pieces, and the windows in the roof area (the ones with the rounded tops and three-step insets). Something worth considering, to bring down the Land Impact cost (assuming it is pretty high), is to reduce the geometry to relatively simple shapes where the triangle counts are expensive - and use your baked texture as shown here to "fill in" the missing geometry detailing. It's a balancing act, of course, between visual look and Land Impact cost, but definitely worth exploring. Still, that's a damned fine piece of modeling work there Kiyo! :matte-motes-smile:

By the looks of this, I would tend to think it's primarily designed for high-end 3D mesh work (high-polygon meshes), specifically because of the way it generates folds and creases in the clothing mesh when it's draped onto the target figure. The amount of mesh resolution would have to be very high for it to work efficiently. I would have serious doubts that it would be suited for the low-polygon, realtime environment of SL - at least, not without some heavy duty optimisation of the resultant meshes it creates. However, it could possibly be useful for creating realistic meshes AS A STARTING POINT for SL optimisation, especially for baking out shadow maps etc (for usage on much lower resolution meshes). Again, there would still be a LOT of work required to get these meshes reduced to a usable, responsible level of detail for SL. Considering the high price tag attached (if you planned on selling your creations, you would have to choose the Small Business Licence to be legal - US$699 ouch)... and the large amount of work you would most likely still be needing to do to get these meshes usable for SL.... I wouldn't bother with it. If you were able to get these meshes properly optimised for SL, it would mean that you already possess the relevant mesh making skills to create these kind of clothing meshes from scratch anyway, which would pretty much negate the need for the big spend on this software (ie: Creating relevant low-triangle count meshes suited for SL). The software definitely LOOKS LOVELY for sure... and definitely would be handy for some high quality results. However, the cost factor and work required I think negates the usefulness of it overall. "Magical" 3D software like this is rarely a one-step solution - merely another tool in a meshmaker's toolbox. Not wanting to rain on anyone's parade here... but I think it only fair to point out the potential limitations of this. Personally, especially for users who don't know how to optimise meshes for SL (or how to create mesh at all), I would NOT recommend this at all. Not for the cost of it combined with the amount of work you would still need to do. IF, however, this DOES have low-polygon conversion capabilities, I would reconsider my assessment. Just that as far as I can tell on the surface, it doesn't appear to do so). I hope my input helps. :matte-motes-smile: EDIT: Going by Gaia's post (below), it sounds as though Marvelous Designer DOES have low poly capabilities, so I stand corrected on that aspect. However, it's still pretty pricey for what it does, considering (again, relating to Gaia's post) it appears you will still have to do the weight-painting for rigging manually in a separate program. So... nice software, for sure, if top tier results are what you are after... but a high price to pay in order to achieve it, not to mention the work still required.

In regards to the Aditi test grid, as far as I know, you will probably need to redo your mesh quiz before you will be allowed to upload mesh there. It's not a problem though - just a quirk with how your account settings work on the separate grid or something. The L$ balance you have there is purely for upload testing usage (figuring out costings etc) - it's definitely not transferrable to the main grid. However, it's excellent for helping you figure out relevant upload costs. If you run short of upload L$ on the test grid, a Linden employee will generally give you a quick top-up at no cost. When you arrive on the Aditi Grid, just to a world search for "sandbox" to find a location to build in (the default landing point from memory is a non-sandbox area). :matte-motes-smile:

I'm not a Blender user, so I can't really help much in that regard... Your UV-mapping process sounds logical, especially since you say it renders fine within Blender. Mayhaps it's a quirk brought about by switching of Blender versions? ("Clutching at straws" guess there). Or mayhaps the problem materials aren't saved out properly for the export? With your mesh rezzed INSIDE of SL... have you checked to see if all of the materials will accept random textures? (Your post doesn't mention if you have done this or not). To try and narrow down the problem, I would suggest just dragging any kind of texture onto each material face, just to see if the UV mapping is functioning for each. The black that you mention kind of sounds like either borked AO textures, or possibly non functioning UVs. (Generally, if you try to apply a texture to a non-UV mapped material/mesh, you will just get a flat colour (which I guess is an averaged colour for the entire image - I have seen anything from black to browns with my own failed meshes)). If your material faces all show recognisable textures (dependent on where their UVs fall on the image), it could mean issues with the Blender baking. If you just get a big colour smear instead... that to me suggests problems with the UV-mapping process. I hope this helps to at least error-trap the problem for you. (Be sure to use the Aditi test grid for tweaking/testing/error trapping work - will save a lot on upload costs in the long run. You probably already know of the test grid, but I thought it prudent to mention it all the same). (A Blender expert will no doubt jump in here to assist you further... Drongle, you there?) :matte-motes-wink: :matte-motes-smile:

I know that meshes and textures with alpha channels don't tend to mix very well - although for flat things like windows/shadows requiring textures with alphas in the SAME mesh, the alpha glitching can be kept isolated via keeping said textures to a separate material. However, for meshes which are more than single flat planes (where you get surfaces "crossing over" depending on the angle of view), any kind of alpha tends to result in glitching. About the only suggestion I can think of at the moment would be to incorporate your cornea shape into the actual head meshes themselves (for each iteration of blinking / emotions), and isolate its associated texture with alpha channel to its own material. By doing this, the alpha channel of the cornea won't affect the rest of the head mesh, and HOPEFULLY will still have the desired effect. :matte-motes-smile: EDIT: Helium's reply (posted while I was typing mine) probably makes my suggestion irrelevant (which is a good thing, mind you!) :matte-motes-smile:

Eloise: Regarding mesh clothing fitting AV shapes, there is currently a third-party project in development (parametric deformation) which, when completed, should allow mesh clothing to be adjusted to fit most AV shapes (if not all). Once that is released, mesh clothing will be vastly more usable to a far wider range of residents. (As far as I know, this will be released as open source or similar, which will allow for it to be incorporated into various viewers (LL and third party) by their relevant developers). For mesh newbies who would like to learn the processes involved in mesh creation, the mesh forum is a great starting point - plenty of people there are happy to help others (myself included). It's a steep learning curve initially, but with a bit of helpful guidance, it's something residents can achieve (and is definitely worthwhile for the creative freedom it brings). :matte-motes-smile: