Kwakkelde Kwak

Meshes can't be flexible, You'd need two things for that, a hinging point and some sort of weight painting. There's currently no way the engine can determine how the object has to flex. This is the case for most objects, not just sculpts and meshes. Only boxes and cylinders will flex. On top of that a detailed object (a mesh will be that in 99% of the cases compared to a cylinder or box) would mean a huge load on the system. ---------- Gravity doesn't mean the weight of an object, it's what you said, the accelleration applied to it when falling. It's a simple formula, I bet the engine does simulate it. I'm not 100% sure though. Density means what it always means. The higher it is, the heavier your object (compared to its volume). So you will notice this when you apply a force to it (cars, planes etc), when you walk into it or when it hits another object.

Porky Gorky wrote: Anyone know how to achieve the same result in 3DS Max? As far as I know 3ds max doesn't understand isolated edges, that would make sense since an edge means there's more (can't have an edge without a face). You can have isolated vertices though. You can remove them in the modifier menu, under "Edit Geometry", "Remove Isolated Vertices". Make sure the object is an "Editable Mesh" or has the "Edit Mesh" modifier applied to it. You can also turn on "Vertex Ticks" under "Object Properties"(right click the object) to find them by hand.

Amras Sparta wrote: Missing faces are different with every reload, and... as curious note... with Phoenix Viewer, I can't see errors during the upload or inworld, but with the latest V3 Viewer, 2/3 attempts of upload shows any face missed. I recently read your (internet)connection could be causing this, I'm afraid I don't know where exactly, but on these forums. You might want to check your router, cables and connections. Are you using wireless? try a cable and see if it fixes your issue.

Did you clear cache from your viewer or did you really clear it? You can clear your cache by hand. I'd suggest removing your viewer first, then removing all Secondlife folders (also from the application data and local settings, this will get rid of any leftovers, but also your settings). If re-uploading doesn't work something strange is going on for sure, you could contact support...

As you noticed yourself, the viewer can't find the sculpt map. I can think of two things, the first more likely than the other. 1) Your cache is borked... clear it and let your inv load again. 2) The server lost your sculptmap... tough luck in that case probably. I hope that helps.

You could use it to your advantage, the LI bahaviour on these bugged items is very strange because one axis is ignored (or bugged at least). I don't think it's too wise to rely on bugs too often though...Definately annoying yes.

Strangely enough...that's how it currently looks yes. You could also rotate the object before exporting so it's 2x0.5x0.5 or 0.5x2x0.5. There's a jira on it which I haven't looked at for a while, if I find it I'll post the link.

We've (we = several forum dwellers) seen this before on several objects. What we could make of it was very weird, something with materials. When there are 3 or 4 materials assigned, the z value of the bounding box is ignored. Did you by any chance change those? Or did you change the rotation of the original object so the axes aren't the same? If not, maybe you can narrow this bug down some more..

Charlar Linden wrote: One note - Kwak, .... eeeh which Kwak? I didn't post here

Yes when the load is as low as this, I am tempted to add geometry in return for better texture control, we don't have to build everything out of 1 prim... You're very right this "only high LoD" building can only be used in some specific cases like interiors or very large exteriors or skyboxes. I used it for a 64 meter long hall (exterior), I never got the LoD to switch on that thing. Windows can be a problem yes, but a nice trick is making the windows go opaque in the lower LoDs, again extra geometry, in the lower LoDs even, but well worth it since you won't see the interior go poof. Windows can be seperate ofcourse, with 2 triangle imposters/billboards for the lower LoDs (plus the extra ones for materials). It's ofcourse a tradeoff, since you'd like the modules to be as complete as possible, you could still use modules, but as a linkset rather than a single object.

Only Maeve can answer it for sure, but I bet this build can do without LoD med, as long as there are some doors to make sure you can't look through several rooms from a single camera position. Afterall you won't get further away from the center of the object than about 75% of its length (and I think LoD highest will show at 4 or 5 times the object size). I also think adding floors and ceilings won't raise the landimpact. That would make modular building like this very rewarding. Edit...read your post wrong ...you said determined by the LoD med, which is in this case 8 triangles... time for lunch maybe?

I wasn't talking about getting rid of any columns... Just merging the walls "through" them. I understand the texture variation... You won't get it any lower than 2 anyway, with three pieces. Nice low poly build! Oh edit on the compomise I described in my earlier post..that won't work, it would require seperate UV islands...

Maeve, I'm not going to guess... good additional info on the lower LoDs, without them it's nearly impossible to take a good guess. But since you're on the superlowprim path, here's something you could do to lower the count some more. I see you have built your walls in a "column-wall-column-wall" kind of way. The seperate walls aren't needed are they? Just make one long wall with a texture repeat, it will cost you a tiny bit in texture efficiency, but only the parts covered by the columns. You can also make a compromise and use two pieces of wall like you are, but push the edges together and merging the vertices. This will save you two vertices each time (2 less than the one wall) and will make you able to show the entire texture. btw you aren't going to cheat the thread by altering the build and keep people guessing forever?

The problem is in order to be able to resize the object, you need mod permissions on the object. That means people can simply delete the script.

Ah in that case we agree it's the geometry and not the UV mapping that makes the difference:) Using one canvas for multiple textures has ofcourse two other advantages, small as they might be for most people in most cases... You can save upload costs by a factor "number of textures on one canvas" and all posters will load at the exact same time. The second one might also mean some savings in streaming time between servers, not sure. I can imagine it takes slightly more time to fetch and process 4 512 textures than a single 1024. btw, you could also just use the entire multi-texture canvas on all posters, then apply a repeat and offset to them. That way you can make variations in layout. This does sound rather difficult when working with more than 8 non square textures though... EDIT (on your edit) I didn't mean to accuse you of texture overkill:)...but what you describe is a possible way to use too many textures.

What you describe with the 64 meter prims can also be done with normal prims. It's the mesh geometry that should get you excited, both in primsaving and shape, not the UV mapping. For rectangular objects UV mapping has nothing more to offer than offset (which is a way of UV mapping).

I won't deny there are benefits of being able to put multiple textures on one canvas, but it's not as exciting as you describe. With normal prims we were able to do that, all you need to do is set the texture repeat to 0.5 and the offset to 0.25 for a 2x2 canvas. or 0.25 and 0.125/0.375. I think the fact you can have one sided posters is more exciting, saving a lot of prims by getting rid of unused geometry. I am a bit worried about the texture load though. If as you say one poster object as described is 5 prims (my guess is as good as yours, so let's go with that) and you assign a 1024x1024 texture to it, every 5 prims cost you 4 MB of texture memory once fully loaded. You could rezz 3000 of these objects on a full sim. 12000 MB won't fit any graphics card currently available. This is not a problem with the technique by itself, but a problem of using lots of big textures. If you settle with 512x512, the texture load is a fraction of that, but still a whopping 3000 MB, go with 256x256 and your graphics card is probably able to load everything on the sim. 256x256 won't allow a lot of posters ofcourse... If you are planning to reuse the posterprim throughout the sim (using the same texture), the problem ofcourse also gets smaller. Or you can ofcourse compensate the thing by duplicating textures for bricks and streets and whatnot on other prims.

I assume exporting to blender would work as 3ds or maybe fbx (not sure what kind of files blender supports), then exporting from there again as dae, but that sounds more timeconsuming than simply uploading the textures seperately.

Ok, here is what you can do. Make sure all materials are set to 100 illuminated (self illumination) texture your box with a multi/sub with the 6 textures apply a UVW unwrap modifier select edit in the parameters rollout make the selection mode in that new window "face sub-object mode" and select all faces click mapping on top of the window and select unfold close the window again hit 0 to bake the texture (render to texture) click "add" and select diffuse hit render now make a new multi/submaterial, again 6 materials (all duplicates with the baked texture) and apply it to your box now export the box as dae in SL upload both the dae and the one baked texture that should be it... The UV map is not at all optimised...so you should do the mapping differently, but these are the basics. Maybe there's an easier way.... My brain is a bit fried atm:) time for bed edit... personally I'd do it completely different. I would use 6 seperate materials for the different faces. (6 unique textures in SL) A box doesn't UV map very well with 6 sides so you will either have distortion or unused texture space.

A multi/sub is simply a collection or group of materials. I still don't understand what you are after:) Do you want one texture to texture your entire box, or do you want six different textures on your box, so you can seperately edit them in SL?

Swobbly Minogue wrote: So can anyone tell me how to upload the textures directly with .dae With the exporter you are using you can't. I dont know of any exporter for 3ds max that does. Failing that can anyone tell me how to export my multi/sub-object material into a single texture so I can manually apply it to the cube. The whole idea of a multi/sub is that it has multiple materials, not a single one. you should be able to turn the multi/sub into one material, but that's not what you are asking. A multi/sub has different UV maps, you can merge those into one. Maybe you can clarify a bit. You can get rid of the lighting and other effects by only including "diffuse" when you render to texture. To bake all 6 materials into 1, make two UVs, one for the source, containing the seperate materials, one for the destination, which will be one texture. You can select the channel when rendering to texture.

The SL uploader will rescale all your LoD models to the size of your highest LoD. So what you can do for the lowest LoD is creating a single triangle for every material, make them very very small and place them far apart. If you then also let those triangle face downwards (or the least looked at direction anyway), people won't ever see them. I read the uploader ignores real small triangles, so that might be something to take in mind. Either way there's no benefit in making an invisible face just for the lower LoD.

Porky Gorky wrote: So I am assuming that if your lowest LOD model is just a rectangle for example and you didn't want anyone to see it, you could put a 100% transparent texture on all the outward faces of the rectangle Am I missing something? If you don't want people to see it.... don't build it. It will save you geometry and it will save you the display penalty Medhue mentioned.

Tenly wrote: I cant figure out what I am doing wrong in the permissions. I want the purchaser to be able to resize, BUT the texture and color LOCKED just like the ones I bought. Hopefully I'm overlooking something, but I'm afraid the only way to do this is by using a resizing script and no mod permissions on the object.

Hmm, at the cost of image quality? I don't think that's what we're after.... an 80% reduction sounds very appealing though.