Drongle McMahon

"The laws are written to only benefit those which already have wealth." And their lawyers! How much do you think Apple and Samsung have spent on legal cost in the last few years?

"...the folks that are older and know to reset their LODs to 4..." As well as the performance issue Arton mentions, there is the question of what these super-users are doing with their content. If it's in their private space, OK, but if it's somewhere they care about enjoyment by other users, then they may not be getting what they thought. Unless they inspect at lower settings, they may be surprised to learn that their visitors are seeing a degraded environment. It is one of the best features of mesh that you can make good lowest LOD models and stll get very low LI, using the billdoard method for example. As you say, it depends on how discerning your customers are. I would generally not want to buy something until I had inspected it's lower LOD appearance. Doing good low LODs does cost effort, but not necessarily LI. Perhaps the best plan is to offer different versions.

Tool panel->Object tools->Shading->[smooth][Flat] {affects whole object in Object mode, selected faces in Edit mode} Edit mode->Menu->Mesh->Faces->Shade smooth/Shade flat {selected faces} CtrlFA CtrlFH {selected faces} Note that you can use the Edge split modifier to make some edges sharp while rest is smooth shaded. This can either use the edge angle or edges explicitly marked sharp, or both. The angle option is a bit like generate normals in the uploader, but superior because it ignores uv seams.

Draw distance per se doesn't affect the LOD switch distances. Both parameters are affected by the graphics quality, but they are otherwise independent. The quality setting and the object detail slider change the debug setting RenderVolumeLODFactor (rvlf) which does control the LOD switch distance directly. By default it is 1.125 except for Ultra graphics where it is 2.0. Many people set rvlf to 4.0, to mitigate the effects of ugly sculpty LODs, but the safe thing to do is to assume that most viewers are using 1.125. You can change it using the Advanced->Show debug settings menu item. You can a;so change the draw distance that way (called RenderFarClip if I remember correctly). The switch distances for an object are proportional to the object "radius", which is the half-diagonal of the bounding box. It can be calculatede as r = sqrt(x*x + y*y + z*z)/2, where x, y and z are the dimensions of the object in meters. The first switch, high to medium LOD, is at: rvlf * r/0.24; the next at 4 times that distance, and the last at 8 times. For objects with major dimensions of 1m, at rvlf=1.125, these are about 4, 16 and 32 m for a cube, 3.3, 13 and 27 for a thin plate, and 2.5, 9 and 19 for a narrow rod. Just multiply these by the scale factor for larger or smaller objects. Multiply bu rvlf/1.125 for different settings of that parameter. In fact there are other small adjustments that affect these distances, but the above is still a reasonable estimate. There is also a delaying effect that causes the distances to appear different depending on whether the camera is approaching or receding from the object.

If it has a large bounding box, then there must surely be something there. Try going to edit mode, boundary select the vertices you expect, then invert the selection. That should reveal the extra stuff. Press AltH first, in case it's hidden!

Yes, if you want to move them relative to each other, then they have to be separate meshes, but you can still combine some, such as a hinge and one of the arms it's attached to - so that the tiny hinge won't switch too soon. When you are mixing things of very different sizes, then you can compensate to some extent by using more "use LOD avove" for the smaller ones. Still, the most important thing you can do is not to use the automatic LODs. Making your own lets you control exactly what happens. This is especially true if you are uploading the parts together. I don't follow what you are sating about iusing less detail on smaller meshes. For meshes less than 1m, the "lowest" LOD has by far the greatest effect on the LI. You can have plenty of detail at the higher LODs without incurring high download costs as long as the lowest is very simple. What kind of physics are you using? The physics costs of small objects can get very high if it is triangle-based (specified but not "Analyzed"). You can see if its' the physics cost that's making the high LI by using the "More Info" link.

And/or the dae file.

"SL does not care about the Origin" Not changed as far as I know. More likely this is accidental inclusion of some extra geometry, that may be invisible if it's facing the wrong way? A careful inspection should reveal that if it's there. Deriv - Yes. It can get very confusing with object origins wandering all over the place disconnected from their objects!

"This is where the ToS covers their backside." It will be interesting to see how well that holds up when they start selling the plethora skyrim content to other games developers on Desura!

For your 0.625 cube, the expected LOD switches are about 2.25 (hi->med) , 9 (med->low) and 18 (low->lowest) metres. If you use auto LOD, the first switch will be to a triangle (per material), which looks like what you are seeing. There is no need for using LODs at all for a sinple cube which has LI = 1 (0.5 if linked) even at 64m. So for that you should set "Use LOD above" for all LODs. More generally, the autoLOD generator is unsatisfactory for anything but organic and other highly subdivided models. It doesn' know what the object is supposed to be and will susually make inappropriate choices in geometry reduction. You are always better off making your own LOD meshes. It looks as if you are uploading the lamp as a collection of objects. If that is the case, it ends up as a linkset in which each linked object behaves independently as far as LOD is concerned. As they are different sizes, they switch at different distances, and for the smaller ones that can be very closer than you expect from the overall size. Furthermore, the LOD generator works on the triangle count of the whole scene, so that some parts might get completely decimated while others are left alone. You can get much more consistent LOD bahavior by joining the objects into a single mesh before uploading. Since it will thejn behave as a single object with the overall combined size, it will all swutch at once. However, the increased size will mean an increase in LI for the same geometry at each LOD as you used with the separate objects. You can control this by selectively removing the small details in explicitly made LOD meshes. As I said in the other thread, using autoLOD with zero triangle counts in the low and/or lowest LODs is a bad idea unless you are sure the object is never going to be seen from beyong the relevant LOD switch distances, or unless your objective is to mak it disappear.

"There is also a strange issue with certain meshes, which will switch to lower LODs ealier than others." Was that never fixed? I can't see the jira - it was removed from the public list (link->"forbidden"!). I'll ask at the cc/mesh group.

It is tempting to use a lot of subdivision in cloth to get the detailed folding you need. However, there are limits (see below) to what can be uploaded, and the LI will become excessive far below those limits. So if you do use lots of subdivision for this, you will have to reduce the poly count afterwards by simplifying the geometry without ruining the folds. That's hard work. (Don't for get to do the UV mapping before the cloth simulation and then use only UV preserving edits. Otherwise it's even worse.) People also use subdivision to get smooth shading, as I suspect you did with your pillows. This is a big mistake causing unacceptable LIs. Use the minimum subdivision to get the basic shape. Then use smooth shading (Mesh->Faces->Shade smooth). You can use the edge-split modifier to keep some edges sharp. Have a look at message 17 in this thread. The blanket on that bed (LI=1) was made with the Blender cloth simulator and then manually decimated. ETA - Now that we have materials, you can add the finer fold detail by baking a normal map from a highly subdivided cloth simulation onto your lower poly mesh that has only the major folds. This will be very efficient for LI, and avoids the tedious poly reduction step. Of course you have to learn about normal maps and baking..... (Technical) There is a (bug that makes) a maximum triangle count per material of 21844. When the count exceeds this, the viewer code secretly makes a new material. When you use subdivision, the triangles are presented to the uploader in an interspersed order. This creates the sort of patterning you are seeing on your blanket. It also causes large LI increases (if you are using smooth shading - otherwise it's already high) because the vertices have to be duplicated in multiple material lists. When the secret material count goes above 8, the uploader carries on, but the extra materials are dropped from the upload. This leaves holes in your mesh where the 9th and further materials were. This means there is an absolute maximum of 174752 triangles that will get uploaded. The effective limit can be lower if some materials use less than their 21844 triangle maximum.

The LOD your camera sees is determined (mainly) by the ratio of the distance from the object to the "radius" ® of the object, multiplied by the setting of RenederVolumeLODFactor. In this context, "radius" means half the diagonal of the bounding box. (This is sqrt(x^2 + y^2 + z^2)/2, where x, y and z are the object dimensions). If RenderVolumeLODFactor is 1, the distances of the switches are about r/0.24, r/0.06 and r/0.03 (there are other factors that change that a bit). The distances are mulyiplied by RenderVolumeLODFactor, which defaults to 2.0 at high graphics or 1.25 at any lower setting. For a 1m cube, r=0.87m. So at high graphics, the distances are about 7m, 30m and 60m. At lower graphics settings, they are 4.5m, 18m and 36m. So if your object is small, you will see the effects of the LOD switch at quite smal distances (unless you use a higher RenderVolumeLODFactor). What the effect is depends on how the LOD meshes are made. If they use the single triangle trick to minimise LI, then they mau disappear entirely if viewed from certain angles. That is one reason why that trick is a bad idea unless the object is inside a building and will never be seen at medium LOD even by people with low graphics settings. I don't know that any of that is relevant in your case, but put it out there in case it may be.

"A cube mesh at 3m size will cost you 8-10 Land Impact" You must be doing something wrong. A mesh cube, complete with six independently texturable faces, and using the same cube at all four LODs is 1 LI whatever its size (download weight 0.1 even at 64x64x64, physics weight 0.1 or 0,4 if analysed, server weight 0.5). How many triangles/vertices does your cube have in the upload dialog? It should be 12/24. If uts any more, you probably have some uneccesary subdivision or something. What does the wireframe view look like? (Develop->Rendering->Wireframe).

As I don't make clothes, you can probably ignore this. However, one of the best ways of describing a UV map is to compare it with the paper pattern used in dressmaking. The flat pieces of cloth (texture) become the 3D shape of the final garment. So it's possible that the best way to make the UV map for this would be to use just those shapes in the UV map, with the same seams, that the dressmaker would use in the paper pattern - I guess four to six pieces around the skirt, each narrower at the top. That might help with the general texture fitting. The seams would be just like the seams in the real object. It won't deal with the gathering at the waist though, or with other details that aren't in the geometry. That you would still have to do with careful texture editing or further seams and distortion of the UV map. If you want exact detail, then you could model the detail with accurate folding of the pattern pieces, texture that, then bake the texture and a normal map onto your lower poly model.

Bridge7_final_best_last_optimised_evenbetter_thisisit_noreallythisoneisbest_omega_zee.dae

You need as many objects in the physics mesh as in the visual mesh. The physics objects needs to be in the same order as the visual meshes in the respective collada files, to get the right one associated with each. In Blender, you can achieve this by suitable object naming and alphabetical ordering of objects, selectable as an export option.

"LL does not have rights to remove logo or alter the product" Why not? Don't you have to waive moral rights (including attribution) and allow them full rights to derivative works? What is to stop them altering the product then?

"They cannot remove all copies of the Renderosity texture from the SL servers, yet they are in violation of Renderosity's ToU by leaving that texture in Linden Lab's possession." Hmm. I suspect the argument might be that they did not have to agree to the ToS. By not doing so (and incidentally thereby relinquishing all access to SL), they would not be granting LL the license requested in the new ToS in respect of the problematic content. Neither would anyone else. So LL would not have been granted that license and any use they made that required it would clearly be their violation, not that of the user. On the other hand, if the user does agree to the ToS, then he is both affirming he has the right to grant the license, and granting it. The user has the choice, agree, and thus violate the Renbderosity license, or do not agree (and leave SL). That violation is therefore voluntary. It can easily be avoided by not agreeing to the ToS. I am not saying I agree with that argument, only that it can be advanced. Whether the choice given is reasonable is, of course, a completely different question.

Furthermore, the mechanisms used in software, including those used for breedable pets etc., can be patented independently of the language used. Section 2.3 of the ToS talks about IP, not just copyright. As the lawyers pointed out the "sell and resell" in that section is meaningless for copyright, but is not meaningless for patents. Patents are IP. Scripts get uploaded from the editor in the viewer. Discuss...

Projectors are only available with Advanced lighting. For residents not using this because of lower power gpus, they will beheve as traditional lights. There is a limitb of six simultaneous traditional lights allowed to be visible - the mearest six. there is no limit I know of for projected lights.You need to take these factors into account if you want your lighting to be satisfactory for both those with amd those without advanced lighting enebled.

It's not the physocs weight, so far. The download weight is higher, so you need to work on that first. Where you say it looks blocky, is that because you are using flat shading? You need to use mostly smooth shading.This has 1052 triangles. It is just rendered in Blender with colours bu no textures. The sack is maybe a bit too angular as you can see its silhouette. It should be possible to get something reasonable with about 2000 triangles. Then set the lower two LOD triangle counts to much smaller than they are by default. The uploader tries to do 1/4 steps. So it will be 2000, 500, 124, 65. Instead, use 2000, 500, 40, 8, and see how that works out. Then is the download weight is less than the physics weight. you can try making a simpler physics mesh. Just make a box that fits the rough shape of the object, but is convex, no indentations. Upload it on the physocs tab alongh with the main model, and click "Analyze". It is probably a good idea to do this anyway, because the low LOD with the low triangle count may not be appropriate.

When you rez it inworld, edit it and click the "More Info" link. That will tell you the download, physics and server weights. The highest of these becomes the LI. From what you have shown us, you have not specified any physics shape. In that case, you don't get the "Prim" physics shape type option in the edit dialog and are stuck with "Convex hull". That uses the default convex hull which is made for all meshes whether you specify a physics shape or not. If you do not, then the uploader uses the convex hull of the Low (3rd) LOD mesh. In this case, that will still have many vertices, resulting in a high physics weight. So that could be the problem. You can change that by either supplying a much simpler physics shape (a box with eight vertices would probably do - then click "Analyze"), or by setting a much lower triangle count on the low LOD. Ignoring all that, and assuming it is the download weight, you still have far too many tiangles for a mesh like this. Cutting it down drastically will reduce the count at all LODs and reduce the download weight. Make sure you use smooth shading and not excessive subdivision for gettig a smooth appearance. If you need to keep the high detail, then just reduce the triangle counts at the lowest two LODs (say, 40 and 4). Unless your object is going to be very big (>10m), these will have the greatest effect on the download weight. Note that the physics weight of the convex hull based physics shape does not vary with size, while the download weight does increase alarmingly rapidly with increasing size. So make sure you adjust the object to the intended size before looking at the weights.

As said by the others - light source is a property of the prim (mesh or other), not of a face. This is true for projected lights in the advanced lighting model as well as for the traditional omindirectional lights. The effective location of the source is always the center of the prim's bounding box. You have to take that into account to get the desired effects.

Sorry, but I can't offer any explanation here. Since physics is done on the server, I can't see why it would be treated differently for different avatars. The differential height explanation does seem a lot simpler than anything like that.