Aquila Kytori

As Pam said you probably need to give the other unselected faces their own material Just adding this because you seem to have alot of unnecessary faces in your mesh. I have pointed to some of them in red in the image below. If you want to remove them : select all the faces of the mesh with the keyboard shortcut key A then from the tools panel on the right ( T ) hit the Remove Doubles button.  If "Remove Doubles " doesn't remove them all you can try, (with all the mesh selected) : Mesh > Clean up > Degenerate Dissolve 

Hi I think what you need to do is : In Top view open the UV unwrap menu with the keyboard shortcut U and select the "Project from View (Bounds) 

I’m guessing you have accidentally created a Blank black image as well as the colour Test Grid image. If so what you need to do is link the Colour Test grid to the other faces. You can do that by : 1: In the 3D Editor window , in Edit mode select all the faces of the cube. (or just select the faces that are black) 2: In the UV Editor window open the list of Images and select the colour "Test Grid" texture.  All the faces of th ecube should now be showing the colour "Test Grid" image UV Unwrapping/Mapping and editing is not something that can be taught in a forum post like this. So i again urge you to check out some video tutorial on youtube. A good example:

Hi I see Drongle has got here before me You need to learn a little about UV editing. Each face of your 3D cube has been mapped to a 2D square. Watch a few videos on Youtube on “Blender UV Unwrapping”. The mapped faces in the UV space can be edited, moved about, scaled and rotated etc. For your cube each face has been mapped to completely fit the UV space. What you need to do is open up a UV/image editor window, add a texture there so that you can see how it is mapped to your cube then select a face on the cube that is not correctly mapped and rotate that face in the UV Editor. In the Edit Mode select your Cube (all faces selected) : 1: Drag open a new window*. 2: Change this new window type to UV/Image Editor. 3: Open up the Image menu and choose “New Image”. In the dialog box give it a name (Test Grid) and change Generated Type to “Color Grid”. Hit OK and you will see the numbered color grid texture appear in the UV space. 4: Go back to the 3D editor window and change the Viewport Shading to “Texture”. You should now be able to see the test grid on your cube.  5: Deselect your cube then Select one of the faces that has the texture rotated incorrectly, go to the UV editor window and you will see this face is also selected there. Now you need to simply rotate the square by using the shortcut key R followed by an angle of rotation, For example “ R 90 “ will rotate it by 90°. Typing R 180 will rotate it 180° etc. check in the 3D Window to see if you have rotated it correctly. Repeat for each face that needs its mapping edited. *To drag open a new window: hover your mouse over the widget in the top right hand corner of a window, hold down the Left Mouse Button and drag it to the left. To close a window do the same but this time drag over the border of a neighbouring window. Again best is to watch videos . It will be so much easier for you to edit your UV's if first you watch a few video tutorials

Hi Chic Chic Aeon wrote: Using the same balconey file and same physics model dae the analyzed one on the left is 2 -- the unanalyzed is 13. THEY ARE BOTH 2 in the UPLOADER; the difference happens when changed into prim which is of course needed here. So from now on -- until I get a better subliminal idea of which is better for different shapes and complexities (or whatever) I am just going to test both ways  That is not a fair example because, if i understood you correctly you didn't make separate physics meshes.:smileysurprised: When using the correct physics the Physics cost will be below 1 for a "triangle based",and not analyzed Physics mesh.    and for a Box type Physics mesh, (6 non overlapping boxes), and Analyzed in the Uploader the Physics cost would be around 2 when set to prim ( 6 x 0.36 = 2.2)

Hi Chic Chic Aeon wrote: ................ WHY would you not analyze your physics model before upload? .............. An example of when it’s necessary to use non analyzed physics: At the moment i'm working on a prim to mesh build with the primary objective to reduce the LI as much as possible. Below is an image of the top third of the stairs for one of the stairwells.  The stairwells/corridor is designed in such a way that the lower LoDs of the actual stairs will never be seen. Restricting the line of view means that the 3 lower LoDs can be dialed down to minimum in the uploader which gives a Download weight of 1.1 for each of the stair sections. By using planes and not analyzing the physics mesh the physics cost is only 0.5 (when set to prim) so the physics is not effecting the LI at all. But if I were to use an Analyzed physics mesh it would need 8 hulls, and give a physics weight of 2.9 ( 8 x 0.36 = 2.9). This would have tripled the final LI cost of each stairs section.

If Drongle has got it right (he usually is) then that only applies to a physics mesh that hasn't been Analyzed. So what you could do is re-upload the floor again and this time Analyze your Physics mesh in Step 2 of the Physics tab. (The Uploader should create only 2 hulls for physics so the physics weight will be below 1). Then when rezzed change Physics Shape Type to Prim as usual.

As Arton has already said it is probably a problem with the lower LODs Obvious statement : The farther an object is away from you the less detail you can see. Not so obvious statement : A mesh object in SL consists of 5 different mesh objects: High LOD mesh ( ObjectX ) Medium LOD mesh ( ObjectX_LOD2 ) Low LOD mesh ( ObjectX_LOD1 ) Lowest LOD mesh ( ObjectX_LOD0 ) Physics mesh ( ObjectX_PHYS ) In SL, to avoid rendering geometry that can’t be seen at a distance LOD meshes are used. LOD stands for Level Of Detail. (The Medium LOD mesh object will have less mesh detail (geometry) in it than the High LOD model, the Lower LOD less than the Medium and the Lowest Lod fewer again.) When you view and object up close you are looking at (your video card is rendering) the High LOD mesh object. As you move away this High LOD mesh is switched to the Medium LOD mesh object. Zoom out some more and this will be switched to the Low LOD object. Finally as you zoom away this Low LOD will be switched to the Lowest LOD object. The distances at which this switching happens depends on 2 things, the size of the mesh object (radius of the objects bounding box) and the value you have in your viewer for the RenderVolumeLODfactor setting. For a small item like a flower vase the switching distances will be quite short. For a bigger object like a building the switching distances will be much greater. You always have to provide the High LOD mesh object, When you open the Build menu and choose Upload > Model and then select a mesh, this mesh will be automatically loaded into the High LOD slot. If you don’t then go on to upload the lower LOD models then the Uploader will automatically calculate the other 3 lower LOD models. This is what is happened when you uploaded your Painting/s. The uploader can’t know what faces are important to keep in the lower LOD slots and which can be removed. Only you can know this. The result is that when you zoom out from the painting the high LOD(s) is (are) switching to the medium LOD(s) which the Uploader has calculated to be a single triangle on the front face of the canvas . So you need to create at least 1 of the lower LODs yourself .  What I would suggest is : Your painting seems to be a collection of 9 separate mesh objects? If this is the case and you don’t really need to be able to reposition each part of the painting when it is rezzed inworld then join them all into 1 mesh object inside of Blender. To do this : In Object mode, Shift select all 9 parts and the join them using the Ctrl J shortcut. This will be your High LOD mesh. In Object mode make a copy of this mesh, SHFT D and move it to another layer. Open that new layer and rename this copy Painting_LOD2. From the images you posted I’m guessing that the painting is using 2 materials, 1 for the front faces of the painting and another for the edge faces that give the painting depth? If this is the case : Rule to remember when creating lower LOD mesh objects, All LOD models must contain the same list of materials. The only faces that are important to keep for this lower LOD mesh are the front faces that hold the image. But as you have to apply the rule mentioned above you must keep one triangle that contains the second material so delete all but one of the faces that are making up the thickness of the painting . That is the Lower LOD mesh done.  For now, to keep things simple I would suggest for this painting you use this Lower LOD mesh in all 3 lower LOD slots in the uploader. To export: In Object mode select only the High LOD mesh object and export as Collada. When the Export window opens look on the lower left hand side of the window and make sure you check the box Selection Only from the Collada export panel. Save this high Lod model as , “Painting.dae”. Repeat for the “Painting_LOD2 and save as “Painting_LOD2.dae” Upload the High LOD “Painting” as usual then in the Medium LOD slot choose “Load from File” and load up your Painting_LOD2. You can re-use this mesh in the 2 lower LOD slots by choosing the “Use LoD above” option. For Physics you can use a cube of any size and Analyze it in Step 2 of the Physics tab. If the LI costs are acceptable you are done. If they are too high then that will be new forum question.

Hi The following is theoretical, meaning I haven't done any test uploads into SL,but this is how I would go about creating such a sign: Once you have the Carmencita font somewhere on your hard drive: Open a new file in Blender and in Object mode instead of adding the usual mesh primitive choose to Add Text. Then you need change from the default Blender Text to the Carmencita font.  This Text next needs to be converted into a mesh object : Alt C choose the Mesh from Curve............Text option. In this example the text mesh now contains 3.800 vertices. To reduce this to lower number use the Decimate modifier.  Change to Planar and then play with the Angle Limit setting. The higher the angle the less vertices will be used in the decimated mesh. Approximately 15° seems ok for this Font. In Object mode apply the Modifier and the vertice count is now 600. Next extrude to add thickness. Give this mesh a material (Material_1) and then UV unwrap. If the back of th sign will never be seen then don't forget to delete all those unnecessary faces. Important note: This will be the high LOD mesh. To achieve a low LI cost for the sign the lower LOD meshes (Medium, Low and Lowest) will be only a simple box shape.The same box shape mesh for all 3 lower LODs. This box shape will use an "Imposter" texture. Instead of using mesh detail the lower Lods will be a box mesh with a texture applied to the front and back faces. This texture is called an "Imposter" or Billboard" texture. This Lower LOD box needs to have its own material. As all LOD meshes must contain the same materials as each other so you will need to add a little triangle somewhere in the High LOD mesh and assign it the same material as you will give the Lower Lod box (material_2).  Next to create the "Imposter" texture. You can do this within Blender by rendering out a top view of the sign with a transparent background or Import the sign as it is into SL and take a screen-shot or photo of it then remove the background in Gimp or Photoshop. To do it in Blender : Set up a camera and a light above the sign as in the image below:  Then set up the Render > Shading panel and Output as below, Hit F12 to render and then save the rendered image somewhere you can find it again. Open this image in Gimp and crop it to fit the text. Thiis Image needs to be saved with its Alpha channel.  Next step is to make the mesh object for the Lower Lods. This is, as mentioned before, a simple box shape the same dimensions as the High LOD mesh. This mesh object needs to be given its own material (material_2) and UV unwrapped.  Don't forget to add an extra triangle to hold the material of the High LOD mesh. Open the UV map of the Lower LOD mesh in Gimp/PS and map the imposter texture/image to the front and back faces. Then save this new image as a png file (with alpha channel). This new Imposter texture will only be seen at a distance (depending on the dimensions of the sign) so doesn't need to be very big. 256x256 or 512 x512 should be plenty.  Make a copy of the Lower LOD mesh box and this can be used for the Physics mesh for the sign. When uploading , for the medium LOD mesh slot choose the Lower LOD mesh box we created earlier. Then for the Low and Lowest Lod slots choose the option "as above" (or something like that i can't remember the correct wording ) . Upload the Imposter texture that was mapped to the Lower LOD mesh box. Then zoom out from the sign and when you see the LOD mesh switching from High to Medium select that and apply the Imposter texture to it . Afterthought : For an even lower LI cost perhaps think about deleting the faces that make up the thin sides of the Lower Lod box.

Hi Could you post a screen shot of how the "painting" mesh looks in Blender in Edit mode and also one of how how it is UV unwrapped ?

Hi I don't think you can call it bug as it is expected behavior for it not to stick after a relog (not persistent) From the SL Debug settings wiki page  ............ http://wiki.secondlife.com/wiki/Debug_Settings Edited to add: If you are seeing all those yellow lines (not just the outlines in yellow but also the hidden ones as in your image) every time you edit you could perhaps check that the RenderHiddenSelections is set to false in the Debug settings. I think that one should "stick". 

Not sure if this will work in your particular case but you may find it interesting anyways, here is a recent video on how to create an asset library in Blender :

Hi Unfortunately what you describe is normal behaviour for non Analyzed Physics shapes of meshes that are less than 0.5 meters "thick" . There are already a few threads on this topic in these forums, the first I found was this one https://community.secondlife.com/t5/Mesh/Making-physics-HELP/m-p/1751753#M18002 Have a look at message N° 15 and 16 .... Quoting Drongle : "It's quite complicated, so I will try to explain the relevant background. First, there is a mechanism whereby the physics shape type used by the server switches from Prim to Convex hull without telling the viewer. This happens when one dimension of an object reaches a certain minimum. This minimum used to be 0.01m, but was made much bigger some time ago. This broke a lot of content, including thin walls like yours. It doesn't show up on the Physics Shapes display because the change is hidden from the viewer. So you can only detect it by trying to walk through (bug #1)" The work around is to add some extra geometry, a single triangle will do, so that the meshes bounding box will be at least 0.5m thick. In the case of your ceiling mesh it could be added at one edge of the visible mesh where it would be "flush" with one of the walls of the building.Have the normals of this extra triangle facing into the wall so that it will be invisible when rezzed. Don't forget that this added triangle will have to be included in the Physics mesh. And as the Uploader doesn't like small triangles in the physics mesh you may need to experiment a little. If you find the physics weight has increased because of the added small triangle in the Physics mesh then try making it bigger (higher in your case) so that you can use a larger triangle in the Z axis in the physics mesh. Best is to avoid this situation completely by including the walls or part of the walls in the ceiling mesh and so increase the boundng box dimension in the Z axis.

Hi Good question, Its not something i have ever tried before so i gave it a go. In a new Blender file I created a "window" mesh object on Layer 1 and a "window_LOD_Medium" on Layer 2 and a "window_LOD_Low" object on layer 3. I Parented the two lower LOD meshes to the "window" mesh and saved the file.  From a new Blender file I opened the Append menu and went to my saved "window" file and opened the Object folder, selected the 3 Objects there and unchecked the Active Layer option from the append menu. The appended objects were brought into the 3D window and the mesh on layer one was still the Parent of the other two objects. Moving and rotating the "window" mesh in Object mode resulted in the child meshes also being moved and rotated. ( Often when I selected all three layers together to see if the child meshes were following the parent object, the child objects had not moved but then, with all three layers still open, selecting only the parent object and hitting the G key (and not moving anything) the child objects snapped to the correct positions).  Next I grouped all three (parent and child objects) to a new group saved as a new file. Opened a new blender file and this time: Append > Group > and selected my new "window group" and it opened just as the previous time with parent and child objects on separate layers as in my original "window" file.  So unless I'm misunderstanding what it is you are trying to do i can say it does seem to work ok for me

Hi Im not a 3Dmax user so can't help directly but perhaps you may find your answer here in an earlier thread https://community.secondlife.com/t5/Mesh/Please-help-in-the-program-3d-max-With-Second-Life/m-p/1706361/highlight/true#M17246 Kwakedle's reply, message number 8 or check through these other threads https://community.secondlife.com/t5/forums/searchpage/tab/message?filter=labels%2Clocation&q=3dmax+scale&location=forum-board%3AMesh

Hi Tc Ruby wrote: I am wondering now what 3d program to use to try the outfit on before uploading as its you all know 10 linden perload and i dont want to pay 10 linden to see a fail hehe. So any info on this would so be appreciated and greatful. Good news for you , SL has a test grid called the Second Life Beta Test Grid ( Aditi). It will still "cost" the L$'s to upload but the L$'s charged on the beta grid are not deducted from the real L$'s you may have on your account for the main SL grid (Agni). So uploads are free. To login to the beta test grid check out this thread from a couple of weeks ago. https://community.secondlife.com/t5/Mesh/Mesh-upload-for-testing/td-p/3002175 For the standard SL viewer the box you need to check is in the Me > Preferences > Advanced > Show Grid Selection at Login

Again I am not able to import the .dae into Blender. The .fbx imported ok with the bones and vertex groups . I uploaded as a static mesh and the numbers in the LOD slots looked ok.  The .obj imported without bones or vertex groups and on upload again the numbers were as expected.  The only thing I can suggest is you try exporting from 3D Max in one of the above formats and then importing back in.

hi The Collada file you linked to, for some reason is not getting imported into Blender. Its just not in the list of objects in the scene. I tried a quick Google search and noted somebody had the exact same thing happen when trying to import a .dae from Sketchup. Anyways .......... can you link to a .fbx and or .obj of the "planes" and "chunk of hair" ("base mesh") then perhaps I will be able to at least se it in Blender. If Drongle, or any body else sees this do you have any idea why the Collada file linked to in message number 15 above, wont open in Blender yet can be uploaded and imported into SL ?

Hi I had problems when i tried importing your .dae into blender, nothing seemed to get imported, so i uploaded directly to SL then exported back to my hard drive with the Singularity viewer. When I tried exporting that mesh from blender, in the Uploader I got similar results to you except all all the lower LODs were the same count as the High LOD. Also your mesh is tiny ! ! ! In Blender I scaled up your mesh then looked at it again. I noticed that the two large planes were in fact four ! Each large plane had a plane duplicated over it. When i tried separating the planes and exporting, the numbers in the Uploader were then correct.  So I would suggest you try removing those duplicated surfaces and see if it uploads correctly.

Hi I had something similar happening in the Uploader a while ago but not sure if it is caused by the same thing as you mention it only happens "when I add flat planes". For me it was the opposite, Closed meshes, smooth shading and not UV unwrapping. The thread can be found here https://community.secondlife.com/t5/Mesh/LI-count-something-not-right/m-p/2256797#M23895 If that doesn't help you could always upload the .blend file with those troublesome planes to here http://www.pasteall.org/blend/ Someone may be kind enough to check it out for you

Hi Try the following: In Edit mode, looking straight at your cylinder in Front Ortho view (numpad 1) Select all of the cylinder and rotate it a few degrees in the Y axis ( A R Y 10 ) Deselect Cylinder ( A ) Go into Wireframe view ( Z ) and Box select the top ring of vertices ( B drag mouse around to select ) Return to solid view ( Z ) Scale the selected vertices along the X axis until they flatten ( S X slide mouse ) Done.

Hi Chic When in Object mode those numbers, Verts, Faces, Tris, etc are the total counts for the whole scene. In the first image (the sphere) you only have one object in your scene. In the third image ( cube) you have 7 objects in your scene. Select your cube then tab into Edit mode and you will see the correct Vert, Face and Tri count for that selected object.

XbabylonX wrote: what do you mean by saying "Then in the Start at box type in mesh sandbox 1"? When logging into Secondlife with the Firestorm viewer: at the bottom on the screen are the boxes Username:, Password:, Start at: and Login into Grid: . The Start at: is the location (sim) you want to arrive at when you log in. Mesh sandbox 1 is the name of a place on the beta grid where you can rez meshes, so if you type in the Start at: box mesh sandbox 1 you will arrive directly there when you log into the Beta Grid . XbabylonX wrote: Is this the test everyone is talking about? Yes that is the IP test tutorial you have to complete before being able to upload mesh. (but I could not say if that is the actual page for the Beta grid mesh test). There are two identical IP tests, one for the main grid (Agni) and one for the Beta grid (Aditi) When you get to the Beta grid and try to upload a mesh it will give an error if you haven't completed the mesh IP test for the Beta grid. It used to be that along with the error message it linked to the Beta grid (Aditi) mesh test page. I don't know if it still does that or not.

Hi XbabylonX wrote: Greetings, trying to understand how to manage and go to these famous beta testing sims for mesh but I think Im a bit confused. Please, could somebody post the steps one by one? Thank you! At the Firestorm Login screen open the Preferences menu then from the Advanced menu enable Allow Login to other Grids.  Before actually logging in open the Log into Grid menu and choose Second Life Beta. Then in the Start at box type in mesh sandbox 1 as your destination. There are a few other sandboxes you could use but that one is as good a starting point as any of the others  It used to be that everybody received the same amount of "play" L$ to use on the Beta grid but a while ago it changed so that you had the same amount of "play" L$ as you have on the main grid. As Chic has already pointed out you will not be able to upload mesh untill you have redone the little mesh test.

Hi Here is a link to an old thread where I did some experiments to reduce the Physics cost of stairs and still have accurate physics. https://community.secondlife.com/t5/Mesh/Problem-with-home-mesh/m-p/1788825#M18558 but the more usual way is to use "ramps", and generally ramps and side panels for the handrails are good enough for most stairways