Optimum Actor

Ah, simple. That would explain some other weird titles I've come across. Thank you!

I've been using the Builder's Buddy. to make sets/builds for Machinima and Machinicomics. It's come in very handing for moving large sets that couldn't be made in a singlre linkset for one reason or another. I've also discovered that you can have a heirarchy of these. That is, you can can have a 2nd Builder's Buddy hidden in a prim of a linkset that's part of another Builder's Buddy. For example you might have a house that is a collection of linksets - one linkset for each part of the house that you you to do a scene in; and each linkset might have a prim that contains a Builder's Buddy for all the furniture in that scene (that way you might rez just the bare house, and/or add furniture only where you need it). The issue I have isn't with the builder's buddy, but with a side effect of experimenting them on the current set I'm building.In addition to the BB compent script which I addent to each linkset, I added the following script in order to clearly identify each linkset: default { state_entry() { // prim's name (not necessarily object's) in green and opaque llSetText(llGetObjectName(), <0.0, 1.0, 0.0>, 1.0 ); } This shows the name of the linset in a green title above the base prim of that linkset. If I change the name I have to manually reset the script, but that doesn't happen that often. The issue came when I realised it was better to have the base prim of each linset at the top, rather than the bottom. I did this by unlinking and relinking individual prims and adding the scripts again. Not as messy as it sounds. But, I seem to left with an extra title. The following image illustrates this: The original base prim was the floor of the Lifeboat, but now it's the roof. That extra title at front is from the builders buddy prim at front (which is just a prim in the DB Lifeboat (Hull) linkset). But the title for the original base prim is still shown, even though there's no script in it any more. How do I get rid of the superfluous title?

Apart from copyright issues, what about Mesh from video games made by yourself, or by yourself from gamemakers like Unity?

The other programs are good as well. I don't have Scatter Show but I do have Manga Artist 5 and Motion Artist. Hope to make a different sort of Machinima with that from SL some day.

OK, try my best to explain the physics. You have 10 prim boxes in a line (linked). That's 10 prims to the physics engine, I,e each box has 6 sides, 12 edges and 8 corners.(convex_hull) Put them in a bin bag and suck the air out and the physics is that, just the outside shape. The best way to understand is make a flat prim with a hole in it and convert it from a prim to convex hull. When you walk on the hole you will not fall thru it. So any build that needs to be walked thru or dropped in has to have the physics set to prim. A build can be set to convex hull and individual prims that need the physics set to prim.. So clearly one has to be selected in converting. I have some places withat's be useless, but others (like a wall with a big window that never opens in the middle of it) it'd be a good thing. Thanks.

Ah! I've seen references to the 10m limit in older SL references but it never seemed to be an issue when I was building Not familiar with convex hull but I just did a search and found this thread, with an explanation by yourself! Shall go read it. UPDATE: Might need to go back and re-edit some builds.

steph Arnott wrote: Optimum Actor said ----------------------------------------------------------------------------------------------------------------------------------------------------------------------- the Slum City is a free prefab from the Market place has a bit over 2300 ---------------------------------------------------------------------------------------------------------------------------------------------------------------------- I reduced the prim count to 1000 just by resetting the bounding physics and that was not all of it just a quick test, also the prims are as was at the time 10 x 10. So, many of the flat faces of the buids would only requir one prim instead of six. How does one do that? Is there a link i can follow for help on that?

This all reinforces the idea of at least making prototypes in-world, revising and refining them in-world until finalized, and then then maybe exporting those out-world as Mesh and re-importing a refined and tested version, after maybe some external editing. For me the reason not to use the originals would be too much lag. For example, this set which I built has has over 1500 prims in it: ...and the Slum City is a free prefab from the Market place has a bit over 2300! Now the second is fabulous for free item but try shooting a machinima or machinicomic there and you have problems of lag, especially if you have a lot of avatars in one scene The top example really only needs to be used for external shots - otherwise only the individual linksets (it's organized in 13 linksets in a Builders Buddy) and set dressing need be used at any one time. But even so, if it's a regular set and going to be re-used (which is the case) simplifying things by using mesh might (I hope) reduce that down considerably. Coming back to in-world tools that might assist in that then, I have a short list of three:Celzius, Mesh Studio, and Mesh Generator. The last two can be applied to already built items (though MG is limited to 32 prims) and all work by having scripts in the base prim which point you to external pages where you can download a COLLADA file. Everyone seems to agree that post generation editing (using Blender or some other 3D program) is helpful to make mesh more efficient (and/or add texture areas). BUT - Singularity and other viewers can download Collada files direct from a linkset anyway (though there is not support for setting up texture area in-world yet) and is free. The thing is that there seem to be improvements in rendering in the pipeline, including and , so maybe the sensible thing to do is wait, and see what the next generation of official and alternative browsers brings up in features. And of course the above listed makers may be improved as well, or new ones be created.Not everyone has the needs that a machinima or machinicomic group does, so what makes sense for me might not apply to others.

FD Spark wrote: I really can't spend anything more until I get handle what I can. Plus I really need make more money before I start investing in more expensive stuff. It just doesn't seem reasonable for me spend more on something I am going struggle and not be proficient at. I spent last 5 years working on my illustration skills on 2d level. I just wish I could translate that to 3D. I understand the financial limitations very well. You are better off learning than buying I guess, because it may be that the tools bought aren't needed. That's why I started that other thread. I've put off investing in in-world tools until I finish my Masters. As regards 2D illustration skills, I should have thought they'd be usefull in creating textures. Good luck with it.

FD Spark wrote: I have trying to figure out different 3d programs like in world mesh studio, just recently Wings 3D and there this browser version 3d program that is in beta called Clara.io but the easiest so far but still struggling with actual texturing part is mesh studio but I don't know how to make textures for the objects or get it to work. I started another thread about this called in-world versus external sculp and mesh makers? because I didn't know either. Have had some interestng feedback on the question. Not mentioned was an in-world maker I discovered late called Mesh Generator by the same folk who make Prim Generator. There's a free DEMO version where you can play around with with two prims at a any one time, but I haven't had the time to look at it properly. In any case it seems that this and also Celzium also you to define texture areas by painting groups of colours on faces. I may be mistaken there but it seems to be the case. I have been trying and failing for 10 years to use Blender, its just not intuitive program to me and even with the book I got on it last month even the introduction says learning Blender is hard most people get discouraged and give up on it. You're not the only one. I have tried on several ocassions to get into learning Blender and GIMP, but in both cases the GUI seemed to be a major obstacle. This may have changed for the most recent versions of both programs. I know the layout for Blender is different now. With Blender the ability to cut seems and make texture maps seems to be a basic function. It's mentioned in the 2009 edition of Blender for Dummies book (luckily in my local library for free) and is probably in the 2011 version as well.. Nonetheless, it's always better to have a choice of tools to do the same choice, as some may have advantages over other in certain situations. For example I have Paint Tool SAI, GIMP, Open Canvas, Photoshop CS6, and Manga Studio 5. They're all good drawing programs but SAI and OpenCanvas are better on my laptop, Manga Studio is a breeze to draw with, but Photoshop is Industray standard. The last time I used Wings 3D it was to edit a model generated by DoGA! It seems capable enough, but all the free 3D design programs (like Truespace, which is now freeware) have a learning curve and the only way to learn them is to spend time fiddling about. - - - - Coming back to what I asked in the first Chosen Few, will the same techniques listed for Photoshop CC also work for CS6?

Am impressed with these scripts. Not only making stuff an art, but tit's also clearly a science as well!

Example of a mesh model with hand made Level of Detail models: This Jerry Can has a download weight of 0.4. So there isn't really any reason to not provide decent LOD models. This and other examples give me a very good reason to learn to build mesh properly for myself and the group, rather than rely on pre-made mesh objects in the Marketplace. With my aircon unit example, it's hard to tell what they'll look like from the marketplace description, even with a sample image. The Prim version was part of a set and only $L 10. The Mesh version was also part of a set costing $L 99. Both look good on those pages but it's not until you rezz them that you find out how they really look in use. I had a similar problem with stairs. Lots of cheap mesh stairs that look good up close, but on the sets I've been building they deform, presumably because they use the lowest LOD. Not that such purchases are total waste, as they can still be used in close up shots, but it's annoying because it means they weren't fully tested before put on sale. Likewise the examples of display weights also support the idea for using Mesh over straight Prims and Sculpties - 15 more display weight units than a simple prim but lots more detail. This has answered an indirect question I had about the worth of building Mesh in the first place. Obviously though the workflow may be different because of my projects - prototypes have to get built first for feedback - and whether or not in-world tools are used Blender most certainly will. Also, I suspect I will be spending much more time in the Beta Grid testing mesh for free when I come back to this properly! This information's been extremely helpful

The problems I can see with Mesh vs Prim are deformations in the viewer. For example, here's two prefab rooftop aircon units I bought for props, close up and at a distance: As you can see the one on the left (3 prims, non mesh + textures) is fine, whereas the one on the right (1 prim, mesh) deforms. If I read one of the previous comments correctly, this is probably because straight prims have more triangles in them. The one on the right probably has less triangles in it! This is probably not critical for general use, because most AVs will be closer to the object. But for webcomic snapshots you need something that looks good at different distances. That's why I opted to use the prim version (mostly) for this set (a modded version of a Marketplace item no longer available):

One thing to keep in mind is that in-world mesh creation tools are no substitute for a solid grasp of a 3D modeling program. They are good for tossing together a quick prototype to give you something more than a "blank canvas" to start with. But for organic shapes and (the necessary!) complete control over both the topology and UV-mapping of your models, a stand alone 3D modeling program is required. I'm not so sure about this. The following is the "Beta Base" set I built recently for SECONDS: It was put together using straight prims, and selected mesh, sculpties and textures from the market place. Not meant to be a permanent structure, but something to be rezzed when needed for a scene (so mostly it exists as 17 linksets in a Builder's Buddy). But I'm not so sure that building it straight from an external program would be a good idea. I was getting feedback from the rest of the group about what features should be where and how it should look, but if I did that using Blender or another program that wouldn't happen, or not in the same way. That's why I've had more interest in in-world tools, because I can get that feedback before I upload and pay for anything. That doesn't mean though that I can't rebuild or redesign something in Blender based on that feedback. And being able to design one's own mesh via an external program would probably reduce the need to buy pre-made ones on the market place (like the antennae in the image above). It seems that most options lead to Blender! I've just learned how to edit video with Blender and found it does just as good of a job as the "Big Boy" dedicated video editors once you've set up the windows properly, with definable/controlled scene transitions, green screen, rotoscoping, titling, special effects options etc. I'll look into that. I already have several video editors, including Premiere Pro, so I'm not short of options. With that in mind, an 80 hour 2-3 week investment in doing Lessons 1-5, 8 and 12 will get anyone serious about modeling started on the right road. I'll make a note of that. Right now in Real Life I'm finishing a post-graduate degree, and that's got to get priority. Most of what I've been doing in SL has been at nights. Once the degree's done I shall investigate all this thoroughly.

...wouldn't you be better off staying with prims and just change the prim type to convex hull? As long as you don't add any hollow or cut prims (unless you change the physics type to none) I can't honestly see much of a saving. That is probably the case when making a set. I did explain what I was building and how it was used earlier, but without an example people may have missed that. Making a single object for a roof or floor has advantages other than saving prims. A single object roof/floor/wall can have a door script in it to move it out of shot, or to be made transparent temporarily. That can be done manually but it becomes messier with more than one object involved. Anyway, with that said - MS lets me choose how detailed I want my model. So it saves the amount of work one needs to do later. That's worth knowing.

I now use a combination of Mesh studio and Blender. I make my build in world so I can ‘feel’ the space and apply the basic textures. Take the build into Blender and make the LOD/Physics model, make any UV maps I need, clean (if necessary). You can of course do the LOD/Physics model in world but I find it quicker (now) to do in Blender. What do you think the advantage to using Mesh Studio would be over downloading .dae files with Singularity? It's just unclear to me what Mesh Studio Does that the Singularity Export doesn't. See my main use for these tools would be making components of sets, for example the following was the base outline of a floor of a sky hanger: The central white prim is a general use are whereas the blue and green prims have more detailed areas with interior walls. The four holes in the corners are for stairwells and lift shafts. Now it would be nice to have this as one mesh with 6 textures (top / bottom and four sides). This might not seem much of a savings in prims but with a 4 story structure including exterior and interior walls it adds up. But as a set is only rezzed sometimes for a shoot, is it even worth making a mesh (or even a sculptie) of this? Not sure if this makes a difference in which tools to use, but thought I'd explain what I'd be using the tools for in this case. This is also why making something whole in Blender is less appealing than constructing something in SL first, and then making it a mesh or sculptie - because I can see how it all fits first.

That's fair enough. Obviously you get results doing it your way. Yes, I'd like a bit more feedback as well before I commit.

OK then, what about using Singularity and Radgast to save .dae files (and then cleaning this up with Blender)? I can see the advantage of this is that you can build a linkset directly in-world and know that it will fit exactly how you want it with other components of a scene. That means I don't have to relearn entirely any building skills I have. Am I missing any traps with this? It seems you can also (also Free, but external) with the help of some plugins and some cleaning up in Blender, which means that it's another program to build in, another option.Of course both produce .dae files which need to be cleaned up with Blender, so gaining expertise with Blender would seem to be a requirement of either.

I would not recommend looking at any of the in-world or external tools for making sculpties. Personally, I have either withdrawn many of the sculptie items I once had for sale completely or have been slowly replacing them with higher-quality mesh items. I did come across one use for sculpties - an animated toilet seat that raised an lowered when you sat on it, and using a technique of swapping sculptie textures (and thereby changing its shape). I had thought that maybe sculpties had an advantage in the making of simple but irregular objects or components, but perhaps not. The videos I've seen are impressive, but there are problems in applying regular textures to a sculptie, and it seemed too that there was a limit to just how accurate, and how much detail one can get with a sculptie. So, assuming I discard Sculpties as an option for the moment, what of the in-world mesh makers, Celzium and Mesh Studio? I liked the fact that with Celzium you seemed to be able to select texture areas (up to 8) in-world.

I'll confess that I haven't been building very long, and that's why I'm interested in people's experience and opinions about using in-world vs external sculpt and mesh makers. After a search I found these in-world tools via the Marketplace: Celzius - Sculpt Creation Tool L$999 (Cel Edman) "You can choose between 30 geometric shapes, and build something with them. Your build & creation can be converted to sculpt-image. About 13 to 16 sub-sculpties can be used to make a 1 prim sculpt image." Prim Generator L$2,611 (ShopNN) "32 prims become 1 sculpted prim. The sculpted map can be easily made in the in-world."Sculpt Studio L$4,999 (The Black Box) "A high-end in-world sculpting tool for beginners, advanced and professional Sculptors, that allows to apply SL building skills to sculpting." This seems to use a "lathe" system whereby you have a column of radiating points which you can edit, each mapped to part of the finished sculptie (more detail here).Celzium L$1,995 (Cel Edman) "You can select from 32 mesh objects and create more complex build with these, Up to 8 different textures/materials are supported, includes UV-texturing and layout. Up to 200 sub-meshes can be rezzed and converted to a collada .dae file. You can upload into SL, or like blender and other 3D software packages."Mesh Studio L$4,999 (The Black Box) From the videos I've seen and what I've read you build regular prims in a linkset, insert a script into the root of that linkset and then run it. It sends you to a page where a server generates a .dae file (collada model) for you. You can then edit that in an external program and/or load it back into SL as a model.Celzius, Prim Generator, and Celzium all have systems where end up at external pages where a sculpt image is generated for you and you can upload. They also all use a system where you select the components of what you're making from a selection of predefined shapes. Celzius has a DNA expansion for plants, and Prim Generator has several expansion packs, each offering extra shapes with variations. From what I know these are not straight prims, so you can't cut, hollow or twist them, but you can stretch and rotate them. Sculpt Studio has a more detailed way of making a sculptie, and it reminds me of 3D printers which print one layer at a time. I'm not sure exactly what advantage there is in using this method. I was very impressed with the videos I saw of Mesh Studio, and then I discovered that Singularity and Radgast viewers can download a .dae file from your linkset directly. The results seem mixed. It seems that whether you do it in-world or in a 3D program, the models made are improved by making faces which don't sow transparent, because that cuts down on the number of triangles used. But in any case, why use Mesh Studio if you can do it in these viewers (not a rhetorical question - there may be some advantage I'm unaware of)? When it came to external tools I looked at the free ones only, because I have a limited budget and high end 3D editors can be expensive). I found: Blender Fully fledge 3D editing tool. The interface seemed to have improved in the latest version. There are tutorials and plugins to help make/edit sculpties and mesh for SL.Wings 3D Described as a Polygon Modeller. I know you can make sculpties in this, not so sure about mesh (but that's probably the case).SculptyPaint, PlopSL, Rokuro, and Tokoroten All of these are specialized applications for making either symmetrical or specific types sculpties. All of these either have commercial versions, or links to commercial programs to make sculpties and mesh.At the moment I build sets/scenes for a group that makes webcomics in SL (machinomics). This means that on the whole they aren't meant to be permanent. Rather they get stored until they are needed, then rezzed for the duration of a shot, and packed away again afterwards. At the moment I'm not looking sell anything. Nothing stops one from using both in-world and external tools. While I'd be nice to have all the tools listed above, a limited budget means that I don't want to buy something I may not use, or have a free alternative to. So, some input please. The questions I have are: When would or do you use the in-world tools, and why?When would using an external tool be better?

I've finished working on this. I reshaped the head slightly, and added a a neck as well (it was simple enough to copy the coding for SKIN and replace that with NECK). I realized something as well about just where the images are for the the eye and mouth expressions, that wasn't apparent to me at the start. The centre line I've been showing actually matches the horizontal divide between EYE and MOUTH prims. That means everything above that line was the whole 180° of that half. This means that the original test texture... ...was well out of place. When I edited it with something like this... ..it worked, but the images were spread too far apart - the mouth would appear at the bottom and the eyes almost on top. Also I notice how distorted the expressions were horizontally. That didn't matter with the eyes because larger eyes simply made them look cuter, but with the mouth I decided squash them to compensate for that. Also, both eyes and mouth images were resized to fit within the front two quarters of the sphere... ...though I left a minimum gap of 3px from the centre line, so that the textures wouldn't bleed over to the other expression. After altering the shapes, so that the head was taller and narrower in relation to its width, I think it looks good: I also modified the script so I could change the linksets for the alien podling's hands to match the head. This was a lot easier to do, though it took me several minutes of searching to find the right command. Notice the skin colours, which represent the age of the podling - they start out solid colour as sprouts, and end up white in old age as the pigmentation wears out): // Skin colorslist skinColorsList = [ <0.306,0.4,0.318>, <0.278,0,0>, <0.369,0.047,0.29>, <0.286,0.114,0.459>, <0.208,0.294,0.549>, <0.318,0.639,0.639>, <0.459,0.729,0.529>, <0.718,0.82,0.616>, <0.91,0.882,0.8>, <1,1,1> ];integer skinColorsListLength;// and the skin colorinteger skinColor;// holders for prim numbers// Dialog Odds and Ends// how long to wait for a response to// the dialog menu, adjust to tastefloat wait_time = 20.0;list buttons = ["Color++", "-", "--Color"];string dialogInfo = "\nPlease make a choice.\n\nAny \"-\" will close dialog when you are done."; key ToucherID;integer dialogChannel;integer listenHandle;// Just your basic initialization functioninit(){ // Set list length holders skinColorsListLength = llGetListLength(skinColorsList); // set pointers into lists to defaults skinColor = 0; // Set default appearance // (NOTE: Use face 0 instead of 1 if using Dimple instead of Path Cut!) llSetLinkColor( -1, llList2Vector(skinColorsList, 0), ALL_SIDES ); }default{ state_entry() { dialogChannel = -1 - (integer)("0x" + llGetSubString( (string)llGetKey(), -7, -1) ); init(); } on_rez(integer num) { init(); } changed(integer change) { if(change & CHANGED_LINK || change & CHANGED_INVENTORY) init(); } touch_end(integer num) { ToucherID = llDetectedKey(0); // sanity check if(ToucherID != llGetOwner()) return; // Present a dialog menu llListenRemove(listenHandle); listenHandle = llListen(dialogChannel, "", ToucherID, ""); llDialog(ToucherID, dialogInfo, buttons, dialogChannel); llSetTimerEvent(wait_time); // Here we set a time limit for responses } listen(integer channel, string name, key id, string message) { // usual kludge for a switch statement if(message == "-") { // finished for now, so... llListenRemove(listenHandle); llSetTimerEvent(0); return; } if (message == "Color++") ++skinColor; else if (message == "--Color") --skinColor; // Set new appearance // Don't be put off by the use of the modula operator (%), // it simply gives you the remainder of a division and allows // us to wrap around the listings is all llSetLinkColor( -1, llList2Vector(skinColorsList, (skinColor % skinColorsListLength)), ALL_SIDES ); // We're not done yet, so send out dialog again llDialog(ToucherID, dialogInfo, buttons, dialogChannel); llSetTimerEvent(wait_time); // Here we set a time limit for responses } timer() { // stop timer llSetTimerEvent(0); llListenRemove(listenHandle); llWhisper(0, "Sorry. You snooze; you lose."); }} This has been an education for me. When in doubt, ask for help, and get get it!

I updated the faces. They all look more like this: The eyes on each are well above the centre line, and the mouths are well below. When applied I can get a lot of variations: The first three images show different combinations of eyes and mouth textures. The last image shows a slight adjustment to the vertical offsets (-.1 on the mouth, and 0.1 on the eyes). Despite it looking like the character has a big mustache, I think this is a success. The code works well with the setup. Thanks muchly LepreKhaun! I still have some adjustments to do but those are mostly to do with the shapes of the head, which should be easy enough to do.

The division works really well as per the image below: I've played around with the shapes slightly, making EYES and MOUTH .5 for X and Y but .75 for Z. The slightly oval look is more comic than a sphere. The SKIN prim was .49 for X and Y, and .74 for Z, and they were all rotated 45 degrees back. I found that any gap smaller than .01 between EYES/MOUTH and SKIN was really too small, because from a distance the viewer tended to show the skin interrupting the textures on the EYES / MOUTH. But shapes and gaps are things I can play around with at my leisure (for example, making it .75 deeper than taller and not rotating it 45 degrees might work better). You'll notice the spit between the two isn't exactly in the correct place. This is because I was using the textures I've already uploaded and they look like this (designed to fit one sphere twice): The red line shows the centre line between two 128 x 128 areas. In the original test a second copy of the texture appeared on the back of the sphere, but was covered up by the neck prim. That isn't an issue with this version though because the second copy of the texture appears on the flat part of the cut, which is obscured by the SKIN prim! I can correct for this by applying the same offset to both the EYES and MOUTH faces: ..And while this works for a static expression, I'd have to either add that offset in the code in the script that sets the texture to the prim's face, or re-draw the faces. I suspect at this point the second option is the better because the textures I have at the moment were all drawn with the assumption of a single face. Also it's obvious that use of greys on the expressions isn't very effective, because it shows up badly against a skin colour other than some shade of grey.. I should really redraw eyes and mouth to use black for wrinkles and outlines, and solid colours for eyeballs and teeth. I can see other possibilities here though as well. Assuming it were set up correctly I could I guess have three prims: EYES, MOUTH and EYELIDS (which would be set slightly in front of the EYES). However, I think that's a project for another day - this works just right for what I want to do right now.

That should give you a 2 prim sphere that will tile seamlessly across its "equator", with the top hemisphere showing the top half of a texture and bottom the lower half. And you've just cut the number of textures required for this by two. In other words, with just a "frown look-left" texture and a "smile look-right", you'd be able to do "frown look-left", "frown look-right", "smile look-right" AND "smile look-left" by simply controlling your top and bottom surfaces. Make sense? Wonderful - this is what I was thinking. I shall have a play with this and report the results.

I had to go off and finish building before doing more on this, and I had an idea that there was something extra that I could do with this, and that is further divide the face, into a lower portion (mouth) and upper (eyes), each of which could have its own texture. The advantage of this is that you can multiply the number of expressions. For example if you have two eye textures, one for looking left, and another for looking right, either could be used with whatever expression the mouth is showing. In the image above I just duplicated the sphere for the face, and dimpled the two into complimentary portions. There is a crease which can be detected at a certain angle, even if you make the inner face invisible. The textures would have to be re-done, but the beauty of this is that if I do it properly, I can still have a set of eyes and a mouth on each texture, and it will do double duty depending on whether it's used on the top or bottom prim (I'd have to move the eyes completely clear of the mouth section to make this work). Also, the colour can be applied to the inside face of each sphere. This works and you get an odd "space helmet" effect (which I like). But I need to go away and play with this before I do any scripting, because it will change what's scripted. Nothing further today, because a heavy thunderstorm's coming my way, and best to logoff while that's happening.