Christhiana

You don't even have to assign a material to the object. A material slot is enough. Extra material slots on an object create more Li though. Something to keep in mind.

I'm working on my first real build (a small house) and I've been working on the LODs for the past week. The other recent thread on LODs gave me a good start but I keep wondering where to put the balance. I've been testing my LODs at LODfactor 1.125 and 2.0 and I find it difficult to decide when there is enough LOD. My house is mainly build with simple mesh objects, which means there is not much to omit for LODs. For the most pieces I have to keep LOD1 the same as the high poly version otherwise when the LOD switches, the shape changes too much. This offcourse results in a lot more LI then the original model I uploaded with generated LODs. I've included three screenshots. One which shows the original size, one with LOD1 (which is the same as the original) and one with LOD2. The images are taken at the switching points with a LODfactor of 1.125 Does the LOD2 factor have to be even visible? Considering the LODfactor and viewing distance. I'd rather leave the LOD2 models at the lowest possible to save at least some LI (I already noticed the most Li increase is because of LOD0 and LOD1 being the same). But ofcourse I also want to be a responsible creator and do the right thing. What do you other creators think? When is there enough viewing distance for an object? The parts in the photo that are colored darker have handmade LODs, the white parts are still autogenerated.

I've been on Aditi exclusivly for the past 1,5 weeks working on my house. I had set the LODfactor to 1.125 (SL viewer default) to work on my LODs (it used to be set to 4). Today I logged onto the main grid to check Im's and such and it really surprised me how bad the LODs are on most products. The nice leathercouch I once bought turned into a couple of tris at 2 meters distance! Tri soup all over the place! I don't get how you can make a really great looking couch and not even bother with the LODs. I swore that moment, I would never sell such crappy products!

Whenever I see an interesting building in RL, I imidiatly start applying seams and unwrapping it in my head

If you pay for the land, you get to put up banlines if you want. And that is a good thing! If you have a parcel somewhere that doesn't make the whole region yours to wander around. I have my banlines up and I'll keep em up too

You can't really blame people for putting up banlines on their property. They pay good money to have their own little p[iece of land in SL where they can do what they please without being disturbed. Imagine having a nice swim in your little piece of heaven and all of a sudden a speedboat comes racing in disturbing your relaxing swim. I do understand your frustration and it's a shame mainland came to be the way it is today. LL should have protected the infrastructure of mainland better from the start. keeping waterways and roads open and contigious for the public. This would give SL a lot more appeal for the vehicle enthousiasts. I would love to be able to drive everywhere on mainland instead of always teleporting. Also it would be great if every sea would be sailable. I'd probably sell my land, buy a boat and spend the rest of my SL life exploring and sailing from sim to sim.

Thank you both for the very usefull tips and insights! This topic has reminded me of how important it is to be a responsible builder in SL. And that's what I'm aiming to be! I do like the idea of the "imposter trick". I can't really use it for my current build (a house) since I plan to sell an AO pack for it, so people can mod the house with their own wallpapers, etc. That would complicate things too much for the customer. Sometimes I upload textures with the model but never for the final version. I like to upload the model (without LOD and with auto physics as a whole first and then start replacing each part with the final version (including the LOD's, physics) and then upload the textures seperatly. This way I have a guide build that easily allows me to swap pieces for the final versions without having to align everything (I can just copy the coordinates). I know this costs me a little extra for upload, but for me the convenience makes it worth the extra L$. How about SPEC and NORM maps, I assume they also work for the different LODs?

One more then..... Do the LOD models, like the physics model, scale to the bounding box of the high LOD model?

Thank you for the link drongle... I'll have to study those graphs tommorrow after I've gotten some sleep Chinray, thank you for the info, I'll experiment a little with the simpler shapes tommorow. I like the tip of using a trianlge for the inner wall! Good thinking! I kept the larger shapes fairly simple to avoid high land impact while retaining a good LOD at a distance, so I guess I was on the correct path I sometimes use mesh cube shapes instead of prims so I can bake AO on them. One more question though, do the LOD models always use the textures of the high poly model or can I upload a different texture along with a LOD model. This way I could just show a simple shape with a rendered picture of the object on the lower LOD's.

When the medium LOD model is the one we'll be viewing most of the time, should I keep them the same as the high LOD verison for simpler shapes? For example, I'm building a house which mostly consists of simple basic shapes linked together. Reducing the polygons on them would immediatly show. There really isn't much to be simplified without destroying the shape of the objects, some of them are essentially just shaped boxes. Would it have a huge impact on Li if I kept the medium (and maybe even the lower) LOD version the same as the high poly one?

A lot can be learned from just experimenting. Make an object in blender, make different LOD shapes and login to aditi and experiment away... try uploading with different LOD files and physics, try different LOD viewer settings, scale your object up and down and repeat the previous. Watch your object from different distances and see where the switching points for LOD are at various sizes. After playing around for a while you'll start to get a feeling for how things work. No magic there, just a learning curve!

Wouldn't that be easy to circumvent by adding a pair of (non-rigged)mesh hands and feet covering the original hands with an alpha?

I usually save the unwrapping for when I am completely satisfied with the optimized model. If I make a change to my model afterwards I always unwrap again just to be safe and make sure my UV map is to scale and optimized. I think you've done an excellent job on optimizing and getting the Li down to only 1 while still having your mailbox look detailed! These kind of optimizations make it worth that extra L$ over the 1L$ versions. I'd rather pay L$150 for an optimized build like yours then take a 1L$ 10Li version. I think when pricing you shouldn't get too extracted by the lowest price on the market. It's better to compare to the average price of the product type you're selling. If you know you've made a good product, charge what YOU think is a fair price for your work. Things like a good personal customer service can also be taken into account when setting a price (are you just selling or also helpfull to your customers when they have a problem with your product). Will you come to their sim to help them out if they get stuck? Also an established brand ussually gets away with higher prices for the same product then a newcommer would. Believe in your skills and product, price accordingly and people will buy despite of all the freebies and 1L$ stuff floating around.

Ah yes, cube measuring....who doesn't use that every now and then

You can always make a big sign you can rez when working on your child avi that states "I'm not indulging in ageplay, I'm creating a skin and/or shape for a child avi.".

In the terraform window it shows the amount of land you have selected.

Making a fitted or even rigged mesh avatar is one of the hardest things to make in SL. I would recommend starting out with something simpler and along the way learn all about modelling, rigging, etc...

That really sucks about the MP reviews. If you as a customer decide against checking out a product inworld while it's availabke then don't complain if it doesn't meet your expectations. But you'll always get some customers like that. You know what they say, the customer is always right

Good point! I have one floor set at .0001 because it's just a plane but get's a boundingbox of .1. I'll make it into a box then.... I was already planning on making the house mod. What fun is a house you can't alter to your liking. I will probably seperatly sell a UV/AO map kit for the texture modders as well

There are two more discussions about this topic going on atm. http://community.secondlife.com/t5/Mesh/Build-scale/td-p/2870456 http://community.secondlife.com/t5/General-Discussion-Forum/Why-do-some-houses-suffer-from-gigantism/td-p/2870614 I personally decided to use approximately 120% of RL scale.

I started a thread of my own about the subject of house scaling about the same time and didn't even notice this on! http://community.secondlife.com/t5/Mesh/Build-scale/td-p/2870456 I will look into the building scale and the camera tutorials! Although I do feel 150% wouyld be a little over the top!

Ow haha, I didn't even notice that other post, reading now... 150% seems a little over the top to me, but I will definetly have a look at that building scale as well as Penny Patton's camera tutorial! I must admit I still use the default camera posistion myself so I see my build the same way as a lot of customers would. Even though I think you're right about the standard viewing angle being unpractical and really distorting the view, I would still like to accomodate the customer that still has his camera angle on default (which I imagine would still be the larger group). The size my build is at atm is still doable from the default viewing angle, albeit feeling a little cozy but looks quite realistic from the mouselook. I think that'll be the result I will be aiming for. For this build a +/- 120% scale seems to work best but I can imagine that may vary with each build depending on the layout. Thank you all for your input on this matter! And I think you're right about the camera angle being based on those older fake 3d games where you viewed the level from above in orthospective.

Thank you! I did notice that the name presented in the uploader is the name of the last selected object in blender (which will be the root prim of the linkset in SL). If I import a single object seperatly it so far always seem to get the name correct so I'm sure the names are set correct in blender and exported along with the object. I guess I'll just have to work around it. The house consists of 41 objects so it's still doable .

This btw is the house I'm working on. Nothing fancy but I think it's turning out nicely for a first real build .

I guess it are the quirks of SL then I like to upload the basic model without LOD and with autogenerated physics first to get a good feel for size and see if all the mapping is done correct. Then piece by piece I replace the parts with a version with the correct LOD and physiscs. This way I can keep every piece in the same place I have them in blender without having to calculate each position. So in the end it won't be to difficult to check each individual piece when I upload them seperatly and correct any wrong names. I was just wondering if I did anything wrong. It would have made texturing the basic model at little easier with the correct names on the objects . I also prefer to lat every piece have it's own physics model. That makes it easier for customers to mod the build to their liking. I think I have the physics thing down enough to make a good low impact physics model. And I actually like tinkering with things like unwrapping and physics details, but that might be because I'm still new at it. As long as I'm learning from something, it's still fun ...and if you have to walk through it, prim instead of convex hull, I'm starting to get it