Rhiannon Arkin

I would have guessed that one texture might be less data handling then many small ones. But you have good points about tiles. they are quite useful at times. I have not yet had issues with mixing transparent and non-alpha parts in one texture, using png and alpha blending. I turn it off for the non-transparent material and leave alpha blending on for the transparent one. Seems to work ok. And i don't think that png is never not having some sort of alpha in it. i could be wrong.

Hi I was just wondering, artistic choices aside, is it load weight wise better to make 1x 1024 texture to cover lets say various parts of a building or to have 16x 256 textures? or is it just the same ? R.

Related to efficiency, if i would use the same texture file in two different items, but rezzed on the same sim, would sl be clever enough to load this texture only once?

That's kinda what I'm doing. So far it's been two textures for the house. I can compress it even more, the glass will be just 2 or maybe 3 variations, each very small, the wood as you say all fits into one texture. there will be some wall structure and some addtional things that all together will fit into another texture. I wasn't planning on baking textures for this one. We'll see.

I will explore this material solution a bit more. I dont think i will need 8 materials. My current way of splitting a mesh into meshes is based on uv space, or better said, based on how much of texture info i can fit into a 1k image. Another thought is transparency needs, so i keep transparent, or alpha blended items separate from meshes i wouldnt need alphas for. I am still exploring the art of packing as much data into a texture file as possible. So it sounds i might be able to make one mesh with only one phys object but still have my split intom3 textures. Very cool R.

Thanks. yes. i think that is what happened. I wasn't aware I need to make a phys object for each of my objects within an dae file. I thought the one phys object I assign to the upload is the phys object for the whole collection. So while the phys object might have worked for one of the meshes, the other two meshes default to convex hull and closed the door again. So it seems that I will end up with as many phys objects as I have meshes within one building. seems like a waste in a way, because now i need to have overlapping phys objects on top of each other. Because I need to have separate meshes for logical (texture) groups of the building. I was successful in creating the door after using the naming convention correctly and upload each mesh individually with it's own phys model. So thanks for all the help so far. R.

Hi I did manage to log into the beta grid yesterday. yay. seems it's a spotty grid. However, other then standing at the welcome station i didn't quite know what to do there. need to find some sort of sandbox i guess. but it worked. so i'll explore. thanks R.

I have not considered the amount of cubes vs the amount of objects. but I tend to combine logical groups of objects into one mesh (combined). so this house is 3 mesh objects uploaded as one dae file. The physics layers is a number of cubes, which do intersect and are then combined to one mesh and exported / uploaded in one dae file. So i think here is clearly something i didn't pay attention to. Are you suggesting that the phys mesh should also consist of 3 mesh objects as the house does?

I did find a few items i'm baffled by: -- some animals... if i look at their LI it says 2, under 'more info' the numbers constantly change, as if these things permanently download. ? -- a disco ball: same as above. dynamic, constantly loading something. not sure what. -- a few palm trees with 8 and more 1k images as leaves. ... -- some terrain images were 1k too. i changed them to 512s. removing those helped for sure to reduce what we perceived as lag. and my red bar in the viewer corner has gotten a lot calmer too. I still feel there's even more i can do. If I have this flashing/dancing red bar in my viewer. is there a way of seeing what actually is loading in a more detailed view?

thanks I do own the sim, but i can't find an option to turn off pathfinding. where would that be?

that's basically how i do it too. And it worked fine, except that I never managed to get doors/openings working. never yet. i must be doing something different. It looks correct in the uploader preview. But I usually do analyse based on info i found here on the forums. Do I have to NAME the phys model in a particular way?

I switched it to "prim" but in my last uploads i didn't have that option anymore. strangely. I am building the collision object out of cubes, is that an issue ?

I know much talked about but still nothing seems to work. My house is a big house and i'm just at the beginning of building it. and it has huge barn door sized openings as future doors. But no matter what size, shape of physics i develop as a seperate model, the barn doors are always closed. When uploading model and physics model it looks ok in the uploader viewer, the physics look correct, even after analyse. and without too. But once uploaded. the physics model has always everything closed. so i can't walk into this building. It's nuts. is there an known issue with the uploader? when turning on metadata--> physics shapes it looks like default and not like the physics shape i selected (and which appeared correct on the uploader window) . also, related to that... how do I 'really' get onto the beta grid. it seems to be always inaccessible when i try. ??? the attached image shows from top down. the mesh uploaded the physics shape which doesn't work the uploader window (bot left) and how it looks in maya. any help greatly appreciated R.

wonderful! Thanks for all the suggestions, i am sure this will help me optimizing the sim performance. Rhia

Hi all so I heard that rumor that a terrain with hills is more laggy then a flat terrain. is there anything to that rumor? I have a new sim, and i am fighting to find the source of a naughty lag. can't isolate it yet, but since i have sculpted the terrain a little bit I wanted to double check this rumor about hilly terrain being more demanding on the client, thus creating more lag. ?? thanks for your insight Rhia

Thanks. this looks like a nice approach. I haven't worked too much with states yet. This seems like a good reason for using states. I'll see if i can update my script to this approach. If it makes reading the nc faster ?

Thanks a lot for the info. Seems like a new concept to wrap my head around. Interesting. Right now i dont see how this permission system relates to an increases ability to dealing with data.

I keep on hearing this Experience thing. Where can i find more info about it. I have not really an understanding about what this is. I have written it in such a way that it reads my notecards only at the beginning once. So it's just a matter of rezz patience. But i am curious about how to script this more elegantly while leave at least the convenience of notecards. The second script approach seems interesting too. But this experience? Cheers R.

The prefetching you are describing sounds like what i was thinking of. How would that be accomplished in lsl?

Hi So i have that notecard reader in my script. And it turns out to be super slow. My notecard has about 100 lines to read, but it seems, that it loads each line from the server.? Is there a way of loading a nc into local memory before reading it in a script? Cheers R.

thanks for the indepth explaination. It sounds like that's where I should go in future versions of my situation. I'll read more into the json thing, als strided lists sound exactly applicable to my situation. thanks for the feedback

in essence I'm just storing a list of questions in each of the lists. the questions come from three notecards. I want to use globals because reading questions from notecards seems to lag more. So the basic solution of usign 'if' for the three cases is working fine. I find the suggested json lists interesting, but need to read into this. as right now i have no idea what the json thing really is.

hi yes you understood perfectly, that's what I'm after. I guess i was looking for a way of doing this with one loop instead of 3 if's. Even more generic so that even the quest1,2,3 are result of a formula. I am looking into stridend lists too.

hi Ok. this is embarrasing but i can't figure out how to do this: i have 3 global lists i want to add information to (quest1, quest2, quest3) looks like a job for a loop, but string nameOfList = "quest"+((string)counter) is resulting in "quest1" . fine. now this is the name for the list i want to use BUT l = llGetListLength(nameOfList) .... doesn't work as it projects the content of nameOfList and not the list called nameOfList. and the same problem on the other side of the = nameOfList += "blabla" ; clearly doesn't work because nameOfList is not the list itself but only it's name in form of a string variable. I couldn't find an example anywhere so I have to ask here, how can I generate the names of existing lists so i can use them in all sorts of loops? any thoughts about this ?? thanks R.

I am not yet succeeding in what you try to do here. This dataserver thing is a bit enigmatic to me. Maybe i need to load all notecards in the state entry so they are loaded when i get to the touch event. i'll play some more. but it's not really logic to me yet. thanks for your help R.