Macrocosm Draegonne

well they do have their own vertices, but they're in the same location as the front facing side. Is that still going to cause the same issue? Does the uploader cull vertices if they're in the same location but have opposing normals? I know it doesnt the other way for onion skin, but maybe the flipped normals cause the culling. EIDT: This was happening with a two sided plant mesh, I ended up putting a 1mm thickness to separate the front and back, and the uploader was much happier it seemed. So the solidify is not good for a two sided plant plane? I will test the copy and join method then, its easy enough, I saw solidify mentioned here in the forums several places as the best way to get the back side of the mesh visible so thats why I was working it that way..

I sometimes get this issue too in a different context. if I do a solidify modifier, with zero thickness, and open ends, it places the mesh in the exact same location but flipped the other way, this is similar to a duplicate flip and join. Sometimes when uploading though a ton of the faces and verts will disappear, even though the normals are correct. I suspect that SL is culling them because the verts are in the same location?

But how? The sound only travels a very short distance before being totally inaudible. I have two 1024 parcels next to each other on the same SIM, if I play a MoP in one lot, it can barely be heard at all once walked over to the other side of that lot, and not at all in the next lot or anywhere else in the SIM. Maybe LL has tweaked it some since you last tested it?

Yes my thoughts exactly, if someone wants to hide there IP while online, why is that LL's job? To hide their identity will take a full system approach, not simply expecting various apps to protect them. So the way to close up the falsely perceived security risk is to put the burden on the user where it belongs, I would imagine people who want to be hidden already know such things and do whatever it is they do. What I am speaking to are those who are not at all even remotely hidden or secure, and yet balk at MoaP as though its somehow going to put them in some sort of additional security risk, its simply using the internet, people do it all the time on various devices, and use whatever security measures they choose. And, that happens at a much higher level than individual apps, their connection needs to be secured. Besides, most people do not have a static IP, their ip address can change many times in a day, they change regularly. People are just scared because all that furry fornication with their alts. ?

That's an interesting idea, I didnt know you could target an avatar with sound just for them, that could prove very useful someday! Oh, so that shows some potential for syncing. Ideally everything would either sync or be out of range so not to clash with sounds not desired in that area of the scene. So if the BPM and modulations are to change in the next zone the traversal, spatially speaking, would create an audio fade effect. Another way to do this would be with a HUD, so only one source of the audio, right where it needs to be, just a wildly different way to do a similar thing. Seems like LL could close up those minor "security" risks that scare folks needlessly. I would imagine if someone is really paranoid they would take other precautions to hide their IP and such, and SL, or anything else, could not expose it right? Eventually hopefully LL could refresh MoaP, allowing other things than displaying a full web-browser. Code only version is the biggie, audio only could be cool too, I know we can use data: to include a good amount of markup, and then connect scripts externally, but if there was more markup allowed, external connections would not be needed. There could be a lot of things done with that type alone, and it uses no networking at all, only processes code and outputs over to LSL to display and/or mesh/prims prepped to be manipulated by it (with their own set of rules and restrictions as to whats allowed/will-work). Something like the Notecard system could get an additional type thats html enabled, with a WYSIWYG. HTML HUDs. Has been done already, but MoaP has some caveats that have made it much less popular than it could be. Code generated visuals and audio feedback that can be displayed in various ways, not as a browser, but rather mesh/particles/textures/sounds AI and algorithmic things better suited to certain codebases optimized for such, that can be passed off to LSL after the heavy lifting. I should stop, sometimes I can just ramble on forever if im in a flow state! lol I will go back to my meshing now.

Besides, I never said I was building a SIM with this stuff, I am experimenting, and contemplating future potentials! ie. Stuff that is not actually possible, we can perhaps request from LL.

There is no other way to do an audio thing im thinking of, LSL alone with the sound clips will not do it. Parcel streaming could give different sound tracks to different areas alright, which is the most straight forward way to bring in sound, then mix in some smaller loops here and there around the sim. Thats all fine and well and the normal way to do it with the current stuff. I am not talking about that though, this is something different and if that scares folks from trying it then that actually pleases me greatly! lol... Imagine the fool who goes to an audio experience and yet will not enable audio. XD I have no time to suffer such fools!

Ahh, well thats never happened to me that I can recall, but I bet it could be annoying lol. Pretty much the only things I ever have to derender are some sculpties, overboard alpha blends, and overboard Particle effects lol. Like I was saying though, I would use this in a special setup, people would be expecting it, with signs and notes and such.

Im not too concerned about autoplay, if someones going to an audio experience, there would be a sign saying to turn it on at least for the experience. As far as I know its only macs that have the bug/feature of not autostarting chromium, I have autoplay on, never bothered me in firestorm, I thought that was default setting? I was reading into the cromium and CEF stuff, there is quite a lot that can be done, and more yet, depending on how SL implements it. I bet syncing is possible, but it could be the way SL makes each surface separate, because its surface not audio or code only focused. Its able to display a browser, but that is only one feature (probably the heaviest) of the codebase, it can also display OS UI, and other UI's and displays based on standards code, and then everything that code can do, it can spin up its own threads, and do all the things separate from the viewer and SL server, which is why its got so much future potentials.

Seems I had to use a plane for the alignment object in the edit normals, I was using two verts but that was dumb, lol. The plane gives me a much better result, now I can get back to the texture.

I have been tinkering with this all day, trying to get the normals right on some grass im making. To get the double side for the grass planes im using the solidify modifier, also using the Normal Edit modifier to try and smooth them out more rounded, but the thing is normals do different things depending on the order in which things are done, ive tried several rounds in an attempt to get the best method before I tweak my textures any more & add variation etc. Ideally they would be pointing straight upwards smoothly, but with each upload I seem to find a different result and havnt found the best possible way yet apparently. Its almost perfect doing the Normal Edit modifier first then apply it, then add the solidify modifier after, but that wasnt always true with every mesh lol. ? Im super happy with the li and lod, but the shadows are still not totally on point. Anyone have any pointers on this area?

lol, I am talking about a real thing though. Graphic of Dan Winter, its not related to data compression exactly, but represents the infinite compressibility of the structure, this has many implications, but I was really speaking to the efficiency of compressing things, which the large geometric dataset folks have also found, in their own way, the same principles at play.

Transfer will always be a concern even when speeds are much higher than now, ive wondered why online games didnt take a cue from the scientific community and use a better compressible mesh like a Fibonacci hexagrid for example, it compresses to smaller size and does so faster, and same for decompression. Fibonacci vortex grid lattice of the "reality" variety are infinitely compressible/expandable, its how the reality hologram works, at a much higher functioning level than humans can yet accomplish. ? But even a little is useful. Scientific communities need to be concerned with IO, storage, and running calculations with such vast datasets, this has fueled some interesting innovations in the area.

Ive read some Jira on the topic today, some of the issues people have had, and requests that have been made. Its an interesting area that I hope gets a refresh someday too, there's a good amount of legwork already done in the requests and jira as to what would be desirable. Its already cool and can do a lot though. I will keep my eye on this area and push all the limits in testing it out with audio and maybe some other experiments.

Yeah I was expecting to only tinker and test for the most part to see what's possible, definitely wont burn too much time, at least not until it stops being interesting and fun lol. Seeing IP address seems like something that should be fixable, though Im not sure why hiding it is important, or if someone who wanted to hide it didnt have other smarter measures to do so. lol I have tested a little already, the audio is great with zero lag for me, and since SL does point sourced audio it mostly does what I want already, trick I want is to expand the sound and/or repeat it somehow without copies of the prim. Then yes, as you say, to link them together when required, preferably with script, to keep timings & such, im fairly sure that might be doable from the javascript side though, it mostly depends on the chromium implementation I suppose. If I keep the LSL to a minimum it should reduce overhead for doing an array of audio outputs hopefully. The LSL will be mostly for setting up the "speakers" and turning them off when nobody is around, or its switched off, the js and external libraries can do the really wild stuff, at breakneck speeds too. My experience in testing the visual side of webprims has been that everything works great, the js is lighting fast, but it has some sort of cap on framerate that can make certain visual effects look choppy.

Yes, preferably there would be a toggle between dark/light, if your outside during the sunny hours or an extremely bright room a white theme might be more legible, whereas being inside at a computer in a normal or darker environment a dark theme is far better, especially if you're the type to spend good lengths of time at it. There's not much of an argument for either/or since its situational really. Screens are not nice light reflective surfaces like paper though, they're still totally emissive unfortunately with our current stone-age technology ? eventually we will have reflective pixels. Many studies show that too much blue led light is bad, after a long time it can damage eyesight, even causing circadian and mitochondrial disorders that lead to all sorts of problems.

I read some Sansar dev's did come back to SL side recently, im sure theres a lot of good cross pollination going on anyhow. You can look and see a lot of the dev schedule and info for SL on the website, they're always hard at work on something cool. I would not be at all surprised to see a mesh uplaoder revision in the coming years either, its been on the list of demands for quite some time. Thing is it would probably need other major things like animesh, flexymesh, new li accounting, tweaks to mesh system and such to make enough excuse to bust all that uploader code wide open and revise it.

Are we allowed to edit the SL WIKI? There is actually a lot of good stuff over there, but it would be nice if were integrated with the content from here on occasion. There really is a lot of valuable information here in the forums. Would be nice if we could reference WIKI pages, and do cross linking to posts that reference it. It really is hit or miss how something is going to have been constructed, which is great, but you've got to be discerning on choices, I will switch to wireframe view and check an object before I buy it usually. As was mentioned above, the worn items are the worst offenders, because they're hidden somewhat, no li, only complexity score, which is not enforced. I dont think the tri's are the normal culpret for lag, at least for me, the particle system, alpha blends, and how shadows interact with particular build methods, those are the fps killers for me when I get them. Most of which can be upgraded eventually, and/or avoided when building out a scene or its objects.

Here's a few examples https://howlerjs.com/ https://tonejs.github.io/examples https://www.borjamorales.com/3d-piano-player/ http://qiao.github.io/euphony/#158 https://www.npmjs.com/package/midi-sounds-react https://webaudioapi.com/samples/ http://webaudiodemos.appspot.com/MIDIDrums/index.html https://alemangui.github.io/pizzicato/

Web on a prim is really cool, the localized data technique inspired me a lot, and especially after connecting jquery and other libraries externally, I am a web developer and dabble in a lot of code bases. One idea about web-audio Ill share here with a few questions & thoughts if anyone wants to chime in.... Id love to build a sound track that is spatially experienced, depending on the area and context of the zone/activity, as well as micro-location within the zone, with differing and some crossover blending audio modulation and feedback. This would be fairly minimal in most areas of course, using the howler.js cache plus procedurally generated sounds will mean very little to no bandwidth needed for most things. Except delivering actual sound files (if used) of course, though... if you check out howler.js & other libraries you'll see a minimal set of sounds can be reused and tweaked in many ways, even synthesized in real time to produce any sound imaginable. Can a webprims audio output be amplified and repeated/extended, spatially speaking? Some things id want to triangulate an audio to amplify the sound in an area like a main track etc, not using prim copies as they could go off in timings. Using a library like howler.js & others we can have excellent caching, looping, orchestrating, streaming, all manner of possibilities, all lightweight & fast webstandards based too. Will making these web-audio web-prims hidden reduce viewer strain they may cause? Along with hiding controls & interactivity, & setting to autoplay. Web prims perform well on my system, but im not sure of others. This use case really all I want is the audio, so that would be less strain potentially right? What are the good the bad and the ugly about SL webprims that will effect such a creation and others? Can LSL work those with good performance? Is there something we could ask of LL to enhance this area for us? ? It would be nice to have a few different types of web-prims, something like im discussing here could benefit from an Audio Only type with relevant functionalities. Generally speaking, this whole web-enabled area being expanded upon, with a standards based approach, could be a boon to in world potentials. Imagine HUD's that are all html/css using no graphics files, but looking great, with visual cues & helpers, popups, mouseovers, animations, feedback. Interactive signs/boards/books, that can be translated & edited, using fonts and little to no textures. Connecting cloud based AI to SIM's and objects. lol ? ------------------ Alright I have to stop or ill keep rambling, I will tinker with all this in time, but I am curious what others think and have done in this area, there's not a lot googling since this area is more technical than I think most would want to bother with, but the potentials are too interesting not to at least check it out heheh. I might post some web-audio examples I find below.

Yeah true, im sure there's more responsible things to spend time and money on, but I think this will help me with my blender3d adventures and finally putting some of my creativity out in the realms. Its exciting! Especially considering all the new features and goodies the Lindens are working on that we will get to tinker with.

Thats very interesting, I will be excited to see what they make of that and how it turns out, it would be great to have more details when sorting out our uploads.

It depends on the body you're making the nails for, if all you want to do is make nail art, pick your favorite bodies, and get their dev kits, and make your own textures, you dont need to worry about the mesh, since you'll be making appliers. Google on it, there's loads of info out there, good luck!

Check out Blender Benders group in world, there are loads of experts and newbies alike, they regularly have classes in world too, for all levels of skill and the different areas of using the app. Check out their schedule its really quite a lot of good stuff. I bet you can find some mentors in there too, and have an array of people to bounce ideas off of to help push you along, the group chat is great too

I wish I knew, ive yet to dabble in that area, there should be docs for copy of weight data though. Another thing to consider for repetitious tasks, sometimes they can be automated with a script if you're up to writing one lol. The Avastar forums might garner more responses to your query though im sure there are some avastar users here of course. Good luck! It will be interesting to hear how you progress.