Jump to content

arabellajones

Resident
  • Content Count

    162
  • Joined

  • Last visited

Everything posted by arabellajones

  1. So far, the roll-out has been going on for about eight hours. Either somebody has forgotten to send out the completion notice, or something is going wrong. If this is the expected behaviour, next Tuesday is going to be hell.
  2. I put together a mesh house for the first time, a couple of weeks ago, there were a couple of things I came away with. 1: Make a mesh physics model, and do not set physics to Convex Hull That physics model doesn't need holes for windows. 2: Because a house is so big, the LOD switching distances are huge. The uploader-generated LOD models can look broken but you can check before you upload. There can be problems with long, thin, triangles. I'd thought the render-weight calculation discouraged them, by including a measure of the maximum triangle dimension, but that seems to have changed. What I thought happened was that if you split a 4x1 quad into 2 2x1 quads it triangulated into twice as many triangles at half the cost per triangle, and now I am wondering if I imagined it.
  3. The textures are still at that link. Robin Wood's templates/UV-maps are also useful at https://www.robinwood.com/Catalog/Technical/SL-Tuts/SLPages/AVUVTemplates.html Since the new Bakes-on-Mesh looks as though it may have problems with mesh bodies that don't match the classic avatar UV map, it seems worth a reminder.
  4. OK, thanks, that makes sense, but why the hell couldn't you just say so? The sticking point is going to be the interaction between the basic avatar and the mesh body. If you can't maintain the current invisibility and also use BoM, it looks as though I'll be using a horribly more complicated mesh avatar if I want to use BoM. And forcing Alpha Blending worries me. Right now, I'm thinking BoM is going to be useless for me
  5. It's probably how my usual mesh avatar body, looks good with low complexity, uses a different UV map, but how much does it matter? My first guess is that the BoM system doesn't give a monkey's about the UV mapping. It's never mentioned, but why should it matter? The samples pictured use a default texture, there's signs of shape-related distortion, but it looks like what you would expect. The demo is a universal wearable that uses classic avatar UV mapping So why does nobody mention this? These universal wearables have to be made with a shape that matches the mesh body, and they don't have to have the same UV map, but the alpha mapping effect is going to be messy if they don't. It's as if the Lindens writing the documentation have never made a rigged mesh, and never had to work with UV maps. I have used tools, made for the Poser program, which will translate textures from one UV map to another, so it can be done, but what does the Omega Applier system do? Again, nobody says, but it's simplest if the target meshes all use the standard classic avatar UV mapping. I have fitted mesh cloths on the Marketplace, but I don't use this classic UV map. I have some sweaters, and they're set up to reflect the parts of an RL sweater, with the direction of the knitting constant, so I can used a tiled normal map, easy to download and using RAM efficiently. It looks pretty good. If BoM is UV neutral, it's not going to be a problem. Yes, I know, you can't use BoM for anything but diffuse channel textures, but you have to use the same UV mapping.
  6. I do know of one mesh body that has the onion-skin clothing layers as a distinct rigged mesh, essentially a piece of clothing. I do make mesh clothing of my own, and the whole business looks a bit tricky to set up but it might be a useful option. I'm picky about this, I reckon there are some horribly-built mesh objects, and it's going to be a while before there's a BOM viewer I can run, but it might be worth the effort. Practically, having skin and tattoo distinct from clothing might work very well. If you can have a set of clothes on an onion-skin add-on, changing them gets easy. And I don't think BOM throws away the idea of appliers. But I have been using a mesh body for so long that I can't really remember how texture baking handles multiple layers. Layer 2 sits over layer 1, so the application sequence matters?. but I can only recall multi-textures that don't overlap. I suppose I'd better mention the mesh body I'm referring to: the Psicorp Avatar Core.
  7. I tried to set these up on an item today. It was a semi-historical paint scheme for an aircraft, a small passenger/transport type. Every time I tried to submit the data, something, somewhere, terrified the artificial stupidity system into changing the rating to moderate. Nowhere did the responses point to the offending text, and it seemed to be a word I had used in multiple places: Delete all the text in keywords, and the error report changed to point at another field. I suppose somebody is going to say that if the system gives any specific information about where some bad word is, it becomes too easy to game. Well, if you're that scared, don't Google for Kenneth Williams singing "The Marrow Song", because I expect that will provoke paroxysms of pearl-clutching all along Battery Street.
  8. The instructions for the Search (the box at the top of the page) look broken. I used two words, entered as "word1 word2" and the results clearly use the OR function, returning every entry containing either or both words. The suggestion to improve the search was "word1 AND word2", but that treated "AND" as just another word, not as a logic function. Google did return a few relevant results to the same basic search string, but may have stopped indexing the forums last year. I did try digging for a JIRA on this, but I only read and write English
  9. For me, all the viewers have a problem with not picking up on the OS defaults I set, such as font and mouse pointer size. I also struggle with some colour schemes, which Firestorm handles well, the SL viewer doesn't. Yes, there is an add-on for the SL viewer, but I've had it broken by a viewer upgrade. The basic colour scheme is pretty poor. I struggle with the JIRA systems that Linden Lab and the Firestorm team use. I am not convinced it is a viable bug-reporting tool, however good it might be as a tool for planning the work being done. I end up feeling that I need an exorcist more than I need a programmer.
  10. This is something I have been watching with my system for the past few days. I have a program that displays the readings for various temperature sensors. I'm running Linux, so the specifics won't be much help. What I do see is the video card temperature rising, so dropping back the video settings could help. It depends on the video hardware, but the Advanced Lighting Model setting can be more efficient. Firestorm can also limit the frame rate, which will help a little. (There are some jobs I would not try running in the current weather.)
  11. It is still lousy documentation. It needs improving.
  12. A Project Viewer for Estate Access Management came out in August 2018. There were a few mentions of it then, but I've not found any definite source for when it merged into the main SL viewer. I assume it must have, because now it's part of the latest Firestorm Viewer. The only thing I can find which might pass as documentation is a pair of article in Inara Pey's blog, which assume everyone knows what an Estate Manager is. I eventually found this old web page on the SL Wiki: https://wiki.secondlife.com/wiki/Estate_Manager I am not an estate manager. I can ban people from the parcel I control. I can see how being able to get at least some of the info, such as a banned resident's last-logged-in date can be useful to me. But I am not an estate manager. I reckon, with the Firestorm-led hype, that it's time to update that web-page to cover the EAM features, making it clear who can, and cannot, use them. In the past, extensions to section-A through feature-B have suffered from only having B->A links provided, so that a search which finds section-A doesn't have a chance of revealing feature-B. I think one web-page, covering Estate Managers and the current tools, is enough
  13. What has started me off is the lousy optimization to the Mole-attributed items of the SL16B freebie outfits, up to five 1024-size textures for a pair of shoes, with two of those textures for the inside of the shoes, not visible while they are worn. It might be the poor mole was only responsible for loading the versions distributed, but somebody at the lab signed off on this crap, and they couldn't even be arsed to upload a smaller version of the usually invisible textures. This solution depends on finding somebody competent. I doubt I would be up to the job, but on the evidence I don't think a Linden would recognise competence if it sat on their face!
  14. It might not be the ideal solution, maybe a last-resort one to do with support guidance, but the relevant setting is likely to be in one of the preferences files which can be edited while the viewer isn't running. Looks like it's in user_settings/settings.xml and search the file for "TextureMemory". Either I am in totally the wrong place, or setting the value to 512 would fix it. Is it really that simple to fix?
  15. I can just remember a few things that RLV made possible which were replaced by other viewer functions. One was a system of photographic filters which now seem to be handled by Windlight. And, as long as everybody involved is having fun, RLV does have a place in sexual roleplay. I can think of things I would agree to do. There are a few things I can think of where RLV can usefully automate something, but somebody could do the same thing manually. What I would run away from, very very fast, would be the sort of guy who can't ask, who can't communicate, and thinks RLV is all that's needed. It's like the AFK places I see. What's the point? The big feature of SL is being able to communicate with other players. If you don't want to do that, why not download a porn video?
  16. I know the place where this happened, but as long as there is nothing more than that automated acknowledgement of an AR, I am not sure that anyone can be sure whether the Lindens do anything. I can think of a couple of places where these creeps could have been standing, outside of Nor's area of control, where they would be able to harass new residents directed to info and teaching facilities. There needs to be some attention paid to 24/7 support cover, because Second Life is global. There are some long gaps, and the daily variations are, very roughly, 28k users to 45k. I'm not sure how you could pick out events such as this from the AR stream to direct to an overnight team, but just waiting for normal office hours starts to seem reckless. I hope the Lindens can do better than that on the computer security side, because they're holding my personal data (and just shoving the responsibility over to Tilia doesn't change the timing problem). I'm in Europe. Does Linden Lab (and it's subsidiaries) want to be in the same legal mess over European data as Google and Facebook?
  17. There are many websites that have been reporting basic stats, such as concurrency, for all the time I have been using SL. But all of them seem to have stopped getting updates at various times over the last couple of months, most recently https://eregion.kicks-ass.net/munin/servers/akari.eregion.home/SL_Online_Now.html which stopped updating on Tuesday. It's less clear when https://e*****up.com/slstats/concurrency/ stopped getting data, but it seems to have been a couple of weeks ago. Point is, it's not all at the same time, so it's unlikely to be some change to the API. And Firestorm is still reporting a changing concurrency figure. These charts did give some sign of connection problems, you could see if there was a sudden drop in numbers, when it had happened, and whether it was continuing. And, unlike a web page from Linden Lab, it didn't vanish when somebody plugged in a vacuum cleaner. A wild guess, but I know some sites watch out for repeated requests from the same IP address, and block automated requests when they exceed a limit. The requests made by the Firestorm viewer don't all come from the same IP address, so that sort of filtering would fit what I see. Add a bit of security by obscurity to the thinking. It doesn't help that I am using Linux: there's not much chance of getting a straight answer from Linden Lab. Also, I'm in Europe, and I am usually tucked up in bed when those in-world meetings with the Lindens happen. So somebody else is going to have to ask.
  18. The viewer colour scheme, and there is no supported alternative, is pretty horrible. Yes, there is Starlight, but the default colour settings make me struggle to read anything, and no viewer seems able to pick up the OS setting for the mouse cursor. This is not good UI design, and my eyes are old enough to notice this sort of thing. So the viewer is efficient, but is that enough to make it good?
  19. This is one of the tricky settings. What I am seeing my viewer built-in Help telling me is that this only affects UDP traffic, which has been shrinking for years. Stuff such as textures have been shifted to http. Just that is a huge difference. There are recommendations for the setting, based on the type of connection, and they haven't changed since the days when everything was UDP. That does make me wonder if I can trust the advice. If the UDP traffic has been halved, it wouldn't surprise me if the minimum recommendation hadn't dropped that much, but no change at all just doesn't make sense. This is one of those things where any definite answer is likely to be wrong for you. All you can do is try. That 1500 setting is the absolute maximum setting recommended, labelled as "cable", and all I know is that I can't get "cable". That tech used to be high-speed co-axial, and my current DSL is faster than some "cable" tech that was being used a decade ago. And using the Wi-Fi label, the standards for that have changed too, it can be much faster, but the number of wi-fi systems I see as I walk along the street with my smartphone, all from router/modems using the same out-of-the-box default channel, I have a feeling that anyone using wi-fi, for anything, needs to do some serious thinking about how it is set up. I think I would stick with the Wi-Fi label because of the bandwidth sharing with your neighbours, a totally different class of problem, but "DSL" and "cable" are looking antiquated as labels I would say you have to try different Connection settings. You might even get better results down at 250 or 300, but I would try 1500, 1000, and 500. There's no point in going higher, there just isn't going to be that much UDP traffic now. And Second Life is rather unusual in how it shares UDP and HTTP. Keep an eye on packet-loss rates.
  20. As you understand it, but why can't they make it clear? The inactive account thing looks to include just logging on as activity. I'm not 100% clear whether that includes logging in to Second Life, Sansar, or whatever else Tilia might provide these services for. I remember some of the past problems there were, with even just buying a few L$, for people outside the USA. The US financial industry seems so primitive, compared to the rest of the world.
  21. The computer I am currently using is about ten years old, with a much more recent graphics card. The old nVidia 750 I used was pretty slow, but OK for standing around at an event. The replacement I got makes other things more comfortable. Computers which are that old can fail for all sorts of reasons. I don't think I am using anything super-powerful. And I am trying, when I make stuff, not to waste the computer power my audience might have. I have seen a fancy dress with an ARC of over 1 million. I did look, once. It did did nasty things to the frame rate that old 750 was giving me.
  22. When I look at some of the behaviour of non-rigged mesh in SL, I do wonder how many creators know what they're doing.. You can make vehicle wheels which don't collapse into lumps of triangles as you move to the side of the road: I've done it. ARC is used by the Max Complexity limit in graphics settings. You know, the Jelly Dolls. It's a poorly implemented pieced of UI and to see anything over about 350k complexity you have to switch off the limit. That's risky, because of griefers with graphics bombs. As far as I can tell, the LOD switching is badly calculated for rigged mesh, making the Medium->Low and Low->Lowest switching mostly irrelevant, because of Draw Distance settings. but it still seems to happen, and the ARC calculation gives plausible answers. We have confusing documentation (and I have ranted about that) but I don't see any signs of things not working as documented. Could things be done better? Yes, but I am not confident that some of the big name, high sales, creators could optimise their way out of a soggy wet paper bag. Rather than learn just how to do their job, they start screaming "It's a bug!", and go on to do nothing. And you know what started me off on this: people using 1024-size textures with pointless alpha channels. Let me check, rez the original version of this pair of shoes, so it does't invoke the rigged-mesh bug excuse: 6264 display weight. Used reduced size textures, without the alpha channels, and I get a display weight of 2998. The LOD switching distances shown by the Firestorm editor look right for a mesh object of that size. Let's try the Inspect right-click... 1024-size textures with alpha channels: 12,288 KB VRAM Reduced-size textures with no alpha channels: 1,536 KB VRAM Yep, it's an insanely high triangle count as well, but textures are so easy to re-scale, and it looks like these people don't even bother to try. The Download/Streaming weight reported doesn't change, which is a bit odd.
  23. Penny, I agree about memory use having more relevance than than texture size, but they still have to be downloaded, and having better use of memory by the viewer isn't going to help download, streaming, or whatever other label the Lindens are using this week. I can already set my viewer to not display larger than 512-size, but there's no way to not download the 1024-size original. Besides, pixel count, in the data or for the size on screen, it what the graphics system works with. There are shoes in these SL16B sets which have the laces modelled and textured, distinct faces around 10cm across. Why is a 1024-size texture even being used for something that small? That works out at 10 pixels per millimetre. I reckon the level of detail in the model is a bit wasted, but as a texture? (The same shoes have a texture for the lining, the inside of the shoe that nobody sees while it is worn, and that's another 1024-size!)
  24. Please note that the texture system doesn't have any connection to the LOD system, calculating mipmap usage independently of LOD switching distances, and maximum texture size, which might never be seen by a user in-world, but which has to be downloaded and processed by the viewer, does affect the ARC number. Since the Lindens have known about this for a long time, and said sweet fanny adams about changing how it works, persisting in calling this LOD behaviour a bug, and ignoring it, seems like foolish wilful blindness. If that's how the system works, why aren't you using it? Why do you persist in using high triangle counts for the Low and Lowest LOD models, which still have to be downloaded and stored locally? But, before you get all snooty about this, look at what you make in wireframe mode. If the edges of the triangles are so close together that the item looks solid, can your opinion on optimised content be worth anything?
  25. I'm not sure they would do that, some would start screaming about content theft, but 1024-size images are all too common, and the way the graphics sub-systems work on the viewer you have to get very close before you see the full resolution version on clothing. For a good example of essentially unoptimised clothes, The Lindens have given us the four SL16B outfits, with 1024-size textures and very high mesh density. The way the LOD-switching distances are calculated for a rigged mesh, graphics are weighed down with sub-pixel triangles, and the download weight rises a lot, hurting sim crossings and teleports. My usual avatar, mesh body, mesh clothes, and all, comes in at around half the ARC of one of these outfits. Just the texture sizes would make a big difference, with minimal effort. These things are choices, I can see reasons for 1024-size on a large object, but I doubt the competence of some creators who use them.
×
×
  • Create New...