Penny Patton

Advisor
  • Content count

    1,110
  • Joined

  • Last visited

Community Reputation

244 Excellent

About Penny Patton

  • Rank
    Advanced Member
  1. Does LL want to fix SL

    Too often, when a new feature or a long overdue bug fix is suggested, people only look at whether or not it will have an immediate impact on the SL experience. Whenever discussing potential bug fixes or new features for SL we need to take the long view and realize that the impact those features are intended to have will not be realized right away, but over the years after the feature is introduced. This has been the case with every change made to SL.
  2. Does LL want to fix SL

    This is a justifiable concern. My view is that LL should have done this in the beginning, before releasing mesh to the main grid, to avoid fracturing the clothing market. As it is, the clothing market is fractured already, unless you support every mesh body on the market AND the old system body, your potential market reach is far smaller than what it used to be. What's more, most of the popular mesh bodies are no-mod, meaning everyone using those bodies are much more restricted in how they can personalize their appearance. For many casual users with little interest in modding, this doesn't affect them, but SL used to have a thriving modder culture. It still does, but you have to go furry or anime to really see it anymore. If LL introduced a new, well made, system body with the features I describe, I predict fewer new users would see a need to invest in a mesh body, and those who feel restricted by no-mod bodies would also be inclined to make the switch. It would take time, years, but eventually I expect you would see most SL users utilizing the improved system body. Unfortunately this would mean less business for mesh body makers (aside from those who create specialized bodytypes, such as the super muscular, overweight, or extremely sexualized bodies, not to mention fantasy bodies like aliens and mythic creatures) but it would be a huge boon to clothing makers. Why would LL want to do this? Well, they have to look to the future or they'll find themselves eventually out of business. A standard avatar addresses several problems. It makes a better impression to new users, meaning more people are likely to actually stay in SL after checking it out. It makes SL more user friendly. The system bodies were always more easy to support and use, especially for new residents. Finally, by being a standard it will reduce the fracturing of the SL clothing market, which will help boost sales for clothing makers. People buy more clothes than bodies and LL gets a cut of those sales, too. Here's my take. The SL userbase represents people of a variety of backgrounds, interests, agendas and desires. Even at the user meetings not all of the feedback LL receives is constructive. Linden Lab also does not always seem to seem to put much faith in user feedback, possibly because they are themselves unable to discern good feedback from misguided feedback. For instance, many SL users, many SL content creators, are in complete denial about how poor use of textures can impact performance. Whenever it is suggested LL do something about excessive texture use, these users become very vocal against it. Many of them are very talented people who produce wonderful looking artwork, but they lack practical experience and fail to realize they are creating content that is the source of the very SL problems we all experience daily. It's not enough for LL to receive good feedback if no one at LL is capable of judging the quality of that feedback. For every aspect of Second Life, LL needs someone in-house, on the payroll, with actual authority, that LL can trust. So when someone like Oz or Vir shows up at a user meeting for feedback and one SL user, a professional game artist with 12 years of experience, makes a suggestion but then another SL resident, who sells shoes or mesh heads or something in SL but has never done any real graphics work outside of SL and has a very fuzzy, almost entirely incorrect understanding of how 3D rendering works, starts contesting the suggestion, Oz or Vir can turn to their in house graphics authority who can say "This person knows what they're talking about. This other person is spouting gibberish. The idea is good and I can put together a roadmap on how to implement it tomorrow."
  3. Does LL want to fix SL

    I've always believed that rather than creating something similar to user created mesh avatars, LL should create an updated set of system avatars. Not replacing the old system avatars outright, they'd still be there so legacy content wouldn't be destroyed, but new users would start off with the new avatars, and we'd all have the option to swap between them as we pleased. Improved models, improved UV, improved system skin and makeup (these were always great features, crippled by dreadful art assets), and providing the userbase everything they need to create clothing (system and mesh), skins, what have you right from the start (creating skins, tattoos, makeup and system clothes would all be very similar to the old system av, just with improved UV and the addition of materials support). LL could use this new set of system avatars to also improve the broken appearance editor. Break shapes for the new system avatars into heads, torso and legs so that people can easily mix and match the three, and allow the new system avatars to make use of bento face and hand animations. Of course, LL has stated they will never do this so...
  4. Does LL want to fix SL

    I believe Linden Lab has always wanted to make Second Life the very best product it can be, however...they can't. From the beginning, Linden Lab was created with the mindset of a software developer, rather than a game studio. They believed they only really needed software engineers. It never occurred to them that they might also need graphics professionals and experience designers. Their belief was "we only need software engineers to create the world and tools, and we'll leave content creation to our userbase". That is a flawed ideology to build a studio on. Because LL lacks graphics professionals on their SL team, the content creation tools were all broken from the start. The appearance editor gives us incorrect information because the software engineer in charge didn't think having a consistent sense of scale in SL was important. The shape sliders are all weighted in such a way that it is far more difficult to create proportionate human shapes, because the software engineer in charge had no understanding of human proportions. Second Life only allowed its userbase to use blended alpha textures for over a decade because no one at LL understood why blended alpha was awful for realtime 3D rendering. Land Impact doesn't take texture use into consideration because no one at LL understood the impact of unoptimized texture use on performance. Second Life lacks any sort of competent tutorial for new users, because no one at LL is qualified to create or approve such a tutorial. The long forgotten system skins only look ugly because the art assets used in the system were created in MS Paint by a programmer rather than a qualified artist. This complete lack of respect for skill sets other than "programmer" is built into Linden Lab's DNA. It colours their view of every bug report, every feature request, every interaction with their userbase. Linden Lab lucked out in that there is nothing quite like SL out there. Potential competitors all made worse mistakes than LL, or never made it past their initial stages due to a lack of investors. So LL has been able to enjoy moderate success despite themselves, just on the strength of the idea SL represents. However, check out the attached chart from GridSurvey. It shows SL concurrency numbers from 2010 to 2017. Anyone asking fr evidence of Second Life's decline, here it is. If the decline shown in this chart continues at the same pace, then in a few short years SL will have only half the user concurrency it enjoyed in 2010. That's fully half of SL's userbase gone. Linden Lab has to be aware of this. Their options are to produce new products to provide alternate income streams, or find ways to revitalize Second Life. Sansar is their attempt at a new product. Will it be successful? Too early to tell, but the outlook is not good. There's no buzz. No hype. No one outside of SL users really care about Sansar, and the SL userbase is, on the whole, unimpressed at this point. I think VR will really take off in the next 5 years, but it's not quite there yet. Maybe Sansar will become really successful when that happens, maybe not. Right now it is not a money maker for SL. Little birds have told me that LL's confidence in Sansar is shaken and this is why we're seeing renewed development for Second Life, when just a couple of years ago the Lab indicated that SL development, outside of basic maintenance, would be over with the introduction of Experiences. Linden Lab still suffers the same fatal flaw I described at the outset. They lack people capable of guiding Sansar and SL development in the directions it needs to go. They're blind to the problems facing them and just sort of stumbling around in the dark, hope something they do works. I really hope they succeed but I can't say I'm confident. I'm not saying "the sky is falling". The sky fell about 10 years ago when the SL bubble burst. I'm just saying that I'm not holding my breath for any significant improvement.
  5. Most of the doors you see on those buildings, the cave and mine entrances? You could enter those. The doors were teleporters. Click on the door and you'd be teleported into a skybox maybe 1000m up. I spread the skyboxes out so there were never too many in close proximity, this also helped keep framerates up and lag down.
  6. I should note that my idea for the feature would be optional, just like the ability to not rez avatars outside your parcel. Something every individual landowner could decide for themselves. It would also be two checkboxes. One would make it so for anyone within your parcel boarders, objects outside your parcel would not be rezzed. The other would toggle whether or not objects on your parcel would be rezzed for anyone outside your parcel boundaries. Between those two options landowners would have a lot of control for both privacy and performance. That latter option would also make it easy to use phantom landscapes that extend beyond your parcel border. You'd be able to enjoy the illusion of a larger world outside your land, and your neighbors would never even know of the innocent encroachment. There's really no downside here. Everyone wins. The key point here is that not everyone wants the same type of virtual space. Some wish to be a part of a virtual community, adding their build to an ever expanding, persistent, virtual world full of diverse creativity and vision. That is the idea behind the mainland. A sprawling, persistent, world made up of a patchwork of thousands of creators. It's not a bad idea, and appeals to many. Just not everyone. It's also held back by the lack of content optimization and lack of good building habits which result in a very laggy and inconsistent experience. Better creation tools which could encourage better optimized content (and discourage unoptimized content) would help this a lot. As would a Linden Lab able and inclined to show examples of good building habits to guide their users. Something the Lab has never done. Their builds instead tend to showcase a lot of what not to do and that is unfortunate for everyone. Others want more or less the same as above, but wish to be a part of a more unified community theme. They want to build a skyscraper or have an apartment in a sci-fi city, live in a castle overlooking a fantasy world, or in a bunker beneath a radioactive wasteland of post-apoc adventures. The chaotic nature of the regular mainland prohibits these people from having that experience, so they have to find it in private estates, where landlords enforce the theme. Some, such as Caldeon, Insilico and The Wastelands have done very well with this. Their estates are not a part of that "persistent world" LL imagined, and when private estates first started getting popular a lot of oldbies and even Lindens complained about this, as if it were somehow doing SL wrong. People described it as "gated communites". It's not wrong. It's GREAT. Maybe it's not for everyone, but neither is the single, chaotic, mainland vision for SL. And then there are those who want a more private experience. They want their own tiny little space. A home base where they can relax and bring their friends to hangout. They don't want to deal with neighbors, but they also don't have any desire to spend $300/mo for a sim. Just a small skybox where they have complete control of their build and their surroundings. I'd go so far as to say this describes the vast majority of potential SL landowners. I say "potential" because this isn't really possible. A mainland skybox is a clumsy alternative and most people aren't willing to pay for that because of the problems that go along with it.* So they simply do not buy land at all. LL has told those potential customers that they don't want their money. LL has cut themselves off from the vast majority of their potential income because I guarantee all those people who want their own tiny $5-$25/mo mini-sim experience represent far more potential revenue than all of the estate owners and landbarons in SL combined. Those in this group who do give in and buy some land are those who live exclusively in skyboxes, or throw up ugly walls around their parcels. They're not doing SL "wrong" either. They just don't want the same experience that the first two groups are looking for. And, unfortunately, forcing those people into the chaotic persistent mainland experience is a recipe for making everyone in both groups unhappy. LL is poisoning the well for all groups by trying to force that one, single vision of what they want SL to be. You're not wrong. The problem is that Prim Limits and Land Impact are supposed to help keep render impact reasonable, and it could work if the land impact of content were calculated realistically with regards to the impact it has on rendering. But it's not. For example, textures have no effect on Land Impact whatsoever, despite the fact that the amount of texture memory SL throws at your computer has a HUGE impact on the performance you experience. Overuse of bloated textures (on both avatars and environments) are the number one performance killer in SL. If you could solve that one problem you would probably see framerates for most users double. It's not the only performance issue that needs to be addressed in SL, not by any stretch, but it is a big one. Another issue is that the way Land Impact is currently calculated, it discourages people from using LOD levels efficiently. Instead it encourages people to create content that can only bee seen at the highest LOD, and then tell their customers to crank object detail to maximum. This isn't just bad, it is pure insanity. I's LL encouraging people to create badly made content, then those content creators telling everyone else to crank their settings for object detail so the highest LOD models are rendered at all times, killing their framerates. And people do it, because they don't know any better. Here is something to keep in mind, though. It's actually not bad at all that we can put more objects in a sim than we should be trying to render at once. People just need to be aware that this is the case and use it more efficiently. If you have a homestead with a full build at ground level, but then another full, sim-wide build 1000m up, and another at 2000m up, the objects in one of those builds have no rendering impact on people wandering the others. Not unless their draw distance is set at over 1000m. In this case your main concerns are scripts and bandwidth. Reduce the amount of texture data being streamed to everyone in the sim and you reduce the bandwidth being used. Reduce the script time being used in the sim and people will experience less script-based lag. I wish my full sim version of my fantasy build "Mjolka Kyr" was still around. It was a full sim build. All building interiors were placed in skyboxes. There was a castle, a ruined castle with a dungeon, a graveyard, a mountain area, a forest area, a Skyrim style village, and an ancient Greece style island area, all spread across floating island, all packed together, and yet because of how it was arranged, people experienced fantastic performance. I got multiple IMs from visitors every day telling me about the amazing framerates and fast rez times they experienced. I had numerous people tell me that Mjolka Kyr was the only place in SL where they could turn shadows on. This was when mesh was brand new, too, so most of the build used sculpts and people still experienced high framerates and were able to crank up their graphics settings to enjoy features they couldn't experience anywhere else in SL. I'll attach some screenshots. *I could write a whole book on those problems, what causes those problems, and potential solutions to those problems, but then this post would be a hundred times longer than it already is.
  7. This is very true. The thing to remember about getting better performance out of SL is that there is no one, single, silver bullet answer to solve the problem. Encouraging optimized content, reducing the amount of content your computer is forced to render at once, and bringing down the memory use and render impact of avatars will all put serious dents in the issue, so every method of improving performance in SL should be encouraged. Let me offer my own situation as an example. I have a 4096sq.m. (that's 1/16th of a sim). This costs me far less than a homestead and doesn't impose the homestead 20 avatar limit on me. I have the building know-how and experience to fit more content on this tiny parcel than most people seem to manage with their homesteads, it is the perfect size parcel for me in terms of cost and what I can do with it. I knew from the beginning I was going to exclusively work with skyboxes. The sim was mostly empty when purchased the parcel. I tried to get a parcel surrounded with abandoned land but despite all of the abandoned land out there it is not that easy to get a piece of land like that simply due to how LL handled abandoned land (it goes up for auction, not direct sale, you can file a support ticket to try and get a specific parcel size directly but there is no guarantee). When I built my skyboxes, I did so at a height where none of my neighbors had skyboxes. I did this to keep my sky clear, and so if any of my builds spilled out onto neighboring parcels, those neighbors would not be affected by it. However, that was years ago, now the sim is full. As a landowner, you have no control whatsoever over who moves in next door, and since it costs money to pick up a new parcel (and you cannot always dump your current parcel in a timely fashion) being constantly on the move as sims fill up is not even remotely feasible. One of my neighbors, about a year ago, decided to build a giant wall, at the border of our two parcels and right at the edge of my main outdoor skybox. They did not actually build a skybox at that height for themselves. It seems the purpose of the wall was simply to troll me. Maybe they hoped I'd move and they could snatch up my land. Another neighbor of mine would constantly return objects of mine they believed were too close to the parcel boundary. This tool is meant to stop encroachment but here it was used for no other purpose than trolling. They even told me as much, saying they did it because the tools let them and no other reason. I had to rebuild my entire parcel so they could not continue to do this. I've had other neighbors come and go. I've had neighbors move in, then contact me complaining that my skyboxes were ruining their view of the sky. I've had others move in and filly up my sky. I reduce my draw distance as much as I can but that only goes so far and if I want to do any major building I need to crank up my draw distance, forcing me to download and render literally Gigabytes worth of data I don't want or need. I have worked on several role-playing sims, from Nexus Prime to Doomed Ship to Minotaur Empire. In each case I had a full sim to work with but I still wanted to make good use of skyboxes to reduce the rendering load on visitors, so everyone can enjoy high framerates. Being able to set aside parcels at the edges of the sim where I could place building interior skyboxes that would never be rendered unless you're in those parcels would have been a godsend. Sure, it's not a tool everyone would use, but that's true of every tool and feature in SL. Many people would find the ability to derender neighboring parcels, as well as the ability to have their own parcel contents derendered, invaluable and use it to great effect.
  8. You know, I went to a Linden office hour today just to push that exact same idea. They told me it was a good idea, that they had it filed away for future use, but that they had no plans for it at this time. So it might happen someday. Here's the Jira I made requesting this feature in 2014: https://jira.secondlife.com/browse/BUG-5671
  9. Texturing Transparent Hair Strands?

    Also, the blended, translucent hair textures in your examples are a bad way of texturing hair. You see it a lot in SL (in fact I've never seen mesh hair that doesn't incorporate this mistake and have to fix it myself on hair I purchase) but while it might look great for photoshoots under certain lighting conditions, most of the time it suffers from all of the common blended alpha texture glitches. It will have render sorting issues whenever you're in close proximity to other blended alpha textures, it's a bigger drain on framerates, and it does not play well with local light sources, often appearing much lighter or darker than the non-alpha portions of the hair. Rigged mesh hair that uses blended alpha also ends up with this strange looking "halo" effect which becomes very noticeable in front of other alpha textures, linden water, etcetera. It's similar to the effect of the no-defunct invisiprims people used to use to mask out portions of their avatar. Instead look into using masked alpha to achieve a similar effect. You won't get the translucent strands of hair, everything will have a hard edge to it, but you can still get loose, whispy strands if you do it right.
  10. I wasn't trying to argue with you, only add context to your statements. I never said that all of the same tricks could be used in SL, my entire point was that they can't and that you have to take that into account. If you leave it at "well modern games use much bigger textures than we can upload to SL" I guarantee there is a content creator out there right now who is reading it and believing that statements justifies them in giving each tiny button on their mesh jacket its own unique 1024x1024 texture, normal and spec map, all with unused alpha channels they don't even use because they don't see the harm. That's why I feel it's important to give that extra context. SL is essentially an OpenGL videogame engine that we can all create content for. We have all the same memory issues to contend with (and a few more, besides), but not all of the scene optimization tools a modern game dev has at their disposal. We need to keep those memory issues in mind, as well as what we can and can't do to mitigate them. The texturing rules you bring up are spot-on. I'd add to that a reminder to make sure textures do not have an alpha channel if the texture doesn't use it, it's just wasting memory. While we can't take complete control of a visitor's camera and have sections of the build purged to free up memory for higher detail scenes, we can use tricks like breaking environments up into separate maps (or skyboxes) and connecting them with teleporters disguised as doors/cave entrances/etcetera. I use this trick in every build I work on, whether it's a small 500 prim club or a full RP sim, and it does wonders for performance. This might become more important to people in the future, I've heard rumblings from LL that they're considering employing more resource management tools for avatars and environments in the coming year. Better to start optimizing your content now, before LL rolls out an improved jelly doll feature or raises land impact costs for memory intensive content.
  11. (SL) The end of days

    Thanks, and sure. Normally I keep it private (It's my fortress of solitude, so to speak) but I've opened it up to visitors for a bit. Fair warning, it's pretty adult content-wise. ^^; IM me inworld and I'll send you a TP or a landmark. I do have an SL blog where I talk about a lot of the practices I use to keep framerates high and get the most value out of land space and many other SL issues. Most of the articles are fairly old (many predating mesh) but I intend to revisit a lot of these topics with better written articles, revised for any changes and new features added since the originals. http://pennycow.blogspot.com/
  12. (SL) The end of days

    It's pretty easy to get SL running much better than most residents have ever experienced. Yes, it requires optimizing content, but optimizing content isn't hard at all. The only reason SL content is so unoptimized is that LL has rarely done anything to encourage it and most people are completely unaware of why SL runs so poorly. Look what happened when LL introduced the "jelly dolls" feature to derender avatars with excessively high draw weight. Sure, many people complained or simply turned off the feature, but others took notice. They didn't want their content to fall out of favour because it would lead to avatars being derendered, so they took steps to reduce the draw weight of their content. I recall one content creator in particular with a popular avatar attachment that originally added about 50k of draw weight to an avatar by itself. Before the Jelly Dolls feature they rebuffed any complaints about this. However, shortly after the Jelly Dolls feature was released they relented and released an optimized version of the attachment that looked identical but was much easier to render and had a much lower draw weight impact. Discourage unoptimized content and content creators will produce better content. And my mainland home in SL looks so good that when I post raw, unedited screenshots, people think the shots are from a videogame, not SL. And yet I get higher framerates at my home than I do anywhere else in SL, just because of simple optimization.
  13. (SL) The end of days

    SecondLife has many problems. Some of these problems come down to LL believing they can develop and market a virtual world driven by user generated content without any professional artists on staff to help guide the presentation, develop the tools, and point out easily avoidable problems relating to content creation. Most of the problems we experience in SL every time we log in, from the lag to the cost of land, stem from this one critical error on LL's part. I have not noticed any new network issues with SL recently, but I'm willing to bet the problems you describe are related to worsening of VRAM and bandwidth bloat SL which has been plaguing SL from the beginning, but getting worse as LL provides content creators with more ways to sabotage their own content. If you want to talk about the cost of sims, I can show you how to turn one sim into four complete sims simply through adjusting your camera, reducing the size of your avatar, and building more closely to 1=1 scale. Nobody does this because LL starts people off with 7-8' tall avatars and then nobody wants to be the short guy in the sim, even if it would save them literally hundreds of dollars a month. Some of these problems are a product of LL's inability to understand the draw of their own product. SL appeals to a lot of people, no doubt about it. But if you ask anyone at LL responsible for guiding SL's marketing or development, the reasons they give will either be woefully vague and incomplete, or total gibberish. This is why SL continues to lack critical features, baseline functionality, and seems to have no direction whatsoever to its marketing. This is why we don't have NPCs, why we're forced to deal with the harassment from awful mainland neighbors if we're unwilling to shell out $300/mo for a sim, why it took almost a decade for LL to give us a proper way to mask avatar bodyparts out (remember invisiprims?), and this is most certainly why LL has not been able to give a clear indication of who their target audience is for Sansar. My point is, LL isn't doing anything to hurt SL now that they haven't been doing (or neglecting to do) for the past 15 years. If anything I'd say there's been more positive changes in SL development in the past few years than we've seen in a long time. Experience tools have added a lot we can use to make our sims more engaging, animated mesh is coming (assuming LL doesn't render it useless by tacking on excessive land impact costs to it), we're finally getting long overdue improvements to Windlight, some of which were promised to us in 2007, and LL seems like they finally might be at least trying to encourage better made content through features like "Jelly Doll" derendering of avatars with excessive draw weight. I'm not saying LL is doing a great job now, just that they're taking small but encouraging steps in the right direction. It's too early to say if we'll see any tangible improvement to the SL experience, or if it will be "too little, too late".
  14. (SL) The end of days

    I've always suspected that when LL says "Sansar is a completely different product aimed at completely different people" it's the same little white lie a company like Nintendo gives when saying "The Nintendo DS is a completely different product from the GameBoy line." They're introducing a new product they fully expect to take the place of the old product, but they're hedging their bets in case the new product doesn't do as well as they hope. It's a common, and sound, business tactic to prevent current customers from panicking. Not that SL users are prone to panicking. Still, if Sansar isn't aimed at the same sort of people who use SL, I'm not entirely certain who their target audience is. They haven't made it very clear in their....I was going to say marketing but I have seen no marketing for Sansar beyond videos and whatnot they've shared with the SL userbase and the odd interview given to a tech blog. All that said, "prim drift" is a pre-existing SL problem. It was a problem back when I joined SL in 2005 and it remains an issue today. No conspiracy. No sign of the end-times. Just yet another bug LL hasn't bothered fixing. In addition, LL is not going to try and force SL users into Sansar. Linden Lab may not have ever figured out the appeal of their breadwinner, or learned how to properly cultivate its userbase, but they are very much aware that SL keeps their lights on, and they will go to great lengths to avoid rocking that boat too much. Linden Lab have gone to such great lengths to keep their existing SL userbase that it has actively hindered their ability to draw in new users. We're talking about a developer that fully realizes their overhead default camera placement is awful for an immersive, third person experience, and yet they doggedly stick to it because they're afraid that different camera settings, no matter how much better they might be, might upset the existing userbase. We're talking about a developer that deliberately created SL's default windlight settings to mimic pre-windlight Second Life as much as possible, because they were afraid too much of a change would push users away.
  15. 2K17 to 2K18 Skin Help

    The Extended version of Photoshop can do this. You'll know whether or not you have the right version of Photoshop because where it states the software name (such as when you click "About" in the Help menu) it will say you have "Photoshop CS Extended". Otherwise it will just say "Photoshop CS". If you have Photoshop Extended, then you can download the avatar OBJ files directly from the SL website itself I believe.