Jump to content

Scylla Rhiadra

Resident
 View Profile  See their activity

  • Posts

    22,657

  • Joined

  • Days Won

    228

 Content Type 

Everything posted by Scylla Rhiadra

  1. Scylla Rhiadra
    Take a look at the LL wiki page on PBR. That is actually sort of what it says. "The term itself is an abbreviation for a collection of complex mathematical algorithms that attempt to accurately represent the ways that light reflects off and interacts with objects in the real world. In the real world, it is the behavior of light on a piece of metal that allows us, the observer, to recognize “that object is made of metal” without actually reaching out and touching it. The way a metal reflects light differs from that of a polished plastic or some other material, and these differences have been quantified by science. By mimicking real-world physics principles in the virtual world it allows for the creation of more immersive recognizable realistic spaces, but also it helps us relate to fantastical worlds a little better too. While we may not be familiar with what a newly imagined creation is, a metal's inherent metal-ness and aged wood's inherent wood-ness remain constant, making it easier to intuitively understand what we are interacting with in a virtual environment."
  2. Scylla Rhiadra
    In practice of course you're right, but it's more "realistic" in the sense that the surfaces of PBR objects are more sensitive and reactive to "light." The qualities of the object itself (except of course, not really: these are a function of artificially purpose designed material maps) are what produces the particular pattern or look of the reflections and shininess. In BP, shine is more generic and less anchored in the object.
  3. Scylla Rhiadra
    Just to make the relevance of this a bit clearer. If we're judging the success, failure, or value of PBR by assessing how "real" it makes things look, we are, in some sense, asking the wrong question. Put a straight-up, unprocessed picture taken in SL at Ultra graphics settings and PBR next to one taken in RL, similarly without processing, and at least 95% of time we'll immediately know which is SL. But that doesn't necessarily make the SL picture less pleasing, if it looks like a really good 3D rendering that follows the conventions we've come to accept from computer games and CGI. "Bad looking" landscape fails not because it doesn't look like real landscape: it is, with current technology, impossible that it should. It fails because it doesn't look how we've been trained to think "good looking" computer graphics look.
  4. Scylla Rhiadra
    I'm neither trashing nor praising it. I'm arguing that it is producing a conventionalized, rather than scientifically "real," version of the way in which light plays on material bodies. The idea is that it reproduces the mechanics of real life, but that's a literal impossibility as it is literally not "light," nor are we seeing an "actual" reflection. It can look very good, and very pleasing, regardless of how "real" it is -- just as an SL photo -- or an RL watercolour painting -- can look good and pleasing through their employment of conventions that we have come to recognize.
  5. Scylla Rhiadra
    They don't actually; take it from someone who has spent a LOT of time playing with shadows for SL pics. It uses advanced mathematics to simulate closely the same effects. The same is actually true of linear perspective in Second Life, which does not work in quite the same way that it does through our eyes in RL; the code uses mathematics to produce a believable simulation. And the same is also true of 2D painting and drawing: linear perspective visual art uses tricks and conventions to produce the illusion of depth, but it's seldom an accurate replication of real depth as we perceive it in the world. This is a somewhat different effect than I'm talking about, but one that is certainly legit.
  6. Scylla Rhiadra
    Indeed there is not. "Orange" wasn't a colour until the 16th century: it literally didn't "exist" as a distinctly individual tone.
  7. Scylla Rhiadra
    I have no idea what you're talking about.
  8. Scylla Rhiadra
    You're missing the point of what I'm saying, and of the reference to theatrical makeup. If you put a real person on a floodlit stage, or in front of a camera under lights, and take a picture, you are getting a picture of what they really look like in that situation. A document that shows how the lights and so forth changed how they look. And we would find that washed out or sickly. The person wouldn't look like we think they "really" look, even though it actually, literally is how they really look. So we put the person in theatrical makeup to artificially make them look like what we think "real" should look like. Let me give you another analogy from architecture. The ancient Greeks knew that, because of way the lens in the human eye works, a series of perfectly straight columns will not look straight, even though they are. So when the Athenians built the Parthenon, they deliberately distorted the columns so that they would look straight, even though they weren't in reality. That's PBR and all the graphical bells and whistles. It's producing a "look," an aesthetic and visual convention of what we "think" real would look like in a 3D context.
  9. Scylla Rhiadra
    That is indeed the theory, yes.
  10. Scylla Rhiadra
    This isn't the place to get into this here, but I think that the analogy with theatrical makeup is exactly right. SL, even with PBR, looks nothing like RL. For that matter, the very best Triple-A computer game, with all of the latest graphic enhancements, looks nothing like RL. A straight-up photo, unenhanced, from SL is almost never going to be mistaken for an RL photo. I've thought about this a lot since getting heavily into SL photography. In theory, the Gold Standard for an SL snapshot is that it looks "real," but the only way to really manage that is to process the image in a way that makes it look like an RL photo that has also been processed in that same way, thereby erasing the telltale signs of SL's alien look. I actually don't think that what most people are getting excited about with regards to PBR is enhanced "realism." I think what we want, even if it's unacknowledged, is really a Second Life that looks more like a really top-notch computer game, or a well-done CGI movie.
  11. Scylla Rhiadra
    Someone has produced an easy-to-follow, step-by-step guide for ordinary residents who want to update EEPs to make them PBR compatible. Spoiler alert: it wasn't LL https://wiki.firestormviewer.org/pbr_updating_skies Mostly, this is pretty good.
  12. Scylla Rhiadra
    Yes, you're right, I suppose there's a lot of potential there, although most of those effects could have been produced used Blinn-Phong normal and specular maps, and most often haven't been. Again, I haven't seen a lot of evidence of clothing that has used normal maps to enhance the look of knits -- although I have seen some pretty shiny sweaters. I think some of the better clothing makers might be able to run with this -- and now that it will actually be visible to everyone (once PBR is universal), there will be more incentive to do so. Furniture, yes. There's definitely much that can be done. PVC and sequins. We're going to see LOTS of sequins.
  13. Scylla Rhiadra
    It's an interesting question. Most clothing doesn't really benefit much if at all from PBR. Shiny stuff, like things made of leather, latex, or satin, or with sequins might (and I imagine will be seeing a short-term flood of such items specifically to take advantage of PBR) will benefit, but a basic cotton tee, blue jeans, or sneakers? Not so much. So I wonder if clothing will be one of the last sectors to really embrace PBR, given that there's not a lot of point to it for most garments? (The other aspect of this is that clothing creators frequently haven't even taken advantage of Blinn-Phong materials. How many garments use a normal or specular map? Not many, I bet.)
  14. Scylla Rhiadra
    Or dishonest ones.
  15. Scylla Rhiadra
    Sure Luna. Whatevs.
  16. Scylla Rhiadra
    I'm not sure what you mean by this. FS has a team that provides support for users in the variety of different in-world support groups they run, and the members of that team are of course residents. There is nothing analogous to that for the LL viewer. No official in-world source for assistance, nor even a resident-run one.
  17. Scylla Rhiadra
    Absolutely. And no "solution" is going to catch everyone. But I can name at least one person who would have benefited from simple and brief bit of documentation on all of this. Me.
  18. Scylla Rhiadra
    I had forgotten that the LL viewer does that! It tends to have a rather "technical notes" look to it, so it might be redesigned to make it at least look a bit more inviting to the average resident, but this is maybe one reason (the most important being of course that most people don't use that viewer) that we heard less anguish when it updated than we are now.
  19. Scylla Rhiadra
    Well, yes and no, I think. There are people "like this" who are "unreachable," of course. But I think there are a great many who could have been better informed or at least alerted had LL made some effort to do so -- using a large popup or whatever on the login screen for instance. I think more importantly, however, what could and should have been done is to have produced a brief and simple explanation and guide for the layperson that people could be directed to. Something that doesn't waste their time explaining that "legacy textures will now be referred to as Blinn-Phong," but that gives the kind of simple explanations and advice that are now scattered almost at random through this forum. Then every upset person rage shouting here, or in a group chat in-world, could be directed to a clear, concise, and coherent document that provides immediately useful information. Instead, LL has left it to the FS support people to deal with this on a one-for-one basis. (Because of course there is no support at all for the LL viewer.) On most of the software or online platforms I use, an update is accompanied by a popup or whatever that explains "What's New!" Because most software companies recognize that pissing off their customers without providing so much as an explanation, yet alone a remedy, for changes is bad business. Not LL, though. They happily outsource such trivialities to us.
  20. Scylla Rhiadra
    I'm sure Arton is entirely correct with regard to the code foundation for our new working mirrors -- but I actually had screenspace reflections turned off when I took this shot (or shots, actually), because I've been told they are a crappy approximation that has more to do with BP than PBR, and that eats up resources.
  21. Scylla Rhiadra
    lol sure. I'm sure the platform will survive; there's not going to be a mass exit or anything. But it's all so stupid, predictable, and avoidable.
  22. Scylla Rhiadra
    Yeah, it's a crappy hack, and there's next to no documentation on how to deal with it, or what is causing it. I may have to leave RL. We all may have to leave RL . . .
  23. Scylla Rhiadra
    I think that was when people were using a PBR hack to simulate crude mirrors by employing materials with high reflectivity. We've got more robust, or at least dedicated code for them now.
  24. Scylla Rhiadra
    Yeah. This describes probably 80% of those I know and socialize with in SL. They might spend 8 to 12 hours a week here. They're not gamers, and they're using multi-purpose off-the-shelf machines that they were told were a good price for what you get at Best Buy. They know exactly as much about the platform, and the tools on their viewers, as they need to do the relatively limited things they enjoy in SL: socializing, dancing, a bit of shopping and dress up. They're not going to waste an hour a week or so popping in here or perusing the SL blog on the off-chance that there's something important there to see (an occurrence that happens, what, maybe twice a year?). All they know is that they updated their version of Firestorm, as they usually do when told, and suddenly found that SL was slower, harder to run, and didn't look as good as it did before. And expecting them to dig around in the forums or on external web sites to find the information about how to fix these issues, information that they might reasonably expect the platform itself would provide, is ridiculous. There's an awful lot of people who fall in this camp, or somewhere adjacent to it. We can tut-tut all we want, but they're investing exactly as much time in the platform as is worthwhile to them, given how they use it. Most of them will probably stick around anyway. Some will get help from friends. Some will leave. If they do, it's not a moral judgment on them, but it's a pretty good indictment of incompetence of those running the platform they're abandoning. And their loss will matter.
  25. Scylla Rhiadra
    Thanks for the info, Henri! I usually use a view angle of about 0.500 for shots (corresponding more or less to a medium telephoto lens), including this one. The mirror resolution was set at 1024; I might experiment again bumping it up, but this was pretty rough on my system as it was. I had been wondering how accurately the mirror was reflecting. It looks like it's at least a decent approximation. Until performance improves (or I get a better computer), I may stick with plan B, which is to compose mirror shots the way I always have, using multiple pics including a separate one showing what is reflected: it'll give me a better quality reflection, and more control over how it looks. In that scenario, the key advantage of these mirrors is that it calculates the angle of reflection for me: I just need to approximate a shot that reproduces more or less the perspective shown in the mirror.
×
×
  • Create New...