Jump to content

Agent Tairov

Resident
  • Posts

    89
  • Joined

  • Last visited

Everything posted by Agent Tairov

  1. Fluffy Sharkfin wrote: Pamela Galli wrote: A bit OT but I use Gimp (and Blender). I finally bought CS5 (Photoshop) a few months ago, I like it fine, but I barely touch it. For the things I do regularly, Gimp is just faster and easier. Personally I'd recommend plugging away with Photoshop, but then being a long-term user I guess I'm biased. I do think that being comfortable/proficient with whatever application you're using is more relevant to creating good content than the pricetag/featurelist of the application itself (within reason of course, you can't exactly compare the old MS Paint with Photoshop). I was a very long time user of GIMP. I bought Photoshop a few years back, and I routinely get work done in 1/2 to 1/4 of the time it would've taken me in GIMP. GIMP isn't bad, per se, it's just that Photoshop is a lot better. (And if you're just digital painting stuff, GIMP is fine. I found that it's when I started spamming vector masks and layer blending effects that I really appreciated the power of PS.) Loving mesh, BTW. The shackles have come off. ...now if only I had enough time to use it...
  2. Dogboat Taurog wrote: i'm gonna be honest, i don't believe you. and quite likely nor do most of SL. you want someone to argue with, thats all. Actually, I don't. On top of that, you're either a troll or deliberately inflammatory, and apparently you don't care for actually backing up your arguments. I don't have time for someone like that. Bye.
  3. Gaia Clary wrote: Then i do not exactly understand why Agents summary about where meshes are better than sculpties receives this answer from you: "you dont want to get the reality do you. its gonna be a lonely world when theres no one buying anything." As Agent doesn't talk about v2 in that text at all but only about meshes... Because Dogboat isn't interested in actually making a convincing or reasonable argument. I'm gonna be honest. I haven't noticed any performance hit with V2 compared to Phoenix, and my rig's not exactly top-of-the-line gaming material. Plus, Phoenix team's fixing up a real nice V2-based TPV, from what I hear, and their current client's pretty great so I'm expecting much of this next one. I don't see a reason to worry about a mass exodus. Besides, the alternative is a slow death for the platform by attrition, as I've pointed out rather clearly. My old laptop that I used to run SL with three years ago can't run SL acceptably these days, but that didn't cause me to leave the platform. It's up to Dogboat to provide an actually decent argument that A.), not putting mesh in the platform will somehow not lead to it slowly fading away or getting eclipsed by a rival, and that B.), viewer 2 is literally so much worse for older computers, that even with solid TPV alternatives, the switchover will result in mass exodus.
  4. I'm gonna cross-post this old post I made on the blog. I'm a part of the crowd calling for mesh, and you know why I do? It's not "dollar signs" that are "blinging in my eyes". It's because I, and my colleagues, are being held back from creating a wide variety of amazing, fantastic content that will give people a reason to come see and play in SL. Prims are too expensive. Prims have always been too expensive. Prims will probably always be too expensive to build anything of competitive geometric complexity. Prims are pretty useful, though. We love sketching out things in a shared environment with them, and they're relatively versatile. But prims are not enough, and they're overpriced. Have a look at my product lineup. (In-world) My flagship product is a sculpted battlecarrier starship. It's 150 meters long, but it's only around 1,500 prims. Despite the low primcount, it still looks amazing. You know how I did that? Megaprims and sculpts. Megasculpts, in fact. Megaprims and sculpts are the ONLY reason it was economically viable. It blows the pants off of people I show IRL. I tell them "this is what Second Life is capable of," and they now have a reason to take the platform seriously. When I show them the old 2006-2007 stuff, they think it's a shoddy platform with shoddy graphics that isn't capable of anything interesting. But sculpts are not enough: * Sculpts use far too many vertices for most shapes that are represented by them, making them render-inefficient. * Sculpts have a 1-1 correllation between faces and texture detail. This means a number of things, including that in order to get sufficient detail, textures must be larger than they would otherwise be; that textures get unreasonably distorted; and that textures cannot effectively be repeating on most sculpts (again, larger textures, longer loading times.) * Sculpts LoD barely controllably. This leads to "spiderweb" effects and other LoD visual failures that you've no doubt seen. * Sculpts require invisible prim collision skeletons for a wide variety of uses, such as my aforementioned battlecarrier. Sculpts are harder to render. Sculpts take longer to load. Sculpts are wasteful for a great many uses, but they are the ONLY way we have to represent complex geometry right now without blowing the prim budget and making the product economically non-viable. Sculpts enable us to make some seriously cool stuff, but we can do better and we know it. How about newbies competing in the scripts market? Well, let me tell you, I and my team are doing things that push the boundaries of what's possible in the platform, and our customers LOVE IT. I have a *real pathfinding* algorithm automatically moving the fighter docking pads on my starship. We have external libraries. We have distributed version control systems. New to programming? You probably won't figure that out. Are we to smash down and limit the complexity allowed in LSL so that new players who have never programmed before can compete with trained software developers with established codebases? I don't think so. I think that's the sort of crazy idea that would be like shooting the platform in the foot with a shotgun. I was a newbie to programming, once. I was when I came here. This virtual world is what changed the trajectory of my very life! I didn't want to be constrained by nasty limits and did everything I could to break through them (without causing lag, of course). I didn't want to be held back so that those who couldn't understand, or didn't have the time to learn, how to think algorithmically, could succeed. I went out and I scripted things that hadn't been done before in Second Life, or hadn't been done well before in Second Life. It took a long time before I broke even, but I could not suppress the raw urge to create. Look at the current markets in Second Life. Most of them are already saturated. To compete in those markets, you already need to know how to sculpt, or need someone on your team that can sculpt. You need the ability to create custom textures. Newbies, for the most part, already can't compete in most major Second Life markets. They have to create or join new or emerging markets. I joined an emerging market. That's what it takes if you want to make decent cash via content creation on SL. If you're a serious business threat in a market that contains builds, you have already learned, or are in the process of learning, to sculpt. If you haven't yet and you're still managing to be a decent competitor, then you probably have the time to learn how to sculpt. (There are a few reasons you might still not be able to. One of my co-workers has a Mac and Blender 2.4x just wasn't ported well for Mac. But Blender 2.54 fixes that nicely!) However, if you can already compete by sculpting, you can already do mesh. Why? I'm going to let you in on a secret. Meshing is *easier*, substantially, than sculpting. Why? Imagine you are instructed to construct a model of the Eiffel Tower. You are given two options: balloon animal-style balloons, or clay. Sculpting is like working with balloons. Each sculpt is of limited length, and you must pinch it to create the illusion of two separate pieces with a single piece of material. It's harder to texture it once you've got your model. Skilled SL sculptors would be like using rubber bands to bind the balloons and exactly the right size of balloon for each piece. Meshing is more like working with clay. You add, you subtract. You can merge things together. It is much easier to texture, just as it would be easier to paint on clay than on balloons! So, suppose that you have someone who has been trained with neither clay nor balloons. Which one would be easier? I would most definitely say that clay is easier for a newbie because it's more intuitive. You add, you subtract. So, which of these will help newbies more easily compete in a saturated market? Which of these already has faaar, far more tutorials than the other? Which of these has fewer arbitrary rules? Which of these is more intuitive to create with, and thus easier to learn? Mesh, mesh, mesh, mesh. Coming from sculpting, mesh is very empowering. Have you even tried Blender 2.54? It's far, far easier than Blender 2.4x. And now, with the Machinimatrix.org video tutorials that have been out for many months, almost anyone that is willing to can learn to use Blender. Now, let's talk a bit about why Blue Mars hasn't been as successful: * Heavy engine. This is the single biggest weakness of the entire Blue Mars platform. I have a decent PC, and it was just chugging. I couldn't even bring the other members of my team there because their hardware is pretty old. This severely limits the number of possible players, which leads it to get utterly wallopped by the next one: * Network Effect. SL has significantly more content and users already here, with emotional or financial commitments. * No in-world build tools. Thus, no easy ability for users to tweak creations and make them more their own. Let me tell you, I am a big believer in users being able to tweak their creations. I sell all products mod-on-prims. * Awkward UI. Maybe this changed, but I felt constrained in the Blue Mars world by the UI. * Low agent update rate makes physics-based play either expensive (host own server) or practically impossible. * Windows only. Two of my team are on Macs. Deal breaker. * LL released mesh beta. Previously, one of the key competitive advantages of Blue Mars over SL was that Blue Mars actually believed in industry-standard dev tools and that SL seemed like it never, ever would. My team and I forsaw a long, slow, overpriced death as the platform lost momentum due to dramatic dev feature stagnation. That, and the turbulence from the CEO thing, were why we were considering Blue Mars in the first place! Well, LL wised up and changed the equation. We see that they're now committed to actually improving and advancing the dev abilities on the platform. Second Life as a whole *does* compete with well crafted games, because those compete for SL users' time. So, graphics matter. Scripted abilities matter. Limits *matter*. As a result, development tools matter. LL cannot afford to implement powerful mesh direct manipulation tools in the client. But they can do something that allows them to take advantage of *years* of software development along that angle by others. They can support mesh import. They can put Second Life on a competitive graphics footing with more modern environments. It doesn't have to support normal maps, specular maps, et cetera. But it *does* have to show that SL is a serious platform that supports serious tools, not just limited proprietary stuff. The Mesh beta shows that. It shows me that the platform is /not/ stagnant, that my chains /will/ be unbound, and that overpriced, render-wasteful stuff that kills client FPS while users and developers lose interest and leave is not the fate we are all doomed to.
  5. Dogboat Taurog wrote: I don't have an actual argument because I don't know what I'm talking about, so I'll just post an innuendo instead. There, FTFY.
  6. Meshes are much easier to render. I and my comrades are excited about all the hardcore optimization we're going to be able to do. Lower primcount, lower render overhead, proper physics shapes, handpicked LoDs! It's actually harder to screw up rendering overhead with meshes than with sculpts, because sloppy mesh work costs more prims! This means more and better content, and we can't wait to show you all the cool stuff we're gonna make. Second Life must progress and evolve, or it will wither and die. Users are arriving and wondering why the graphics are so low compared to other, more modern games. They're wondering why our stuff isn't as cool. Well, we're working hard on that, coming up with new and innovative ways to get more for less. Megaprim sculpts with separate megaprim bounding skeletons, LSL preprocessor libraries, coalescing of scripts to reduce memory overhead, et cetera. Fewer prims, lower script time, more cool stuff. Well, mesh allows us to take that to the next level. We can do even better once we're out of handcuffs. From what I've seen of him, Dogboat seems to be either a troll or a fool. Maybe both. His position is ungrounded. Blue Mars had some viability going for it *because* LL was averse to mesh. SL must implement mesh, or it *will* fade away. LL knows this, Dogboat, which is why your posts here will do nothing to change the course necessary for SL's long-term survival.
  7. So now the only way residents can show support for a JIRA issue is by getting spammed with notifications? You are asking for better communications with residents, and then throwing away one of the most clear, efficient, easy feedback indicators you have. Why should I believe the communication will improve at all when it's being shot in the foot so blatantly? This looks like a smokescreen to abstract your actual customers away from your developers.
  8. My post was mostly intended for the other people reading this post, and also to show that the anti-mesh sentiment isn't the only viewpoint out there. ;p Phoenix's new viewer project's gonna handle that complaint.
  9. I'm a part of the crowd calling for mesh, and you know why I do? It's not "dollar signs" that are "blinging in my eyes". It's because I, and my colleagues, are being held back from creating a wide variety of amazing, fantastic content that will give people a reason to come see and play in SL. Prims are too expensive. Prims have always been too expensive. Prims will probably always be too expensive to build anything of competitive geometric complexity. Prims are pretty useful, though. We love sketching out things in a shared environment with them, and they're relatively versatile. But prims are not enough, and they're overpriced. Have a look at my product lineup. (In-world) My flagship product is a sculpted battlecarrier starship. It's 150 meters long, but it's only around 1,500 prims. Despite the low primcount, it still looks amazing. You know how I did that? Megaprims and sculpts. Megasculpts, in fact. Megaprims and sculpts are the ONLY reason it was economically viable. It blows the pants off of people I show IRL. I tell them "this is what Second Life is capable of," and they now have a reason to take the platform seriously. When I show them the old 2006-2007 stuff, they think it's a shoddy platform with shoddy graphics that isn't capable of anything interesting. But sculpts are not enough: * Sculpts use far too many vertices for most shapes that are represented by them, making them render-inefficient. * Sculpts have a 1-1 correllation between faces and texture detail. This means a number of things, including that in order to get sufficient detail, textures must be larger than they would otherwise be; that textures get unreasonably distorted; and that textures cannot effectively be repeating on most sculpts (again, larger textures, longer loading times.) * Sculpts LoD barely controllably. This leads to "spiderweb" effects and other LoD visual failures that you've no doubt seen. * Sculpts require invisible prim collision skeletons for a wide variety of uses, such as my aforementioned battlecarrier. Sculpts are harder to render. Sculpts take longer to load. Sculpts are wasteful for a great many uses, but they are the ONLY way we have to represent complex geometry right now without blowing the prim budget and making the product economically non-viable. Sculpts enable us to make some seriously cool stuff, but we can do better and we know it. How about newbies competing in the scripts market? Well, let me tell you, I and my team are doing things that push the boundaries of what's possible in the platform, and our customers LOVE IT. I have a *real pathfinding* algorithm automatically moving the fighter docking pads on my starship. We have external libraries. We have distributed version control systems. New to programming? You probably won't figure that out. Are we to smash down and limit the complexity allowed in LSL so that new players who have never programmed before can compete with trained software developers with established codebases? I don't think so. I think that's the sort of crazy idea that would be like shooting the platform in the foot with a shotgun. I was a newbie to programming, once. I was when I came here. This virtual world is what changed the trajectory of my very life! I didn't want to be constrained by nasty limits and did everything I could to break through them (without causing lag, of course). I didn't want to be held back so that those who couldn't understand, or didn't have the time to learn, how to think algorithmically, could succeed. I went out and I scripted things that hadn't been done before in Second Life, or hadn't been done well before in Second Life. It took a long time before I broke even, but I could not suppress the raw urge to create. Look at the current markets in Second Life. Most of them are already saturated. To compete in those markets, you already need to know how to sculpt, or need someone on your team that can sculpt. You need the ability to create custom textures. Newbies, for the most part, already can't compete in most major Second Life markets. They have to create or join new or emerging markets. I joined an emerging market. That's what it takes if you want to make decent cash via content creation on SL. If you're a serious business threat in a market that contains builds, you have already learned, or are in the process of learning, to sculpt. If you haven't yet and you're still managing to be a decent competitor, then you probably have the time to learn how to sculpt. (There are a few reasons you might still not be able to. One of my co-workers has a Mac and Blender 2.4x just wasn't ported well for Mac. But Blender 2.54 fixes that nicely!) However, if you can already compete by sculpting, you can already do mesh. Why? I'm going to let you in on a secret. Meshing is *easier*, substantially, than sculpting. Why? Imagine you are instructed to construct a model of the Eiffel Tower. You are given two options: balloon animal-style balloons, or clay. Sculpting is like working with balloons. Each sculpt is of limited length, and you must pinch it to create the illusion of two separate pieces with a single piece of material. It's harder to texture it once you've got your model. Skilled SL sculptors would be like using rubber bands to bind the balloons and exactly the right size of balloon for each piece. Meshing is more like working with clay. You add, you subtract. You can merge things together. It is much easier to texture, just as it would be easier to paint on clay than on balloons! So, suppose that you have someone who has been trained with neither clay nor balloons. Which one would be easier? I would most definitely say that clay is easier for a newbie because it's more intuitive. You add, you subtract. So, which of these will help newbies more easily compete in a saturated market? Which of these already has faaar, far more tutorials than the other? Which of these has fewer arbitrary rules? Which of these is more intuitive to create with, and thus easier to learn? Mesh, mesh, mesh, mesh. Coming from sculpting, mesh is very empowering. Have you even tried Blender 2.54? It's far, far easier than Blender 2.4x. And now, with the Machinimatrix.org video tutorials that have been out for many months, almost anyone that is willing to can learn to use Blender. Now, let's talk a bit about why Blue Mars hasn't been as successful: * Heavy engine. This is the single biggest weakness of the entire Blue Mars platform. I have a decent PC, and it was just chugging. I couldn't even bring the other members of my team there because their hardware is pretty old. This severely limits the number of possible players, which leads it to get utterly wallopped by the next one: * Network Effect. SL has significantly more content and users already here, with emotional or financial commitments. * No in-world build tools. Thus, no easy ability for users to tweak creations and make them more their own. Let me tell you, I am a big believer in users being able to tweak their creations. I sell all products mod-on-prims. * Awkward UI. Maybe this changed, but I felt constrained in the Blue Mars world by the UI. * Low agent update rate makes physics-based play either expensive (host own server) or practically impossible. * Windows only. Two of my team are on Macs. Deal breaker. * LL released mesh beta. Previously, one of the key competitive advantages of Blue Mars over SL was that Blue Mars actually believed in industry-standard dev tools and that SL seemed like it never, ever would. My team and I forsaw a long, slow, overpriced death as the platform lost momentum due to dramatic dev feature stagnation. That, and the turbulence from the CEO thing, were why we were considering Blue Mars in the first place! Well, LL wised up and changed the equation. We see that they're now committed to actually improving and advancing the dev abilities on the platform. Second Life as a whole *does* compete with well crafted games, because those compete for SL users' time. So, graphics matter. Scripted abilities matter. Limits *matter*. As a result, development tools matter. LL cannot afford to implement powerful mesh direct manipulation tools in the client. But they can do something that allows them to take advantage of *years* of software development along that angle by others. They can support mesh import. They can put Second Life on a competitive graphics footing with more modern environments. It doesn't have to support normal maps, specular maps, et cetera. But it *does* have to show that SL is a serious platform that supports serious tools, not just limited proprietary stuff. The Mesh beta shows that. It shows me that the platform is /not/ stagnant, that my chains /will/ be unbound, and that overpriced, render-wasteful stuff that kills client FPS while users and developers lose interest and leave is not the fate we are all doomed to.
  10. Now wait just a minute. A lot of us have been asking for mesh, and for a very long time. In order to remain competitive, SL needs mesh support. This is probably something he's realized when experimenting with the SL build tools.
  11. No, they aren't. Saying "I think he should check the Third Party Viewer Directory before upgrading his graphics card" because he specifically asked if there was a way he could reduce his lag without upgrading his computer, and without even specifying a specific third party viewer, does not amount to an "ad board" for third party viewers. It's like getting asked the question "Hey, I'm using a minivan and I'm having a lot of trouble hauling a lot of mulch" and responding "Have you considered using a pickup truck?" Honestly, Suspiria, I usually skip reading your posts in viewer threads. You like the Viewer 2 UI? OK, good for you, but that is an aesthetic preference, like electronic music vs country music. But then you run around being paranoid the moment someone even suggests someone having lag trouble even look at other viewers. Right after you suggest they do specifically what they asked to circumvent having to do. "Look through Hubble instead of a straw." Really? Do you know what a preprocessor is? Do you know what includes are in programming? Do you know that LSL doesn't natively have includes, preprocessing, or constants? You know what's like looking through a straw? Using a viewer that doesn't have the necessary features to do advanced content creation for the platform you make spending money from at a fast speed. You know what's like looking through Hubble? Getting features in a constrained and limited programming language that cut your maintenance time on a non-trivial codebase somewhere between 25-50%. Getting features that allow you to cut your scripts' memory usage noticeably in a memory-limited environment. I'm sick of your Anti-TPV schtick. You don't like TPVs? Fine. But don't pretend like your client of choice is the best one for everyone, because it just plain is not. The OP has a right to make up his own mind, and who's to say he even knew TPVs existed?
  12. He specifically asked if there was a way he could make it not lag without upgrading ("amping up") his computer. I agree with the other poster that he should check the TPVD; it's far cheaper than a new graphics card.
  13. Are you noticing, say, randomly high ping times in excess of 500ms?
×
×
  • Create New...