leliel Mirihi

Syle Devin wrote: So I am slightly confused from the last few posts on the topic. So I understand now that I can upload my own objects for low and medium LoD settings. Is it bad to have them all as the same model? The reason I as is because I already create objects with as low LoD as possible and so there is nothing more to remove. Is it safe to set them all to the highest my model is at? That depends on the object. Some things really are so simple that they don't need the high LODs (note I said high LODs, not low LODs). Though all I mostly care about, aside from upload costs and lag on behalf of the model, is that setting my own LoD would not have objects disappear immediately like they are. Would it save me from that? I take it then moving away to where medium LoD would usually show up, if I have the same model set for medium as I do high, nothing would change? Whether or not an object will look deformed at various LODs is entirely up to you. When building the LOD models think about what parts will be visible at that range versus what won't and only remove what can't be seen or would be too small to notice.

Stanton Sinister wrote: It's interesting to know that there are addspaces that track what sites you view..... I was unaware of that. How do you think Google, Facebook, et al. make their money? Seriously that's how the entire internet works.

Justyn Blackburn wrote: I dont think its lag-wise to set all three LOD´s to high when you feel you have an LOD Issue. [...] Sure LOD is made for view distance, that is giving more space and air for the everybody RAM and Graphic Card and LOD is in every other game. Geometry LOD is for lowering the load on the rasterizer and fragment shaders, it has nothing to do with ram. Depending on the object what you're doing can have a massive impact on performance. I think the prim count and the server count is giving me right. My experience is that when i do that and the last is to 0 i have no problems and the land impact is like nothing. Yeah you don't have any problems with your one object, try filling up a sim with objects like that an comparing it to a sim full of the same objects with proper LOD. You will see a significant difference in frame rate in favor of the properly LOD'd objects. The problem is when LOD is changing many Creators feels just bad about it, when they see a lower LOD, LOD means lower Visible Vertices and Triangles, just the whole Creation suffers. It makes the customer feels like that there is something not ok. That's because they make bad LOD models. Its no problem to create four LOD levels of a mesh but that is not solving any issue caz the meshes of the lower levels will still be lower than the first higher LOD. If it's no problem then why aren't you doing it? It is solving something and the lower LODs are suppose to be lower than the high LOD, that's why they're called lower. There's a reason every game uses LOD and it's not because the developers like wasting time making multiple models of everything.

WolfBaginski Bearsfoot wrote: But one HTTP request involves 6 packets, and three round trips, minimum. Um, HTTP is a stateless protocol, there is no handshake. Perhaps you were thinking of the TCP handshake, but that is only 3 packets (SYN, SYN+ACK, ACK), and the client can include its request in the SYN packet and the server can include its reply (or the first part of it) in the SYN+ACK packet. So only 1 round trip is needed to start getting data.

Jenni Darkwatch wrote: They work well, but depending on where you want the building to be it might not yet be worth it. More and more people can see mesh, but nowhere near enough to make it feasible for a public place. There's a few mesh buildings on the MP already. The last "official" number given by LL was 16% adoption rate around christmas, which seemed about the same as my in-world experience. These days it "feels" like more people can see me in my mesh avis. That was 16% of sims that had mesh items rezzed, that number is now up to 29%. The number of daily sessions with a mesh viewer is 64%, in other words the majority of the grid can see mesh. LL publishes these number every week at the mesh usergroup, the transcripts are posted here.

Some of what you say doesn't apply, the rest are economic reasons Onlive wouldn't do it, not technical. Anyway I see cloud rendering as a solution that's behind its time. It just barely makes financial sense to do it now, after most hardware is fast enough already. I think cloud rendering is on it's way out before it even got started.

There's no technical reason Onlive couldn't do it. It would, however, cost them more money to run than a normal game because sl is a bandwidth hog itself. They may have just not thought of it, try asking them.

I'll break my word and come back for one post because I think I see where part of this misunderstanding came from. Phil Deakins wrote: I have a question for too? Jo said she has 3x2m hotel rooms to rent. Do you think it's possible for an rl-sized avatar to navigate well enough in such a furnished room? Phil Deakins wrote: Even if you do, which I genuinely doubt, grabbing your av on the back isn't the way to navigate in a room. It's only a way to see. Phil Deakins wrote: How do you move by grabbing the av on the back? Even if you can move, it's no way to work thing in SL unless you really want to. The normal way of moving is to press cursor keys. Moving and navigating aren't the same thing. How did it change from one to the other so quickly? I thought you meant navigate because that's the word you used in your early posts (admittedly not the word I used in my replies), but I see now that you quickly changed it to move. You're right that you can't move with the mouse (aside from single or double click to move, which means you're wrong), but have a look at one of the meanings of navigate. "to ascertain or plot and control the course or position of" One of the ways to ascertain or plot a course is to look around which you can most easily do with the mouse. You say this discussion is not about the methods of moving (nee navigating) around yet what's this right here? Phil Deakins wrote: Even if you can move, it's no way to work thing in SL unless you really want to. The normal way of moving is to press cursor keys. Maybe you mean to press WSAD keys and swing your av's view around at the same time. It a bit contrived, isn't it? That shure looks to me like you're saying one method is better than the other. Not that it really matters tho, topics change over time. The title of this tread is "How tall are you?", not "Should buildings in sl be rl size" or "Can you navigate in a 12' by 12' room". I won't bother asking you to clarify what you truly mean this time. And I'm done.

Yes you're right that many games just work around it instead of solve it, it's a lot of extra work tho. The whole point of order independent transparency is it makes the problem a nonissue from the artist's point of view, they don't have to think about or even know about it. Maybe you're right that hardware will never be fast enough due to ever increasing demands, but we can dream can't we. Maybe this year I'll take a crack at adding A-buffer support to the viewer. Of course once I see the performance in sl I'll probably run for the hills screaming.

Tiberious Neruda wrote: Charlar, the only way you're going to be able to really squash both issues at once is to fix the alpha sorting glitch. Yes, that one. Since I'm hearing that's something that's existed long before SL even came about, that's not in the cards, and thus what I have will be broken and can't be fixed. Wait a few more years for hardware to catch up. Solutions (yes there are more than one) to alpha sorting have been around for a while (decades in some cases), however the amount of processing and memory they require is quite high and scales with the amount of alpha content visible (which is way too much in sl). The current best solution that I know of has unbounded memory requirements, it would need close to 500MB of vram to render the average forest sim in sl, that's just for sorting all the alpha fragments. Many professionally made games are already using these solutions, but well, they have professional artist that know what they're doing or get fired if they don't. We don't have such luxuries in sl where any one can upload anything, thus we have to wait for hardware fast enough to handle our user made (amature) content.

Phil Deakins wrote: My point is that you should not quote somebody, add something to the quote that they neither wrote, meant, nor implied, and then try to make a point based on that. That's not the point I was talking about and you know it. Thanks for dodging the question. I'll back out now since there's obviously no point in discussing with you. Phil Deakins wrote: Whether or not you still wrongly think that only the cursor keys can be used to move around in my opinion, is irrelevant. You are wrong about it. I just told you that I've used the WASD+mouse (long before this thread, incidentally), so why you want to hold on to a wrong belief is puzzling, to say the least. WTF are you even talking about then? How can I admit I'm wrong when you won't tell me what you actually ment? Note: just because I disagree with you about this doesn't mean I hate you for life.

Phil Deakins wrote: You are mistaken, Leliel. It wasn't implicit at all. I have been well aware of the WASD+mouse method - I've even used it. So, as I said, making a point by changing what somebody wrote into something that they neither wrote nor implied, must mean that you don't a point to make at all. I'm sorry, but that's the way I understood you and still do, thanks for giving me the benefit of the doubt and not just accusing me from the start. And I'm pretty shure you already knew about mouse look, at the very least because you were replying to some one that had just brought it up. What is your point exactly? I read it as only the curser keys are the normal way, you claim that's not your point but have yet to provide a more thorough explanation. So are you just going to continue to tell me I have no point or are you going to tell me what your point is? Btw, SL isn't a game, so 'seeing how long I live in a game' doesn't have anything whatsoever to do with SL and this discussion. BTW, SL isn't a game, so telling people they need a fast GPU with up to date drivers doesn't have anything whatsoever to do with it. The discussion is about moving around in 3d environments, I'd say 3d games are a valid comparison. If you don't like the "how long I live" part then I'll just bring up games with 3d puzzles like Portal and Amnesia.

At one point SL supported textures up to 4096x4096 but LL lowered the limit after rampant [unintentional] abuse and performance problems.

Phil Deakins wrote: leliel Mirihi wrote: This same method is used in third person games as well, it is in fact the normal way of doing movement & camera, which is the reason sl supports it. Phil's assertion that [only] the cursor keys are the normal way is an archaic belief that dates back over 20 years, no modern games actually do things that way because it doesn't work in a 3d environment. You added a word (in square brackets) that I didn't use, and converted what I said into something that I didn't say. Square brackets are used like that for something quite different. If your point needs you to change what someone else said, then perhaps you don't have a point at all. Curser keys (or the WASD keys) are the standard method of moving in SL. Curser keys alone don't give you the fine grained control you need to maneuver in a complex 3d environment. Just try it yourself in any modern game and see how long you live. You may not have used the word but it's implicit in your entire argument that you mean only the curser keys and not the curser keys plus a mouse which is what you're arguing against. If there is any other way to interpret what you're saying please enlighten me.

Baloo Uriza wrote: leliel Mirihi wrote: This same method is used in third person games as well, it is in fact the normal way of doing movement & camera, which is the reason sl supports it. Phil's assertion that [only] the cursor keys are the normal way is an archaic belief that dates back over 20 years, no modern games actually do things that way because it doesn't work in a 3d environment. The first game that made WSAD+mouse movement the default (instead of purely keyboard movement by default) was id Software's 1996 release of Quake. So more like over 15 years. WASD + mouse look wasn't actually the default in Quake, but everyone quickly enabled it since it gave you such an overwhelming advantage. There were plenty of other games that experimented with various forms of mouse look before Quake tho.

Kelli May wrote: It might sound unnatural when described, but I found it almost completely instinctive to pilot my av this way, and now rarely use any other method. I've taken part in several activities where fast, assured, accurate movement is necessary and this is incredibly useful. It closely mimics the way the mouse and keyboard are used in First Person Shooters: mouse for tilt/pan, and movement in the direction you are looking. This same method is used in third person games as well, it is in fact the normal way of doing movement & camera, which is the reason sl supports it. Phil's assertion that [only] the cursor keys are the normal way is an archaic belief that dates back over 20 years, no modern games actually do things that way because it doesn't work in a 3d environment.

Hellespont Hoorenbeek wrote: I am getting so tired of some people here and am not goint to reply everything just do us all a favor and research some things befor posting your thoughts it gets twisted and missing the point compleetly. my original post was "Hands of viewer 1.23 / Phoenix" Maybe because your initial premise was ridiculous, 1.23 and Phoenix/CoolViewer/et al. are not comparable in any way. The former is outdated, broken in several ways and missing many key features, the later are modern viewers with retro interfaces. Attempting to put them on equal standing is nothing but bad joke.

Hellespont Hoorenbeek wrote: - I will not use viewer 2 or 3 interfaces because i like my viewer 1.23.5 and i respect you to use v2 or v3 interface viewers you need to start respecting others aswell did you for example know 45% of the SecondLife residents is using a v1 based viewer? They're using a viewer that has the v1 interface but has been updated to support modern features. The number of people still using 1.23 has been in the minority for a long time now. Go ahead and still use 1.23 if you want, nobody cares. But every day that passes less and less of the grid will feel sympathy for you. Disclaimer: My best friend in sl still uses 1.23 yet she doesn't seem to have a problem with me standing next to her wearing mesh clothes. We both made our choices and respect the others.

Sorry but you kind of sounded like a little old lady calling the cops on some teenagers that were walking down the street just because they were dressed funny. Your first assumption probably shouldn't have been that LL was trying to hack your computer and steal all your info.

Sidney Dionysus wrote: Let me put it in perspective for you. If they are port scanning me, they are probably port scanning YOU. You might not have even known. Are you ok with that? As some one that knows how TCP/IP works and has done a few port scans themselves all I can say is so what? Port scans can be used for more than stealing your tax records. Read up on what nmap can do. Don't worry, they are going though your pockets and checking all your doors and windows to see if they are unlocked, but it's LL. So don't worry about it. It's just a huge organization of people who you don't know, hiring strangers THEY don't know, and giving them the keys to YOUR house. If that doesn't concern you, it certainly should! And I know they are doing it because Intego VirusBarrier X6 keeps catching them at it. Not once, but four times so far. A commercial product who recomends 'permanatly blocking' all traffic from that address. All I want is an expalnation. Maybe I should open a ticket, but I thought others might want look into this as well. How do you know that LL is trying to steal your tax records? How do you know they're trying to steal anything? So far all you know is that machines owned by LL are attempting to connect to your machine, nothing else. Have you used a packet sniffer to see what LL is sending? Or are you just jumping to conclusions and making up alarmist hyperbole?

Sidney Dionysus wrote: Yes, but... Port scanning? I'm no expert, but the function of a post scan is to see if there are any openings to your system though which once computer have access to another computer without the owner of the 'victim' computer nessessarily even knowing it. Most people I know of consider port scanning an attack -- includeing me. Their software is free to LIMITED data, but I did NOT grant LL free and easy acess to my tax records and Quicken data! They need to stay off my computer. I suppose it could be argued that they are looking for weaknesses in my firewall so that can read my files and learn about my video cards, etc -- but just ask!!! I'm happy to provide any of that stuff. But don't try to trapse around my system uninvited and look for any damn thing you want. Not even remotely cool. Talk about jumping to conclusions. Port scanning is the equivalent of standing on the sidewalk and taking note of where the doors and windows are on a house. Nobody said anything about LL trying to break into your computer and steal your tax records. What makes you so shure it even is a port scan? Last I checked the viewer and the grid had some pretty legitimate reasons to talk to each other and as you said yourself this only happens when you're logged in. There's no rule that says the viewer that has to initiate all communications.

MoiselleErin Teardrop wrote: Anyways, with the low FPS, what bears more responsibility for that? Is it lack of RAM or just a crapy graphics card? My card is - Graphics Card Vendor: ATI Technologies Inc. Graphics Card: ATI RADEON XPRESS 1100 Series It's the graphics chip. The xpress 1100 is just a renamed xpress 200 which was just a mobile version of the X300 which was just a PCIe version of the 9550 which was a cut down version of the 9700pro which dates back to 2002. The lack of ram doesn't help tho. How would I know if a better graphics card would even work with my machine? Tell us the manufacturer and model number of your machine.

Kwakkelde Kwak wrote: leliel Mirihi wrote: Video games use dirty tricks like this all the time to cut down on resource usage, especially on consoles. I'm not that familiar with video games, not at all actually, but I am pretty sure it would do wonders for SL if LL makes it possible, as you edited yourself, it would mean a 66% decrease in video memory use if the highest res could be left out and up to pretty much 100% if the object is in the far far distance. That could work, would have to do some testing to see how well it would fit in with the viewer tho. I think using texture compression would also be good, it would give 75-87% savings across the board at the cost of image quality. It would eat up some CPU time to recompress the textures but most librarys are pretty fast.

Kwakkelde Kwak wrote: leliel Mirihi wrote: As the edit to my above post shows it's beyond academic to the point of being a waste of time to talk about. Yet you kept talking about it where I wanted to leave it a page or so ago.... lol I wanted to drop it a few pages ago as well. I guess we both just like to argue. ^.^ In a video game textures are used as many times as possible, in SL they aren't. So working with a trick or solution or technique that only works for non duplicated items would suit SL, not an optimised video game. Video games use dirty tricks like this all the time to cut down on resource usage, especially on consoles.

Kwakkelde Kwak wrote: leliel Mirihi wrote: To what end? What are you trying to save by doing this? It's the full size image that's taking up all the ram, cutting back on some image half way down the mip chain isn't going to do much of anything. We're talking about extensive modifications to the viewer to over ride how the GPU does mipmaps to save, what 5%? Who cares? There are much lower hanging fruit. Isn't that exactly what I said 10 times already? It not being worth the efford and it being academic... As the edit to my above post shows it's beyond academic to the point of being a waste of time to talk about. Kwakkelde Kwak wrote: leliel Mirihi wrote: This just highlights the differences between sl and a real game. In a real game you can make these sorts of guaranties about objects so doing these kinds of optimizations makes sense. I'd say quite the opposite. Your examples I quoted say otherwise. And how does the viewer know what LoD to show? Because it already keeps track of both object size and position. I already said it's possible to get it to work, just not worth the effort.