Nalates Urriah

Also, in Preferences->Move & View you can set the single and double mouse click on land action. One of the choices is to walk (move) to that 'clicked' spot. I use teleport on double click land. It is handy but has a gotcha... if the land is set to only allow TP to a hub, no matter where you double click you end up at the HUB. Mouse steering is how I have used SL for years. In third person view (not mouse look) left-click-hold on the avatar, press W or up-arrow. You go forward. Drag the mouse left and right to steer. This works very well for me. I do tend to wear out my mouse left button. There are variations. While holding the left-click A & D keys make you walk parallel with yout screen... in that frame of reference W walks you into the screen away from you and S walks you out of the screen, toward you. W & A or D walks you at 45. So, when shopping I can left-click-hold, press A or D and walk parallel to the wall of stuff I was looking at. While doing that I can walk around a corner by pressing and holding W while left-click-hold and A or D hold then as I progress release the A or D key. Try it. Easier to do than say. Also, with left-click-hold and A or D or W I can roll the mouse forward or backward to change the camera angle up and down. And... there are ways to change the default camera angles so a more enjoyable view. We'll save that for later.

Probably not... The Lindens have been working on removing the need for QuickTime and improving CEF rendering in the viewer. The original work for MOAP was a 2011 thing with changes in 2014, 2015, and late 2017 and some more just recently. So, older tutorials are likely wrong... anything pre-2017 is suspect. Most of the SL Wiki stuff is 2011... The original thread on CEF-MOAP is here. (2015) I think I last wrote about in June 2015. Second Life: Chrome Embedded Framework Update So, you may have to figure it out. I couldn't find any current videos or other tutorials on getting YouTube on a prim. I know it works because I have my blog site on a prim in front of my SL home. I just tried playing a YouTube video I have on my site from in-world using MOAP and it works. IM me in-world and I'll send you a LM. The site link is in my signature. You can look at the page source to see what display code WordPRess is using.

It isn't hard on the viewer... it breaks down the viewer's workload to smaller chunks. Which allows the viewer to spread out its load and concentrate on what you need to see first. Progressive DD Stepping is a good thing for the viewer. The idea can throw extra load on the servers. The Progressive Draw Distance HUD's in the marketplace do that. They improve the individual user's experience at the cost of degrading server performance for everyone. There was a great deal of debate about progressive draw distance HUD's a few years back. The Firestorm team was in the middle of it. I forget what they did to mitigate the problem. I think the way they avoided asking the server to supply an interest list multiple times was to get the max DD list and then handle downloading only the nearest objects viewer side then work their way out. The final result was reducing the region PEAK server load... well... spreading it out over time, lowering the peak, without requiring additional work from the servers. Then a bunch of Linden work was done on the Interest List and how items in the list are downloaded. The effect was to have the LL viewer do something similar to the FS Progressive DD. With the addition of CDN services the load on the servers was further reduced. So, the progressive DD in FS shouldn't be much of an issue. It was designed to offset the impact of marketplace HUD's. We could probably ask Whirly about the current state of Progressive DD Stepping in Firestorm and get a definitive answer on how it works now.

Caches do not load... they lay there on the storage device until a file in the cache is needed. Then only that file is loaded. So, I doubt you see any slow down from a large cache. The viewer caches are indexed. So, file retrieval from a 1,000 files or a million files only varies by a couple of milliseconds at most. I have a 4GB FS cache on SSD and a 4GB LL in a RAM drive and a number of other viewers' caches in the 10GB size on a mechanical drive. Regardless of which cache is in use. the viewers launch and come up at about the same speed. Larger caches are faster. The current limit of 10GB is to be removed in the next cache redesign. Larger, faster caches are coming. Scene render is noticeably faster with the LL Viewers using the Ram Drive cache. But, the viewer caching is pretty lame so it is not consistent. The Lindens have started an update project for the cache. Which is good because some scene renders for places that are already in the cache are way slow, whether HD, SSD, or RAM Drive. What is consistent is that an empty cache on my machine is absolutely the slowest at rendering a scene. Especially if I visit a region no one else in my area has visited and I hit a cold CDN cache. Firestorm has a progressive load feature... Preferences->Firestorm->Extras->Enable progressive draw distance stepping... Try adjusting it. What size is your cache and where does it live? 15 minutes for a scene to render is a long time. I think 5 minutes is a long time. However, the video card is not the likely problem. If textures load, get clear, and then go blurry and repeat the process then you have texture-thrashing. A newer video card can help with that problem. Scene render is dependent on your cache, network connection, and disk speed. The CPU and Graphics Card have a significant effect on render speed once all the parts are there. If you have good FPS (>15 FPS) and the scene is slow rendering look in the Texture Console (Ctrl-Shift-3) to see what is happening. While watching the Texture Console also watch the viewer stats, Advanced->Render->Object Cache Hit Rate. You'll get an idea how it is or isn't working.

Try to ask your questions in only one thread....

Your numbers aren't all that bad. The longest round trip was 200+ ms, which while not great isn't horrible. You aren't being disconnected from a speed issue. The likely problem is packet loss. You can test for packet loss by entering ping -n 218.51.###.### Of course replace the #'s with the rest of a region IP address. You can troubleshoot the problem with this information: http://blog.nalates.net/2011/10/26/troubleshoot-your-sl-connection/ You can of course use the Viewer Statistics to see what it thinks the Ping and Packet Loss are. But, server factors get into the viewer's numbers. So, they are not as clean as the Command Window numbers. Also, you can look in the Second Life log file. At the end is a summary of the network connection stats. I don't know how you are connecting via your wire and how that compares to an uncle's WiFi. Do they both go to the same gateway/modem? Lumyia uses way fewer connections to SL. That your desktop connection is failing and Lumyia working suggests a router/gateway problem. Shut everything down. Then bring up the gateway/modem, next any routers, switches, or relays, and then bring up the computer. Allow each one to fully reboot and stabilize before starting the next. Older hardware has a hard time handling the multiple connections used by the viewer. The hardware can run out of channels. Some viewers do not to a good job of closing Internet connections when they finish using them. Restarting the hardware can clear orphaned connections.

What you see is called Bake Fail. See https://wiki.phoenixviewer.com/fs_bake_fail If this doesn't solve your problem come back to this thread.

Any program you try to install that is not Microsoft Certified will bring up this set of warnings. It is a recent change in Windows.

The Lab repeatedly claims they read EVERY Abuse Report. Lindens I know and trust attest to this also. If your reports are factual, have a precise time, and list a ToS violation... they are pretty certain to draw action. You won't hear what was done, but something was considered. The totally ignoring them is about the worst thing you can do to them. It may take a while, but even the most persistent annoyance eventually gives up. Plus, it is the one course that is ToS legal. Almost everything else you might want to do could land you in trouble. I can close chats faster than they can open them. I can also block avatars faster than they can create new ones. Just hang in there. I also tend to hit back. If a particularly annoying insult gets through I may save it until I cool off. Then I'll come up with a comeback that I believe will be especially stinging to them. This prepares me for those iRL situations where many are left in a 'wish I had said' senacio. I get particular satisfaction from knowing I used them.

@sthashan The CPU temp is high enough that the chips will throttle to reduce temperature. In general Intel says max operating temp is 75C and general use is 65C. At 100C it is over max limit spec and likely to burn out, melt. So, your 94C is dangerously hot. This typically has 2 causes, dust or a lose heat sink. If this is a laptop, vacuum out all the cooling ports. Use a crevice tool. Avoid over revving the fans. So, quick swipes across the ports with 10 to 20 seconds between passes. Laptops seldom have a loose heat sinks as they are bolted in place. On a desktop, take the side off the case and look. If it is a dust problem you'll be able to see it. Use a can of air to blow it clean. Do NOT let the fan over reve. You can use a soft paint brush to dust the fan. If it is clean, then check to see it it is loose. The fan/heat sink is pretty robust, so don't be afraid to take hold and try to move it. If it is loose, it will have to be reset. There are YouTube videos on replacing a heat sink. They will show you what you need to know. If it is neither loose or dirty then it is too small. You can get great heat sinks & fans online. I use New Egg, B&H, and eBay. You want NEW. Even great heat sinks are cheap. Packet Loss - That 1.2% may be high. At startup the value spikes up. If you have been sent 2 packets and lost one, that's 50%. 4 packets on and still 1 lost, your down to 25%. The longer you are on the lower it goes, unless you keep losing packets. The ping time is long. Not disastrously so and possibly tolerable for numerous SL activities. But, enough to warrant looking at and certainly too high for any combat or vehicle play. Use this information: http://blog.nalates.net/2011/10/26/troubleshoot-your-sl-connection/ None of this round of info shows anything to make the video problems you have. UNLESS the computer is for some reason using the CPU's HD Graphics rather than the NVIDIA card. The CPU is running HOT and likely having issues. So, make sure you have a game setup in the NVIDIA panel for each viewer you run. The how to is here: http://blog.nalates.net/2016/06/05/nvidia-settings-2016/

We are going to disagree. You are right the servers are not bandwidth bound. They are however CPU bound. Time Dilation and Server FPS are the primary measures of how well or poorly they are doing. You can watch Ping time change as TD and S-FSP degrade. The server has 22ms to get everyone updated. Exceeding the 1500 max-b by 20 or 30 people will burn up the time devoted to network processing that is needed for delivering UDP data. One can Show the Updates to see how much info is being sent via UDP. It is that flood of UDP data going to ALL the clients that can lag the server. The portion sent to the individual client is a tiny percentage of what a workstation can handle. I can run 2 or 3 viewers, stream a radio station, and have a bunch of web pages open or stream video to YouTube gaming and and still not get my network resource over 50% usage. My workstation is certainly not a dumb terminal. I don't know of any Win-10 computer that is used as a dumb terminal. The viewer certainly does not behave as a dumb terminal. Anyone can open the Win-10 task manager or Resource Meter and see how much work the viewer is doing. High CPU use shows an active system. I'm not speculating on how this works. I'm using what I've heard the Lindens and various developers at UG meetings talking about. 10 years +/- of UG meetings has given me lots of information on how the SL system works. I think you'll be hard pressed to substantiate your statements.

My main channel region did not get restarted last week. It did get restarted this week. It is running the same version as last week, 18.06.14.516450.

You are starting out. So, the basics are: There are System/Classic avatars. These are more less the original avatar built into the viewer and have been around for a long time. There are Mesh Avatars that are the newer iteration of the avatar. Basically these are a custom user made mesh body attached to the system avatar that covers it. A second skin... There are system clothes and mesh clothes. System clothes are like decals painted on the system/classic body. You essentially only have to make a diffuse texture to make these clothes. Mesh clothes are attached and cover the body, mesh or classic. These are custom user made 3D models with weight painting and texturing. There are two types of mesh clothes, rigged and fitted are our jargon. Rigged clothes are rigged to the animation bones in the avatar skeleton. They are NOT affected by many shape sliders in the Appearance Panel, like Body Fat. The result is these clothes are made in 5 standard sizes and users fit their body to the clothes. They are not very popular these days. Fitted clothes are rigged to the collision bones and are affected by most shape sliders. The Body Fat setting works. These are generally made in sizes related to the mesh body brand being worn. So, sizes are for Slink, Belleza, Maitreya, etc. To make fitted clothes one gets the Dev Kit from the body maker. They generally provide a model usable in Blender or Maya that lets you design for that body and match the weight painting to the body for good fit and conformance during animations.

One of the basics not mentioned is the Chip Midnight and Robin Wood's templates (UV Maps). In a general sense these work for both system and mesh body tattoos. You can Google for Second Life clothes templates to find the various templates available. Mesh body makers usually have templates available in their Dev Kits. It is probably easier to start with an Omega Applier as it will work with most mesh bodies. Saves chasing down all the dev kits. The coming Bakes-On-Mesh feature will hopefully simplify making skin, tats, and clothes for mesh bodies. The process should then be similar if not exactly like making skin, tats, and clothes for the system avatar. So, learning how to make system tats will be time well spent.

A couple of things I didn't see when skimming the thread... Regular reliable events will build attendance. Various contests will build attendance. Best dressed, costume, dancer, etc. There are people looking to make some L$ that follow contests and bring all their friends to vote for them. So, compare the Lab's advertising fees and results to the cost of your events awarding prizes. You can probably talk some designers into sponsoring events that feature their products. Having a group to remind people of the event start time and provide a link. I will say that I REALLY have to like a place to commit one of my group slots to it. I am much more likely to do a Subscib-O-Matic commit than a group join. Also, multiple notices all day long will drive me out of a group. Also, there are DJ's that have followings. So, see about DJ'ing at popular clubs. You may pick up some fans. Most clubs will allow you to promote yourself but not your club. So, a group based on you rather than your club may work better.

Vivox is big in providing voice services to a number of games. Do a Google search. One of the big problems for SL and Vivox is the griefers from all the various games. Both SL and Vivox get attacked. There is no reason to think it is just SL griefers hitting Vivox.

The UDP and HTTP are separate communication protocols/channels. Coffee incorrectly says setting it too high sends packets faster than you computer can receive them. But, the max value, recommended by Firestorm peeps experience, is 1500. That hasn't changed because the data load problem is server side. At 1500 the servers can almost always keep up. Above 1500 and a busy server can get behind and the user sees problems. Basically the server is too busy to pack all the information for all the users and get it sent. The client side with multi-megabit network speeds and handling capabilities is seldom going to get overloaded. If one often visits busy regions, they may get better performance by reducing the setting. The idea being the server is likely to get the important info into their UDP packet and get it sent in the time it has available for each person in the region. But, the Lindens like getting ride of settings and this is one they are depreciating. Much of the UDP info bouncing around in SL is out of date. Even the Firestorm wiki still says set Max Band to 80% of your download speed or 1500 whichever is smaller. With 100 to 300mbps speeds common 80% is never less than 1500. Also, most people are unaware of Monty Linden's work in 2016 and 2017 to move SL server to client communication over to HTTP. You can press Ctrl-Alt-Shift-U (Show Updates to Objects) to see the things affected by UDP information. Almost all textures and other assets in SL are now delivered by HTTP. So, changing Max Band no longer affects render speed. Once the scene renders how things change is affected by Max Band. Rubberbanding is an effect of lagging/lost UDP data, to and from the server. The experience of the FS people is this loss/lag is "LIKELY" to increase at settings above 1500. Also, even before the move to HTTP the SL viewer caching system was updated and made more robust. With the move to HTTP it is even more robust. Clearing cache is a LAST RESTORT as clearing cache often makes problems worse, as sthashan found out. But, the old 'clear cache' recommendations persist. The Lindens are now working on yet another cache overhaul. So, the effect of cache-clearing's negative effects will likely become even more noticable. I don't see where you have mentioned your download speed or how well your connection is or isn't working. We know the SL side works because it works for all of us. We know it is not your viewer as two different brands of viewer have the same problem. That leaves your hardware and your connection. The image you posted does not look like any of the video card failures I've seen. So, about the only thing left to check there is to make sure the video card and computer case are dust free and not overheating. You can use the free HWMonitor tool to check temperatures and not have to open the computer up. While you may have a good general Internet connection, you need to check how well you are connected to SL. Open Viewer Statistics, Ctrl-Shift-1 - it is a toggle, open and close. Packet Loss and Ping are the basic measures of your connection quality. Smaller numbers are better. Packet loss is tricky. Generally <1% is livable. But, if packet loss is scattered, intermittent, now and then higher rates are tolerable. If it clumps, even 1% can get you disconnected. Ping as measured by the viewer has a portion of SL server performance in the numbers. So, if ping is high, >250ms, check how the region server is doing. Time Dilation and Server FPS and Physics FPS should be 1.0, 44fps, and 44fps... ideally. As these numbers degrade ping goes up and thus can be misleading. So, test ping on a well performing region. Above when you say "back to the same" I assume you mean the visual appearance of the render was messed up. But, check your Max Band setting. Make sure it stuck. A silent crash on exit will often allow settings to revert to their previous value. I suspect the relog had more to do with the temporary improvement than the bandwidth setting. I'm curious to see how this resolves.

I'll often take my main avatar to ADITI for script testing. I can escape all distractions. The only drawback is the inventory update, but even that is seldom a problem. Within 24 hours everything I have in AGNI is available in ADITI. Just remember, it is the login to ADITI that triggers an update. The update comes after you log out of ADITI. Exactly when is a bit fuzzy. It seems the updating process takes some time to get to all the accounts needing an update. At this point it may only be the operations people that know exactly 'when' updates are likely to happen. The bigger problem on ADITI is inventory-swell. Whatever I delete on AGNI is not automatically deleted from ADITI inventory. So, I use ADITI folders for all I do on ADITI. I can delete everything else and only the good stuff comes back at the next update. Firestorm has round trip script editing. You can set FS to use your favorite editor. Open the script in Firestorm and it opens your editor. I haven't used it as I usually start in my editor and then move to the viewer for additional testing. I use LSLEditor for most of my work as it has an amazing IDE allowing me to work with multi-script projects. So, I can debug one script's output and see the input in the target script. The BIG drawback is the editor is getting out of date. Also, the debugging system is not well documented.

The avatars we WEAR are outfits we put on our account's avatar. My avatar is Nal in-world. I can put on a vampire or werewolf. She is still Nal. If I delete Nal I've deleted my account. There is nothing to put anything on... So... What do you mean you "deleted your avatar"? How did you delete it?

Many problems are transient. The problem is not your computer in such cases. It is the connection or the servers. Vivox provides voice for SL. So, you have to make a connection to SL, then SL tells the viewer how to talk to Vivox servers. There is lots of room for things to go wrong. General advice is to give voice 24 hours to self correct before doing anything drastic, reinstall... whatever. If you are impatient (who isn't?) check Grid Status to see if a problem has been noted and/or go into Firestorm Support to ask about whether others are having a problem. If you are Premium you can ask SL Chat Support. You are looking to figure out if the problem is on your side or their side. With voice it is common for the problem to be on THEIR SIDE. So, giving it time is the usual fix. I hate it when that is the fix... Me WAIT!?! Sheeeeeeeeeeee.....

I'm not sure what you mean by "Now I see that all the items that I blacklisted (permanently) are back." Back where? And what kind of blacklisting are you doing? Why are you clearing your cache? The wiki link you provided goes to a page that says basically don't clear cache. That items are gray for 5 to 10 minutes is likely because you are clearing the cache. Don't. I think the wiki page is very clear one should ONLY clear cache if there is a known texture or object file problem. Do you think you are having such problems? Cache files rarely corrupt. The textures and objects transfer via the error correcting HTTProtocol. So, viewer crashes are the more common cause of corrupted files and the viewer has to crash as the file is being written. A rare thing. I suppose, you could have a bad enough connection the viewer was having a hard time assembling the files from error correction data. But, I think that way unlikely and file corruption would be a small problem compared to what you would be experiencing in lag and disconnects. The lets be sure we are talking about viewer and browser caches as seperate things. The wiki page you pointed to is about the viewer caches. Browser caches are a term we use to refer to Internet browser caches. These are unrelated and separate things. Clearing one or the other has no effect on the other. Windows 10 has its own caches and those are hard to get to and best left alone. They are separate from viewer and browser caches. If you visit the SL websites, like this forum, your browser will have history and cached pages from the sites. With CCleaner you can totally destroy a computer. As of Windows 10 all the stuff one does with CC is built into Windows. Such programs are made for people that do some word processing, email, and Internet browsing. Once one moves to more sophisticated use of their computer CC is likely to be more of a detiment than a help.

I think you are missing my point... There is no anything that is BoM ready as BoM isn't released and may still change.

It seems that in certain circumstances a viewer crash can kill the HUD's attached during the crash. The 5.0.11 FS Viewer does crash silently on exit. I have no idea how frequent this is. My personal experience is about 33% of time. I only know because I get the error reporting popup every now and then. A poor connection can stall a HUD. But, a relog usually brings it back to life. The known fixes are; detach-reattach, relog, relog in a different region, and finally get a new copy of the HUD.

There are some different ways to get into SL. Chic is giving you the BEST basic way. When you download and install the viewer it will make minor changes to your Win-10. THen the alternate ways to get into SL will likely work. If you have not installed the SL viewer and you use an alternate away in by clicking on the website Destination Guide's Visit Here button or any SLURL (unique URL's directing one to regions in SL), Windows (any version) will launch an Internet Browser, e.g., Firefox, Chrome, Edge, something, and try to get you to download and install a viewer. I think this is where you are having problems. So, use Chic's advice. Whenever you are using the Internet your computer and some server somewhere are talking. Your Internet Browser knows how to speak several 'computer languages', which are called protocols. The browsers do NOT know how to speak the 'protocol' used by the SL servers. We use a specialized 'viewer' to see the virtual world the SL servers are telling our computers how to draw. If your computer has not been taught how to speak the SL Protocol (by installing a viewer) it is as confused as you when someone speaks to you in a language you don't know. The i7-7700HQ is capable of displaying SL. The 'H' denotes it is a CPU with additional graphics capability. Thus a better gaming machine, but NOT a 'good' gaming machine. CPU graphics is mostly designed for streaming video. HOWEVER, it should not matter as the machine's specs advertise the unit as having an NVIDIA GeForce GTX 1050 Ti chip. BUT, you have to setup a game spec in the NVIDIA control panel so the laptop knows to use the graphics card with the SL viewer. Otherwise, the laptop is very likely to use the CPU graphics, which suck - relatively, in an attempt to save power. The instructions for doing that setup are here: http://blog.nalates.net/2016/06/05/nvidia-settings-2016/

We have lots of different opinions... The Max Bandwidth setting is in general being depreciated by the Lab. The setting is a bit of information the viewer sends to the region server. The server uses it to determine how much UDP data is sent. The UDProtocol is used less and less by the SL viewer-server. More and more information is sent via HTTP. So, the setting is having less and less impact. Whether higher or lower is better depends on your connection a bit but more on the region server and whether it is lagging or not. So, finding your best setting can depend on the region you are in. Obviously that makes it a challenge. At this point about the only data traveling between servers and viewers is update information... where the avatar is standing, turning, direction, speed, and when stuff in world changes. You can press Ctrl-Alt-Shift-U to see the things that are getting/generating update information. Max Bandwidth and Lost Packets are strongly related to how those things behave. Deeper in the viewer stats are ack-noack stats that provide a bit more information on how things are working... or not. The Max Band @ 1500 is a value the Firestorm team determined empirically. A combination of developer experience and FS Support problem solving contributed to their decision/recommendation. They suggest this as the maximum value for this setting.