arabellajones

I have mixed feelings about the whole deal. High Fidelity seems to have an infestation of Silicon Valley Tech-Bro Blockchain types. Some of the company assets, such as their 3D voice system, could be a big benefit to SL. But are my problems tracking who is talking on voice due to the voice system, or to my not-so-young ears? It's not simple, but blockchain and crypto currencies are dominated by con games and scams. And you can argue that Philip Rosedale has only managed one success.

I use Cool VL Viewer and Firestorm. Currently, Firestorm has a much lower frame rate, but 120fps would be wasted on my monitor. Firestorm also has a frame-rate limiter option, and I use the cinema/TV trick of setting that to half my monitor frame rate, which gives me a little bit of headroom. It's an indirect measure of the work being done, but my CPU and GPU both run a little cooler with that option It's part of a long-running SL problem. We don't all have the same hardware. And some of the colour schemes that Linden Lab use in their viewer are not meant for the eyes of mortal men.

Can pie be counted as pudding? (I was involved in an argument about that yesterday, which also tried to deal with cookies.)

I did notice, Sunday, some erratic behaviour with alpha-maps and classic clothing on a BoM Mesh body. It settled eventually and could have been merely very slow texture delivery over the network, either AWS-specific or the wider internet. I suppose it could be related to the very long sim server uptimes we have had since the last restarts. I have no answers, but there are times the Second Life system does things wrong.

We do now have the status notifications for rolling restarts, Tuesday and Wednesday, though they could be automated. It'll be three weeks since a restart, and if the automatic system works, as it seemed to the week before Christmas, there will much much relief here. The restarts were pretty fast, possibly faster than the map could respond to. So I am fairly hopeful. The seasonal holidays should be predictable. Is Christmas really a surprise? Not making plans for them doesn't look competent for any company. There are other problems this year.

I hope Linden Lab can get their act together on notifying their paying customers about planned work on the server systems. Tuesday next week will have to see rolling restarts, it'll be three weeks since the last. The restarts on the 21st December went pretty well, but not the notifications. Is there going to be new code rolled out to the RC Channels on Wednesday? There's the same reason for rolling restarts, and while I have not seen any signs of significant problems from the long uptimes, I expect the Lindens have reports. There is some lag. I suppose user group meetings will also restart next week. Some apparently relevant wiki pages are horribly out of date. Is the illusion of competence forever out of reach?

It looks as though they are simple rolling restarts, as little as 7 minutes for the restart cycle and no change to the server version, which fits with other info such as Inara Pey's report on last week's SUG meeting. The in-viewer map data looks unreliable: the up/down indicator doesn't update so rapidly. A region could go through the cycle, though it's hard to tell.

These restarts were first announced on Status three minutes before they started, and I have found nothing else.

The general kerfuffle on Wednesday, which produced an enormous collapse in concurrency, was attributed by Linden Lab, in a status page, to internet problems. Other public news media reports an AWS failure, at the right time and in the right geographic region. Inara Pey, in her report on the SUG meeting. gave a version number for Tuesday's MC release, 566406, described as necessary for the change to IM/email options, but while something new was deployed the RC channel, no version number was reported. This should be the last server deployment for this year The SUG meeting nest week is apparently intended to be a Solstice Party, but even if it is transferred to an alcohol production facility, I have doubts about Linden Lab planning. (NB: I am British...)

They're doing rolling restarts, but no sign of new code. Have they finally got that working as planned? Who knows...

Do you have a URL to a web page, definitely controlled by Linden Lab, that documents this?

I would just like to note that the LOD switching distances are calculated differently for rigged Mesh, and that ripples through in display-weight/complexity calculation and what people can see, but too many Creators seem to neither know nor care about this. I don't claim to have got this right on my clothing products, the High-LOD can look a little poor if you zoom in pervy-close, but I try to have things looking OK.

Overnight, I received an email purporting to be from Second Life, which fits a classic malware delivery pattern. 1: It claimed to be from "secondlife@stellaconnect.net" 2: It claimed to be about a service-quality survey. 3: Responding to the survey requires me to click on a link to "stellaconnect.net" I have never seen this domain linked to Second Life before, and step 3 is a common point to deliver and install malware, by taking advantage of a software bug in the OS or email client. There is a genuine stellaconnect but my preliminary check only revealed a "stellaconnect.com" I would urge anyone receiving such an email to be very very careful.

I don't have strong feelings about the theme, and the echoing of the old Meadowbrook style. Looked at as architecture, all the themes feel like American fakery and speculative building. I moved from one of the old Tahoe homes to a Chalet home. The change from prim tech to mesh could be a driver for this theme, and could make for other themes, but will we ever see the big numbers? I'd be unsurprised by an echoing of the Tahoe theme, it's still American, but can Linden Lab ever get away from an American feel again? As a matter of architecture, Linden Homes are always going to be awkward, echoing Pete Seeger's "Little Boxes". It's inevitable, in essence it is mass-production, nearer to an apartment block than individual constructions. And headline architecture is going to be closer to Falling Water, a unique product for a single customer. But the water and the overhangs, it suggests possibilities for a future theme.

The basic problem is centred on the 8-bit range available to cover darkest to lightest visible tones, and how what we can see is affected by our monitor settings and the lighting where we have it. Whatever we do when we create a texture, we can't be sure what the user will be seeing. It's an old problem, ask an old-time photographer. 8-bits is what they would call 8-stops, neither is the brightness range of reality, and even today, with a very good computer monitor, it is hard to show a full 8-stop brightness range. I am old enough to remember analogue TV broadcasts, and TVs that used cathode-ray-tubes, and how fiddling with brightness and contrast knobs on the back can make huge differences to how the picture looked. At least modern hardware doesn't drift away from how you set it as easily as it did then. But, whatever you do, get your local display as good as you can. How? I use this web page. which has test images and instructions. http://www.lagom.nl/lcd-test/ It can be a bit overwhelming, but the black-level, white-level, and gamma images are enough for a quick check. What makes film different is that the gamma, the relation between input and output, isn't a straight line. And it could be affected by the chemicals you used to process the film. We can do the same sort of adjustment with Photoshop, and similar software. I use The GIMP. Whatever software you use, you can make the darkest part of the texture a little lighter than totally black, and the lightest part a little darker than totally white. And why would you do that? This is where EEP comes in. The highlights are a little brighter than the texture, and the shadows are a little darker, and if you have a pure, full-brightness, white, there's no room for the extra white of a highlight. Small white bits of the texture might be OK, but the white of a snowscape is all close to white. For film photography it's difficult, you have to make the snow less white than it really is. You can do this with an environment setting, but if affects everything. It darkens the clothes people wear and the buildings. The old-time photographers did hours of work to get from the brightness range of the landscape to something that worked for the brightness range possible for a photograph printed on paper. And the EEP setting is just one tiny part of that total. Just dropping that ultra-white snowscape a little bit only affects that texture. It doesn't affect the clothes somebody wears, or the textures on a building. And you can still use an environment setting. But an EEP environment setting is for everything you can see. We have a lot of snowscapes around the grid, and you can see some of the differences on the maps. There's a large area, the Snowlands, where the map tiles are not solid white. You get a sense of the variations of the landscape. And when you visit, a personal environment setting that works in somewhere such as Bay City or the Blake Sea is likely to be usable. Other places, on the map they look flat white with buildings. It might be a totally flat region but it's not a good sign. And then you use the in-Second-Life camera, and throw Photoshop work at it,. That's really where you would do what the old-time photographers did, dodging and burning, choosing the contrast grade of the paper you are printing on, waving your hands in a beam of light like some magician. I had the chance to try that. Never got it to work well. I can do it in the digital world, slowly What we use Second Life for is live pictures. Think what that sort of correction would do to the frame rate. Do it as early as you can, and be as specific as you can, before you ever get into Second Life. (I think I drifted away from hard technical details. But maybe the key point is that creating scenery is not like taking a photograph. For a moment, you are closer to God than than to Ansel Adams. "Moonrise, Hernandez, New Mexico" needed both.)

The Status reports show the restarts as running but I have not seen any of the usual signs of restarts happening. I have been doing other things but I know of in-world indicators I can check.

I had a couple of RL distractions, but the concurrency numbers are now showing the expected drop.

Just the Status Page messages.

Has it finished yet?

I think it is possible that the "big river" through the Chalet and Stilt regions could be part of any northward extension. Some rail-compatible theme to the north, with the "big river" as a boundary? We are seeing defined-edge regions appearing for the Chalet and Fantasy sub-continents, very narrow waterways modelled rather than a sharp cut-off.

It's been a while since I made any train runs, but it is worth remembering the the SLRR/VRCC scripting that is generally available and used is very old, and there have been a lot of changes in how LSL works. Some people have moved on, some I am not sure about. I have written some fairly efficient to animate mesh components for a steam engine, but it depends on the ancient code to move and guide the vehicle. I have tried to understand the code for that, and maybe one day the headaches will stop.

While my situation is different in so many ways, running the Firestorm viewer under Linux, there does seem to be something risky about the latest nVidia driver versions. I have a suspicion I am seeing a side effect of the viewer's dependence on OpenGL, but, whatever your system, I would recommend being prepared to roll back any nVidia driver upgrade.

It's not a total waste, but that 90.000 figure feels ancient. It's described as a "record for concurrent players" and I think that must have been before my time. The numbers Linden Lab used to report, concurrent players and total sign-ups, could come back and bite them. The Roblox comparison seems to be about the Prim-era Second Life, and the pictures are at least modern. But the reporter doesn't seem to realise that SL depends on graphics power. It doesn't need the exotic, and it maybe needs more RAM than a current games machine, but something that benchmarks at the level of an NVidia 1050 card is what I would suggest. Current hardware prices and availability are afflicted by worldwide supply problems. And, frankly, the talk about the Facebook/Meta plans to recruit 10,000 staff in Europe alone is likely misleading. That's how many Linden Lab equivalents? How many programmers? How many Mole equivalents? I'd count myself as a Creator, you can find my stuff on the Marketplace, but how many of those 10,000 staff are going to be at my level? I think I have done some stuff pretty well. and I think some SL Creators are misunderstanding some parts of the job. No, I don't really trust the BBC. But that's more of a problem with their reporting of UK politics. This feels more like a knowledge problem.

OK, that all explains the behaviour of the concurrency numbers for Tuesday and Wednesday. There's a very regular pattern, every day, a consistent shape to the curve. A restart shows a very large dip in concurrency (bots, I suppose) Nothing much on Tuesday, a typical smaller dip on Wednesday. I get the data from here: softhyena.com but it only shows the detailed concurrency for the last 24 hours.

As with any other computer hardware, these days, SSDs are not always what they seem. The big fake hardware has been GPU cards, and there are SSD brands which I have never heard of before. I have been careful, and I do get decent results for SL caching. There are a lot of traps in how the caching works, but the bulk of the data arrives at a rate limited by your internet connection.