Qie Niangao

Probably the copy of the animation in your local cache is corrupted. You can try the normal cache-clearing thing from Preferences, but you may need to manually remove all the local saved cache and settings files. (That's pretty close to a "clean install" so if you're comfortable with that for whatever viewer you're using, go for it.)

This is what the manual says: For dual channel configuration, you always need to install two identical (the same brand, speed, size and chip-type) memory modules in the DDR2 DIMM slots to activate Dual Channel Memory Technology. Otherwise, it will operate at single channel mode. I honestly don't know if you can get by with a slightly different model by the same manufacturer and the same frequency. They're probably also testing the detailed timing specs, too (latencies, etc), so it's most likely to work if you can find the exact same model. Incidentally, that graphics card ("Asus" rather than "AUSS", I assume), looks reasonable to me. I mean, it's not exactly top-of-the-line, but it only draws 75W, and it doesn't require a separate power connection, so it looks likely to work for your application. It benchmarks 3D about ten times as fast as the graphics card you have now.

Wow. So, the way memory works in those machines is that a single slot runs much slower than two slots both with identical memory installed. Speed of memory access may not be the main bottleneck for your machine, but nonetheless, if you can find another 2GB DIMM that's identical to the one you already have (preferably exact same model, but definitely all the exact same timing values), it is very likely to make a real difference for general use of the machine.

Bobbie Faulds wrote: It probably uses a PCI rather than a PCI-e slot, ... I worried about that, too, but this machine is not quite that old. That old geforce card is actually PCI-e. First generation, but still a lot better than AGP or non-express PCI.

emapen Juliesse wrote: i don't want to spend money at the moment I don't game I only use sl for english improvement. do you think my pc would benefit from adding 2 more ram even not considering (apart from) sl? Depends what else you're doing with the machine. If it's routinely swapping to disk, then yeah, more memory would help. And as Sassy suggests, one might expect Windows 8.1 to want to swap quite a bit with only 2 GB of physical memory. The problem is, your motherboard has just two DDR2 memory slots, so to increase to 4 GB, you'd need to remove the two 1 GB sticks that are in there now, and replace them with two new 2 GB sticks. Unless you have that new memory already, it will cost maybe US$60, and you will not be able to re-use it when you eventually replace this machine (because a new computer will not use DDR2 memory).

Mircea Lobo wrote: If it's a new project, starting off as closed source is of course Linden's choice, and not for me to complain over. Note also that so far we have only heard that plans for it are exactly that: to start as closed source. The SL viewer was closed source for years before it was opened, and personally I would choose much the same path for a new generation: start closed (although this time develop with the knowledge that eventually it will be open-sourced). It might be a different story if there were already fully-specified requirements, or full-specified changes from an existing implementation. Then third-party developers might be able to contribute to a shared open source tree (maybe -- but it would require vastly more discipline than the current TPV development approach). Of course, the prospect of ever opening the source does impose strict limits on what kind of licensed software can be included, so that would likely mean that any possible open-source approach could only use older components, so that may not be good either. As we've seen with some of the current viewer's componets, it can take years for open-source alternatives to appear, if ever.

That's an important distinction; I'd just further add that LL's interpretation need not necessarily permit all and only what the law allows. They have a pretty big incentive to allow as much as possible, and another big incentive to disallow stuff that actually violates some RL law, but they also have only a limited amount of revenue coming from in-world gambling ("skill games"), so they can only afford to spend so much on enforcement. Hence they necessarily will cut corners, and that may mean that they will not be willing to investigate whether a particular instance of a product has been configured "no-lose" whereas other instances of the same product are configured as skill games. If the particular corners they cut aren't what we'd prefer, we can b!tch and moan about it and they might change their minds, or not.

This may not be the place for this to be addressed, but somewhere now might be time for additional clarity about possible pricing of the Experience feature. I'd forgotten about any fee for Experience Tools, until yesterday when the subject came up at a Server User Group meeting. The way it was described by the resident who raised it, the prospect sounded scary as hell, as if Experience owners would be charged a lot. I've subsequently read that the original idea was to charge some nominal fee -- perhaps comparable to a Mesh upload or something -- just to discourage possible griefing / spamming use of the tools. There's a big difference in perception between paying some nominal one-time fee, versus hefty ongoing charges to finance a database of large-scale persistent store. In fact, I don't have a problem with a separate product offering for a larger or higher transaction-rate persistent store, but I'd really like to know that LL is committed to a modestly-scaled "entry level" Experience service with no recurring fees and only a nominal setup cost (order of magnitude L$100 - L$1000, say). Is anybody able to rein-in speculation about this, to some broad-gauge estimate of what we'll need to pay? (and/or any other information about how availability of Experience Tools may be constrained?)

Michaele Avelino wrote: Hmmm I think nothing because I have 7 days bevor the rent end... dont answer because I had things to do. What could possibly go wrong?

Power supplies such as these things. The reason it could be an issue in this case is that good graphics cards use a lot of power (and dissipate a lot of heat), more than the average desktop machine needs, so the power supply is one component of a refurbished desktop that's likely to need upgrading when a graphics card is added. This used to be even worse. For example, the nVidia 4xx series was notoriously power hungry; successive generations of cards have delivered better performance for less power, and therefore less heat, and therefore less fan noise. Oh yeah: cooling. It's possible you'd need to replace or add fans to a refurbished computer's case. I think that can wait until it proves to be a problem, if ever.

zombiegal wrote: the land must remain set to the rental group for our tier bonus that helps us stay competative on the market. Well, it needs to remain deeded to a group, and that group must be controlled by the rental company for land management abilities. But the rental company need not have been the founder of the group, it just has to be the sole remaining Owners. So if the RP group (or a successor RP group) is willing to make Owners of your rental managers and then leave the Owner role themselves, you could safely deed the land to that group, and then let them have whatever Roles with whatever Abilities they want, short of selling-off the land or buying more. It does also mean that your rental manager(s) will need to reserve a slot for that extra group. That's not a big deal for one big important tenant, but can be a pain at larger scale.

A MacBook Pro? I don't think so. I'm no Mac expert, but as far as I know, those notebooks all come with embedded Intel HD4000 graphics subsystems, with no way to upgrade (and certainly no way to use a desktop graphics card). Generally, notebooks are very considerably more expensive for comparable performance especially for graphics. If you don't need a notebook, you'll get much more bang for the buck with a desktop. Those 680s are very nice cards indeed. I'm perfectly happy with a (slightly overclocked) 660; I use nearly Ultra settings but with short draw distance and no anti-aliasing because I much prefer that, but your requirements may make use of the extra capacity of the 680. One caution: if you'll be adding a graphics card to a refurbished desktop, be sure it has (or you add) a big enough power supply to keep that graphics card happy.

Gadget Portal wrote: They can get away with doing it without falling under gambling regulations because no matter what, you're getting something from the pack that they said you would, even if it's not the Sword of Awesomeness. That's a good analogy with gachas, but not so much with sploders. Even though the Sword of Awesomeness might have some RL market value, it is not the same as L$s, which walk a very fine line between game token and fiat currency. The current debate, however, is even more specialized, and has to do with whether a device that can be configured to be a game of skill (and hence subject to those new rules) will always be treated as a game of skill even if it's configured to behave differently in some cases. At present, the smart money seems to be on "yes", and that the device creator will need to generate two distinct products, one of which is in the skill gaming category and the other doesn't handle L$s at all (or is something altogether else, like a gacha vendor or something).

I own a little plot with roughly 536 sm and 132 prims. The plot is about 23x23 mtr. If it supports 132 prims, it's 576 sq.m., and if it's square, it's 24 x 24m. (It can't be 536 sq.m. because land is always an integer multiple of 16 sq.m.) If it happens that you're neighboring Linden protected land, you'll usually be fine encroaching on it, especially if you do it at a high altitude. In practice, the same may apply to abandoned land, at least until it gets auctioned off, which may be never. (As already mentioned, you'd need to keep everything rooted on your own land, using your own prim allotment, otherwise it will get returned.)

but i do not know about the test avatar thing Yeah, it's complicated because it's in different locations for different viewer versions and TPVs. In recent Linden viewers, it's in the Develop menu (search for Test Male or "Test Female" on this page). Getting the Develop menu to show is another hurdle, a preference dependent on choice of viewer. Anyway, the point is just to get rid of all attachments and see if it helps; the test avatar is just a shortcut for doing that (and replacing all worn textures with basic Library stuff).

A couple of easy things you might try. First, reboot your router and any other local network gear including your computer and anthing else between it and the ISP's entry point. If that doesn't help, then remove *all* attachments from your avatar and try again. (That means removing everything including hair, HUDs, mesh/prim clothing, etc.; in fact, the easiest way is to use a "test avatar" if you know how to do that.)

Not to justify this script behaviour (I suspect the scripts are just stupidly protecting themselves from being redistributed), but there are related complexities in securing from prying eyes textures and their UUIDs.* I agree that it's pretty clunky to pass out actual texture assets just to permit retexturing of an object. In the previously-cited case of OpenCollar, of course, those textures are intended to be freely distributed -- hence "OpenCollar" -- and of course it's a win if an item's textures are already in wide use elsewhere so they're already loaded in many viewers' caches. But smart or not, most creators will want to keep their textures more private than that. Instead of needing texture assets, retexturing scripts can use texture UUIDs, but still somehow the scripts must know those UUIDs, and that's not as simple as it sounds. If the UUIDs are stored in notecards, besides being annoying for the creator, any notecard readable by the intended scripts is also readable by a trivial script that dumps the contents to local chat. (Only products that are both no-mod and no-copy are immune here; otherwise that notecard is an open book.) A more sophisticated way to do this is to embed the UUIDs in the retexturing script itself, custom-generated for each product. That's how I'd do it myself, so the creator just copies the generated text as their own script. Even this, however, has a vulnerability: the script can be used to put those textures on other objects. In practice, that's not likely to matter; I mean, who really cares if a customer can paint their couch and lingerie with the same set of textures? But if a creator did care, the script would have to be fussy about only running in objects created by that creator. *Ignoring that some TPVs merrily blurt out texture UUIDs because they can, and so they must, because SL just isn't complicated enough without moar viewer features.

This thread is instructive for me. After seeing this, I'm thinking more and more that we need FOSS scripts that do this retexturing function efficiently and conveniently, maybe accompanied by an optional resizer, all equipped with the newest LSL functions (e.g., materials setting as well as full-object rescale in a single call). I don't know when I'll get time to work on this myself, if ever. (I'm kind of buried in the Experience Permissions beta -- which is extraordinarily cool, by the way. Stuff in here with all kinds of applicability beyond what we thought of "Experiences"... stuff of scripters' dreams, I tells ya! But I digress.) What's so instructive is the kind of requirements that creators would have for such scripts. For example, the problem of not resizing a "DEMO" prim, that's so practical and so simple -- and something I'd never think of myself in a million years. This part confused me: "What me as a merchant would like to see is a script that wont ask me to put my textures or my texture id on the object im making." Now, if this just means that the product shouldn't expose those textures nor their UUIDs, that's clear. But if it's saying something else, something about how the scripts should get loaded with the texture information needed to operate, then those are requirements that need some further specification.

1. Is there "Mainland" thats an Isle or Ocean? I want it too look like that but it should be "Mainland" for the above mentioned reasons. For what you want to pay, you will necessarily have neighboring parcels in the sim, whether it's Mainland or Estate. That said, here are a couple of currently for sale parcels in your target range (remember, a 4096 tier gets and additional 512 "bonus" tier from premium membership). L$ 45K a 4064 water parcel (if you wanted an island, you'd have to fake it with mesh) ... overpriced, and with ugly neighbors About L$20K, a 4496 beachfront with some ugly neighbors, but not dreadful. Obviously I'm not recommending these parcels, I'm just pointing out that such things exist.

[...] 2. If I have like 4000sqm of "Mainland" do I have full rights then with no one above? So I can terraform (not much) and set the rules without asking someone but linden lab? This is the second time you've used "set the rules." There are certainly land permissions you can set as owner, but it's a little worrisome, this "rules" term. Anything specific you have in mind there?

DarranDax wrote: Second life has no quests, no dungeons, no storyline, it´s about interaction and building. And that´s what I want. So I assume the only possibility for me is getting a mainland parcel? Are there parcels considered mainland, who meet my needs, like tropical terrain and maybe 2000-4000 sqm? Will I be able to terraform and set my own rules when just owning such a small fraction? That would still be like 25$/month plus premium membership, but would be ok with me. You can absolutely do this on Mainland, and it's not very expensive (about which more below), but the problem is that your neighbors can do the same, so the overall experience is a combination of all those neighboring parcels, and that can be pretty discordant. And controlling just how wildly discordant that gets is why the Mainland constrains some things. Most of Mainland, for example, can only be terraformed +/- 4m from it's reverted terrain height. That's generally plenty to shape the land for a build that fits the terrain, but it's not enough to make an island from a deep water parcel. (Some rare and expensive Mainland in "city zoned" sims cannot be terraformed at all, and a few of the nearly oldest sims have a +/- 40m terraforming limit -- and a visit there will reveal why that's a disastrously bad idea for sims with multiple landowners.) By "not very expensive" I only mean "less expensive than it might appear at first," and that's mostly because Premium membership (on the US$ 72 annual plan) is as near to free as makes no difference, once you take into account the value of the L$300 per week stipend and the 512 sq.m. of "bonus" land tier. The painful part is the monthly tier needed to get a workable amount of land. You've seen the fee schedule, so you know it's US$ 15 - 25 to get the kind of space you describe. You also have to find and buy the actual Mainland parcel, and as you say, you want what most people want: water, beach, or island, and it's a free market so you'll have to pay something for that. (On the plus side, the search is kind of fun. Just open the Map with the For Sale parcels showing, and look around the continents for interesting spots. Be a little careful near the Blake Sea because there are some Estate sims there, and an Estate parcel "sale" price is pretty meaningless.) On the other hand, it's also true that there are reputable Estate owners who offer comparable alternatives, probably a bit less expensive, with the downside that you're completely at their mercy. But, you know, it's a market, and they have every incentive to treat their tenants well. You just have to be sure they're reputable, and they're offering the kind of permissions you think you want on your land.

If About Land shows you as Owner of the parcel, you're all set. It can take a day or more for the Map to update. (It's even weirder than that. Sometimes it will show the yellow color gone from the parcel but still have the tiny "$" icon, and sometimes vice versa.)

Sathya Breil wrote: The scripts I use are really well known. Its not the first one I tried. I ended up with these because they were easy to work with. 90% of the scripts I tried had thjs feature. The scripts itself are no mod, when I build the product i need to use a licence script, that will ONLY work if the object is set to no mod, and then i can put the master script for the hud or slave one for the garment. If the object perms are modify the script delete itself from the object, so it cant be used at all. The only one Ive tried that didnt had this issue had a worse one. It had a resizer included and even when set to disable the resizer was workig; this must sound like who cares, but if were talking about a demo, its a huge issue because the demo sign could be resized to disapear without afecting the mesh, if it was riggged. This scripter in particular knew about this, and the solution I was offered was to use Mega prims to attach at demos...... And trust me, it wasnt a cheap set of scripts This situation is just crazy. And even worse, in an earlier post, you mention that the scripts don't even include a way for the end-user to delete the script from the item. (I grant that a basic re-texturing / re-sizing script need not be so large that it's grid-threatening not to be deleted, but it's such a standard functionality now, so expected by customers, one would think the scripter would need to include it just to keep any business at all.) What I'm finding so weird about this is that no creator should want their scripts to work differently from any other creators. That would be counterproductive because the more common a script, the more likely the buyer will already know how to use it without the merchant getting as many support requests. So it's not simply that the scripters suck, it's that economics doesn't seem to be working. Very odd.

I'm not sure which mechanism this HUD is using to detach when it leave the region, so it could be that anything that's supposed to run at that point is necessarily unreliable -- as in the case of teleporting into a no-scripts region, perhaps. If that's a real problem, it may be better for the HUD to actively record its continued attachment, and when those updates cease for too long, conclude that it's detached. If this kind of reliability isn't all that important, you could have the HUD send a regular old non-Experience message (llHTTPRequest) to a server that is on an experience-enabled parcel, and have that server update the Experience Persistent Store.

Thanks!