Jump to content

Jennifer Boyle

Resident
  • Posts

    2,121
  • Joined

  • Last visited

Everything posted by Jennifer Boyle

  1. If Karolina wishes, I'd be happy to share more detail about making alpha layers. I have not learned how to use BOM yet. So, if I do, I can make alpha masks that cover exactly the parts they need to and aren't constrained by the areas defined on the HUD?
  2. Did you pay with a credit card? If so, dispute the charge. The seller defrauded you. That computer isn't adequate for most modern games. Fraud should be ample justification for a chargeback, which should be a strong incentive for the seller to accept a return. I, personally, would not buy a computer of unknown provenance from a stranger online. It's way too risky.
  3. As I said in an earlier post on this thread, the reason I have such an insane amount of RAM is that I hoped that putting the SL viewer and SL data files on a ramdrive would help performance, but it didn't. At least at my level at technical competence, that seemed reasonable, since, in one of the only two sims that I frequent, I often wait, and wait, and wait for textures that should be cached to load when I logon.
  4. If you are using a classic body, it's pretty easy to make your own alpha layers. In my pre-mesh-body days, I used to do it pretty often, and many times when the one that came with the clothing didn't work well, the one I made worked perfectly for me.
  5. If that's the case, it isn't very effective. I have three alts who are premium members. It's not a secret that are my alts; their profiles say that they are. ETA: I don't play Linden Realms.
  6. Yesterday, I watched the names on calling cards in my inventory change.
  7. I should be more restrained. I wasn't as restrained as I should be because I'm frustrated that poor performance has been allowed to continue so long, that LL hasn't been forthright about the cause, and that, due to that I have wasted so much money doing everything I could think that I could do to improve it. It feels like LL purposely withheld information because they had rather let me waste a lot of money than admit their shortcomings. Actually, the most instances of a standard graphical viewer that I can run and get decent performance is four. According to Windows Task Manager, with more I run out of GPU resources.
  8. "Add more RAM to boost average speed by 10% until you need to upgrade your GPU for further gains," I have 128 GB RAM, so adding more not only won't help, but isn't possible. "Improve the network to speed texture downloads before any rendering improvements will be noticeable."Nope. I have 300 mbps fiber-to-the-house, so that isn't my problem, either. My real problem is that LL won't optimize their software to provide a better experience to users.
  9. We have a new example of LL's misguided priorities:EEP is Out! Introducing the Environmental Enhancement Project I need better performance a lot more than I need that. Yes, it is really nice. Yes, I would absolutely love that they did it, IF SL PERFORMED DECENTLY WITH THE FEATURES IT ALREADY HAD. Why do they keep adding bells and whistles instead of giving us better performance?
  10. What monitor re*****ion are you using? ETA: Why does the forum software not want me to say the word that sounds like REZ-O-LOO-SHUN? Let me rephrase. What is the product of the number of vertical pixels and the number of horizontal pixels on your monitor?
  11. Since my CPU speed is limited by CPU temperature, would I get better SL performance by disabling all but a few cores? I'm think a few could run faster if the others were disabled. I suppose it would help, too, if the viewer was the only thing using one core. Right?
  12. Mine is custom-built. It is overclocked, but by software. It has software that adjusts the CPU clock speed according to the CPU temperature. I am not technically-adept enough to know where bottlenecks are. I can say that each instance of Firestorm uses about 20-25% of my GPU's power, so I can always run three without much change in performance, often four, and never five. CPU load is never very high.
  13. Addicts will stop at nothing in their (in my case apparently misguided) quest to get SL to work better.
  14. Great idea. But why not have the server optimize content that needs it? Surely algorithms exist, or can be created, that can do that. I am not (much of) a coder, but, surely, it's not an insurmountable problem for a program to evaluate the size and shape of an object, evaluate the number of polygons, evaluate the size of the textures on it, and reduce the number of polygons and the size of the texture to the number needed to produce a nice image. I imagine it would be easy to provide the user with settings for quality of images that would govern the number of polygons and the size of textures sent to their viewer. What am I missing? LL could also monitor the quality of content using algorithms, and ban the sale of content that doesn't meet announced current standards.
  15. Sorry. It was posted while I was typing. Damn, I sure wasted a bunch of money. It angers me that LL didn't give me good advice to the effect of, "Don't buy a high-end computer hoping it will help SL performance. It won't, so don't waste your money." I'll bet that, in the aggregate, residents have wasted millions of dollars buying better computers that they hoped would run SL better, because they didn't know that it wouldn't help, money that, otherwise, would have been available to spend in SL, increasing LL's revenue. It's hard to believe that SL couldn't be rewritten to utilize multiple cores. Practically everything else has been. If the problem is that content is dynamic, why did people writing software for video editing, facial recognition, etc. manage to overcome it, but LL can't? Perhaps LL needs to redirect their emphasis from adding bells and whistles to just making the product work better.
  16. I’m glad this thread was revived. I, too, am frustrated by poor performance of SL viewers. I have tried everything I could think of that I can do to improve it. I got my current computer specifically to (hopefully) get better performance, but I was disappointed. My computer: CPU: AMD Ryzen Threadripper 2950X 16-Core Processor (3493.43 MHz) Memory: 130946 MB OS Version: Microsoft Windows 10 64-bit (Build 18363.778) Graphics Card Vendor: NVIDIA Corporation Graphics Card: GeForce RTX 2080 Ti/PCIe/SSE2/11 GB The reason I have such an outlandish amount of memory is that I hoped that running SL viewers from a ramdrive would help. It didn’t. The OS is on an SSD. The viewer program files and chat logs are on a different SSD, and the cache, which is set to the maximum possible, is on a third SSD. I have a 300 mbps fiber internet connection. I use a wired Ethernet connection directly to the six-month-old router. What else can I possibly do to make SL viewers perform better? I use Firestorm 99 percent of the time. Two things are more frustrating than any others. One is slow loading of textures. Many times, I teleport to one of the two sims where I spend 90 percent of my online time and, even though most textures should be cached, loading often still takes a very long time. It varies a lot from time to time when I have changed nothing, so it’s not on my end. The other is low frame rates. I recently spent several hours experimenting with graphics settings and comparing frame rates. This was not a scientific experiment. I stood on the roof of my house and looked east, over a developed area where there were very few avatars. What I found was interesting, and, I think, may be useful to others, so I’ll share. Only three settings had a noticeable effect. They were monitor resolution, shadows, and draw distance. None of the other graphic settings made enough difference to notice, but these had large effects. I tested every monitor resolution available to me that had the same aspect ratio as the highest. They were: 3840 x 2160, 2560 x 1440, 1920 x 1080, 1366 x 768, and 1280 x 720. I used draw distances of 16, 32, 64, 128, 256, 512, 1024, 2018, and 4096 meters. After every change in resolution, I restarted the viewer (Firestorm). I used the wide range thinking that extreme values might reveal more. At the highest three draw distances, the performance was poor at all resolutions, with no frame rate higher than eight and none lower than three, and little change as the resolution changed. At the highest resolution frame rates were about equal at 11 to 12 up through 256 m; at 512 they dropped to around 6, were the same at all higher resolutions. At lower resolutions, frame rates at lower draw distances were very good. As draw distance was increased for each resolution, the frame rate didn’t drop much until the change from 256 m to 512 m. At all resolutions but the highest, it didn’t drop much as draw distance increased from 16 m to 128 m, but then it dropped dramatically at 256 m and again at 512 m, but not much with further increases. Turning off shadows with all other graphics settings maximized, which I tested at 1920 x 1080 because I had decided that that resolution gave me the best trade-off of appearance versus performance, resulted in a dramatic increase in frame rate. My “sweet spot” is 1920 x 1080, at which I get a frame rate of 47 at a draw distance of 128 m, with shadows off and all other settings maximized, which doesn’t increase with further decreases in draw distance. I get an OK frame rate of 36 at 256 m, so that’s useable if I need to see farther. I love shadows, but not enough to endure the hit on performance they cause. In the future, I’ll use them only for photography. I hope others find this helpful. I have two other performance-related questions for people who know more than I do. Would adding another graphics card help? Why is the cache limited in size? There are directions on the Firestorm web site for increasing maximum cache size, but they are over my head. Added after seeing new posts: Multi-core processors have been out ten years or so. Is it too much to expect LL to update its software to be able to utilize their power after that long? It seems to me that that should be more important than adding features like experiences.
  17. I would do that in a minute, if I could put links to outfit folders into outfit folders. As it is, I'd rather put in the time at update time than have to add two or more outfits each time I change. However, since the recent Maitreya update, I've had problems with autohiders not working, possibly because they're getting added before the body. So having a scheme that assured that the body was added first may have an additional benfit, and I'll have to give the idea more thought.
  18. I don't disagree with the other answers. For me, the major obstacle to changing to a mesh head was the difficulty of keeping the same face. I had had essentially the same face since 2007, and it was very important to me to look like the same person. I don't think I could have achieved that on my own, because it would have taken such an enormous amount of time experimenting with demos. I was very fortunate to have a lot of expert help, and I still spent a day tweaking the shape. Before that, I had repeatedly spent a few hours trying demos before giving up in frustration. The appearance of the body matters a lot more to some than others, depending their Second Lifestyle. There's really no reason for someone who always wears long pants, long sleeves, and boots to have any mesh parts other than hands and head. OTOH, it would be more important to someone who wears scanty, see-through, or no clothing frequently. Another disincentive for having mesh heads and bodies is that they are much more complicated to use, assuming that one makes full use of their features. In addition, every time their is a major update, I can expect to spend at least a few hours updating my many outfits. The "replace links" feature is a godsend for this, but it still takes considerable time. My three alts have mesh bodies and system heads, so I also need to invest time for each of them, though it's mush less because their wardrobes are much smaller than mine. They don't have mesh heads mostly because of the cost, but also because of the additional time it would take to add and maintain them. Finally, Most people I encounter who are on SL regularly and are not new have mesh bodies and heads. Most people I encounter who don't are either new, are infrequently in SL, or are returning after a long hiatus.
  19. I have an object that I want to wear all the time when I am in one sim, but not elsewhere. I would like to have a script that I can put into an object that I wear all the time that would automatically add it when I enter that sim and remove it when I leave. Does anyone know of such a script?
  20. I think I used to be able to do that. Now, the closest I can come to is is a search that returns posts by others in which my name is mentioned.
×
×
  • Create New...