Jump to content

Flea Yatsenko

Resident
  • Posts

    489
  • Joined

  • Last visited

Posts posted by Flea Yatsenko

  1. 20 hours ago, Henri Beauchamp said:

    Wine got bugs (at least two of which do impact viewers), making it a very bad option to run a viewer under Linux. I got workarounds for those bugs in the Cool VL Viewer, but even so, the stability is at best fragile, and the frame rates and smoothness very bad...

    Just run native Linux builds.

    Get lots of mesh and material upload errors in Firestorm alpha. FS will give me MISSING_NAV_BLOCK yet official client uploads fine. FS Alpha won't bulk upload materials either, always gives an error. Sometimes I have to use it, though I haven't tried yours in WINE yet. Plus FS doesn't have HAVOK.

     

    23 hours ago, Ardy Lay said:

    Do publishers of viewers for Second Life provide lists of Linux distributions and versions on which their viewers run reliably? 

    Linux software should work in all (distributions) versions of Linux. Publishers just list specific versions because they don't want to get stuck troubleshooting someone's Gentoo installation where they have obscure setups. Some distros are a lot easier to use than others, and some a lot easier to fix or troubleshoot.

    Anything based off Debian or Red Hat is always a good choice, though I vastly prefer Debian (or Devuan since systemd is awful)

  2. 56 minutes ago, Ardy Lay said:

    /me comes in and starts a fire:

    Linux is not THE solution, it is, or, should I say they are, a sea of solutions looking for users.  Do publishers of viewers for Second Life provide lists of Linux distributions and versions on which their viewers run reliably?  The person that chimes in with "compile it yourself" might be in for a ride.

    I remember pitching a fit when Second Life would no longer run on Windows 2000 without tricks.

    Now we are living in dread of the death of Windows 10, are we not?  I'll hold out for a while on a few machines I use around the office because they still do their little duties okay, but, for Second Life, I am one of the weird ones that buys new hardware every few years just for Second Life Viewer to run on.

    Wow, we need Python to launch Second Life Viewer?  Was that really necessary?  I suffer constantly at work because of thousands of systems running R.A.D. puke instead of compiled-for-target applications.  The stacks of interpreters and the chains of scripts they run, all passed off as commercial products make my work life hell.  Now consider that we put the target platforms in the hands of people that want an appliance to "just work" and witness the complete failure of a rickety R.A.D. demo released into the wilds.

    I would suggest replacing that version checker with something that doesn't add unexpected dependencies.  Second Life Viewer is very much already a dependency nightmare without it.

    Firestorm works perfectly fine for me in Gentoo Linux. I just download the tar, extract it, and it's fine. The official client in WINE loves to crash on me though. It's a pain.

  3. I have a similar problem under different circumstances. I resolved it by removing the SL updater exes from where SL is installed. I forget exactly which exes to delete, but if you search for Linux WINE crash on start you should find something. My problem wasn't SL client, it was the SL updater and it caused the whole show to not run.

    I think that's something LL needs to fix, because if anything goes wrong with the updater the client fails to load and they have a user who is upset they can't run SL. SL starting shouldn't depend on the SL update services succeeding when they can just log in, get the client blocked message, then download the new version.

    Anyways, try deleting the update exes and stuff in the install directory (MAKE A BACKUP!!!) and see what happens.

  4. Deleting in world stuff owned by people who haven't logged in in a long time makes no sense. Someone is paying for that land and they are keeping it there for a reason. Unless I'm mistaken there's no way to have free land in SL that could be abandoned for years. You have to constantly be paying for it. If someone is paying for a sim and it's filled with stuff from someone who did leave SL, they have the ability to return that stuff if they don't want it anymore. So that's not really a problem. I don't think any inactive in world stuff is a problem because someone is paying to keep it around and they have the ability to make it go away with a few mouse clicks. I don't understand why people are getting so heated about this. I have an in world store, if I decided to leave SL forever and just abandon my store as it, the sim owner would return all my stuff and rent it to someone else. Two cases of people who left SL with in world stuff I can think of and none of it is a problem. If someone is the sole land owner and they leave and stop paying, how long do you honestly think that land is going to stay in the state they left it? Old in world stuff isn't a problem, at all. Because having to pay for in monthly ensures whoever has in world stuff is active enough to pay for it.

    However, SLMP is a mess and this horse has been beaten to death. But I'm gonna share another angle since the only people who really complain are merchants.

    It's not fair to customers who buy things from people who aren't active merchants anymore. Customers have no way of knowing from a marketplace listing if a merchant is active or not. People are gonna feel a lot better about buying things if they know they'll get answers. Merchants have a lot to lose from all these outdated listings but customers have a lot to lose too. But no one ever stands up for the fact they basically get ripped off buying some outdated product because of the listing and how SLMP works. All they can do is leave a review. Who here hasn't bought something off SLMP because it looked good in the pictures, realized it was outdated, then went to the profile of the creator and see they've left SL? It's happened to me more than I want to admit. Do you think the average new user or casual SL user knows what mesh, PBR, etc is? Or do you think they just buy by pictures? This old content is hurting customers, especially new users who are buying old outdated stuff and don't know any better.

    Imagine knowing nothing about SL other than it's some game you found online or a friend told you to play. So you set up your stuff to buy L$, spend some real life money on L$, then go buy some random old stuff from the first page of search results. You honestly think keeping around a bunch of old stuff on SLMP is worth putting new users through that?

    SLMP is LL's chance to show a new user the exciting and cool content SL has. We all know SL's content is ridiculous in scope and quality compared to other virtual worlds. Right now it's completely failing to do its job. And the MAU is shrinking. It's not hard to understand.

    If reddit was filled with old posts from 2010 and people started freaking out because they thought about removing old stuff or burying it in an archive or something, no one would be using it.

    • Like 2
  5. 6 hours ago, Mistress Morbid said:

    How regular? Items are already removed from the marketplaces of merchants who don't log in, under the following conditions -

     

    Those conditions are not very good. If it sells, it stays. And SLMP search has been borked recommending ancient and outdated products which people are buying because they can't find newer stuff since search has been so bad. People are buying old stuff and not realizing it. It's happened to me because sometimes you can't tell if something is mesh or whatever just from the pictures.

    SLMP only ranks the top 50k selling items. They need more requirements to filter out active merchants and modern content from the older stuff.

    The whole point in wanting to clean up old listings and stuff is to make it easier to find newer stuff. Gaslighting about saying this is deleting history or censorship or whatever. It's not. It's about making SLMP search not have to rank the 15 year old prim shoes with keyword spam and heavily edited promo pictures against some brand new rigged mesh material shoes and trying its best to show you what you are looking for.

    If you're really so upset about this "deleting history" and giving LL "censorship powers" then put your money where your mouth is and move into a 15 year old prim house, get rid of your mesh AVs, and support historical SL by making sure you are wearing all sculpty and prim clothes. Get yourself some nice flexi prim hair and support vintage builders and make sure you keep SL's content history alive and well! Or are you going to tell me you don't actually want to buy that stuff?

    • Like 2
    • Haha 2
    • Confused 1
  6. On 10/5/2022 at 2:15 PM, Arielle Popstar said:

    For all we know they may have been rejected already considering their stopped development on the Mac mobile they were working on.

    It was never mentioned why the sudden flip flop but that seems the most likely cause. If they were rejected once, it is doubtful they would succeed in a future submission without some major changes that in all likelihood would drastically change S/L as we know it.

    I don't want to get too off topic, but they had a mobile viewer before the current Unity engine one. And they stopped development because they realized it wasn't what anyone wanted. IIRC it was more like speedlight, which already exists, and wasn't anything like the new mobile viewer they're supposedly still working on. They've been very quiet about it lately so I'm not sure what's going on.

    But there was a mobile viewer before the current one in progress and people were not excited about it. AFAIK the Unity mobile client is still being worked on. I would not give up on a mobile viewer yet. I think LL knows laptop and desktop is a shrinking market and they have to do something to get on other platforms.

  7. Just needs a secret key given to real world users, which gets sent to the servers on login. If it's a valid key and client, you are flagged as a trusted agent or something and you count towards traffic stats and all that. If you don't have a valid key, you are flagged as a scripted agent or unofficial client or whatever and you get restricted in specific ways like having your presence not count towards traffic. You still have to deal with people who want to run 10 instances of a viewer, but most people can barely run a single instance of SL. How many have a computer good enough to run 10 SLs at the same time?

    You aren't ever going to completely stop people from gaming traffic numbers and stuff. Just like no one can stop spam emails. You can only minimize them and their bad effects.

    There are coding libraries you can use to write your own bot software for SL. It's not like all these people with group inviters and stuff are using the official SL clients.

    • Haha 2
  8. Would be interesting to see if people running these bots are using a VPS/remote hosting or if they're running them off their own computers. No doubt all the serious ones are using IPs from datacenters.

    LL doesn't need to try and stop them, they need to make it a pain in the butt for people to run bots. I.E. every new client has a unique key tied to the version number. Third party devs get a login key from LL for each version. If you don't get a key, you aren't a valid user. LL can block clients from logging in. Bot people could try and spoof the keys or whatever but LL and third parties can keep updating them. It's going to be a lot more resource intensive to run 15 viewers at one time than it would some custom program running in a terminal. And you couldn't use a datacenter or another 24/7 hosting service. You'd need a complete desktop. There is no guarantee all these people are running these bots on their own computers. They could easily afford a VPS and run their bot software on the VPS and it would be very cheap. VPS are very cheap right now.

    Having a bad key wouldn't stop you from logging in, it would just limit your account and treat you like a bot. Your AV wouldn't contribute to traffic stats, you'd have a special label on the minimap and in your profile, etc. Bots can be very useful in SL and there are some really cool use cases, like store mannequins. But SL isn't accommodating that and it's letting those valid use cases poison their data for determining traffic.

    The problem is if your competition is going to start using bots to boost traffic, you get no reward for playing by the rules and registering as a scripted agent. In fact it hurts you. Using bots for traffic needs to be stopped. If someone has a valid use case for a bot, they won't care if the traffic doesn't get counted for that bot. In fact I think they would be pretty happy because they more than likely have competitors filling their land up with bots for higher traffic numbers. And I don't just mean competitors with stores. I mean two competing social venues. One has 3 real people in it who are willing to chat and make friends. The other has 25 bots. People go to the one with 25 bots and miss out on actual humans.

     

    • Thanks 1
    • Haha 2
  9. 31 minutes ago, AmeliaJ08 said:

    That isn't my experience at all, I still meet people. I'm 2017 intake though, I don't know what it was like but I meet people every time I go looking for them.

    Not everywhere is social but the places that are seem full of interesting characters.

     

    17 minutes ago, Zalificent Corvinus said:

    That's the REAL cause of Social-Fossil Ranting.

    Way back in 2006/7/8, the Grid was a LOT smaller, private estates were new, and not that numerous, and the Grid was infested with wannabe speculators, who had heard about getting rich without doing real work, in some game ( a pre cursor to the Crypto-Crap Craze ), who swarmed across the Grid like lice, engaging in "buy a micro parcel for more than it's worth and flip it for 100% mark-up" because they seriously believed that "there's always a bigger sucker than your self"

     

    So yeah, concurrency 80k, on a lot fewer regions, imagine 80,000 logged in on 10,000 regions? Average 8 per region, compared to today's less than 2, but 6 of those were failed amateur hobbyist land flipper wannabes who would lie to your face, so yeah SL seems emptier to them, but realistically, most of those 80k, weren't WORTH talking to, they were just proto-crypto-crappers trying to scam a fast buck.

    Back in the so-called "Golden Age of SL" when you could make L$2000 an hour standing on a street corner holding a non functional "toddler build" grade primcrap guitar lookalike in glorious "default plywood" and spamming a gesture saying "Linden love for a starving busker?"

     

    We "filthy noobs under a decade old" didn't actually miss much. Nothing worth trying to reclaim, certainly.

     

    I've been here for 15+ years, and yeah, that describes it very well. It was a lot more dense before. Which is what's wrong with SL right now, too much land and builds and not enough people to go there.

    It was also a lot more unique. A lot of us old timers came from AIM, AOL, Yahoo or MSN Chat, etc. We were used to nothing more than text based pop up windows, and SL at the time was absolutely mind blowing. Right now, it's not. It's just another drop in the virtual world ecosystem.

    I don't want to sound salty and old, wasn't my intention. But SL has definitely changed and it needs more density, which is what I'm hoping the mobile client can bring. Right now current SL is vastly a "I'm going to sit around AFK and show off my AV and hope someone cool IMs me." That doesn't mean that's all SL is. But it's the majority. There are still awesome people in SL, just as there were in the past. Land owners using bots are just trying to fake having density that's lacking. SL has intense competition now. 15 years ago, first signing on was one of the coolest experiences I ever had using a computer. Today SL isn't close to that.

    And back then, it was nearly impossible to make an SL bot. And this is someone who was very familiar with chatbots on other platforms. Now it's pretty darn easy if you just have basic coding skills since there's libraries that do all the hard work for you.

    I know as long as I've been here there's been people who think SL is going to die next year. I've been hearing it for 15 years. I don't think SL is going to die. I think it has some problems they can somewhat easily fix and make SL a major player. I want this place to do well. I have a lot of memories here, live with someone I met in SL. For all it's problems it's still amazing. But it needs growth and it's been on a slow trend downwards. Nothing wrong with that, it can be fixed, which is what I'm suggesting and I'm just trying to point out why SL might feel a bit different for us old farts.

    • Like 1
    • Thanks 1
  10. People who are in SL don't focus on it like they used to. They log in, make their AV look good, then go do something else while they wait for an IM. As others noted it's difficult to tell who is a bot and who is just AFK. And most people are just chatting in Discord now. Heck there's even a thread about wanting SL group chat to show up in Discord. Why? Because they're using Discord as their main chat program.

    That said, there are lots of places that fill their sims with bots and TPing there with aspirations of exploring and meeting people is massively disappointing.

    SL has definitely changed in the last few years, it's not what it used to be, at all. The scene of meeting people at random hangouts and just searching for interests and being social is more or less over.

    SL also feels completely empty because the hardcore users just want to make stuff, either content to upload or to make hang outs and in world builds. The ratio of content creator to user in SL skews heavily towards content creators. Usually social internet things are backwards, not enough content creators. But the whole bot thing and faking traffic is a common problem with starting social things on the internet. No one wants to show up if there's no content and people, and there's no content or people without people showing up in the first place.

    I think the real sad pill for you will be realizing a lot of the internet is fake, AI articles, posts, etc. Admins using multiple accounts pretending to be different users. Reddit jump started itself this way, they even had a way for admins to toggle multiple accounts when making comments, and then they'd have conversations with themselves. I think it's just very obvious in SL, but if you look hard you will find a lot of the internet is fake. Fake votes, fake comments, fake articles, etc.

    I really hope the SL mobile client is successful because all the phone users are a bunch of people who are only going to be able to consume content, it's going to be horrible to create anything for SL on a phone, and phones aren't very good at multi-tasking.

    My hope is that something changes soon, because it definitely needs to. People are slowly trickling away from SL and they have been for a while.

    • Thanks 1
  11. 19 hours ago, Henri Beauchamp said:

    Well, currently, Linux (and its forks, such as Android) and BSD (and its forks, such as macOS) are immensely more common than Windows for ARM hardware. So... 😛

    As for ARM vs x86, here is my stance:

    As a Linux user, should we get ARM-based desktop PCs with the same performances as existing high end desktop x86 PCs, it won't bother the least to switch to the ARM solution, much to the contrary !

    x86 is an antediluvian, poorly designed ISA (*), which is today impaired by all the compatibility modes it must keep to run old software, causing bloated CPUs full of (mostly) useless parts needed to run those legacy 16 and 32 bits modes, support segmented memory (eeek !) and whatnot... ARM would at least allow to get rid of all this old cruft, once and for all (and yes, at the cost of using emulators for the old software, but those won't suffer from the emulation since they were designed to work on waaaaay slower hardware).

     

    (*) If only IBM had chosen Motorola and its gorgeous m68k ISA (32 bits registers, flat memory model, I/O in memory address space), instead of Intel and the 8088 (8 bits, segmented memory, separate I/O ports), we would have gained 10 years of advance in OS design for PCs, instead of waiting before Windows 95 (1995) and Linux (1992) could finally offer true multitasking 32 bits OS for the x86; back in the 80s, we already had genuine 32 bits OSes on the Macs, the Sinclair QL, the Amigas, the ATARIs, and genuine preemptive multitasking on the QL and Amigas...

    The only Adobe software that runs natively on Windows ARM is Photoshop and Lightroom. x86 isn't just about all this old software you're talking about, there's lots of modern proprietary software that doesn't run on ARM at all. A lot of modern content creation software does not work on ARM. I imagine content creators, or people who want to get into it, buying ARM and realizing they only have Blender and everything else needs to be emulated or whatever. I'm sure there's more software that I'm missing, autodesk, adobe, probably a huge list of content creation stuff you'll miss going to ARM. And that's ignoring games. LL won't even support Linux because there's not enough users, you think they're going to make a Windows ARM version?

    Modern x86 is just an x86 instruction decoder with an architecture that has nothing to do with the old CISC archs from back in the day anyways. But for serious computer work x86 bringing things like AVX is totally worth it and it's something CISC ARM will never really bring to the table because that's just not what the architecture is for.

    A switch to ARM is difficult, most people won't want ARM because the software is missing and most companies won't want to make ARM versions because the customers and users are missing. Apple got away with it because they control the ecosystem. But something like Windows stands hardly any chance.

    Linux is probably the best way to use ARM CPUs since there are arm distros. Some distros like Gentoo it wouldn't even matter. In that case, though, you'd have to run qemu with WINE to get Windows x86 programs to work. Yuck.

    • Like 2
  12. 18 hours ago, Zalificent Corvinus said:

     

    It's simple.

    Imagine somebody with a Steam account, a decade old, who has spent thousands of $ on recreational software, out of their own pocket.

    Then some marketing clown comes along and tells them they should spend 50% more than the cost of a brand new top of the line REAL PC with a grown up OS, on buying the "Wakibaki 9000 with Metal Arms", a machine that will NOT run ANY of their 200+ back catalog of well loved games, games mod loaders, games modding tools, etc.

     

    All they are offered is the "amazing new SirLagALot REAL PC to Metal Arms emulator" which runs like molasses in wintertime, and doesn't support ANY of the graphical API's used in the software, and results in degraded graphics.

    The moment the PC user sees the word "emulator" they will back away quickly. so, you LIE and call it a "translator" hoping they won't spot that its basically the same damn thing.

     

    And then of course, you'll get cultists saying "But why would you want to play games that are MORE than 3 months old? If the Wakibaki takes off, eventually there will ne NEW games, delete all your old faves with over 1000 hours of gameplay each!"

     

    It's simple, if recreational PC users who pay for their own software, rather than having it rented by their employer, can't run all their preferred existing apps, they will NOT pay through the nose for some overpriced Futureness Propriatory Lock-In- Fail-Tech.

    They didn't buy NEXT, they didn't buy Amiga 500, they mostly haven't bought Metal Arms Mac, and probably won't buy this new thing, unless that "run your old software at least as well if not better" issue isn't solved.

    Telling people "We know you love [game title] 4, but IF you all buy this new thing, maybe the publisher will release [game title] 6 on ARM in a year or three" just does not cut it.

     

     

    People don't really understand how big of a change it is to shift architectures. People won't run Linux full time because there's software that doesn't work as well in Linux. I love Linux, and I still dual boot and keep Windows around for games and other stuff because I don't want to deal with WINE or whatever. If you are using a Windows PC and you want to jump to ARM, uninstall Windows and go Linux only and see if you can still run every single program as easily as you could in Windows.

    And just my two cents, but Windows has been getting worse and worse with each version. 11 is terrible, you can't even more the taskbar to a different location. The window manger is missing the most rudimentary features you can find in KDE or Gnome. With Windows going downhill, the last thing I want is to be a second priority for MS and Windows. I can't imagine how much stuff is going to be broken or not work.

    The only viable way forward for ARM is Chromebooks and Linux, because you're going to be giving up all that proprietary x86 software anyways so it doesn't really matter if you lose it or not.

    Nvidia is just paying for marketing articles. They have always been super salty their Tegras have never really been successful besides Nintendo Switch. I think Jensen Huang is really bitter about AMD having GPUs and CPUs and Nvidia being only a GPU company. Just my two cents but Nvidia has been trying to be relevant in CPUs for over 10 years and it doesn't happen. Much like Intel trying to make serious GPUs with Arc. It's just not going to happen, and the few people who do make the jump will end up stranded and abandoned. I think Nvidia is kind of scared they are only a GPU company, and it's also why they are always making software that does more than play games for their products (AI, etc) and they're always creating software for their GPUs (hairworks, physx, etc)

    • Like 1
  13. 1 hour ago, AmeliaJ08 said:

    I really think in the PC market this is going to succeed/fail based on backwards compatibility. People are probably not going to want to give up on x86 apps entirely at least during the transition period which will probably be quite a long time in the PC world.

    Something I don't understand though, Microsoft in their Windows on ARM marketing always talk about "x86 emulation" and yet Apple always describe Rosetta as a translation layer. Are they functionally the same thing though?

    https://learn.microsoft.com/en-us/windows/arm/apps-on-arm-x86-emulation

    https://developer.apple.com/documentation/apple-silicon/about-the-rosetta-translation-environment

    I'm still not holding my breath, the benchmarks they used isn't even named. Typical marketing. if they had a valid benchmark like Cinebench, Blender, a game, productivity software, etc. they would be more explicit than just "single thread performance is more points". It's probably just geekbench. I guess we'll have to see, but I know people who switched to Apple M1/2 and they aren't happy because they have stuff that doesn't work, or doesn't work well.

    https://www.techradar.com/computing/qualcomm-launches-snapdragon-x-elite-for-windows-pcs-and-promises-on-board-ai-and-days-of-battery-life

    That said, yes, you have to use something like qemu to translate CPU instructions. WINE just translates API calls to something that Linux can understand. A program is just a list of instructions for a CPU to execute, x86 has their own instructions, and ARM has their own. To run x86 on ARM, or vice versa, you have to translate those instructions to something the other CPU can understand since the architectures aren't compatible.

    But anytime you add a layer like that, you make an opportunity to cause problems that wouldn't exist if it was native. I use Linux for all my work stuff, SL viewer in Wine loves to randomly crash, especially after uploading, and SP in WINE loves to not start up until I relog, especially if I'm using Wayland. Problems I wouldn't have if I were just to use Windows. Even with that, I use Linux because the other benefits are worth it. But switching to ARM doesn't offer much. I'm skeptical, again. I remember when AMD was going to go all in on ARM before they released Ryzen. Good thing they didn't.

  14. 13 hours ago, Zalificent Corvinus said:

    Problem is...

    There are not enough people using Awful Mac's with Metal Arms, to justify LL wasting time, effort, and MONEY, making a special snowflake viewer just for them.

    Mac users are only about 5 % of SL in total, and Metal Arms are a fraction of that.

    Same reason they don't waste time, effort, and money, pandering to the 1.4 % of SL that use Linux.

    It's harsh, but that's economic reality.

    Pretty easy to make one with Unity Game Engine, which they are using for the mobile viewer. I don't think it'll ever be much of a content creation platform for SL, though. Highly doubt there's going to be much effort to get the proper viewer running on it, and it'll always be missing key content creator software.

     

    16 hours ago, AmeliaJ08 said:

    Not sure it is different to be honest. I would assume whatever they're planning will be built off Tegra.

    Of course it does seem to fall down on Microsoft to actually get users interested though. I guess Nvidia think with AMD and Intel supposedly jumping into the ring this might be a good time to give it another shot?

    I think the answer to who wants that is most laptop users and presumably most desktop people who pay for their own electricity :) it does feel very like the 90s again though, the death of x86 was assured back then but of course it never happened since we PC people are pretty stubborn. This is that all over again but the rise of ARM everywhere else does make it seem slightly different and could mean the time has actually come, who knows.

     

     

    Yeah, this feels like deja vu. I don't know how it'll end up. Last time they tried this laptops had 1 hour battery life unless you bought a netbook. Now you can buy a nice x86 laptop that goes for 8+ hours and is pretty fast. And ARM didn't make it back then. In fact AMD made Jaguar architecture and it was pretty darn good compared to ARM.

  15. How is this different from Tegra? Tegra was announced in 2008 and the only time it's been in a successful commercial product is the Nintendo Switch. And only because Nintendo was able to bully Nvidia into giving them a good deal since Nintendo knew no one liked Tegras (too hot and inefficient for mobile but too slow to compete with proper desktop chips).

    I feel like this is an article from 2010. But I checked the date and it's not. So they're just going to try and make a Tegra that competes with Apple? Who even wants that?

  16. https://www.theverge.com/22684730/students-file-folder-directory-structure-education-gen-z

    There were a lot of articles talking about how younger people had problems understanding files and folders because they grew up on search. Older people grew up when search was terrible or didn't even exist, you had to organize your files because search was super slow or missing and didn't work half the time. But with iOS and Android you don't even really use folders other than selecting a place to give permissions for an app to read and write to.

    Just my two cents but a simple directory tree layout that the inventory browser uses is totally foreign to a lot of people. There are a bunch of articles covering this subject, and the only criticism people seem to have is "I know lots of older people that don't understand folders and files either."

    People's inventories become a mess because they don't organize. I do it too. Nothing wrong with people who do that. It's just that the structure and behavior or folder based inventories cause problems. SL has always attracted more tech affluent users, and even those "well above average" users struggle with a lot of stuff in SL. I hope they do something cool with inventory for the mobile viewer. It's far from ideal in the mainline viewer but I think if it were changed it would make a lot of people frustrated.

    • Like 1
  17. 9 hours ago, Coffee Pancake said:

    I spent most of the last couple of days poking the search trying to see why my objects had just vanished (and my sales with it).

     

    Search now heavily favors minimalist low effort submissions and seems to punish verbose listings with detailed feature guides and broad keyword coverage. Under some headings it's actually  recommending products so old you can count the prims just by looking at the bad snapshot.

    For example .. 

    mKjE9wh.png

    Searching for "fence" sorting by relevance. 

    The top 3 are sculpts. The bottom 3 are apparently "mesh".

    Notice anything else?.. the objects are all named "fence", no brands, no embellished names, no fences of type or purpose. just "fence" .. the descriptions and features are equally brief, and yes .. it should be no surprise that all these purchases ship a single object named "fence".

    This is the new normal and it is a slap in the face to merchants who go the extra mile.

    This is also being gamed already.

     

    It's been like this for months. I'm sure a power user will show up and tell you that you just need to add a bunch of boolean operators, check some boxes, etc. Yet the average user has no idea how to do that or just doesn't care enough to fight search. It's a serious problem, and yes. There's no reason to make detailed listings anymore.

    These are probably still selling because they're at the point of "who cares, it's L$2 and if it's a terrible product it's not a big deal", so all the data to sort by sales is borked. There's also apparently not much effort given to reviews in sorting, either.

    This hasn't been fixed in a while. I don't think it's ever going to be fixed at this point. Those merchants are probably long gone from SL and those L$ are just going to sit in an account forever and not go anywhere.

    My sales have been more like normal the last few weeks or so but it's obvious search is still messed up. I think SLMP is too big for its own good, 10+ items for a fence and I'd wager 90% of them are outdated. No sorting that properly unless some sort of drastic actions are taken.

    • Thanks 1
  18. 13 hours ago, Polenth Yue said:

    A lot of the issues being raised here aren't new (outside of the marketplace search changing, which I've given advice in other threads for how to rewrite listings to get them back up in search). Rather than looking for signs that it's bad for everyone equally, I'd suggest looking at people who are doing fine and seeing how you can apply that to your business.

    It's a good idea to have some cheaper products that are aimed as fun things for end users. Not only because that's extra money, but they will bring people to the shop, where they can see the more expensive goods and services. The market is much more limited when it comes to expensive things or things that are mainly for builders. If you already do that sort of little stuff, some items outside the usual can bring more people in.

    Events can be great, but not all events are equal. The same for hunts. I'm surprised more people don't ask about which events have worked for people. Once you find events that work, they're the best marketing. You make money and advertise at the same time, in a way that doesn't annoy the socks off potential customers.

    Adboards, marketplace adverts, and malls are frequently suggested by people because they've seen other people suggest them. Few giving those recommendations have personal success with them. Those who do find them useful often have circumstances that don't apply to most. Personally, these things have never worked for me. A shop hidden behind a secret door did work for me, but that's also one of those unusual circumstances things.

    The new search seems to favor gacha items. I've also seen residents who have stores with 10k+ gatcha items. It doesn't really matter what content creators do. When people are making stores with thousands of gacha, and it's taken me over 10 years to make 300+ original mesh and texture items, gacha is going to drown out original content.

    I don't think people understand just how bad the flood of gacha is and how it's flooding search results. The lack of ability for merchants to stack color variants into one listing is also massively terrible for SL and flooding search results. Merchants love it because it's very easy to make color variations of a product, just change colors of the textures. Customers love it because they get more choice. People who search hate it because if they don't like the product, there's a ton of listings clogging up search for something they don't want.

    As a merchant, it doesn't matter what kind of products you make or what advertising you do or whatever. Like @ValKalAstra said, there's basically no escape from gacha on SLMP. And if you just want to buy what you want and don't want to leave it up to chance, it's a huge pain.

    Gacha makes a lot of money though, anywhere on the internet. and it's no doubt making LL a lot of L$. But it's definitely doing damage to people who don't like gacha and use SLMP.

    Things are a lot worse than what you think. I know a content creator who has been making stuff for a long time, kind of niche products, but very popular and very successful. He's always worked a part time job to make SL content. SLMP has been so bad, no matter what he does, that he's looking for a full time job. SLMP's problems are so severe that content creators are starting to reduce what they make (I have) or practically stopping with new content. These problems have always existed but they're so bad it's causing things to change in SL, by a lot.

    • Like 1
    • Thanks 1
  19. One other thing I thought of last night. SLMP search results have been bad for months. I think a lot of people have given up on SLMP and finding things to buy. There's only so many times you can go to SLMP and try and buy stuff and not find anything modern or related to what you're looking for before you give up searching on SLMP. LL said they made more money with the new algorithm. My thoughts are that the results were bad so SLMP wasn't as stressed as much so it didn't have down time and random errors so it had lower, more consistent sales of older stuff. But eventually a lot of people gave up on the older stuff and moved on.

    As a merchant it's really easy to write off how bad search results just might be something contributing to bad sales. But for a consumer, that means trying to find something to buy and not being able to find it. It's like going to a real world store and the store is a huge mess with nothing organized or in a nice place with any sort of signs. Just a big pile of merchandise. And the customer is told to go there and find what they're looking for. And if they can't find it, it's their fault for not searching properly. But for customers, that's extremely frustrating and discouraging. They are about to spend a good amount of RL money on stuff, and every frustration is a paper cut that makes them want to spend their money elsewhere.

    • Like 3
    • Thanks 1
  20. 16 hours ago, Cristian Hirsch said:

    Nvdia GeForceNow and Boosteroid works nice and smooth with other games in any browser or device .... would simplify things much easy and faster ....just saying

    Given that the new iPhone is running AAA games I don't think remote streaming and the extra latency and stuff is really worth it. SL already has so much latency because every action is sent to the server, server handles it, then sends the update.

    Imagine the latency in SL when you send your keypress to the streaming service, which then sends it to the simulator. The simulator processes everything, then it sends it to the streaming service, then the streaming service sends it back to you.

    SL latency is already horrible compared to modern games who update your screen as soon as you do something and then checks with the server to make sure it's what actually happened.

    Now imagine that on mobile internet where latency can be 200ms or more and now you have half a second between pressing move forward and your avatar actually moving, then you let off the movement button and you don't stop for half a second. Not everyone lives in a city with fast 5g and all the bells and whistles.

    But I think those remote streaming services are doomed and will never take off. Mobile GPUs and CPUs are going to catch up before internet gets good enough to make it work.

     

    iPhone 15 Pro

    https://browser.geekbench.com/ios_devices/iphone-15

    Intel Core i7 10700F

    https://browser.geekbench.com/v6/cpu/2900345

    AMD Ryzen 3700x

    https://browser.geekbench.com/v6/cpu/2900385

    Yes it's not that great to compare ARM to x86 and yes it's just geekbench. And yes there's better Intel and AMD CPUs. But these scores being so similar are something not many expected. And honestly that Ryzen and Core i7 is probably better than what a lot of people use to get in SL on their desktops and laptops.

    Eventually that iPhone 15 pro performance will trickle down and your $700 phone will be just as fast. And eventually the $500 phone will get there too.

    Hardware wise I think LL is going to launch the mobile viewer at a very great time. Mobile gaming, and the hardware and software, is reaching levels it's never been at before.

    • Like 4
  21. 22 hours ago, Katherine Heartsong said:

    Where I'm seeing the larger corporations being in a larger metaverse (if not trying to control it outright) would be as major tourist attractions/destinations in a sense. They have the capital and content to create large, well designed, realistic, all encompassing pieces of the world that the normal person would never be able to create or host. Extravagant experiences backed by large amount of money and meant to hook you into their world to drive their profits.

    But that means there will still be out of the way destinations and worlds on a much smaller scale to offer unique and intimate (emotionally) experiences. I liken the corporate metaverse worlds to Las Vegas, and the smaller ones an equivalent to the small landscaped gems that a few people create here to share (but will be larger than a 8186 parcel).

    https://www.thepoke.com/wp-content/uploads/2022/08/meta-2.jpg

    Just because they have the money, doesn't mean they can.

    • Like 1
    • Thanks 1
  22. 2 hours ago, Istelathis said:

    As far as a metaverse, I dunno how far it will take off, I've always envisioned it more of just a different iteration of the web.  Facebook may want control over it, but ultimately I think it is going to be decentralized as the web is.

    Decentralized web already has one foot in the grave. Look how many people only visit the major sites (reddit, facebook (and their other sites), Twitter, Amazon, etc). Google search only really links to blog sites or SEO optimized garbage written by people who don't actually know anything about the content. The more these major companies take control, the more they're going to want to slurp up your internet life, control it, and harvest your data.

    • Like 1
    • Thanks 1
×
×
  • Create New...