How to NOT buy bad stuff in SL

Everybody knows they should demo things before buying them… but how many people know what to even look for when demoing.

Sure there is the obvious “do I like it”, “does it fit my avatar”, “does it match my theme”, “is it Copy/Mod or is this store’s owner a jerk?”

But there are three much more important tests that most people don’t know about – even though we all know about what happens when the results of these tests are bad… Most of us just don’t know how to find out before we’ve shelled out our precious loot for the latest fad…

So… Here is what you should really be looking for, how to look for it, and why you need to look for it.


The three steps in the try things out that way too many people never do…

  • Check out its script burden – very important.
  • Check out its avatar draw weight – the most important.
  • Zoom way the heck back and see what it looks like – situationally important.

Script Burden and Avatar Draw Weight will determine whether or not the item contributes to slowing down your SL, maybe randomly crashing you or others, and so on… ie: client-side appearance of lag that is actually your video card freaking out…

Zooming back will tell you how well they optimized stuff. It relates to what happens at a distance far enough to trigger a new ‘LOD’ (level of detail)…


 

So here are some screenshots and advice to show how to check and set all of these:

Script Burden:

Script Burden is my term for two things, Script Memory and Script Time. As users of SL we can only see Script Memory Allocation, and not the actual memory, nor the script time. We “CAN” use the LSL scripting language to find… script time allocation, but not the actual time…

Allocation is like making a reservation at a restaurant that only seats people at tables of 4. If you have 5 people, that is 2 tables, if you have 7 people, that is 2 tables… If you tell the restaurant you expect 5 people, it allocates two tables, and then only 2 of you show up… you still allocated 2 tables… but only used 2 seats…

Script time allocation… that’s like google maps telling it will take 23 minutes to drive there… Even if once you hit the road, it actually only took 19 minutes… you still allocated ’23 minutes’ of your life to expecting to be sitting in traffic… and the restaurant won’t serve you until 23 minutes have passed because that is when you made your reservation…

So its all about guessing here, when we can only see one leg of the elephant…

This is why you will often hear people say checking scripts is useless. Because its guesswork. But an educated guess is still better than just ignoring all evidence…

Checking the script memory allocation of things is STILL VALUABLE. The allocation tends to be a good estimate for how much resources something is going to eat up. If it is high it tends to indicate something that will slow you and everyone in the same sim, and any other sims on the same server, as you down. Yes, each sim in SL shares a server with 4 other sims – you have no idea where they are, but some jerk on one of them could be running a 700mb scripted hair piece from 2009 and totally killing your SL… That jerk might even be you… so here is how to find out if it is you so you can get some control back to your experience… (you can’t do anything about that other person except link them to this blog and pretend you’re being friendly and didn’t notice that giant flashing lag beast over their heads) 😛

Here is a screenshot showing where to find your script cost:

ScriptCount

So: World -> About Land -> Script Info -> My Avatar

Older items tend to be very bad about this. Many older items have been repackaged as freebies… so always check this if you get a freebie…

Yeah, my AO is a minor beast… Its the biggest thing I use, by a large amount. But a few years ago it would have been seen as small. Things are getting better in SL… Basically if your total here is ‘under 3 to 5mb somewhere’ you are “probably” not the problem. I say probably because we haven’t checked the other issues yet… 🙂

(My AO is reasonable, BUT consider that it limits what else I can also wear… especially if I had an older system… or if a friend of mine did, and many do… and I don’t want to be crashing my friends while I’m trying to chat them up…)

No-Mod hair is the most common cause of high script issues – because from 2009 to 2011 there was a popular ‘resize’ system going around the hair shops that required putting an individual script into every single strand of the hair… The linden’s added a new linkset script system in 2010 or so, I made it a freebie, as have others – and since 2011 versions of the linkset system have spread, each working a different way, but all finally low on script memory. However every now and then there is someone still using the old method.

Avatar Draw Weight:

This is the most important test you can run. Its a big performance issue, and a generally unknown factor.

While most people today know about scripts, almost nobody knows about this one. And many who do confuse it with the old system (ARC) that it replaced. The old system… was broken and gave meaningless data. The new system, is accurate and pretty much paints a red sign over the head of all the people around you who probably just aren’t quite aware of the impact of what they’re wearing and why they and/or people around them are feeling so much lag in SL. Keep that in mind… most people don’t know the causes of this stuff. So once you do, lay off the blame game, and just try to help people find solutions.

AvatarDrawWeight

Advanced -> Performance Tools -> Show Draw Weight for Avatars -> Look over your head for a colored number.

And note the 72,604 number over my head there – that is just high enough to start causing lag. People are often way over this. If you bust past 100,000 to 200,000 or so – you can start causing random people around you, or yourself, to crash… I’ve seen people over 400,000; sadly not uncommon.

Ever wonder why you’re happily hanging out somewhere, and then SL just up and dies on you? Now you know… Something with a very high Draw Weight just got noticed by your graphics card.

I saw “Something” and not “someone” on purpose… This issue is even WORSE with rezzed things like houses and furniture… Anyone who’s been happily walking about and hit upon a new scene made in mesh or sculpty and suddenly notices their computer starts crawling or the fan kicks on… that’s bad Draw Weight items around you… Its VERY COMMON in the Gatcha scene… and I see it a LOT in some fancy shops and clubs that are burdening down their build with ‘cute but badly made’ items…

You can’t do anything about it if somebody at a venue you like has rezzed all the bad stuff… except try to tell them and hope they’re not one of those weird folks that freaks out anytime somebody talks to them… BUT you can control it on your own stuff…

I don’t tend to wear that hair anymore as a result… It was about 30,000 or so of the number you saw up there…

So… you just bought some fancy new mesh kitchen and want to know this for it… how do you tell? Well… you actually CAN wear your kitchen in SL… 🙂 So that is one way – wear it, and then check your scripts and draw weight.

Another way is to go into wireframe mode and look at all the lines… This one is guesswork. If the lines are “all up in my business like some crazy cat lady” – so thick you can’t see Jack from Jill… or rather, so thick that you CAN see Jack… if the lines in wireframe are so thick the object almost looks solid… that’s your nightmare object right there. The wireframe test can be done in the store, before you buy it… But its a bit of a guess. I HAVE seen a low wire item that was laggy, and a densely wired item that was low lag… and these were both mesh bodies by the way…

Here is where you can find wireframe mode, with obligatory cat lady in your business:

Wireframe

Develop -> Rendering -> Wireframe

So this is a hint of where to start looking, not the final proof… MOST of the time the more solid the object in this view, the more laggy it is being… but not always. I wish I knew how to tell what kinds of shapes in wireframes were bad… I’ve seen that very point discussed on a Poser forum back in 2000 or so… the shapes of these polygons matters, but what shape is good… Not sure…

And yeah – my tail there is pretty solid. Its the only mesh tail like it I know of… it IS laggy, but it is true animated. There are another few mesh tails that animate by changing transparencies on different parts of the mesh – they are even MORE laggy…

Notice my hair is also pretty solid: but, it is actually a LOW lag hair from booN (booN also makes some cornrows I LOVE on my human alt – best looking hair in SL, but they ARE very laggy, so I had to pack them up and no longer use them… /cry). Shoutout here: booN has the best looking hair in SL in my opinion, half of it is very laggy, the other half is very well optimized. I don’t know why because its not about is it new or old… But the shop has many styles and I highly recommend them. Just test before buying.

See the pink heart inside of me… that’s from the nipples… I wish this modeler made those mod so I could remove that thing – the nipple set is low lag, but it’d be about half the lag it does have if it didn’t have to render a high detail mesh heart inside your body…

But the highest render cost in this entire screenshot? My eyes. Maybe you didn’t notice that they’re basically solid… If I could replace those eyes I could probably double the number of other things I wear. FATE Mesh eyes did get a script reducing update about a year or more ago, but the mesh itself is still too high detail.

Zooming back test, renderVolumeLOD, and correcting popular bad advice.

The last test for seeing the quality of something is to zoom yourself way the heck back, and see what happens. How far back can you go before the item breaks apart visually, turns into a triangle, or some other weird thing happens…

And is this distance too short for you, personally, and how you plan to use the item. For example… does it really matter if that new mesh toilet you bought breaks up into a giant single triangle 90m away, if its going into a bathroom in your little skybox on a small plot somewhere? Probably not…

But if the same thing happened to your hair at only 90m away… then you’re going to end up looking bald to a lot of people. And a few months back I bought some beautiful hair for an alt and had this happen at only 5m away because I forgot to zoom back before hitting buy… /cry+delete…

So this one is “PARTIALLY” a subjective preferences and “how you use it” test… But its one many people don’t know to do because they’re not aware of how mesh is both made and ‘rendered’ (drawn on your screen).

Advanced -> Show Debug Settings -> rendervolumeLODFactor

AdvanceMenuShowDebug

rendervolumeLODFactor

To run this test it is first IMPORTANT to set your “renderVolumeLOD” to 2 (or lower):

Notice the 4 in that image. That’s an old screenshot. Many people, self included, used to recommend 4. I actually used to recommend 9… I don’t know why… It was BAD ADVICE.

Remember that at 4 or higher you have disabled SL’s optimization code… so SL will lag a LOT more at 4 than at a lower setting. You want it at the default (1.00 – official viewer), 2, or at most 3…

A common hack to getting the “Avatar Draw Weight” to be low is to optimize an item by turning it into a triangle at ‘far away LODs’… and then telling people to ‘break’ their LOD setting (rendervolumeLOD) by setting it to 4, so that they never see this… but that breaking actually comes at a VERY HIGH cost… and is one of the biggest reasons users of SL on ‘normal computers’ and not ‘suped up gaming machines’ think its laggy…

Ideally you should have your ‘renderVolumeLOD’ debug setting at either default or 2. NEVER the commonly recommended 4, because at 4 you have disabled SL’s optimization code… slowing down your experience and creating… client side lag…

But once you drop to 2, and zoom back, you quickly learn how badly some stuff is made…

And you want to do this BEFORE testing items so that you can wear them, zoom back, and see if they “break”. If they do – its a badly optimized item. Buy something else.

So… renderVolumeLOD now set at 2, it it time to take your camera up to something silly – camera distance 512m (DO NOT do this normally – you “ping” everything in your camera distance, the farther it is, the laggier your SL, and everyone in your range that you need to ‘ping’ and get an update from also takes a hit. Normally I keep my camera between 64m and 128m – 64m is ideal for avoiding having to see that your neighbors put a building with improper texture repeats on top of a beach that has fullbright trees… AND without using support pillars… 🙂 ).

You might notice that I also have advanced lighting off. I usually have it on – but it has no impact on this test. You don’t need it on or off for testing this kind of thing.

Here I am close up, everything looks good:

CloseUpCamLOD

Just a little bit back, my coffee machine has vanished. I am fine with this because it is usually in my skybox in a small kitchen that’s only about 2m wide by 3m long. Out here it’d be an issue, so I may need to rethink using it here:

MediumCamLOD

A bit more back, the platform has deformed:

FarCamLOD

This kind of is a problem… and as soon as I find a replacement, this platform might need to go… Because honestly that is not that far back.

What is going on here is improper optimization done in order to cause these items to have a lower ‘Land Impact’ cost. This also happens with worn items – but I just didn’t remember any worn things, that I still own, that break so easily – I tend to not buy them, or delete them once I find them (like the alt’s hair I mentioned above)…

Notice that the trees, the art canvas, the table behind me, and the campfire below all never broke. They are all actually LOWER land impact than the platform that DID break… Why?

Because the ‘hack’ to break things at distance LOD is just that – its a hack done instead of doing proper optimization. It is a sign that the 3D modeler is likely a hobbyist who has learned from other hobbyists and has never had to model for a paycheck at something like a gaming company – where having low polygon good LOD models is vital for a video a game to work online…

The stuff that didn’t break – much of it, like that art canvas from Dysfunctional (same shop as the trees – Dysfunctional tends to be quite functional 🙂 ) is actually 1LI items… but they are very well optimized by people who were either professionals, or just hobbyists turned SL merchants that learned from the right training advice (because often hobbyists are way better than professionals… Hobbyists tend to be either very bad or very good, professionals work to standards and end up ‘in the middle’ in most fields… in my opinion)…

The table that didn’t break is made by the same person that made the platform that did break – so just because a builder gets one item wrong, don’t discard their entire brand. Test every item individually unless you start to find a very consistent bad pattern – this platform builder’s stuff is usually on my list of good examples – here is just an exception.

Now remember to turn your camera distance back down after doing this test…


So… I hope the above manages to be useful in informing people on what to test for before making that purchase.

Getting these things under control will greatly improve your SL experience – particularly if you not on a top-end computer.

 

Yes I have called out two or three brands that I like in this. I’m not getting anything from them, I just like them, this is my blog, and I’m going to say my piece on it. 😛

 

Footnote on LOD:

Some may wonder where I got the idea that a high LOD is bad. If you think about how LOD works, how your graphics card renders detail, and how optimization works it ought to be obvious. You can see it in any video game really – settings to lower the detail in order to improve performance because the more you tax your system, the harder it gets on that system.

But here is a quote, from a meeting the Lindens had with the Firestorm team some time back. What they say in this, is something you will see said in any real-time 3D graphics platform. This advice is not at all new:

Here is a quote, from a meetings with the Lindens, that covers why a high LOD is bad:

http://modemworld.me/2014/05/20/may-17th-firestorm-meeting-audio-and-transcript/

12:02 JL: If you turn-up your LOD, you’re going to see things really nice … it’s also more complex because your video card is rendering more things … So if you want to take really nice pictures in SL, crank up your LOD; but it’s going to have an effect on the performance of your viewer.

12:55 JL: So what a lot of content creators do – and if you’re one of the content creators that do this, yes I’m pointing at you … is they want their stuff to look really good, so they’ll have their LOD cranked up really high …. and then they’re making their necklace or their shoes … and it looks really great to them, it looks really perfect and they package it up and they send it out. And people who don’t have their LOD cranked up, get that item and it looks like crap. And then the content creators says, “Oh, it’s because you have your LOD down too low; turn up your LOD and it’ll look fine.”

13:40 JL: Let me tell you something. If you’re that content creator, if you’re that person, you are doing it wrong. You make your content at a low LOD, and you make it look good at a low LOD level, and it is guaranteed to look good at a higher level … That’s the way to do it, because that’s how you make sure it looks good for everybody. not be making it to settings most people can’t use … you make your content based on a low LOD level, and that makes sure it’s going to look good on a low LOD level and on a high LOD level.

JL: Don’t tell people to turn their LOD up … it’s not good for them … because they forget they’ve turned it up and then they have really bad performance because they go to a region full of all kinds of prims and avatars with mesh and all these things, and their viewer performance is just utterly crap And they come to us and say, “Firestorm is a piece of crap. I just ran Singularity in the same situation and Singularity runs really well in the same situation!” But it’s not the same situation, because the LOD isn’t turned-up on Singularity …

JL: And that’s very common. People will compare one viewer with another viewer, not realising that they’ve changed a setting in one viewer that they haven’t in the other, and that setting has a huge impact on performance … Just leave the LOD alone in the viewer.

15:36 EM: The ideal setting for most people is between 2 and 4, which is why our LOD slider only goes to 4.

15;44 TS: And in fact, LOD is restricted in the viewer, there is an upper bound above which, no matter how high you set it, won’t make any difference anyway. My fuzzy memory tells me that number is four, but i could easily be wrong. i haven’t looked at that chunk of code in quite some time.

16:03 JL: If you want to do pictures, go into Preferences and crank-up your LOD. But when you’re done taking pictures, bring it back down again. Or just jiggle the graphics slider, because the LOD is affected by what setting you have in the performance slider. So if you choose low, it’ll have a lower LOD, and if you choose Ultra it will have the highest LOD that we would possibly recommend. But for content creators, drop your LOD when you make your content, make it look good on a low LOD and you’re going to be putting out a really good product.

16:57 TS: Which is why the defaults are good choice, because they’ve been tweaked pretty well … for most systems.

17:38 JS: Just generally speaking, we have had arguments, big arguments, internally on the team on what we should make default LODs. We try to follow along with Linden Lab, although seldom do we agree with Linden Lab’s defaults. We’ve spent a lot of time trying to choose what we feel are the safest LOD levels for each quality setting on the slider.

16:13 TS: Linden Lab has a lot more tolerance for low frame rates than we do and what our users tend to do.

So they say from 2 to 4. The default is actually 1.00 (official viewer). Why did I pick 2? Well, a bit of intuition really… I’ve been trying to get it lower and lower over time and trying to find the point where lowering it isn’t helping me anymore, but is instead breaking too many things. I think that is 2 to 2.5 maybe… Now if you’re obsessed about min/maxing what your system can deliver out of SL, you can play around with this and try to find a sweet spot in there that is better than where I’m at.

Most quality mesh still holds up at 2… so I’d rather not go any higher, because I want to maximize every last piece of performance gain I can get…

4 though… at 4 NOTHING breaks no matter how far I go out. That’s a pretty good indication that the optimization is no longer in play… So 4, 4 is too high. And at 2 – yes, you will see a performance gain.

 


See also this more recent article for added information:

https://catnapkitty.wordpress.com/2015/12/09/avatar-draw-weight-transparency-flipping-second-life-render-speed-tricks/
– You should test with all of that in mind when doing the tests here.


 

 

ps: Just because a person is exploring nudity in SL, does not mean its appropriate to toss your random weird freak sex fantasies at them…

Advertisements

17 Comments (+add yours?)

  1. zzpearlbottom
    Dec 07, 2015 @ 06:49:18

    if you set LL graphics preferences slider to ultra = lod 4.
    Mac lod that 1 should use if wants = 12, from there nothing really improves.
    But at 12, zooming out using 1024 draw distance, can make a diff.
    Still the advise for creators is a wise 1, to create with Lod2 in mind.
    Lod 1, sorry but that it only mean an impossible mesh land impact on most.
    As a builder, if i bould some to be used inside a house, i make sure it can be seen with med and low lod but i would set lowest to 0.
    on the contrary, if i want a small object to be seen, i must ensure that the lowest lod it still makes it good (reason why many jewels are so high on Draw height).
    In the end, some si spussed to have a sort of minimal hardware specs, a builder can not ever create, if it is thinking on low end graphics vga or old cpu of 6 years ago.

    Reply

    • Pussycat Catnap
      Dec 07, 2015 @ 08:09:54

      Setting it higher will improve your graphics – but more greatly tax your system as you are not allowing your system access to lowering priority of detail on further objects – forcing them all to render fully, thereby resulting in a notable performance loss and greater burden on your graphics card. For many on low to moderate end machines this becomes the point where that card can heat up, and even crash SL or their system.

      Reply

      • zzpearlbottom
        Dec 07, 2015 @ 08:24:51

        Yes i agree, but many now have systems at least as good to play fallout 4 or sims 4, on those for sure they can handle Sl in ultra.
        But still Lod set to 1 for building is to low, i would recommend using Lod factor 2 for all builds.

      • Pussycat Catnap
        Dec 07, 2015 @ 16:46:29

        Another way to look at this is that no matter how good your system is, you will gain performance by using lower settings – so it is best to go for the lowest settings that do not negatively impact your experience. Thus using a lower LOD. Lower renderVolumeLODFactor will give any system a performance gain. It is important for a lower end system – which is unusually common in SL. But it is still a gain to a high end system (like my own).

        It is however, the least important of the 3 factors in the blog – while the other two are concrete, adjusting your LOD is just the setting for helping determine the third factor – your comfort level between how far away an item breaks up and how far away is far enough for you.

        I get that there is resistance due to so many years of bad advice… but this is not the factor to focus on for the blog…

  2. Pussycat Catnap
    Dec 07, 2015 @ 08:11:48

    I do have some jewelry with very low draw height that still retains detail at moderate distance. I’ll have to look up the brands.

    Reply

  3. zzpearlbottom
    Dec 08, 2015 @ 01:44:23

    I agree, one of the 1st things i knew on second life, was to bought only mod items to be able to remove scripts. I did race nascar on Sl for some time and it was mandatory to be unscripted in order to enter the sim.
    Also the Draw weight is a must, i found amazing how many are well over 100000kb.
    I only use mesh items (but head) and due to the fact that i can mod all the pieces i wear, not only i can be 0 scripted without AO (and i use the free sub ao, that is not a lot laggy, only 256kb), but also my dw only goes a lil over 50000 when i use a specific dress that i really enjoy to much to get ride of it. (btw i don’t buy mesh clothes, all i wear are free gifts or finds on marketplace, so there are great creators that know how dw is important, in fact some already advertise it on their products).
    As you already know, i do disagree to the fact that most recent mesh builders, mainly of mesh bodies, don’t allow their items to be mod (with a few and admirable exceptions as Kitties lair mesh bodies, hand and feet and others i never tired so i can not advertise), so most are not able to remove not needed layers (like the layer for a tattoo if one does not use any, the skirt, shirt or whatever if you use onlt mesh clothes and so on), my lena perki body come with 5 layers by default, if you wear it straight of the box is Dw is over 50000kb, but if you just make a copy, remove all but the skin layer, for when you want to bre naked it goes down to 7000kb, So i do have a body that is only 1 ,layer for when naked, and another with also 1 layer that has faces that one can select and hide to use with mesh clothes. Both well down 10000kb.
    Imagine if Matreya or Slink (the main brands that already had the choice to delete scripts on their huds) allowed us to do same!
    (Can not mention belezza has i dont think it even allows to delete scripts via hud, much less mod, lol).
    Another issue, and that is a well known one, is also the size of textures being used.
    Many use oversized 1024×1024 on many builds that would only need 512×512 or even less..
    My skin for example, is only 512×512 but i wonder, by the time so many avatars i see around take to rezz, how many are using 1024×1024 skn textures.
    Still, there are in fact creators that care, not all, but we do must emphasis does who do!
    I dont have the time, but i would love to see a blog that would mention only the creators with good practices, like allowing their items to be full mod, by rezzing in world that use a respectable Dw and so on.
    Perhaps then, many, most of all, those that are proud of using their old avatar looks from the 2007 period (cause there are quite a few that look like that, not cause they can not afford, but by option) could move on to mesh that if optimized performs much better (compare a good mesh hair dw, like truth, to the insane Dw of flexible hair).

    Reply

    • Pussycat Catnap
      Dec 08, 2015 @ 22:43:02

      The whole new thing about no-Mod is really insulting. They say its to protect us from ourselves, to prevent people from breaking things… its really because they don’t have faith in the ability of their fellow SL users to learn things.

      That is somewhat justified – there are a lot of very thick headed people who can’t handle change in SL… But that doesn’t justify placing such limits because one doesn’t want to deal with basic customer support. It is not that hard to tell someone to read the notecard, to get a new copy, and that they’re on their own if they modify and break it.

      Lena Lush/Perky are great examples of why Mod is so important. What out of the box appears to be a ‘wasted throw away purchase’ suddenly becomes one of the best Mesh avatars out there once you realize you can mod the thing and strip away the pieces you don’t need.

      Furthermore, being able to mod it means you can mess with how it is textured. If you have a fullperm skin (admittedly a rare thing unless you grab the old 2009 free photoshop ones that are… lacking in quality compared to modern skins)… but if you do have a fullperm skin, you can make your own alphas by just putting them in at the photoshop level.

      (and those old 2009 photoshop kits work well if you’re looking to make a fantasy skin like a martian or goblin or something).

      Reply

  4. zzpearlbottom
    Dec 08, 2015 @ 01:48:10

    Reblogged this on ZZ Bottom misadventures in Virtual Worlds! and commented:
    A very good post about how all can make Sl a mcuh better place to be in.

    Reply

  5. Trackback: How to NOT buy bad stuff in SL | Pussycat Catnap’s thoughts | A Kat and A Mouse
  6. Trackback: How to NOT buy bad stuff in SL | Second Life - ...
  7. Sat
    Jan 14, 2016 @ 03:38:49

    Great post. I’m an oldie (from 2006 – and like an old person I like to wag my finger and be all “back in my dayy!!” LOL) and I didn’t know about the secret little red number. Turned it on to find my NikotiN cigarette gives me a solid extra 15k! Switched to Vista Fem.Cig which is VERY low in script AND draw weight. Win!
    Thanks so much, a good read, now I’ll be judging those around me even more >:)
    Was just standing next to a lady with 450.000! Let the games begin.

    Reply

  8. Viperilla
    Jan 17, 2016 @ 04:10:00

    Hello, inspired by your post and other similar, on the blogs of Nalates and Penny Patton, I decided to put my hand on my “weight”. I’m not a fat woman (<100k) 🙂 but I like to optimize things, because, both for work and RP in SL, I have to attend crowded places, loads of lag and it is important people have no trouble visualizing my AV quickly and well.
    BTW in terms of scripts I put myself on diet years ago and I am now very "light", both in numbers and in memory footprint (<256k).

    So I spent the weekend "weighing" my various accessories. Without surprise I horrified to the cost of some, in particular of some lovely flexi hair that I loved (some with rendering costs of 30 / 40k), although the best flexi in some cases were below certain mesh hair…
    So some things went straight in a box and in the closet of memories, other in the trash, for others I managed to improve texture and linkset.

    But, and it is the reason I write waiting for your comments and opinions, I was surprised when I addressed the body mesh. Referring also to the comments made about the body by KL (love…), with which I agree COMPLETELY regarding it being mod.
    Let me give you a picture of my situation. I own three bodies: TMP, an error of "youth" ( I hate it being no mod AND closed system, K$L in the trash). I do not use it anymore, thus I did not even weighed it. MY bodies are instead KL and Maitreya.
    Love the KL body, both for the fact of being mod ( a great plus to me eyes, I am VERY biased on this) and for the shape, which really it's …me 😉
    But my default at present is Maitreya. I had to buy it for a number of reasons (KL initially had a layer less, Maitreya became a reference, etc etc ..). I am very satisfied with it, but …
    Reading your comments I wondered if having as objective the optimization, I could put back into play KL. I guessed: "separating the layers I would get very light !", "I could wear them up to the need!!", " in this way I could even use RLV for "interactivity", very useful for my job and the RP!!!".
    But…what a surprise, here are the number I discovered:

    Maitreya adds 6298 points to my base rendering cost.
    KL Perky adds 61548 !. My god!! 10x the number!!!
    It pushes me stright into red numbers!!:((((((((((

    Had to investigate.
    Texture? No way…
    I disassembled KL, kepping the two up and low skin layers and nipples.
    Well…now reading a reasonable 7020.
    But it is nice only to be…naked. And usually I am not stright naked…LOL.

    I investigated more. Let's see the wireframe. Maybe too much triangles in KL…
    Ctrl+Shift+R… The KL triangles appear to be not so regolar as in Maitreya. But for sure they are fewer. The mesh in not as dense as Maitreya. So what?!
    Confused… What can it be? Maybe KL messed something in the LOD, i see it is considered into the calculation of the render weight.

    Let's see the land impact if I rez 'em both. Again: Maitreya 60, KL 180. Not 10x, but still 3x…
    But here I noticed a difference, don't know if relevant, maybe you have an opinion.
    While KL once rezzed appears to be a well formed manikin, Maitreya it's a stacked pile of body parts. A compact blob of pieces.
    Could this cause any difference?

    I see VISIBLE triangles are counted in the weight calculation. Thus having most of 'em hidden in the pile of pieces would count them out?!
    If it is so, it would mean that KM and Maitreya actually "weigh" similar, but are evaluate different. Because the triangles are hidden! Obviously this would lead to assume that the calculation is applied to the mesh as downloaded, BEFORE rigging to the cbones happen!

    I guess to be wrong in this assumption. But I admit to not being able to advance a different argument to explain a SO different weight (10x !!!), being triangles and layer similar and with the same textures applied.
    Any opinion?
    About my fanciful hypothesis.
    About the great weight of the KL body.
    I would really love to be able to prefer it, for the same reasons that you stated above.
    But if the reason had to be what I have suggested, it would be hard to argue with the owners of the clubs where I work, explaining them that I am not that "fat", that the red number they read above my head is due only to how it is calculated …

    Reply

    • Pussycat Catnap
      Jan 17, 2016 @ 12:48:15

      Wow at how low you’ve gotten your scripts to be. My AO alone is too high for me to pull that off.

      KL is interesting – because it is Mod you can take it to a sandbox (or somewhere you have about 200 extra Land Impact. Edit it with ‘Edit Linked’ – turn on the option to see transparent things, and just drag away layers you don’t want, then unlink them.
      – Some full time nudist friends use KL with just the body layer because once you get it down to that it is very low lag. But even just body and tattoo will be handy (except that ONLY KL makes clothing for KL’s bodies… so your fashion choices get ultra limited… again, KL is a great body for nudists, but not much of anyone else).

      On Flexi:
      Everytime I make a blanket statement about the horrors of Flexi, I find a counter example. I am not sure why some flexi is a lag beast and other flexi is just fine. I wish I knew. Even my theory that it has to do with ‘cut prims’ and ‘hollow’ is wrong… (saw a flexi gown with this yesterday that was almost lag free, I’ve seen other similar flexi gowns that would in the 100k draw weight. Why… I don’t know).
      – I wish I could find the “thing” that makes the difference…
      Because Flexi gowns often don’t have demos, they are thus risky to buy. Flexi hair usually does have a demo though.

      On the whole thing between Maitreya and KL you note there… rez a copy of Belleza (get the demo and see). Edit it – absolute nightmare to have that thing in your edit box (its no-mod so this isn’t something you will do often anyway). And yet once worn, its pretty low lag like Maitreya.

      Note to fellow Belleza owners: wear your body on the spine, not the right-hand, then edit it once and drag the little transparent box that is behind you inside yourself – this will prevent you from accidentally going into edit mode on that body. Its a pain to do, but once you do it, you will avoid those ‘lock ups’ that happen when you accidentally edit that mesh.
      – This is a good thing to do with any item you don’t want to be accidentally clicking on.

      Oh… reading your comment again, I think I see the reason why default KL is so messy:
      “The KL triangles appear to be not so regolar as in Maitreya.”
      – If I recall right from my reading of a Poser forum back around 1999… The pattern you lay out the triangles in is extremely important for how much a mesh will tax your system when rendering and posing it for fine art. I strongly suspect this is even more true for ‘real time graphics’.

      I think ‘visible triangles’ just refers to those triangles that have a texture face. Otherwise wearing a skin on KL that was just the default transparency texture would lower your render cost, and it doesn’t. Likewise using alpha cuts would lower you, and it too also doesn’t do that. In a lot of mesh you will find triangles that have no face – no UV map covers that part. It is thus always invisible.

      Reply

  9. Viperilla
    Jan 18, 2016 @ 23:53:24

    Hello and thanks for the reply.
    I guessed to be wrong LOL.
    My guess come from this portion of the specification from wiki:

    “Base cost is a weighted average of the number of triangles times 5. Triangle count is weighted by the count of triangles at each LOD and the area each LOD is visible from, giving the average number of triangles visible within the visible range of the object. This uses the same distribution function as the streaming cost code, but left in terms of triangles, not scaled to prims. ”

    In any case it suggests that the LOD is involved.
    Maybe KL has mismanaged it. Or maybe Maitreya uses a trick between different LOD (I read of the practice to rig the low LOD making a simple triangle).

    In any case, elements that I think are still involved to justify such different weights (10x!) may be:
    – Management of LOD
    – Alpha type. KL uses blending instead of masking. And I understand has a higher multiplier (4x per prim)
    – Size of the object. I think, however, the algorithm calculates the weight as the object downloaded, not placed on the skeleton. So the more compact size of the object becomes significant.

    Topology do not know. Certainly your statement is true. Only I do not find the topology as a variable in the specification of the DW. Definitely a bad constructed object costs more to the graphics engine (decrease fps). But it seems doesn’t LL consider it in the algorithm for DW calculation. In a sense what I’m assuming it’s a bit to doubt the methodology of calculating it as an absolute reference.
    Knowing KL is less laggy compared to other bodies, heavyer in DW, suggest this.

    Thanks for your hospitality and bye.

    Reply

  10. Trackback: New Graphics Preferences Now Help Auto-hide Laggy People | Pussycat Catnap's thoughts
  11. Trackback: “OMG – it’s so lagggy here…” – ShenaniganS Music Cafe

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: