The Big Prim Problem

A small quote I know lol, its from a few pages back...

I dont believe that, as other 'problems' such as the pivot LSL and so on, that cause Server and also Physics problems, they got rid within a few months, granted, huge prims arn't as server consumptive as pivot was, but the issues are the same here.

You can say its because of art, creation and so on, as I would agree fully, but there was a few things I saw years ago, that made pivot just as creative.

I think the only reason they are still here is because of the previous comments, maybe... and I STRESS 'maybe' as its only speculation, I'm not talking about conspiracy, its common tactics that have been used before to diagnose and present a 'problem', but with different media (No I'm not talking about LL with this one).

But like you point out about all Lindens as an individual... me, you and the rest of this community are too, with our own opinions.

The only difference is that LL is a company not run by individual ideas, its run of cumulative. If it was, it wouldn't of lasted as long as it has done.

Do you work for LL? No. Do you Talk to Lindens about projects and what not? Probably, Yes, you're in Dev Group, the same as me. You seem to state things in this thread like you know the ins and outs of how LL works, but they come down to being opinions of your own, and should be treated as much.

I am sorry though Seifert, I don't mean to sound rude or discoutious, as I like your work and the things you've done in SL, but, one of MY peeves is residents that sometimes act a little more than what they are, which is... one of the same as the rest of us.

I'm not gonna name any names but i agree with this statement, as I've had to deal with my share of these people. One even tried to damage my reputation for reasons i still don't know to this day. They pretend to be professional, but their nothing above the rest of us. Some are even below standards.

If both of these are causing physics related issues then chances are it's one of the things LL would like to fix by removing mega-prims. Having prims heaped in the corner like that should only effect physics if you walk into them, but if it affects all physics then it implies that we need optimisations in that area. This would be really easy if mega-prims didn't have to be taken into account, since you can just break the sim into sections and completely ignore ones you're not in.

This would account for Yiffy Yaffle's claims that the mega-prim actually performs better in their case than lots of smaller prims, however if LL were able to put optimisations as I describe (or following the same principles) then the reverse would be true.

I'm not big on the Havok engine, but chances are it's something along these lines that LL would like to work with. The problem is that if mega-prims are still around then this can't be done, as such a system would cause only the centre 'chunk' of the mega-prim to work with collisions, the rest would essentially become phantom.

I think those that are big fans of mega-prims need to recognise the greater potential of a system with limits. If you know prims are going to be no larger than 10m x 10m x 10m then you can start making assumptions and use these to optimise performance, which is good for everybody. But mega-prims are a freak-occurance that as a result mean that such assumptions can't be made without breaking them or coding in lots of special cases that defeat the purpose.

Assuming then that those of you who support mega-prims as a building tool are not exploiting them to get around prim-allowances; then you should recognise the merits of system that provides the ability to work as with a mega-prim, but that uses smaller pieces to achieve an identical, or even superior result. Such a system as I described in my previous posts would then be able to allow builders the greater flexibility of larger prims, and allow LL to do whatever performance enhancements or new features that they have in mind, giving us the best of both.

Hmm cool idea for a test exccept delete the irrelevant step 3 as I don't think you can torture a MegaPrim and most are used as flat floor or wall surfaces anyways, not making giant coil springs.

Not quite the same issue here, good sir. Ye olde joint system was an implemented feature, removed because it was simply too strenuous. Megaprims are not an implemented feature, they are a user exploit that has been tolerated.

Let's all remember why this is an issue now, and to those who were doubting my argument about Havok4: megaprims are on the verge of death thanks to problems that came up in Havok4. There's no need for all this discussion about how strenuous they are, or whether anyone can prove they cause server problems, etc. It doesn't matter if it's absolutely proven beyond a shadow of a doubt that megaprims slow down servers, or that they cause no impact at all; whatever the result is, that's how it's always been since they were invented. LL has tolerated it until now, and the reason they are stopping is because of the problems resulting on the beta grid under Havok4. Thank you.

Side note: don't tell me Havok4 can't handle a new joint system, please bring back joints

It is my understanding that joints were removed because it was impossible to develop Havok 4 whilst still supporting them. That is, they use such different parts of a physics engine from the usual prims, that it would take a huge amount of work writing new code to have them work. As far as I am aware, it was not about cpu time on the server, it was about having to write lots of new code to get them running.

The same is not at all true of megaprims. The same code that works for normal prims also works for megaprims.

BS.... it's ALL about money!!! more prims = more land = more tier = more money.

Oh! and by the way,
655 posts and more than a week later,
and still no input from the lindens.

Money indeed.

Wait until several dozen corporate builds, sims that people paid 30k USD and more to be developed, are trashed horrifically.

Sims that would likely cost near as much to fix, if they aren't just canned.

Sure, no feature is guaranteed. But goodwill and customer dollars don't come with guarantees either.

* * * * *

It's a philosophy.

We can either have sharp scissors to work with, or the safety kind with the dull edges to protect ourselves from ourselves.

So what will it be?

Take our megaprims, but then take all classes of scripts that can be used to harm the user experience (almost all LSL can cause trouble somehow).

Then take away the option to texture too many prims and lag a server to death - cut our prim limits to 5000 per region so we can't hurt ourselves that way too.

Something to think about.

When I said bring joints back, I didn't mean the old system, I just meant something new along those lines. As for your statement about megaprims, that's exactly the issue with them in the first place; any prim larger than 10x10x10 only has the resources of a 10x10x10 prim, which results in all the physics problems. From what I've learned, implementing natural large prims would take up more resources, which is the big limiting factor on prim allotment. In other words, even allowing us to make 20x20x20 prims might be cause to drop prim limits to the point where it's not worth using them for their prim-saving factor. At the same time, people who don't use larger prims would suffer from the reduced prim allowance.



Be less shallow please. The problems with megaprims on the beta grid aren't a myth, and megaprims aren't just for circumventing parcel allotment. LL isn't going to magically profit from removing them.



Megaprims are not a feature, they are technically a TOS violation. No corporation that spends 30,000 USD on a build would be stupid enough to have megaprims, and if they are, they certainly would have zero ground for complaining about their removal. Corporations are not like independent users, they have specialized employees or lawyers who breathe the TOS for them and prevent them from doing anything that would endanger their investment (e.g. use megaprims for a build), at the same time making sure LL doesn't do anything beyond their capabilities that could also endanger their investment (as stated, megaprims are a TOS violation, and LL has complete authority to remove them without question). If you offered to build IBM some offices with megaprims, they'd laugh you off the grid.



Oh I'm thinking, quite hard actually, and I just can't figure out where you got this safety analogy. Megaprims are having problems under Havok4, and since they are nothing more than an old LSL exploit tolerated to this point, some Lindens are interested in completely removing them. This has nothing to do with us hurting ourselves. Yes there are plenty of ways to harm the user experience, as you put it, but again, megaprims are a TOS violation and something easily done away with at no liability. Sure, megaprims have their uses, but this is nowhere near comparable to your examples of "safety" limits.

Quite a few people in this forum seem to believe megaprims are an entitlement. While my previous arguments hold that they should either not be removed or come with some alternative, I'm still completely aware that they can just be removed and no one will have any right to complain. It's really a tough situation, and I think most of us believe that the Lindens handling this issue are leaning a certain way, but it's important not to cloud up the issue with misperceptions and false information.

I haven't heard this before. Is use of them really a TOS violation? Can you point me to the relevant section? I could believe that the way they were originally created was such a violation (although I'd want to see a reference for that too), but the use of them?

So how does having to own 4 times the land (based on the use of 20X20X0.5 megas) to complete the same build not benifit Linden labs? Simplicity is not shallowness.

/me laughs... Good one

I know lol I was just merely pointing out that 1) mega prims and pivot cause server and phys problems, like I said I know that its not as intensive as Pivot was but this issue is the same as for deletion. 2) Creativity. and 3) I seem to remember that a few people were abusing them too, I cant remember... Ah well lol QUOTE QUOTE QUOTE!! See what you've done, you've got me quoting myself lol I'll restate my last post... "I Mean - Sort of the same, indirectly the same and sort of the same issues" lol I'm not comparing prims to scripts, just the ideals. 'Cuz that would be stupid, aye?

You're joking though? Huge prims arn't a feature?!?! - joke lol.

Theres loads of exploits in SL that have been tolerated, mega prims, invisi-prims, invisi-shine, physical flexi prims, null-prims... lets just get rid of all the good stuff shall we? Actually, if they're getting rid of one exploit, the same should apply to all of them. We can't pick and choose now can we.

I know someone will come back and say - Well it doesnt matter does it, from what alot are saying, an exploit is an exploit. Lets take our only way of making larger prims for prim conservation, out... our only way of making the base AV invisible, Nice shiny windows and so on... (Even though I don't think I'd ever use phy flexi )



YUS! You can torture huge prims, some of them look quite pretty, I have a nice 40x40 torus coil for a multistory car-park



IBM was one of the first major companies to use Huge Prims in SL. There are others... Its against TOS to make exploits from SL malfunctions or 3rd party, but again, its a matter of picking and choosing, like copybot, its completely against TOS, but Huge prims have been allowed to be used on private islands, just not on the main grid.

You said that its our own fault for using such an exploit and if we dared using them that we should of known they would of been deleted... well, my answer to that is LL are being a little bit of an 'indian giver' with this one (No offense to indians ), the original post to this thread (and I remember them saying it originally) was that LL allowed huge prims to be used and on private islands... now a year later, its become a problem. Its like giving a baby icecream, complaining that he/she has it all over their face and then taking it away because he/she was trying to eat it.



YUS! lol I wanna make nunchucks! lol GIMME!

On another note, I may of over quoted...

You're saying that LL removing megaprims means people will absolutely be forced to own four times as much land, thus quadrupling LL's profits. This doesn't figure. Obviously if you already own an entire sim, you can't add any more prims to it even if you buy more. So you must be talking about smaller parcels. However, let's say you go from a quarter of a sim to a full sim. The only case where LL is going to make more money this way is if you're buying more mainland; if you're doing this on an estate, LL is already getting as much money as they can from the island. I'm going to assume that most mainland is already owned, and LL couldn't possibly make any more from small tier payments. Whatever small margin exists in which they could, it is far overshadowed by the profit machine known as the estate system. So granted that maybe there is some mainland left in which expansion could lead to Linden profits, what makes you think that people capable of expanding are even going to do so? Not everyone uses megaprims, and some of those who do might never buy more land anyway; if they had the money to, they probably would have had more to begin with.

To recap, LL doesn't profit if you decide to quadruple your land in a non-Linden-owned estate, and in fact the only method by which they will profit is so limited that it doesn't warrant the complete removal of megaprims just to make that small amount. On top of this, there are already technical reasons and issues which have led to this debate, so you must be quite shallow indeed to think that this is somehow a money-making scheme.

You forget that if person (A) has to take 4 times the land, no matter if mainland or estate, person (B) can't have it anymore and has to look elsewhere. So to give person (B) land, there has to be more land added, either mainland or estate, and LL will profit from it.

We tested replacing one small floor in my club on Mainland with 1 40x20 megaprim, the lag doubled and the prim had texture issues. I will not sacrifice lag for a prim shortage!!!

I have another spot on an island , was lag free until some bozo built a entire large glass office building nearby using them, all 20x20's as far as I can tell... you can come to my space on the island now without crashing on arrival, sinking into the ground or wearing newsprint, the SIM owner is going to allow them to stay, because it would be "TO HARD FOR THEM TO REBUILD?" So they couldn't read the covenant and I have to suffer because I did?

I might just make 1400 20x20's in a pile to prove a point. How are 4 20x20's less problem than a 40x40? the math for the argument seems real bad...

Anything created by GENE REPLACEMENT should be DELETED!!! EVERYTHING!!!
If everyone is doing CRACK, do we legalize it?
Allow bigger prims ONLY if there are NO performance Issues.
Bad enough I have to see little UGLY buildings, do we need HUGE ones?

Put as many normal prims in your sim as you need to fill up a megaprim space. Then compare the lag with doing the same with a mega prim.
Then come again.

Done it, on Linden Mainland, more lag...
Done it on a Private Island, more lag...

The change was way more obvious on the Mainland...
Before megaprims, no lag at the Mainland, neighbors start using them, lag, nothing has changed at my location since May. Although, the lag seemed to increase about the same time voice was added to the client. We tested before voice was added at that location.

Hmm well I stand corrected the, but from what I've seen the primary use for megaprims is flat floors/skys and walls.
Ther OP suggested 14999 megaprims vs 14999 standard prims in a sim test which is silly it's like comparing the performance of a light truck loaded with 10x baby elephants to a truck with 10 adult elephants.
Better would be to compare 10000 10x10m flat sheets to 2500 20x20m as this is closer to a real use test.
And perhaps compare performance of a single 256x256m deck to the same area decked with 10x10m?
Over 256m is silly, if they want a never ending sky they can put more 256m ones up in the sims next door or probably link them to the centre one and slide them out if it's an island.
Perhaps LL need to perform these tests for us with various size megaprims 256m or less with some independant residents present and publish the results if they truely show drop in sim performance the numbers will show it in black and white numbers.
Anyone with an empty sim game to do these tests and publish the results in numbers? If we can see numbers clearly showing us that 100 20x20 degrade performance more than 400 10x10, many of us will be convinced to a need for their removal.

A 600 prim (Regular size prims) sculpted wheel I made, lags me more than my entire sim that consists of over 3000 huge prims, ranging from 20x20 - 10x30 - 50x50 - 100x1x1 - 100x100x100 and the remaining 12000 are regular sized. Regardless to say my sims been up for a year with no problems.

I'm not going to go over the build I made again, its a few pages back if you want to know more about the 1200 huge prim race track. It wouldn't of been the same using regular prims.

More people seem to be just focused on if theres a really crappy house next door.

Anyone that rents land is entitled to make what ever they want, as long as its not a violation of TOS. Even if you don't like the build.

If they made the building with regular sized prims, you'd still have to deal with living next to a crappy house. But you wouldn't have anything to blame BUT the resident. The same should apply to people using Huge Prims in their builds.

Another comparision is -

Sculpted Prims -
Physic Problems - Yes (Especially if you do not create them as a maximum scale)
Server Issues - Yes
Thus Cause Lag - Yes. In large amounts - God Yes.

Huge Prims -
Physic Problems - Yes (Some more than others)
Server Issues - Yes (Can crash sims when not used correctly)
Thus Cause Lag - Yes

Regular Prims -
Physic Problems - Torus = Yes
Server Issues - Minimal, depending on what you're doing with them
Thus Cause Lag - Large amounts in small areas can cause large amounts of lag and crash sims.

Heres something you can do at home - check your console in the client menu and see what causes the most problems, because Sculpted prims cause way more than any other type of prim.

Try filling a sim with sculpties at HALF the sim prim count and see what happens, never mind filling a sim full of huge prims.

** My argument is any type of prim has its problems, even actual featured ones **

That sounds like a bug with sculpties; server-side they are just spheres, everything else is a client-side effect only.

i think LL should give a bit more prims to people...if we had a bit more prims to fill our land then people wouldnt have to use maga prims...I own 4096 sqm and what LL gives you is not enough to fill you land properly....I'm sure if LL gave a bit more prims then nobody would have to use mega prims.

I said it before and i will say it again. if the system is unable to handle larger prims then REPAIR THE SYSTEM so it can! It's quite simple. Doesn't take a friggin genius to understand that...

When linden lab made accounts free, they didn't repair their system to handle the load of people it would gain but as time progressed they sorta got off thier tuffs and repaired the grid to handle more players. This is just another step of evolution.

Yup, true... in which only enforces that every prim has its problems.

Flexi being the same (Client-side), attach 300 flexi prims and see the problems that lay there. It may not be server-side, but it does effect everyone viewing and can cause the viewer to crash. 20 flexiprims are enough to effect surrounding viewers.

Granted client side can be 'tuned' in preferences and the debug menu, but theres still a problem there.

I know its off-topic, but people are complaining that Huge Prims are possibly causing problems to the servers, but this only becomes a possible problem if you're in the sim or in view.

Personally, I don't see there being an urgent need to delete them, I've been using them for over a year now and only see them as a great tool for overcoming the 'Regular Prim Problem'.