Any chance for Animesh taking up Less prims?

[Rob Huntsman]

It's a real shame that most animesh pieces are 30 - 40 land Impact. Now some may call me stingy for this, but I prefer most my objects around the 1-7 Land Impact range or I don't set it down. Is There any chance that Animesh is going to get an update to take up less prims/usage. 

[Lucia Nightfire]

Maybe people have just got used to incorrect/inefficient land impact calculations all these years.

When Project ArcTan is fully implemented, get ready for even more screams/tears when most existing mesh has its land impact increase and avatar complexity values increase as well.

Edited June 12, 2019 by Lucia Nightfire

[Erwin Solo]

1 hour ago, Haselden said:

It's a real shame that most animesh pieces are 30 - 40 land Impact. Now some may call me stingy for this, but I prefer most my objects around the 1-7 Land Impact range or I don't set it down. Is There any chance that Animesh is going to get an update to take up less prims/usage. 

Animesh gets calculated by a special formula just for Animesh, and the result is there is a land impact penalty for using Animesh.  There's a whole forum for that topic and @Lucia Nightfire is a significant contributor there. 

https://community.secondlife.com/forums/forum/354-animesh/ 

 

[steeljane42]

3 hours ago, Lucia Nightfire said:

Maybe people have just got used to incorrect/inefficient land impact calculations all these years.

When Project ArcTan is fully implemented, get ready for even more screams/tears when most existing mesh has its land impact increase and avatar complexity values increase as well.

There won't be much screams and tears, because it's apples and oranges. One is completely new product that you can avoid buying if you think it's too heavy li wise, it was like that from the very start. The other will affect every single rezzed object (or mesh only, no idea really, but it's rare to see non mesh things these days, unless you explore ancient-abandoned mainland places) on the grid.

It would be commercial suicide for LL to tell people "welp, now you can't use things you bought, because... reasons. but it's good for you! now go buy L$ and replace your things with new ones". But LL won't do it anyway, they have been very careful to not break any existing content. Also see the recent thing about completely free 7 groups reducing, which didn't cost anyone anything, yet people were THAT unhappy so LL decided to revert the change before it even happened. Now imagine it would be something affect things people have bought for 1.5 decades (a bit less, if it's only for only mesh). And I remember Oz saying about ArcTan, that they want people to have more stuff, not less, so perhaps even if some stuff will become more heavy LI wise, it will come together with another LI upgrade. It surely will be after moving to cloud is done, so I won't be surprised if it was the case. 40-50k LI full sims might be a reality then, especially if they'll sort the scripts issues.

But yeah, I think heavy li is very big limitation for the animesh things. it's been what? Half a year+ since full official release and I barely see anything "animesh" if don't count an odd attachment here and there (mostly tails and ears, sometimes some weird npc followers). I thought I'd see cute rezzable animesh cats and other critters ages ago, but there's almost none on the MP and ones that are... let's just say I don't like them, to avoid more negative comments. I can totally see why 30li at the minimum is one of the main reasons for it. If someone has a small parcel with around 300li, then it's 10% for just one thing. Less possible sales = less reasons to make them for creators, since it's not a direct upgrade over existing static/animated ones thanks to such high li.

To compare, bento stuff started to show up even before official LL viewer was updated with it, half a year later there already were bento hands for all major brands except belleza, there were plenty of bento heads and numerous wings/tails that you could see everywhere in-world already. Because it was direct upgrade, while animesh is more like a questionable sidegrade, thanks to its li. At least when it comes to rezzable things.

Edited June 12, 2019 by steeljane42

[ChinRey]

5 hours ago, Haselden said:

It's a real shame that most animesh pieces are 30 - 40 land Impact. Now some may call me stingy for this, but I prefer most my objects around the 1-7 Land Impact range or I don't set it down. Is There any chance that Animesh is going to get an update to take up less prims/usage. 

Yes and no.

Remember that land impact isn't just an arbitary limitation LL came up with because they're evil, it's an attempt to estimate and control the load objects put on the computers and the network. They did quite a lot of work to determine how heavy animesh is and we can be reasonably sure the LI is fairly realistic - if anything it may be a bit on the low side.

What you can hope for, is that content creators find ways to make more efficient animesh as they become more familiar with the feature.

 

1 hour ago, steeljane42 said:

I thought I'd see cute rezzable animesh cats and other critters ages ago, but there's almost none on the MP and ones that are... let's just say I don't like them, to avoid more negative comments. I can totally see why 30li at the minimum is one of the main reasons for it.

Moving animesh characters like pets and NPCs have a special problem there. They use pathfinding to move around and that will always use almost 15% of the total sim resources for a full sim and more than half the resources of a homestead. There's no way around that.

Edited June 12, 2019 by ChinRey

[Selene Gregoire]

6 hours ago, Lucia Nightfire said:

Maybe people have just got used to incorrect/inefficient land impact calculations all these years.

When Project ArcTan is fully implemented, get ready for even more screams/tears when most existing mesh has its land impact increase and avatar complexity values increase as well.

Wait... so all the effort and work I've been putting into making my creations as low LI as possible has all been wasted since, at some point in the future, LI is just going to increase after LL increased region/parcel allowances? 

Sounds an awful lot like the recent blowup over decreasing basic account group allowances (7) in order to be able to increase premium group allowances (10) in order to reduce the "load" on the servers. Just how is decreasing by 7 and increasing by 10 supposed to help when you end up with 3 more groups per premium account than what you started with? Increasing the load is supposed to help? 

Rhetorical questions. 

Still the same old Linden Lab Dog and Pony Show.

[KanryDrago]

21 minutes ago, Selene Gregoire said:

Wait... so all the effort and work I've been putting into making my creations as low LI as possible has all been wasted since, at some point in the future, LI is just going to increase after LL increased region/parcel allowances? 

Sounds an awful lot like the recent blowup over decreasing basic account group allowances (7) in order to be able to increase premium group allowances (10) in order to reduce the "load" on the servers. Just how is decreasing by 7 and increasing by 10 supposed to help when you end up with 3 more groups per premium account than what you started with? Increasing the load is supposed to help? 

Rhetorical questions. 

Still the same old Linden Lab Dog and Pony Show.

I think that was an opinion 9n what Lucia thinks they should do rather than something ll have said

[Selene Gregoire]

15 minutes ago, KanryDrago said:

I think that was an opinion 9n what Lucia thinks they should do rather than something ll have said

I have the feeling it isn't just an opinion of what should be done when I take into consideration LL's past track record that I know of...

When I have that feeling that a bombshell is about to drop, it's rarely wrong. Just a question of when. It wouldn't surprise me if this were the bombshell I've been waiting for. It's already falling. It's just a question as to what it is and when it hits.

It's Linden Lab. There's always a bombshell. 

[Theresa Tennyson]

2 hours ago, Selene Gregoire said:

Wait... so all the effort and work I've been putting into making my creations as low LI as possible has all been wasted since, at some point in the future, LI is just going to increase after LL increased region/parcel allowances?

A lot of time, the "effort and work" involves making something less efficient/useful in order to game the land impact system.

Jurassic style - using a cut and hollowed prim to form multiple walls of a house and actually making the physics engine work harder than it would to treat those walls as one prim each.

Cretaceous style - using extreme drop-offs in polygons for lower LOD's of a mesh object and telling people to crank up their viewer settings to compensate.

[Selene Gregoire]

30 minutes ago, Theresa Tennyson said:

A lot of time, the "effort and work" involves making something less efficient/useful in order to game the land impact system.

Jurassic style - using a cut and hollowed prim to form multiple walls of a house and actually making the physics engine work harder than it would to treat those walls as one prim each.

Cretaceous style - using extreme drop-offs in polygons for lower LOD's of a mesh object and telling people to crank up their viewer settings to compensate.

You might want to give that information to people who haven't been building in SL for 15 years and don't know any better.

Oh wait...

you just did. 

That is, if they know of the existence of the forum or even read it.

[animats]

There's a whole topic on this, as someone pointed out. The short version:

• Minimum animesh: 15 LI.
• Optimized character which looks like a classic avatar: about 30 LI.
• Good mesh avatar, nude, as animesh: 80-130LI.
• Good mesh avatar, as animesh, with good mesh clothing: 130LI to "Parcel Full"

30 LI. Bento animesh, baked on clothing, mesh hair.

My test animesh character. Around 10,000 triangles. 666 tris = 1 LI, plus 15 LI for being animesh. This is about as low LI as a good looking animesh character can go right now.

Edited June 12, 2019 by animats

[Lucia Nightfire]

Project ArcTan's goals have been brought up in the Server and Content Creator meetings several times.

Animesh's "charged triangles" protocol is a main part of the calculation change ArcTan will use. The other main change hinges on texture memory an object uses.

For these two facets alone, it has been speculated by developers that existing mesh content will have land impact gains grid wide.

I've proposed a dual land impact implementation method where existing rezzed mesh will keep using the current system and only newly rezzed mesh will use the new system.

Oz Linden stated that he does not want to support a dual system.

This means there will have to be some kind of accommodation in place before the switch is thrown, else you're going to see mass auto-returns of objects.

There currently is no work being done on ArcTan, nor is there any schedule for it to start.

Feel free to participate in the Server and Content Creators meetings to bring up your thoughts/concerns.

The matter has always been open for discussion even though there are few details in place at the moment.

I still feel we have a long time before any serious work starts. It will probably be as late as next year.

[ChinRey]

2 hours ago, Lucia Nightfire said:

Animesh's "charged triangles" protocol is a main part of the calculation change ArcTan will use. The other main change hinges on texture memory an object uses.

For these two facets alone, it has been speculated by developers that existing mesh content will have land impact gains grid wide.

That's my impression too. The triangle weight of the LoD models for animesh is very "generous" indeed and if that's what they're going for, there shouldn't be any increase in the download weight for any existing meshes, quite the contrary.

I have mixed feelings about this myself. On one hand animesh style weight calculation doesn't exactly encourage optimised content but on the other hand it should reduce the amount of butchered LoD models since there won't be much to gain from it anymore.

Of course it's possible that they'll include render cost in the new LI calculation but if so, it'll only be symbolic.  Any limits with practical significance there will break Belliseria and that's not going to happen.

Edited June 12, 2019 by ChinRey