What is the script penalty?

[Pamela Galli]

I had thought the script penalty for mesh had been reduced but not sure now.

First I imported a sofa made of three meshes -- 12 LI.  Then I imported the same sofa as 11 meshes -- 6 LI. So far so good.

Then I added three invisible "sit prims" loaded with the sit animations for the sofa, plus a MLP prim, and a removeable shadow prim. These 5 transparent prims = 5 LI linked t each other and set to Covex Hull.

Yet when I linked them to the sofa, they added 7 LI, for a total of 13. AND when I copied the sofa, both original and copy changed to 14 LI.  So the 5 scripted prims added 8 LI to the sofa.

 

Weirder still:  This sofa is an update of one of the first mesh sofas I made. That sofa is in 11 meshes for 9 LI, but the 5 scripted prims add only 6 LI, for a total of 15.

 

This all just seems so random.

[arton Rotaru]

The official formula for the server weight is num_prims*0.5 + num_scripts*0.25. So each script should add 0.25 to the server weight.

How many scripts in total has the sofa in the end?

Always check "More Info" to see which is the highest weight. Download, Physics, or Server Weight. LI alone doesn't really tell anything when debugging.

[MIstahMoose]

Adding the invisible prims could have pushed it over the size threshhold for LI and bumped everything up a tier, So it ended up adding a lot more LI?
Just a thought.

Also, So many sit scripts, why still doing poseballs? If removing poseballs pushes it down then throw on a newer sit script that can handle all the positions and avi's you need?

 

ETA: Welp looks like Arton beat me to it! (Though maybe the single script with positions etc would decrease it still)

[arton Rotaru]

Indeed, reducing the script count will decrease the server weight. However, we stll need seperate pieces in a linkset to set individulal sit targets. Sitting multiple avatars on a single sit target has still the "No room to sit here" issue, when the object is surrounded by another prims bounding box. When you sell these sitting devices, you never know where the customer will place them. So it's best to have a sit target prim for each avatar. The 11 mesh pieces should be plenty though. Unless Pam wants to sit up to 16 people on that sofa. Who knows?

[Pamela Galli]

I dont use poseballs for the sits, but I do also have a MLP engine in there, which does have a lot of scripts.  So that answers one of my questions.

[arton Rotaru]

The way it used to be in the beginning was, when adding a single script to the linkset, it had pushed it from 0.5 per prim to 1.0 per prim.  The 16 pieces would have been 16 LI (Server Weight) with just a single script. So yeah, it's still reduced.

[Pamela Galli]

Yes that's a lot better than it was.

Still mysterious -- I have a bar, part mesh, 30 prims total, considerably larger than the sofa, with dozens of scripts, and it = 17 LI. 

[arton Rotaru]

Yeah, the script counting is kind of weird indeed. It also looks like the server weight won't go any higher than the number of prims in a linkset, no matter how many scripts you add on top of that. So there is a maximum server weight of 1.0 per prim..

[Drongle McMahon]

The official formula for the server weight is num_prims*0.5 + num_scripts*0.25.

Arton, is there an official source for this? I remember reading about the changes, but I can't find it anywhere. The wiki still has the old calculation in. Somebody should change it.

[arton Rotaru]

Right, the wiki hasn't been updated. The most official source I could find are the server release notes here: http://community.secondlife.com/t5/Second-Life-Server/Deploys-for-the-week-of-2012-08-06/m-p/1626331/highlight/true#M6581

 

[Drongle McMahon]

Thanks. I think that may be where I read it. I'll quote the salient extract here, for the record...

Changed prim accounting for legacy prims which use the new accounting system All legacy-style prims have their streaming cost capped at 1.0 (except for sculpts, which will be capped at 2.0). This provides the benefit of not penalizing prim-based creators for optimizing their content by opting into the new system and will make the streaming cost more reflective of the true network cost of the objects.

Server cost will be adjusted to MIN{ (0.5*num_prims) + (0.25 * num_scripts), num_prims }. This preserves the current value for unscripted linksets and reduce the cost for linksets containing fewer than 2*num_prims scripts. It provides the benefit of rewarding creators for reducing the number of scripts in their objects.