Jump to content

detecting avatar lag with llGetObjectDetails


ziicutie
 Share

You are about to reply to a thread that has been inactive for 4160 days.

Please take a moment to consider if this thread is worth bumping.

Recommended Posts

I am having some issues with the command llGetObjectDetails and finding out what number values I need to watch out for when this command is used on an avatar. For example: using OBJECT_SCRIPT_MEMORY, I know that if an avatar is over 3 MB total, then they are concidered overweight on their scripts by most people's standards and some places will boot them out. I am tring to find some reasnable standards for OBJECT_SCRIPT_TIME, OBJECT_PRIM_EQUIVALENCE, OBJECT_SERVER_COST, OBJECT_STREAMING_COST, and OBJECT_CHARACTER_TIME.

Link to comment
Share on other sites

Not all of those constants return sensible values for avatars: PRIM_EQUIVALENCE, SERVER_COST, STREAMING_COST and CHARACTER_TIME have no meaningful results when used on avatars.

SCRIPT_TIME works, but has one major flaw: When someone enters a region, their script time will be very high while their scripts get transferred into the simulators' engine. Depending on some factors it can easily take a minute or more for script time to settle down.

About the best thing you can probably do is gather some statistics at your desired target region and then set your own limits. One possible approach: Check what idle time your region has, then divide that by the number of avis you'd like to support on the region.

Link to comment
Share on other sites

What you think is "avatar lag"?

It's not that simple.

Scripts are only one minor threat here. Scripts increase the server load. You can get:
- the number of scripts (very high numbers indicate possible trouble)
- the max allocatable memory (but 99% of the scripts only use a fraction of that so this number is worthless)
- script time (thats an indicator but you need to track that number for quite a while after an avatar tp'd in to get a usable result)

The other issue is the client - your client. If there are many avatars wearing many prims/sculpties/mesh with many textures and flexies then your client is probably too weak to render the scene with a decent framerate anymore while others on big machines dont have any issues.

There are no script functione to measure that.

 And:

If an avatar teleport's to a place he starts to load geometry data, textures (probably loads of textures), all data from other avatars around, sounds, etc. If they have a high view range they load more. That adds network lag to the scene, especially when avatars tp in and out all the time. So maybe its the landing point with millions of prims and textures that causes the lag :)

Link to comment
Share on other sites

You are about to reply to a thread that has been inactive for 4160 days.

Please take a moment to consider if this thread is worth bumping.

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
 Share

×
×
  • Create New...