Avatar Rendering Cost: Change Your Habits?

LOL -- OK, fair enough. I certainly wouldn't approach you that way. It was a poor choice of phrase.

My point remains: nobody here said or meant to imply, that avatars over some arbitrary limit would be peremptorily banned.

Great post, Kitty; thanks for solid facts and good analysis. Very interesting.

BTW, I have a trivial freezefiew HUD that could be useful for this kind of thing; if anyone wants it just IM me ingame (I'll see it in email, no IM caps. Please don't drop me a notecard, which I can't read until I'm ingame).

It's just a little translucent square in the lower left (or wherever you attach it) that when you click it, it freezes your cam and turns solid white. There are a number of GUI things that you still can do to move the cam (which ones changes from release to release), but if you walk or hit escape you zip back to the freeze point. Click again and your cam is freed back to the normal case. Handy when building or shooting a photo sequence and you want a reference cam point.

not in this thread but in others and to boot: http://jira.secondlife.com/browse/VWR-7074

Kitty, how does the FPS effect from these prims change if you zoom out to some nominal viewing distance, like 5M or so ... ideally, the average avatar distance in a group. (Or perhaps the average non-impostered avatar distance, for those of us who use that feature.)

The av imposter feature is great, but too bad it kicks in so close. It would be great if that distance were configurable. (Applying it based on ARC would be cool too, as someone suggested above.)

/me reminds himself to look for JIRA entries concerning avatar imposters.

Thanks for pointing that out. I posted a NO vote.

My only point was showing that geometry comes into it as well really, it's hardly a real life test .

The biggest problem is that everything on screen needs to be exactly the same except the different attachment you wear. If you're looking straight ahead in one try and sideways in the other then you're rendering different things (the front but not the back in the first, half a front and back side in the second). I think I tried each hair 5 times just to make sure that what I wrote down was still accurate .

If I had zoomed out, level of detail would have come into play on the hair, but something else in the room would have come into view that wasn't visible before, adding something new to render and making a comparison much harder .

Common sense is going to be the best tool of all anyway too . If you know your puter grinds to a halt when it has to deal with fog particles, you can turn off rendering avies and prims and you'd still get 2fps while if you turned off only particles you might jump to 8fps. Or if you know your puter can't handle crowded places but you want to go to a party, don't look into the crowd but ask your friends to gather in a corner and have your back to the crowd and face only your friends.

I think this is a great little tool. I'm not going to completely give up my favorite alpha textured hairs and flexy skirts, but I have been more careful about which items I combine in my outfits.

I did a little test a while back in my skybox and with great effort, I was able to get my ARC up to 20,916! Not an outfit I would ever want to wear again, but a fun way to test out various items in my inventory. If you're curious, here's a pic and some details:

http://iheartsl.com/2008/05/13/how-high-can-you-go/


I'm not about to bash someone else who has a high score. I would also be sad if I could no longer admire the amazing highly detailed avatars that some people are able to create. But for times when you have large gatherings of people, I think this could be very helpful. I went to a "U2 in SL" concert last year that forbid anyone to wear any prim attachments, including hair (gasp!) They managed to pack in over 80 (somewhat ugly) avatars into that sim, and I was amazed by how easily I could move around in there.

OMG that made me remember that concert hehehe I was one of the bald security there hehe

OK. I totally believe those people who say that ARC doesn't take into account the amount of scripts someone is running. I was at this store the other day. It was one of those really big sim stores and there was no one else there. I was happily shopping around when this person TPs in who looked like ruth with no attachments. The whole sim just turned super laggy and started to freeze every few seconds. After several minutes of standing where she TPed in Ruth leaves... everything went back to normal. Now what the heck was that?

An example of just how useless this "tool" is.

Oh...................I'd say that was a "scriptard" (as opposed to a "blingtard".