Are temp rezzed structures allowed?

Phil it sounds to me that you are talking about client side lag. I think temp rezzers are a problem because they lag the server. I think server lag would be the same no matter where you are in the sim. I've seen it happen with my own eyes.

Sindy and Eli. You are both right. When I think of lag, I think of the personal experience of it rather than the continual load that the server is coping with, and that effect is usually much less up in the sky, where there are few objects/textures and no other avs around, than on a 'busy' ground.

You don't seem to have got what I'm saying. I'm not that what you describe can't cause problems for a sim - I'm saying that it doesn't necessarily cause problems for a sim. It depends entirely on the circumstances in the sim. So it cannot correctly be said that "using temp rezzers to create a couple hundred prims on a sim once a minute, 24/7, is something that actually causes sims to lag". What can correctly be said is, "using temp rezzers to create a couple hundred prims on a sim once a minute, 24/7, is something that can causes sims to lag". That's all I'm saying. Anyone who says anything different is not seeing it correctly. You can test it. Go to the beta grid with a temp rezzer that will rezz a couple of hundred prims every minute or so, find a completely empty sim (no scripts, no other avs, no objects) and see if the temp rezzer causes the sim to lag. It won't. It will require more than that one temp rezzer to cause the sim to lag.

It's not a case of people not liking temp rezzers "they actually do cause problems". It's a case people not liking temp rezzers because they *can* cause lag, but in many cases, people don't like them just because they are used to get round prim limits and nothing to do with lag. It's that that some people are against.

Just for fun, I made a "Temp Rezzer" that rezzes the famous 27 prim "Low Prim Beach House" every 60 seconds and used my own "Performance Meter Wand" to sample Simulator Time Dilation every 0.5 seconds with a cutoff threshold of 0.97. The Low Prim Beach House was set to delete itself 0.5 seconds after it was rezzed.

Here's the result:

[11:11] Object: Rezzing Low Prim Beach House
[11:11] Performance Meter Wand: 0.846344
[11:12] Object: Rezzing Low Prim Beach House
[11:12] Performance Meter Wand: 0.851851
[11:13] Object: Rezzing Low Prim Beach House
[11:13] Performance Meter Wand: 0.857352
[11:14] Object: Rezzing Low Prim Beach House
[11:14] Performance Meter Wand: 0.849294
[11:15] Object: Rezzing Low Prim Beach House
[11:15] Performance Meter Wand: 0.844192
[11:16] Object: Rezzing Low Prim Beach House
[11:16] Performance Meter Wand: 0.849259
[11:17] Object: Rezzing Low Prim Beach House
[11:17] Performance Meter Wand: 0.862778
[11:18] Object: Rezzing Low Prim Beach House
[11:18] Performance Meter Wand: 0.848820

As you can see, every minute and as soon as the Low Prim Beach House was rezzed, Time Dilation CONSISTENTLY dropped by about 0.15. That is LAG.

I'd call that a negative spike - not lag. To my way of thinking, lag is a continuous thing.

I would call it a squarewave lag incident, sampling even faster @ 0.1 seconds and everything else the same in my experiment
:
[11:35] Basic Temp Rezzer: Rezzing Low Prim Beach House
[11:35] Performance Meter Wand: 0.835280
[11:35] Performance Meter Wand: 0.839218
[11:35] Performance Meter Wand: 0.837531
[11:35] Performance Meter Wand: 0.832526
[11:36] Basic Temp Rezzer: Rezzing Low Prim Beach House
[11:36] Performance Meter Wand: 0.839090
[11:36] Performance Meter Wand: 0.832276
[11:36] Performance Meter Wand: 0.832784
[11:36] Performance Meter Wand: 0.834462

Shows a 1/2 second drop in Time Dilation which is enough for an AV walking along to experience the rubberband effect.

(Edit: after letting it run a while, up to 0.8 seconds of drop occurred)

Similar results found on the Beta Grid on Scheduler Test sim with no other AV's objects or scripts.

It's a negative pulse train, and the pulses occur once every minute and last for half a second or so. That's 1/120th each minute. I don't call that sim lag any more than negative spike/pulses once in a while, caused by any number of things, is sim lag. If the negative values lasted, then it would be sim lag, but they are very short-lived.

I have the stats bar permanently on-screen, and I see such pulses quite frequently, but the sim doesn't lag. If it did, I wouldn't be able to walk around smoothly in it. To me, lag is a continuous thing - not a train of very small, widely spaced pulses.

I believe that is what is known as "server lag," which is different from "client lag" and "network lag." Each is considered "lag" in SL terminology.

(At least that's what the knowledgebase and SL wiki tell me)