Phil Deakins

When setting a limit that's over 64k, I believe it quietly sets it to 64k. If I'd written that bit of the system, I'd have sent the owner "cheeky buggar" message and set it to only 4k, and laugh while I imagine them trying to figure out why they can't get rid of the stack-heap collisions

I just wanted to say that I laughed at the reply to the 2 spam posts that were posted in this thread (and deleted a minute ago, along with the reply). I clicked to reply to the reply but the posts were gone. So to whoever it was who posted the reply... Too true! lol

PeterCanessa Oh wrote: Yep, that's it. I knew you knew that really. Except. My llSetMemoryLimit(4096) Mono script will report as a 4k script, behave as a 4k script and be a 4k script only as long as it isn't at a party. As soon as it is it just says llSetMemoryLimit(65536), switches from llWhisper() to llShout() and goes wild. Shouldn't that be 65535? There are 65536 bytes in 64k but the numbering is 0-based

lol. No I didn't know. I did know that a new thing that LL introduced a while back only showed the sum of maximums, and I thought that was rather useless. It was obvious that what Phoenix shows is also the sum of maximums. But when Qie said in a recent thread that customers can now see how much memory an object uses before buying, I thought it was something different that would actually show the correct value, which is why I started this thread. I wanted to know, and now I do - there is no way for a customer to know how much memory an object's scripts use before buying. Anyway, just for the sake of it, rather than for any practical value, I now have the memory limit of each of the 4 mono scripts in the object I'm working on set to a little above what they are actually using after initialising, and the object now shoes 75k instead of the sum of maximums (256k) that it showed previously.

I was hoping that Qie was referring to another way than just the sum of max possible amounts. Unfortunately, most script memory useage is misleading but I'll set the memory required, if only to check that Phoenix reports it correctly, as it makes no difference to a script's actual useage. Thank you for your help, Jenni. It's appreciated.

Will do.

Ok. I think I have my answer (unless Qie was thinking of another method). The script memory information that Phoenix and Jenni's HUD return are the maximum that *might* be used by an object - the maximum that an object is able to use whether it gets close to using it or not. Pity.

The problem with that is that you're saying there is no way of getting the more accurate script memory useage of an object unless the scripter has set the memory limit, and even then it would be set higher than is actually used. So you're saying that HUDs can't do it, but not in so many words. And yet Qie said it's possible, so my question about that type of HUD is still open. I see in my emails that Jenni has sent me her HUD so I'll be able to see for myself The reason I'm asking is because Phoenix reports one of my objects (one I'm currently working on) as using 192k of memory (3 mono scripts) and if people actually believe that, they will have been grossly misled because it actually uses less than a third of that. If the HUDs can't do any better, then it would be bad for people to think they can tell how much script memory an object uses before buying it, especially if they judge an object's impact by it.

Thank you, Jenni. I'd appreciate a resend In the other thread, I'd already not accepted the HUD before you posted that you'd sent it, and I was too embarrassed to say so at the time. It was Qie who posted in that thread that we can see how much script memory an object uses before we buy it these days. But, if we can't get any more accurate than the sum of all maximums, we can't see how much memory an object uses at all. That's what I'm trying to find out - waht exactly these HUDs show.

I'm not winding youup at all, Peter. If you read my first post again - slowly - you'll see what I'm wanting to know.

Amethyst is right. Complainers are not the most important people in the world - I am. And LL caters to me. On this occasion they're doing the work at a time when I need to out, and should it take longer than expected, I told them that today is good because I'll need to be out again not long after I return.

I know about llGet Used memory() Peter. In fact I use it. I'm asking about a HUD that was mentioned in a recent thread and I want to know if it can report the correct memory useage on any scripted object rather that the simple (but incorrect) sum of max sizes.that is reported by Phoenix. I specifically want to know what the HUD reports.

It's not the script time that I'm after, Peter. It's the amount of memory the scripts in an object use. Incidentally, I'm reliably informed (by Qie) that setting a script's memory limit is irrelevant, and only any good for reporting. A Mono script can use up to 64k but it only uses the amount of memory that it actually needs, so setting the memory limit makes no difference, except for reporting.

When a parcel is set to disallow the running of scripts, scripts in HUDs don't work on it. If you have a radar, you'll probably notice that it gives you no information on some parcels.

In a recent thread, somone posted that it's possible know how much memory an object uses for its scripts before buying it. It was stated that a HUD is needed and someone sent one to me. I don't accept objects from people I don't know so I'd deleted it before reading in the thread that she'd sent it. But I have a question about these HUDs... With Phoenix, and presumably with their other viewer, you can right click on an object, click More>, More> and then S. Count, and it will tell you how many scripts are in the objects, the size of the scripts, and the time of them. What I'd like to know is whether or not those HUDs provide the same size information. The reason I ask is because the number that Phoenix shows is wrong. All it does is add up the maximum filesizes of the scripts. So 16k is added for a tiny script that actually uses, say, 1k of memory, if it's compiled with LSO. Similarly, if it's compiled with Mono, 64k will be added, even though it only uses 1k of memory. Consequently, the total reported is no reflection at all on the actual total memory the scripts use and is, therefore, a waste of time showing it. If the HUDs show an accurate number, and don't do what Phoenix does, then I'd like to get hold of one.

Pussycat Catnap wrote: So um... how did they manage to create that popup which looks a bit different from the sort one can usually send? Some kind of script able to do that, or is this a server hack, or are the victims using a malware-TPV that pretends to be the official viewer? Out of necessity, I'm just getting a bit used to the V3 viewer, and that popup message doesn't look to be out of the ordinary. In fact, I think is was discussed in the other thread.

I've only ever had good support from the support staff, and I've said that a number of times in this forum - including again right now. Where I've had bad customer service is with the AR team. Once in a rare while, my ARs are quickly dealt with, like recently when objects places on abandoned land were spewing particles over the surrounding land, including mine. Apart from such rarities, I've only had good customer service from the AR team when I've got the AR team's manager involved, which happened only twice - once with an encroaching prim, and once when an AR team person, plus the AR team person who handled my appeal (might have been the same person), failed completely to use their brains; in other words, they didn't bother checking anything, but issued a warning instead. The AR team people are really bad for SL but the support staff are generally good, given the limitations of the outsourced staff. When the support people were in-house, I always found them to be excellent.

Sorry - my mistake. It's not opensim, it's OpenMetaverse - previously LibSL.

Marigold Devin wrote: The *movie star* bots are showing as being on Viewer 2, because most avatars are showing as being on Viewer 2. Phoenix often shows bots as being on V2 - or did until LL intentionally broke Phoenix's ability to show the client. It showed V2 as a default when it couldn't determine the client. I don't think the movie star bots will be on text-only clients. My bots run on my opensim-based bot programme, and Phoenix often showed avatars on opensim-based programmes as using V2.

1. Rude is rude, and nothing to do with the things you said. 2. I was not rude to anyone in this thread. 3. I can remember just one person posting an agreement that I was rude. I can remember several posting that I wasn't rude. You are only the second to post that I was, so who are your imagined "many"? 4. I don't consider it a rudeness to point out that the way someone writes makes it difficult for others to read. She may have been blissfully ignorant of the fact but after the comments in this thread, and the comments in the seperate thread that it spawned (not by me), she knows very well what plenty of people think about such a way of writing posts. If she didn't know before I mentioned it, she certainly does now. It's her choice to continue doing it or not, but at least she knows what plenty of people think of it.

They didn't try at all. They posted the current, updated, status each time but not the reason for it all happening. They only did that in the later status post that was copied above. What I would have hoped they'd do is post something like what they later posted (see above) within the status post at the time - just like adding parts to it which they did. But it's LL, of course, and LL is very well known for not considering their customers at all.

Maybe someone thought it wasn't really necessary, which it isn't if we only consider necessities and not what's usefulness, but it was very useful. I suppose we should be thankful that we can still open a new inventory window and do it that way, but it's not as convenient or as quick as the real thing was. I wouldn't be using LL's V3 at all if it wasn't for the fact that I can no longer open scripts and notecards with Phoenix - not even my own creations. I'm writing a system (scripts and notecards) so I need to be able to open them. When I'm not doing that, I still use Phoenix. If they'd make the camera controls a decent size, so that they use very little of the screen (like the V1) I could stay with the V3 because it's quite good. There's no reason to waste so much space in the cam controls floater.

With the V1 it was possible, and extremely useful, to drag an item from the Recent Items inventory listing to the actual All Items tab itself, when the All Items inventory would open. It made it much easier to drag new items into the correct folders than dragging them from a folder in All Items to another folder in All Items. I'm using the current LL V3, and dragging from Recent to All isn't possible. I didn't use the horrible V2 so I don't know if it went missing in that viewer. Is this an oversight or intentional? Either way, it's a change for the worse.

It's good that they told us what went wrong on Thursday. It's a shame (read 'shameful') that they chose not to tell us, their customers, while it was happening.

Ann Otoole wrote: Maybe they were just doing some cleanup and restared the services resulting in a short outage. Seems to be there for me. More worrisome is: Domain name: secondlife.com ...Whois Server: whois.ascio.comCreation Date: 29-APR-2002Updated Date: 07-APR-2011Expiration Date: 29-APR-2012 And the 29th is on a Sunday (tomorrow). Wouldn't it be something if someone forgot to renew it. I'll be tickled pink if it happens :smileyvery-happy: