Jump to content

Full sims - queue?


Orwar
 Share

You are about to reply to a thread that has been inactive for 1949 days.

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

Recommended Posts

5 minutes ago, Fionalein said:

Then stay away, and send your jailer to shop for you, duh!

Or how about you realize that people don't have to do what you want about how they use their items or their time in SL just because you don't really like it. if they want to go shopping with their collar/hud/ao's on it is their choice to do so unless specifically said otherwise by the event manager. It is not the right of anyone else to tell them they can't go unless you are their owner. So if their owner lets them go shopping it is their right and choice to go and shop even with their collar still on, even if you do not like it, nor do they have to hide it if it is not something they want to do or allowed to do by their owner.

for really a sub/slave could have their collar placed in stealth and then you wouldn't even know they had one on. unless you go around perving their avatar and scripts instead of shopping and moving on.

Oh, they might be a switch too and both they are their owner might wear collars, so I guess by your logic neither should be allowed to go there now because it is not what you like or want.

Link to comment
Share on other sites

1 minute ago, Drakonadrgora Darkfold said:

unless you go around perving their avatar and scripts instead of shopping and moving on.

/me raises hand.

I do that all the time. To me, that's way more interesting than shopping. I'm a scripter, so I watch the script times of attachments, and try to notice trends in popular scripted attachments. It's not possible to see data for HUDs, but the "external" attachment points reveal some interesting trends. Germane to this immediate discussion: usually a collar uses about as much time as all the other external attachments combined, including mesh bodies and heads, genitals, clothing, even cuffs. (I don't hang out where weapons are common, so they may or may not be significant.) As discussed above, all these times individually show huge spikes just after a teleport and then fade down to more stable numbers after a minute or so.

Incidentally, from watching my own HUDs (the only ones for which I can collect data), I don't think AOs are much to worry about, even though they seem as if they should be a big deal. Maybe older ones were or something; folks seem to think they've solved world hunger when they migrate to viewer-side AOs.

  • Like 1
Link to comment
Share on other sites

14 minutes ago, Qie Niangao said:

/me raises hand.

I do that all the time. To me, that's way more interesting than shopping. I'm a scripter, so I watch the script times of attachments, and try to notice trends in popular scripted attachments. It's not possible to see data for HUDs, but the "external" attachment points reveal some interesting trends. Germane to this immediate discussion: usually a collar uses about as much time as all the other external attachments combined, including mesh bodies and heads, genitals, clothing, even cuffs. (I don't hang out where weapons are common, so they may or may not be significant.) As discussed above, all these times individually show huge spikes just after a teleport and then fade down to more stable numbers after a minute or so.

Incidentally, from watching my own HUDs (the only ones for which I can collect data), I don't think AOs are much to worry about, even though they seem as if they should be a big deal. Maybe older ones were or something; folks seem to think they've solved world hunger when they migrate to viewer-side AOs.

some older collars are more script intensive yes, but newer collars are not. they might have several scripts in them but not all of them active and listening all the time. where a combat hud set in defense mode might be, or something like emdash running the simscan constantly scanning the region ever so often. so I have to disagree that all collars use as much time as everything else combined that someone else might be wearing or using. in fact it is often one of the addons in the collar and not the actual collar scripts that make it seem like it has a higher script usage; such as a spy or garble or if the leash is active and if titler or label is active and if being currently posed or animated by the collar. so its really dependant upon whats in the collar and if its being used or not.

I have seen mesh body huds/ao's have far greater impact then any collar would, or the slex system like project arousal or some robot controllers that redirect chat to appear to be talking like a robut.

and yes any scripted item cause a huge spike just at the entrance or leaving as the scripts are being initialized and getting the settings and all from the last sim they were just in keeping them in sync. which often lead people to blame the item for the lag, when really its just the nature of how scripts work upon teleport. so in a busy shopping area with a lot of people coming and leaving there will be a lot of spikes making it uncomfortable for some people and frustrating to them.  But as you said, after few moments and no more people come or leave, it settles down. 

Link to comment
Share on other sites

15 minutes ago, Drakonadrgora Darkfold said:

some older collars are more script intensive yes, but newer collars are not.

I checked one that was updated this week with a new release. Same thing.

[EDIT: That's not to say they didn't used to be even worse. And in fact, some older cuffs are very, very bad indeed. But don't underestimate the script lag from collars. Measure; you may be surprised. I was, because I honestly don't know what they should be doing to burn so many cycles.]

Edited by Qie Niangao
  • Like 1
Link to comment
Share on other sites

2 minutes ago, Qie Niangao said:

I checked one that was updated this week with a new release. Same thing.

was it a vd/p9 collar? if so wouldn't surprise me at all. there is a lot happening under the hood of that collar that a lot of people don't know about. unlike a stock OC 7.1 with just the basics and no extra addons and not everything on or in use.

then again anything dealing with rlv and having active rlv will naturally have higher script times than anything else. its the nature of the beast on how rlv works. but there are a lot of things that use rlv and are not collars.

but then agian it is still their right to go there with those things on unless forbid by the event manager. no one has to remove anything just so someone else has the experience they want at the cost of that person have to degrade their experience for the sake of the other person. it has nothing to do with being polite or fair at all.

it would be akin to someone in rl telling another person that they are not allowed to dress or drive a car faster than them because they want or think they should be dressed better or in front of the other driver at all times.

  • Like 1
Link to comment
Share on other sites

13 minutes ago, Drakonadrgora Darkfold said:

was it a vd/p9 collar?

Yep. That seems to be most of what I see in the wild -- which could be an artifact of where I lurk with my pervy script meter.

15 minutes ago, Drakonadrgora Darkfold said:

but then agian it is still their right to go there with those things on unless forbid by the event manager.

No disagreement from me. For one thing, I imagine the share of collar-wearing avatars hasn't much changed in a decade, in contrast to the number of scripted avatars and heads. (Or, in the opposite direction, shoes and hair which at one time commonly had one script in each of many linked prims. Before most folks' times, no doubt.)

I would just add that scripts are typically not the pain point for sim lag. They are acutely felt in shopping event sims because vendors are so dependent on scripts, and also folks are generally more aware of script lag because they try to operate their mesh body and head HUDs (e.g., to adjust alpha cuts) to find that they take ages merely to open, let alone to have an effect on the part they're trying to control.

  • Like 1
Link to comment
Share on other sites

My stance is this, yes you are gonna wait to get in. But we have always had to wait to get in when the sim was full. We went about our business, but to implement a "Waiting list" as it were. I think would fail, one with the scripting, and two with the amount of time each person takes at an event. Some people turn it into an all-day excursion, some people just park their avatars, and sit there AFK and doing stuff for 8 plus hours RL. Instead of actually trying to create a queue, have event creators, use their tools that are put in the viewer. To put a time limit, on how long each person is there. So after that, they get kicked. It would be a better alternative than to add another scripted system.

  • Like 2
  • Sad 1
Link to comment
Share on other sites

18 hours ago, Drakonadrgora Darkfold said:

was it a vd/p9 collar? if so wouldn't surprise me at all. there is a lot happening under the hood of that collar that a lot of people don't know about. unlike a stock OC 7.1 with just the basics and no extra addons and not everything on or in use.

then again anything dealing with rlv and having active rlv will naturally have higher script times than anything else. its the nature of the beast on how rlv works. but there are a lot of things that use rlv and are not collars.

This is false information.

For once Peanut9 is not a fork of "OC 7.1" but they forked our 6.5.5 code and it basically stayed that way, nothing really was changed but headers.

Where we developed further to Peanut 9 which is in no way heavier scripted at all.

RLV is not using higher script times if it is not in use which it normally is not, unless specific functions, mainly inventory related are being used.

Some half-knowledge gets blown up here and things and people get blamed for simply untrue and false facts.

It seems to be popular and common these days though on every possible occasion to hurt the ones who gave all the users for year and years what they use and are happy with.

We had to remove people from our help groups already because they caused confusion and deliberately gave false information.

Also reading collars use so much script time more than other things, well look at this:

WIth Collar:

Script info: 'Garvin Twine': [49/49] running scripts, 2944 KB allowed memory size limit, 0.047939 ms of CPU time consumed.

Without Collar:

Script info: 'Garvin Twine': [29/29] running scripts, 1664 KB allowed memory size limit, 0.034929 ms of CPU time consumed.

 

For me the 20 collar scripts used only 1/4 of the script time, which makes the remaining body/head scripts use 3/4 though they are only 9 scripts more.

Please just do not spread wild guesses without real knowledge and blame it on the wrong people.

Link to comment
Share on other sites

If you put in atime limit forget your paying designers - no one will bother to come as everyone has to compete with top tier sellers, so in order for a customer to come to your newly created store you will have to wait for your customers to come back 2nd or even 3rd time after checking out all the big names - not many will do, so it's no worth to participate anymore for all those 2nd tier designers, sorry.

Edited by Fionalein
Link to comment
Share on other sites

5 hours ago, Garvin Twine said:

For once Peanut9 is not a fork of "OC 7.1" but they forked our 6.5.5 code and it basically stayed that way, nothing really was changed but headers.

 

 

No, oc is not a fork of the 6.5.5 they have updated it and added features to it. The 6.5.5 never belonged to vd at all, it was always still OC until vd was kicked from OC because they tried to turn OC into their own personal cash-cow selling premium no-mod scripts and items. Causing the original owners of OC to have to step in and take the group back, kicking the members of vd out forcing them to start their own group. which the founder of the new group has never gotten over that it was done. You can read the history of OC on the oc website, not the vd run oc website. maybe you should read both and then decide which you choose to believe.

They added in a new dress feature that works better for mesh wearers than the older system that was there. where its easier to strip without possibly stripping off the mesh body as well.

P9/VD is a fork of the 'original' open source opencollar. It has been heavily modified. It has scripts that will delete its scripts if someone tries to make changes in them breaking the collar. It once had a hidden account within the scripts that the collar would contact if certain criteria were met without the user of the collar knowing this. Most of vd scripts are no-mod so not open source or OpenCollar for all OC scripts are full permission. So vd is not the 'original' or 'official' oc collar scripts.

its even been said in the group that they could disable the updater that someone has, its even been said that they can disable the creator kit they sell if they don't like the person. Goto the vd sim with a certain other groups tag on and you will get kicked and banned. Have that group in your groups and visible and you will get banned from the group.  Either you are for vd or against vd in their eyes, so either you use only their items or gtfo.

the collar revisions even have a 'cute' naming system where it can be used to discover where the collar came from(supported/liked distributor or not and if is a vd approved collar or an oc collar that was changed to a vd collar.  such as doe, fox, wolf, snake, frog. and if you have a frog or snake they will frown upon you in the group.

5 hours ago, Garvin Twine said:

RLV is not using higher script times if it is not in use which it normally is not, unless specific functions, mainly inventory related are being used.

 

 

rlv itself doesn't cause higher script times but the scripts that use rlv can because of how they work especially at initialization at login or teleport when the scrip syncronizes its settings with the rlv api to apply whatever restrictions that were the last set or saved. and with the listeners in the scripts, so they can listen for any commands applied to through the rlv relay in the collar especially if it is set and left on auto. There are other functions besides the inventory functions that will cause the script times to rise in a collar. the chat re-direct scripts, the little hearts it creates when near an owner, the leashing and particle effects from the leash, the animation overrides such as making an avatar crawl or walk like a pony through the collar.

all of these can and do add up to added lag in a sim.

5 hours ago, Garvin Twine said:

Some half-knowledge gets blown up here and things and people get blamed for simply untrue and false facts.

It seems to be popular and common these days though on every possible occasion to hurt the ones who gave all the users for year and years what they use and are happy with.

We had to remove people from our help groups already because they caused confusion and deliberately gave false information.

 

not untrue facts at all, its been well discovered by other users about what happens in that group and with the products from them. I personally wouldn't trust vd products for anything rlv related at this point. the only positive thing from vd was the hardvanilla mode they created. their attempts to 'streamline' the collar's rlv and relay basically turned vd collars into 'collars for dummies'. the mod's and even the founder will go ballistic and ban anyone that even suggests or says anything they don't like at all. especially if it is about how someone might choose to use their items, contrary to how the creator/developers think they should only be used as. and no false information was given, just it is claimed to be false to try and save face and deceive other users about the truth about their group or product and the developers of the product. a very common practice by any developer or creator here. "I don't like what you said, so you are wrong. so begone thot." "I don't like your review so will have it removed". which ironically the commentator made a post in another topic about that happening to them and complained it was not fair, yet vd does it all the time to keep only positive reviews seen.

hurt the ones that gave? more like didn't pander to the ego or vanity of the ones that created things and thus thought it made them 'snowflake's' or above anyone about what they think or say or feel. that you have to praise them for their effort, kneel before their superiority complex. vd and any creator here should be grateful that any users ever chooses to use their items free or paid for and leave a review or not, positive or negative. and above all don't question them about what they say ever. if you have any suggestions about using the collar especially their pride and joy 'heartcore' in a manner they didn't like or want other users to do, they got very defensive or even offensive about it and would either threaten to kick or would kick and ban from the group and sim and their discord to keep you shut up so they can continue to enforce their view of how things had to be done using their product, which is not how it normally works in world. a user buys a product they are free to use it how they wish and do not have to follow the rules of someone else about what it's for or not. but not with vd. a car seller would not try to tell a buyer how to drive the car after it was bought, yet vd does about its products.

I could go on about other things that vd has done in the past but it would probably be pointless and would all be claimed to be false and untrue or said just to 'hurt' the ones that gave.

5 hours ago, Garvin Twine said:

Script info: 'Garvin Twine': [49/49] running scripts, 2944 KB allowed memory size limit, 0.047939 ms of CPU time consumed.

Without Collar:

Script info: 'Garvin Twine': [29/29] running scripts, 1664 KB allowed memory size limit, 0.034929 ms of CPU time consumed.

For me the 20 collar scripts used only 1/4 of the script time, which makes the remaining body/head scripts use 3/4 though they are only 9 scripts more.

1

according to the number you posted there are 29 scripts still running, not just 9. So either you have some huds on or ao's on that have been already said can and will cause higher script times. so it is not just the body/head scripts causing that higher number like you tried to claim. its a combination of items. which has been noted before that excess scripted items will add to the to lag. remove everything but the collar then post the number. then remove all the huds/ao's/slex systems and just leave the body then post the number.

that 1/4 difference can add up if there are lots of people in a sim with collars on.

the point that was made is that collars and the scripts in them can and will add to lag of a sim if enough people have them on and are using them at the time with whatever functions they have active in the collar or within the addons of the collar.

But none of this is the point of what this thread was really about, it was about if a wait queue would be good for when teleporting to full sims. I think it could be beneficial for it has been used in other servers and worked well within them. Some people don't like it because they wouldn't be able to tp hammer or lose their spot in the queue.

Link to comment
Share on other sites

6 hours ago, Fionalein said:

If you put in atime limit forget your paying designers - no one will bother to come as everyone has to compete with top tier sellers, so in order for a customer to come to your newly created store you will have to wait for your customers to come back 2nd or even 3rd time after checking out all the big names - not many will do, so it's no worth to participate anymore for all those 2nd tier designers, sorry.

But if you put a "waiting list" you would lose customers too. Why don't we just leave things as they were then, that is the best way of doing things. We have tools in the viewers, we could use. However, no one uses it. Using a new "waiting list" system. Would just introduce more scripts, constantly listening, which would add to the lag. So as I said before in this statement, leave it the way it was. 

Link to comment
Share on other sites

I am saddened to see this interesting thread hijacked by the "hate Fireflower" faction.  Apart from anything else this individual appears not to be able to do simple maths.  I give the opinions expressed no credibility whatsoever.

Link to comment
Share on other sites

32 minutes ago, Drakonadrgora Darkfold said:

But none of this is the point of what this thread was really about

That's for sure. I don't think the forums deserve to become a battleground in the collar wars -- at least not this thread, which has enough points of contention of its own.

Perhaps a transition back to the topic. I sent an alt to one of the spots where some collar and updates are distributed to further test my assessment of collars typically burning half the script time of visible attachments for those who wear them. Turns out it's all over the map -- and that variance is really the story I want to get to -- but just to get an idea of the dispersion, here's a table of some scripted-collar-wearing avatars' script times for the collar itself, the sum of all script times of visible attachments[1], and the corresponding percentage:

collar_μs

total_μs

share name____________________________________________________________________________
64 70 91%  Dancer's Collar by Wendy Starfall
8 40 20%  [Forge] Noir Collar (Rust)
55 120 46% Hartgummi Collar by Tapi Karu
8 116  7%  .random.Matter. - Kitten Collar - Plain - Silver/Black
19 47 40% <avatar name>'s Collar
16 52 31%  Strict Collar by Grey Mars
57 161 35%  S&P Collar Freya metal (OC 7)
171 444 39%  [Tethered] Collar - My Heart is Caged-Wyvern Dragon OC6 (copy)
107 108 99%  OpenCollar v7.1 FE
7 23 33%  OpenCollar Six

[1] Note that this is the total time of all visible attachments, not including HUDs. I realized as I was collecting these data that I could also get the total times for all of each avatar's attachments, despite not being able to see the individual times for HUD attachments. Obviously those larger totals would lower the percentages some, and there may be even more variability in those HUDs.

So, the variability. Most striking isn't the spread of percentages, per se, it's the wide separation in script times for these collars. Now I'm no expert in collars -- I've barely looked at the source code, and that was years ago -- so maybe some are busier than other for good reason, but I bet a very large share of that variability is due to the amount of time the wearers (and their scripts) have been in the region, and thus how long ago those scripts started loading into the sim. We discussed that point earlier in the thread. I'd reiterate that those initial times are both messy (may not be completely accurate) and truthfully elevated (as we've all watched sims performance dive when highly scripted avatars enter the region).

One thing I think is happening is that when scripted attachments arrive, they consume a lot of time per script, even if the scripts aren't doing much. I mean, they may have on_rez, CHANGED_TELEPORT, or CHANGED_REGION event handlers to run, but I suspect most of the time is spent simply loading the scripts into memory, registering their event handlers, and slotting them into the scheduler.

Collars have a lot of scripts. I suspect they have more scripts than are absolutely necessary because distributing the code into separate scripts can sometimes reduce the total processing. It can take extra logic, for example, to multiplex several streams of unrelated asynchronous events in a single timer queue, and non-timer handlers may need repeated tests to determine which is the relevant stage of processing to perform. So, steady-state, more scripts might sometimes be more efficient -- and yet be significantly more work to start when loading into a sim. It's an interesting architectural problem whether to optimize for steady state (where most of the time is spent) or start-up (where the worst pain is inflicted).

(I know, I know: It was a long way 'round the barn to get to that point.)

  • Like 1
Link to comment
Share on other sites

1 hour ago, Aishagain said:

I am saddened to see this interesting thread hijacked by the "hate Fireflower" faction.  Apart from anything else this individual appears not to be able to do simple maths.  I give the opinions expressed no credibility whatsoever.

heh, one of the fireflower defenders come to defend wendy's honor no doubt. which makes your opinions expressed of no credibility as well. everyone knows how narcissistic wendy is and how egotistical otto can be. Its proven daily in the group by how they respond to anyone that doesn't respond to them how they think they should and how quick they are to ban anyone that dares question them. very few people are really being fooled by these behaviors for long. wendy is even notorious for posting as her alt to pat herself on the back at times to feed her ego as was recently done.

I can do simple math, he tried to claim that his collar was not adding to the lag in any significant manner and that is false. he tried to make it sound like the 1/4 difference was a major difference when it is looked upon in the larger context of total users it makes no difference. the collar still adds lag to the region. lowering the experience of others.

57 minutes ago, Qie Niangao said:

That's for sure. I don't think the forums deserve to become a battleground in the collar wars -- at least not this thread, which has enough points of contention of its own.

1

which was never my intention by my reply, but as simply to clear the air about some of the things that were said by the commentator; those within that group like to fabricate their own batch of lies and try and get people to believe them and that they are the victims when they are not and are often the one victimizing others and think they can get away with it without anyone ever saying anything about it.

for yes this thread has its own problems between those who want such a feature and those who do not. if implemented correctly it would work very well. 

oh, by the way, you were at Arcadia, for that's about the only place you will ever find wendy when she is on. usually sitting at her desk. Wendy usually doesn't have much running in her collar. so is why her numbers were a little lower. for really it depends upon what scripts each user has loaded in the collar and if they are active. and yes some collars have several unneeded scripts for just simple usage, but then again some people like having lots of options and choices aka addons about what their collar can do. and this has varied effects on script time and lag. so like how the above commentator Gavin showed his collar being lower could simply be he has very few addons loaded and running compared to someone else who might be running a full blown out collar with all the premium apps and having them running.

and yes it is related to how long they have been in the region and the last time the scripts were polled for actions needed to be done or if the collar was recently interacted with by the wearer or the owner set on the collar. which is why some regions lag really bad when several people enter and then calm down a few minutes later. but at shopping events where people might be coming in often, it means the server never really has the chance to settle back down again for long. causing a laggy and sometimes unbearable experience for some people.

script usage is also affected by if the user is afk or not and if the viewer is foreground or background. which is why you cannot really ever trust any script monitoring system to be accurate about what is really going on at that time. you don't know who is there how long they might have been there compared to when you arrived, and who is afk, or who might be tabbed out thus using fewer resources.

I would think it would be better to optimize scripts for steady state, for unless a person is constantly teleporting around, not something most sub/slaves would probably be doing if their owner is online. over startup, to just try and get them up and running sooner. yes, some will disagree and think quicker startup is better since it begins to lower the effects sooner, but won't have as wide an effect at lowering compared to a steady state optimized script would.

Either way, I still think that having a queue would work for no one would really ever encounter it often unless the sim was full. Which not a lot of sims stay full all the time. So it would have a negligible effect since it wouldn't run or be active unless the sim was full or near reaching full capacity.

Or the other way would be to update viewers with a timed auto repeat teleport where someone didn't have to try and tp hammer to get in. similar to a phones auto-redial function. ie if teleport fails to try again in x minutes for y number of times. and then let each user be able to adjust those settings for how often they wanted it to retry and how many times before giving up and if succeeding maybe playing an alert sound to let them know encase they were away while it was trying.

Link to comment
Share on other sites

I won't repeat false accusations and facts here but I think it is clear who tries to use any opportunity here to accuse and shame the ones that brought everyone what collars are these days in SL.

Just to make clear here, that Drakonadrgora Darkfold was one we had to remove and now he is using other forums and places to continue his agenda to harm us (no clue why). But you see how he knows everything and gives advice to scripters and viewer devs just never did anything himself.
BTW, the new strip feature (which filters already certain folders by default and can be setup to filter out things you do not want to be stripped at all) is called Strippy and was developed by us where they only added a detach feature which simply shows all attachments, including your head, body, eyes etc . Please just stop to blame because we do not like you stating false info to our community.

To add something to this thread:

Of course Collars are relative "heavy scripted" with 20+ scripts, but these do most of the time nothing.

RLV is pure viewer related (Restrained Life Viewer) and only very view functionality of RLV goes on sim resources.

One of these is used by a tool that switches group tags, which is not a function collars make use of, specially not in a short timed fashion.

Also do many popular AOs use still overly scripted 10 years old scripts with rapid timers that go on script time as well.

Please do not forget these different combat huds for vampires and demons roleplay who use the whole grid as they play ground which are always active scanning for possible opponents and unlike collars, which mean something for the people and are therefore locked in place.

The high impact an avatar has when entering a sim is for sure the biggest issue but that can only be asked to LL to reduce this problem or at least give general guidelines or better detailed guidelines for scripters how to keep this at a minimum. (I develop in a high traffic sim, so I know very well what lag we talk about here, but so we also test right away in such environments and know what impact we do and always try to be as low as possible) 

Just to mention it was again us who introduced the "Highlander" rule which allows only one collar instance to be worn, before that you could meet avatars having multiple items attached where each of them had a fully scripted "collar" inside. (and we get more critic and hate for that than thanks)

  • Haha 1
Link to comment
Share on other sites

11 minutes ago, Garvin Twine said:

Of course Collars are relative "heavy scripted" with 20+ scripts, but these do most of the time nothing.

Yes, but I'm now thinking those would benefit from being refactored to reduce that count and to completely turn off scripts -- llSetScriptState(.., FALSE) -- for features that the collar isn't actively using. (That's a little tricky -- hence "refactored" -- because non-running scripts do not preserve state across sim handoffs, which is good, but those scripts would need to be written as stateless services.) It's just optimizing for a slightly different performance characteristic than most attachments have been, historically.

17 minutes ago, Garvin Twine said:

... only very view functionality of RLV goes on sim resources.

One of these is used by a tool that switches group tags, which is not a function collars make use of, specially not in a short timed fashion.

I understand that function (or rather, the changing of active group role title) is going away completely from RLVa. I asked about that in these forums a while back, prompting Kitty to remove it, and I gather the Lab to throttle it. The way an avatar's active group has such far-reaching effects on the sim is... surprising.

21 minutes ago, Garvin Twine said:

The high impact an avatar has when entering a sim is for sure the biggest issue but that can only be asked to LL to reduce this problem or at least give general guidelines or better detailed guidelines for scripters how to keep this at a minimum.

Indeed. It seems to me that comfortable avatar capacity for a region could be substantially increased if scripters really knew the trade-offs in crafting low-impact scripts. We have a lot more profiling tools than we had when I wrote my first script, but they don't really help at the system engineering level. Of course, there's also a problem of scripter -- and consumer -- education.

Sometimes I'm amazed that SL works at all. When you think about it, the Lab's "system engineering" role regarding user-generated content is so meta, providing functions that may let us shoot ourselves in the foot, but being careful that the wounds aren't fatal to the platform.

Link to comment
Share on other sites

9 hours ago, Solar Legion said:

Take the bloody collar system drama elsewhere - seriously, it has no place here.

I totally agree.

But if people start to bash on what I do, I have the right to correct wrong statements, I did not bring it here at all.

Anyway for once collars are only a little part of scripted items  ( I see regular avatars with way more and higher impact scripted stuff than collars (example just taken now: 

Script info: 'xyz Resident': [329/364] running scripts, 15712 KB allowed memory size limit, 0.402572 ms of CPU time consumed.

(taken after the avatar was settled in the sim, in the first minutes the CPU time was around 4ms!)

I seriously doubt any other product was so hard maintained in the last years (specially when we started with version 4) and focused on less impact than these collars you like to blame here. (Main Brand AOs still use ZHAO 2 scripts from 2008, sometimes multiple instances!)

Yes Collars are scripted items and add to the lag, as everything you add to your avatar but they are for sure not the root of lag issues and then blaming our brand to be the worst is just plain wrong.

 

9 hours ago, Solar Legion said:

As for the automated group swapping ... Lovely, an actually useful tool is being removed thanks to a bunch of nits ...

This is exactly the attitude of many if not most SL users.

I do agree it looks cool to see the group tag switching every few seconds, and so they love this and do not care about the impact it has.

Same applies to many other things, users do not care about script or any other impact things they use have as long as things go the way they like them for themselves.

This leads to products that just bash it out (because SL has so many flaws and to overtone them, it is easiest to just slam with a sledge hammer rapidly on it so things look smoothly) and in the end the lag gets worse but your group tag switches soo lovely. 

I hope I could add some info and view point on the issue. 

Edited by Garvin Twine
Link to comment
Share on other sites

6 hours ago, Solar Legion said:

As for the automated group swapping ... Lovely, an actually useful tool is being removed thanks to a bunch of nits ...

Although I tried to make it clear in my post, I may have created confusion by responding to Garvin's comment about changes to active group with information about changes to group active role. I believe the former will remain. As far as I can tell, the use-case for the latter is to spam all viewers of the scene with scrolling unicode and produce untold lag effects on sim and viewers alike. If scrolling unicode is the application, best stick to titlers (although scripters know that llSetText() isn't nearly as innocent as it looks, either).

  • Thanks 2
Link to comment
Share on other sites

1 hour ago, Qie Niangao said:

Although I tried to make it clear in my post, I may have created confusion by responding to Garvin's comment about changes to active group with information about changes to group active role. I believe the former will remain. As far as I can tell, the use-case for the latter is to spam all viewers of the scene with scrolling unicode and produce untold lag effects on sim and viewers alike. If scrolling unicode is the application, best stick to titlers (although scripters know that llSetText() isn't nearly as innocent as it looks, either).

Actually the rlv command does both, and there is only this one command that does it. (Either it never was there or it has been removed from the sl wiki where all other rlv commands are documented).

It sets the current active group and the group role in one command and was meant as one time thing, to set this and lock it with another command.

This can be also troublesome in other ways, I talked to someone who abandoned by accident their own group when trying to leave another one, because the timed script did set the active one back to the one set there and so they had to file a support ticket to get back their own group again.

 

 

 

Link to comment
Share on other sites

11 minutes ago, Garvin Twine said:

Actually the rlv command does both, and there is only this one command that does it.

Yes, but I believe the problem was the ";<role>" part of the "@setgroup:<UUID|name>;<role>=forcecommand, which I thought was a recent RLV/a extension. As far as I know (but I'm no expert) the old "@setgroup:<UUID|name>=force" / "@setgroup=<y/n>" syntax will continue to work -- although I seem incapable of finding current RLVa documentation.

Link to comment
Share on other sites

49 minutes ago, Qie Niangao said:

Yes, but I believe the problem was the ";<role>" part of the "@setgroup:<UUID|name>;<role>=forcecommand, which I thought was a recent RLV/a extension. As far as I know (but I'm no expert) the old "@setgroup:<UUID|name>=force" / "@setgroup=<y/n>" syntax will continue to work -- although I seem incapable of finding current RLVa documentation.

Correct.

I should have said, the command can do both, where the role part is optional.

In the documentation (sl wiki rlv api) it is (at least right now) not even mentioned to have the option for the role.

Link to comment
Share on other sites

You are about to reply to a thread that has been inactive for 1949 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...