Content Creation/Scripting User Group/Transcripts/2011 08 15

From Second Life Wiki
Jump to navigation Jump to search

List of Speakers

Cale Flanagan
Fancy Greeter
flexi campfire
Helena Lycia
Leonel Iceghost
Liisa Runo
WolfGang Senizen


[09:02] Koli Contepomi (sahkolihaa.contepomi): Hey kelly.

[09:02] Liisa Runo: hi

[09:02] Helena Lycia: Hiya Kelly

[09:02] Cale Flanagan: hi

[09:02] Kaluura (kaluura.boa): Hello...

[09:02] tehKellz (kelly.linden): Good morning

[09:03] F L I P (flip.idlemind): Oooh a chart

[09:03] Cale Flanagan: may i ask, i was searching for transcripts in the wiki.... and the archive lead me to mesh archive...

[09:03] tehKellz (kelly.linden): How was everyone's weekend?

[09:03] tehKellz (kelly.linden): I have a large backlog of transcripts I have been meaning to upload.

[09:03] tehKellz (kelly.linden): Sorry.

[09:03] Cale Flanagan: so there is hope, good:)

[09:04] Cale Flanagan: was only irritated to find me in mesh:)

[09:04] Helena Lycia: Spent most of it trying to track down a bug with llRegionSayTo - not sure if it is a real bug yet, still need to repro it

[09:04] tehKellz (kelly.linden): Helena point me or Simon to the jira once you have a repro

[09:05] Helena Lycia: Will do

[09:05] Helena Lycia: Was hoping to have done it in time for here but got distracted by Google+

[09:05] tehKellz (kelly.linden): Any go to SLCC, physically or virtually?

[09:05] Helena Lycia: I saw the web feed of Rod's speach

[09:06] Koli Contepomi (sahkolihaa.contepomi): I don't actually remember what I watched from the live stream yesterday.

[09:06] tehKellz (kelly.linden): "memorable" I guess.

[09:07] tehKellz (kelly.linden): Anyway! Welcome to the content creators user group, scripting meeting.

[09:07] tehKellz (kelly.linden): News up first.

[09:08] tehKellz (kelly.linden): The chart behind me shows the sim performance improvement grid wide for "DRTSIM-75" which was the simulator / scripting performance fixes that went out a couple of weeks ago.

[09:08] tehKellz (kelly.linden): It is always nice to be able to see a visible improvement.

[09:08] RaithSphere Enyo Darkstone (raithsphere.enyo): thtas site a difference

[09:08] RaithSphere Enyo Darkstone (raithsphere.enyo): *quite

[09:08] Helena Lycia: I'm gettting transparency issues from the prim behind it

[09:09] Helena Lycia: Thanks

[09:09] F L I P (flip.idlemind): Umm whats with the giant dip

[09:09] tehKellz (kelly.linden): neither is transparent. hm

[09:09] RaithSphere Enyo Darkstone (raithsphere.enyo): PNG Transparancy

[09:09] tehKellz (kelly.linden): release day is just wonky

[09:09] Liisa Runo: the texture has alpha channel, use TGA next time

[09:10] tehKellz (kelly.linden): ah. my screenshot tool does png, I'll have to convert it

[09:10] Cale Flanagan: png works fine for me..- normally

[09:10] tehKellz (kelly.linden): I dunno why a screenshot tool would add transparency to a desktop screenshot

[09:10] tehKellz (kelly.linden): anyway

[09:11] Cale Flanagan: aero?:)

[09:11] tehKellz (kelly.linden): Moving on

[09:12] tehKellz (kelly.linden): On the BlueSteel channel this week is the Scripting Maintenance which has a bunch of scripting features and another set of performance improvements.

[09:12] Liisa Runo: ?

[09:12] tehKellz (kelly.linden): I don't see Moon here, so I hope she doesn't mind but I'm gonna quote the jira comment she made this morning

[09:12] tehKellz (kelly.linden): This server version can handle double the amount of scripts, compared to pre-mono2 versions.

A very impressive result, this is a major breakthrough.

[09:13] Qie (qie.niangao): wow

[09:13] Liisa Runo: neat

[09:13] Kaluura (kaluura.boa): Yeah... Interesting...

[09:13] Helena Lycia: Yikes. Does that mean I'll have to write twice as many scripts now?

[09:13] Liisa Runo: also the new LSL stuff is awesome. Even my most negative friend was stunned and jumped in joy

[09:13] tehKellz (kelly.linden): That is a 3-4x improvement over current versions, if she stated that right, since the current server seems to be doing around half what pre-mono2 did

[09:14] tehKellz (kelly.linden): Her comment is here:

[09:14] flexi campfire: [#SCR-120] Idle scripts still take server time

[09:14] tehKellz (kelly.linden): 11k idle scripts / 19k sleeping scripts.

[09:15] tehKellz (kelly.linden): Honestly though her tests found results much higher than mine did. I only saw ~8-9k idle

[09:16] Sammy (ayamo.nozaki): P:

[09:16] tehKellz (kelly.linden): Which is still a few thousand more than reported for pre-mono2 (5-6k idle)

[09:17] Liisa Runo: so, will this go grid wide this week?

[09:17] F L I P (flip.idlemind): If not I'll cry

[09:17] tehKellz (kelly.linden): It is a good possibility. There are some high priority fixes that may get first shot, but as of right now the script maintenance is looking good for it.

[09:18] Koli Contepomi (sahkolihaa.contepomi): Woo.

[09:18] tehKellz (kelly.linden): And we wouldn't want to make flip cry

[09:18] F L I P (flip.idlemind): Especially since tomorrow's my birthday. Scripting maintenance would be a great gift

[09:18] F L I P (flip.idlemind): Gyahahhah ooooh

[09:18] tehKellz (kelly.linden): heh

[09:19] tehKellz (kelly.linden): I think that is all I have as far as news goes.'

[09:19] Qie (qie.niangao): Wondering... Mesh test sims will pick up the scripting maintenance stuff, too? if it goes to general release, I mean.

[09:20] tehKellz (kelly.linden): Yes, if the script maintenance gets promoted it should be on all regions by the end of the week.

[09:20] Qie (qie.niangao): cool

[09:20] Kaluura (kaluura.boa): Yay!

[09:20] Kallista Arliss (kallista.destiny): wohoo

[09:21] tehKellz (kelly.linden): And for anyone not up to date, the release notes for the script maintenance are here:

[09:21] Kallista Arliss (kallista.destiny): the performance degradation drop is very goo

[09:23] Brins0 (brins0.blackheart): so all the things on that page for the 11.08.10 are going to be coming to the main grid tomorrow?

[09:23] Qie (qie.niangao): Yeah, this release is just full of crunchy goodness!

[09:23] Kallista Arliss (kallista.destiny): indeed, NOM NOM NOM

[09:23] F L I P (flip.idlemind): Hopefully it will

[09:25] tehKellz (kelly.linden): All right, any super hot issues before we start triage?

[09:25] Kaluura (kaluura.boa): When will we get slice in prim params?

[09:25] Kaluura (kaluura.boa): ...If ever...

[09:25] tehKellz (kelly.linden): Not soon.

[09:26] tehKellz (kelly.linden):

[09:27] tehKellz (kelly.linden): First up looks like SCR-98

[09:27] flexi campfire:

[#SCR-98] Scripts not loaded create script errors when interacted with

[09:27] Brins0 (brins0.blackheart): is there a chance we may see a function to change the frequancy of a sound?

[09:27] Brins0 (brins0.blackheart): like, to make a better engine?

[09:27] tehKellz (kelly.linden): Brins0 I haven't seen any requests like that pass my radar before. If you want to write up some details in an SCR jira we can take a look.

[09:28] Brins0 (brins0.blackheart): i'll add it at some point then

[09:28] Dantee Eliot (dantee.soulstar): What about PRIM_SOUND in llSetLinkPrimitiveParams?

[09:28] Brins0 (brins0.blackheart): i'm surprised it's not been brought up before

[09:29] tehKellz (kelly.linden): All right. Today is triage day. Next monday is bring-your-own topics / general discussion. I'd file a jira if you can't find one that already exists, or comment on an existing one and bring it next week.

[09:30] tehKellz (kelly.linden): SCR-98

[09:30] tehKellz (kelly.linden): Any opinions?

[09:30] Liisa Runo: i think i hit on this once. could not repro, but this is the only reason i can imagine that would have caused the error

[09:30] Kallista Arliss (kallista.destiny): Seems like this is indeed a serious , if transent, issue

[09:31] Leonel Iceghost: that happens sometimes in homestead sims

[09:31] tehKellz (kelly.linden): Hrm. Okay. I'll add it to the backlog. It may not be too hard with some of the other recent mono changes.

[09:31] Ash DeAllura Weststar (ash.qin): I've never encountered it in the wild before, but never tested for it either.

[09:31] Kallista Arliss (kallista.destiny): and it explains some odditys that occurr on login

[09:31] tehKellz (kelly.linden): SCR-26

[09:31] flexi campfire:

[#SCR-26] llGetAnimation constant transparency

[09:32] Leonel Iceghost: I happened to me, in homestead sims when something is just rezzed

[09:32] tehKellz (kelly.linden): A documentation request?

[09:32] F L I P (flip.idlemind): Hold on...gotta wait 20 minutes to connect to google-analytics ?.(?)

[09:32] Kaluura (kaluura.boa): Looks like it...

[09:33] Helena Lycia: I must admit, I've wondered if there are some undocumented retrun values from that

[09:33] Sammy (ayamo.nozaki): ?(????) google y u so slow?

[09:33] tehKellz (kelly.linden): I'll do that one just to close it

[09:33] Qie (qie.niangao): " it's essentially impossible for the community to do this definitively."

[09:33] tehKellz (kelly.linden): SCR-64

[09:33] flexi campfire:

[#SCR-64] llKey2Name returns avatar names with double space between first and last (sometimes)

[09:33] Kallista Arliss (kallista.destiny): Sometimes it;s easier to look at the source code...

[09:34] Ash DeAllura Weststar (ash.qin): That seems pretty bad...

[09:34] F L I P (flip.idlemind): We can't expect google to mine our personal information in a timely manner

[09:34] Kaluura (kaluura.boa): Never seen...

[09:34] Qie (qie.niangao): not seen it either.

[09:34] Ash DeAllura Weststar (ash.qin): But I have never seen it.

[09:35] Helena Lycia: me neither

[09:35] Liisa Runo: i have not seen it eiter, but it would surely breaka pile of content

[09:35] F L I P (flip.idlemind): For the record by display name actually has 2 spaces between each letter...but as far as that JIRA, no never seen it either

[09:35] F L I P (flip.idlemind): my*

[09:35] Kallista Arliss (kallista.destiny): I think I saw that once... when I was working on the llSensorRepeat issue

[09:35] Fancy Greeter: Charlar Linden has arrived!

[09:36] tehKellz (kelly.linden): I actually think we've cleaned up that code since the last comment. I'm gonna ask for more info to see if it is still happening

[09:36] Helena Lycia: Hello Charlar

[09:36] Charlie (charlar.linden): hi all!

[09:36] F L I P (flip.idlemind): I wonder if, when it happens, it only happens of "Resident"s (or if it makes a difference)

[09:36] exNeko Kohime (exneko.andel): Nice outfit.

[09:37] Brins0 (brins0.blackheart): afternoon

[09:37] tehKellz (kelly.linden): Possible flip.

[09:37] tehKellz (kelly.linden): SCR-93

[09:37] flexi campfire:

[#SCR-93] llDetectedType() does not return AGENT for avatars sitting on objects

[09:37] Charlie (charlar.linden): sigh, wish ambleside was mesh enabled.

[09:37] Koli Contepomi (sahkolihaa.contepomi): :p

[09:38] F L I P (flip.idlemind): Tons of problems are caused by sitting avatars ?.(?)

[09:38] Liisa Runo: fix this one please, it is annoying

[09:39] Qie (qie.niangao): it's kinda a bigger issue, though, as Strife points out in the comments

[09:39] Liisa Runo: there is workaround, but i rather waste limited juice to something else in the script

[09:39] Kallista Arliss (kallista.destiny): Yeah, annoying at the least

[09:39] tehKellz (kelly.linden): Actually I believe the behavior currently is as designed

[09:40] tehKellz (kelly.linden): Collisions always detect the root prim, which for sitting avatars is the object they are on.

[09:41] tehKellz (kelly.linden): I'm trying to refresh my memory on the collision system.

[09:41] Liisa Runo: hmm, not sure, but i think it might affect to other llDetectedType things, maybe even sensor. cant remember

[09:41] tehKellz (kelly.linden): I think you can still find sitting agents with llSensor

[09:41] Ash DeAllura Weststar (ash.qin): That's correct

[09:42] Liisa Runo: find, yes, but then to get proper flag for detectedType might be wrong

[09:42] Object: Hello, Avatar!

[09:42] Liisa Runo: as sometimes we want to sense agents and objects in same sense call, and then separate them

[09:42] Kallista Arliss (kallista.destiny): This is true, because GM weapons uses llSensor to detect who they are near ad they work on sitting avatars

[09:43] Leonel Iceghost: Kelly I'm just reading the script with the SCR-98, it does llResetOtherScript() and llSetScriptState() right the next line, and only happened in a homestand, never in a normal sim

[09:43] Qie (qie.niangao): right. I mean, the most consistent thing would be to return multiple num_detected for collision with a seated avatar, but god only knows how much existing content that would break.

[09:43] Leonel Iceghost: llResetOtherScript() and llSetScriptState() to a not loaded script in state_entry, that is

[09:44] Kallista Arliss (kallista.destiny): it would't break properly scripted items (that itterste across num_detected

[09:44] Liisa Runo: ill experiment with this one and see if it affects to other llDetected thingies

[09:44] Object: Flip Idlemind is of type AGENT PASSIVE

[09:45] F L I P (flip.idlemind): But Im trying to be more assertive

[09:45] tehKellz (kelly.linden): so sensors still count agents found as type agent

[09:45] Object: Hello, Avatar!

[09:45] Sammy (ayamo.nozaki): yes

[09:45] Kallista Arliss (kallista.destiny): AGENT_PASSIVE_AGRESSIVE

[09:45] Object: Flip Idlemind is of type AGENT PASSIVE AGGRESSIVE

[09:46] F L I P (flip.idlemind): This is how rumors start

[09:46] tehKellz (kelly.linden): SCR-94

[09:46] flexi campfire:

[#SCR-94] llVolumeDetect collision_start/collision_end undependable

[09:47] Ash DeAllura Weststar (ash.qin): This has always been the case

[09:48] Brins0 (brins0.blackheart): I've been affected by this in the past too

[09:48] tehKellz (kelly.linden): Since the jira is nearly 3 years old, that is probable.

[09:48] Ash DeAllura Weststar (ash.qin): I had had to implement workarounds that force collisions to fire off again. Such as changing some parameter of the prim to do so.

[09:48] Qie (qie.niangao): yeah. dunno how possible it is to fix, though.

[09:50] tehKellz (kelly.linden): llVolumeDetect is a tricky beast

[09:50] Kallista Arliss (kallista.destiny): If it can be fixed, it should be

[09:50] tehKellz (kelly.linden): We are asking the physics engine to do something it really doesn't want to do

[09:51] Cale Flanagan: for physics?:)

[09:51] Sammy (ayamo.nozaki): report collisions to something nonphysical

[09:51] tehKellz (kelly.linden): One of the primary jobs of the physics engine is to keep things from interpenetrating.

[09:51] Kallista Arliss (kallista.destiny): Speculation: is it detecting another surfce crossing the volume boundries

[09:52] Kallista Arliss (kallista.destiny): tht is anoter surface of the bounding box?

[09:52] tehKellz (kelly.linden): I can import this as it seems legitimate, however it is unlikely to get much attention.

[09:52] Helena Lycia: metaphysics

[09:53] tehKellz (kelly.linden): s/import/acknowledge

[09:53] Leonel Iceghost: also when using VolumeDetect cause object to be pushed a bit, I wonder if there is any to improve efficiency there

[09:54] tehKellz (kelly.linden): Maybe, I don't really know.

[09:54] tehKellz (kelly.linden): Next is SCR-108 probably the last we have time for.

[09:54] flexi campfire:

[#SCR-108] llRezObject() fails when owner not present

[09:54] Liisa Runo: you seen this happen lately Leonel? i have seen it isn the past, but not lately. i think it is maybe fixed

[09:55] Leonel Iceghost: Liisa, I didn't test lately no

[09:55] WolfGang Senizen: i experianced this about a year ago

[09:55] WolfGang Senizen: but havent since

[09:55] Leonel Iceghost: I'm glad then, I'll start to use it again :)

[09:56] Helena Lycia: Isn't this the same issue that I queried last week? That the documentation in the wiki wasn't clear?

[09:56] Kallista Arliss (kallista.destiny): Perhaps we could let people do some testing and then readdress it next triage?

[09:56] Ash DeAllura Weststar (ash.qin): llRezObject() fails on land that is set to group only

[09:56] Ash DeAllura Weststar (ash.qin): And the owner of the group is in the same group, but not on sim

[09:56] Ash DeAllura Weststar (ash.qin): But when they are on sim, it rezzes

[09:56] Helena Lycia: Unless the owner is in the same sim or a child agent thereof

[09:57] tehKellz (kelly.linden): Yeah this is expected behavior

[09:57] Liisa Runo: seems to be the same issue than last week

[09:57] Helena Lycia: Which was just a documentation problem

[09:57] Ash DeAllura Weststar (ash.qin): That's the only reproducable case I've seen of rez failing

[09:57] tehKellz (kelly.linden): Okay.

[09:58] tehKellz (kelly.linden): 3 minutes left.

[09:58] Cale Flanagan: i would kill for scr-21 :)

[09:58] flexi campfire:

[#SCR-21] llResetOtherScript() will not reset other scripts that have stack-heaped

[09:59] Leonel Iceghost: or the llResetOnCrash() flag

[09:59] Liisa Runo: i would kill anyone using script to reset bugged script instead of fixing the bugged script

[09:59] Leonel Iceghost: if it is easier, now that we are going to work more with the limits with the new memory limit function

[09:59] Leonel Iceghost: Liisa, LL may change something to use more memory in the future.. like it happened in the past

[09:59] Leonel Iceghost: it's not always bugs

[10:00] Leonel Iceghost: it just happen when user managed data

[10:00] Kallista Arliss (kallista.destiny): sometimes it's external bugs, or odd events

[10:00] tehKellz (kelly.linden): There are more features now to detect memory usage and hopefully prevent stack heap collisions

[10:00] Liisa Runo: yea, LL breaks stuff, we just have to adapt, not wrap a boken script in duct tape

[10:00] Kallista Arliss (kallista.destiny): duct tape is a useful tool

[10:01] Kallista Arliss (kallista.destiny): ans is bailing wire

[10:01] F L I P (flip.idlemind): Still seems like llResetOtherScript()'s job is to reset other scripts, that should apply even (or, especially) when the script needs to be reset because of stack heap collision

[10:01] Liisa Runo: orbit is usefull toool too when someoen is spamming script errors

[10:01] Cale Flanagan: well if ind it usefull, bec memory chekcing isnt easy, we will see wht the new function makes possible, but if i delete in a list it can axplode and i dont know how to forsee that exctly

[10:01] Qie (qie.niangao): Well... there is llRemoteLoadScriptPin()... which might get a freshly reset script into the prim, if you've got another prim and some patience.

[10:02] Kallista Arliss (kallista.destiny): lols

[10:02] Kallista Arliss (kallista.destiny): that is a nifty hack, that should not be required

[10:02] Leonel Iceghost: also I'm sure we are going to see a lot more crashed objects now.. we never needed a way to work in the memory limits before.. and that is trouble for no mod objects

[10:03] tehKellz (kelly.linden): All right, thank you all for coming. Next week is general discussion. Have a good week everyone.

[10:03] Liisa Runo: thanks everybody

[10:03] Qie (qie.niangao): thanks Kelly.

[10:03] Nal (nalates.urriah): Thx Kelly

[10:03] Koli Contepomi (sahkolihaa.contepomi): See you, Kelly & Charlie.

[10:03] Ash DeAllura Weststar (ash.qin): Thanks Kelly

[10:03] Kallista Arliss (kallista.destiny): thanks kelly and Charlar

[10:03] Cale Flanagan: is there a my list uses x bytes memory function? that woulod help in detectig that

[10:04] Kallista Arliss (kallista.destiny): llGetListSize( (list) foo)

[10:04] Cale Flanagan: or a enough space todo delte function?

[10:04] Cale Flanagan: thats count not byte?

[10:04] tehKellz (kelly.linden): It looks like llRezObject has the right caveats for group land in it.

[10:04] tehKellz (kelly.linden): (in the wiki)