Chat log from the meeting on 2019-03-05

[11:00] Bill Blight is online. [11:00] Gavin.Hird @grid.xmir.org:8002: I think [11:00] Ubit Umarov: yeap bullet module has native code also [11:01] Telehub: Welcome back to, Dev Outreach, Andrew Hellershanks. [11:01] Vincent.Sylvester @hg.zetaworlds.com:8002: Now it running on riscv that I would like to see [11:01] Gavin.Hird @grid.xmir.org:8002: so the region is running on grid.xmir.org:8002:Dayturn [11:01] Andrew Hellershanks: Hello, everyone. [11:01] Gavin.Hird @grid.xmir.org:8002: Hi Andrew [11:01] Selby.Evans @grid.kitely.com:8002 is online. [11:01] Ubit Umarov: you got the last blabla-test from opensim-libs? [11:02] Selby.Evans @grid.kitely.com:8002: hi everyone [11:02] Gavin.Hird @grid.xmir.org:8002: no it is not quite there [11:02] Ubit Umarov: hmm? [11:02] Leighton.Marjoram @grid.outandproud.life:8002: Hi Selby [11:02] Ubit Umarov: what is not there ? [11:02] Gavin.Hird @grid.xmir.org:8002: the lastest 2 weeks of commits [11:03] Gavin.Hird @grid.xmir.org:8002: on the Pi [11:03] Ubit Umarov: ubode does not have commits in sometime [11:03] Ubit Umarov: i mean ode library [11:03] Gavin.Hird @grid.xmir.org:8002: that has the latest version [11:03] Ubit Umarov: ok [11:03] Andrew Hellershanks: Gavin, you are running OpenSim on a Raspberry Pi? [11:03] Gavin.Hird @grid.xmir.org:8002: it runs pretty good [11:03] Gavin.Hird @grid.xmir.org:8002: I do [11:03] Ubit Umarov: more updated than current master :) [11:04] Gavin.Hird @grid.xmir.org:8002: on mono 5.18 [11:04] Andrew Hellershanks: Pi 2 or Pi 3? [11:04] Ubit Umarov: think bill is running that here also [11:04] Gavin.Hird @grid.xmir.org:8002: 3+ [11:04] Andrew Hellershanks: ok. I know nebadon has run it on a Pi. I haven't tried that yet. [11:04] Gavin.Hird @grid.xmir.org:8002: I added PoE to it so it is powered from a switch [11:04] Vincent.Sylvester @hg.zetaworlds.com:8002: So long as you don't throw a lot of physical stuff at it it will run just fine [11:05] Ubit Umarov: i should try power my i7 like that also [11:05] Ubit Umarov: (hihihihi) [11:05] Gavin.Hird @grid.xmir.org:8002: fat ethernet cable? [11:05] Vincent.Sylvester @hg.zetaworlds.com:8002: I smell burning plastic lol [11:05] Ubit Umarov: well 31w now, not much [11:05] Gavin.Hird @grid.xmir.org:8002: hehe [11:06] Gavin.Hird @grid.xmir.org:8002: I have the region HG accessible to see how stable it is [11:06] Vincent.Sylvester @hg.zetaworlds.com:8002: Is that an invitation? [11:06] Ubit Umarov: opensim stable? hmm [11:06] Gavin.Hird @grid.xmir.org:8002: hehe [11:06] Gavin.Hird @grid.xmir.org:8002: go ahead and crash it [11:07] Vincent.Sylvester @hg.zetaworlds.com:8002: I don't really condone violence against defenseless regions xD [11:07] Gavin.Hird @grid.xmir.org:8002: biggest issue is running out of memory as the ystem only has 1 Gb [11:07] Ubit Umarov: its 32b or 64 that arm ? [11:08] Vincent.Sylvester @hg.zetaworlds.com:8002: You should get an odroid, they come with up to 4gb [11:08] Gavin.Hird @grid.xmir.org:8002: processor is 64 bit but kernel is 32 [11:08] Gavin.Hird @grid.xmir.org:8002: it runs debian stretch ++ [11:08] Ubit Umarov: 32b does use less mem on opensim, if you can run it in32b there [11:09] Ubit Umarov: mb 2/3 the ram ? [11:09] Ubit Umarov: or less [11:09] Gavin.Hird @grid.xmir.org:8002: right now it use about 440 Mb with a 2x2 var [11:09] Ubit Umarov: you had to mk it a var on a tiny machine :) [11:10] Gavin.Hird @grid.xmir.org:8002: acid test [11:10] Ubit Umarov: well terrain patchs compression is the main issue [11:10] Ubit Umarov: and sending [11:11] Gavin.Hird @grid.xmir.org:8002: if it works and is stable it is a dirt cheap region server [11:12] Ubit Umarov: so what more news about opensim do you bring today? [11:12] Ubit Umarov: :) [11:13] Ubit Umarov: (silence) [11:13] Vincent.Sylvester @hg.zetaworlds.com:8002: I started hardcoding all the config variables into the code [11:13] Vincent.Sylvester @hg.zetaworlds.com:8002: Found a bunch of them that have no defaults set in code along the way [11:14] Vincent.Sylvester @hg.zetaworlds.com:8002: I always kinda wondered why we have OpenSimDefaults when the code provides the exact same defaults [11:14] Ubit Umarov: well it is named defaults :p [11:14] Ubit Umarov: and i do use that to tune some parameters without changing code [11:14] Gavin.Hird @grid.xmir.org:8002: so they can be overridden if you run on a PI ? :-) [11:15] Ubit Umarov: and it is also a template for opensim.ini [11:15] Ubit Umarov: users should avoid changing it [11:15] Vincent.Sylvester @hg.zetaworlds.com:8002: I know, but should it not be removable entirely aka all its variables set in code as well [11:16] Ubit Umarov: but we may, as i said avoidind a full compile [11:16] Jeff.Kelley @88.161.20.136:9000: it's a way for users to know they exist [11:16] Ubit Umarov: code should survice missing things on ini files [11:16] Gavin.Hird @grid.xmir.org:8002: I think it is good we have than in that file [11:17] Ubit Umarov: the cases where values are as in code are mostly commented out [11:17] Vincent.Sylvester @hg.zetaworlds.com:8002: Defaults does not even have all possible config options in it [11:17] Gavin.Hird @grid.xmir.org:8002: so we can test settings without recompiling [11:17] Ubit Umarov: some are not for you to know :p [11:18] Vincent.Sylvester @hg.zetaworlds.com:8002: lol [11:18] Ubit Umarov: well you go see code ... details :p [11:18] Vincent.Sylvester @hg.zetaworlds.com:8002: Okay so I am not saying hardcode them, rather make OpenSimDefaults capable of being deleted and OpenSim reading from defaults set in code entirely [11:19] Vincent.Sylvester @hg.zetaworlds.com:8002: Most things in Defaults also have default variables set in code, but not all [11:19] Jeff.Kelley @88.161.20.136:9000: isn't hardcoding usually bad ? [11:19] Vincent.Sylvester @hg.zetaworlds.com:8002: It would be nice if all variables described in Defaults had default variables set in code [11:20] Vincent.Sylvester @hg.zetaworlds.com:8002: Defaults may still override them, but it should not fail to start without it present [11:20] Vincent.Sylvester @hg.zetaworlds.com:8002: It's an inconsistency at the end of it [11:21] Ubit Umarov: guess some older code may not have it right [11:21] Vincent.Sylvester @hg.zetaworlds.com:8002: That seems to be the case [11:22] Ubit Umarov: phone call again andrew ? [11:23] Ubit Umarov: wlel we had opensimfest last weekend [11:23] Vincent.Sylvester @hg.zetaworlds.com:8002: It would be nice if Defaults could just be deleted and it not causing a startup failure is all I am saying, I only mentioned hardcoding, because that is what I did last week and I came across a lot of older variables that I had to look up in Defaults rather than just deleting the config reader. [11:23] Ubit Umarov: a few minor glitchs, but guess went ok [11:24] Bill Blight: had issues with floods but most people had a great time [11:25] Ubit Umarov: that's what matters [11:25] Vincent.Sylvester @hg.zetaworlds.com:8002: I went to bed yesterday thinking about ways to improve inventory loading, fell asleep however, guess that topic was more tiring than I anticipated [11:25] Ubit Umarov: :) [11:25] Leighton.Marjoram @grid.outandproud.life:8002: a few lessons learnt for next year but people seemed to enjoy it despite the issues [11:26] Bill Blight: yes everyone I talked to who toughed it out, had a great time [11:26] Ubit Umarov: like not having blinking lights all over :p [11:26] Bill Blight: yeah [11:26] Vincent.Sylvester @hg.zetaworlds.com:8002: Our default data structures don't hold up all too well if they get larger, something that eventually needs a looking at seeing as mesh means ever faster increase in sizes [11:26] Bill Blight: killed most of those [11:26] Ubit Umarov: well i did reduce they bandwith cost to half [11:26] Ubit Umarov: but thats still 2 much [11:27] Vincent.Sylvester @hg.zetaworlds.com:8002: My main concern at the moment is with database response time of inventory tables, I just cannot think of a better way to arrange that stuff to reduce query times [11:27] Ubit Umarov: old issue also at sl [11:27] Selby.Evans @grid.kitely.com:8002: A continuation of the werewolf story (writer's workshop) demonstrating how a story can also be used to support learning. [11:27] Ubit Umarov: ( hte lights) [11:28] Leighton.Marjoram @grid.outandproud.life:8002: spreading out the region across a few will help as well but nothing too difficult to address and improve the experience next time. [11:28] Selby.Evans @grid.kitely.com:8002: https://virtualoutworlding.blogspot.com/2019/03/2019-muve-event-opensim-fest-rises.html [11:29] Ubit Umarov: 28k prims.. well [11:29] Leighton.Marjoram @grid.outandproud.life:8002: the phoenix was such an appropriate image for this years event [11:29] Bill Blight: yeah, new plan for next year [11:29] Bill Blight: tighter constraints on builders too [11:29] Leighton.Marjoram @grid.outandproud.life:8002: I dont know what I was thinking having people land on that region *facepalm* [11:29] Vincent.Sylvester @hg.zetaworlds.com:8002: Unfortunately I could not mobilize anyone to take a look at DSG yet, would be interesting to see if that would improve anything these days [11:30] Ubit Umarov: gezzz [11:30] Ubit Umarov: we are DSG already [11:30] Leighton.Marjoram @grid.outandproud.life:8002: Distributed Scene Graph [11:30] Ubit Umarov: and see how crossings worh [11:30] Leighton.Marjoram @grid.outandproud.life:8002: I think and hope [11:30] Ubit Umarov: imagine "crossings" at all levels... [11:31] Ubit Umarov: and you get a idea of how a more complex DSG will be [11:31] Andrew Hellershanks: Ubit, no phone calls today. Just letting you all talk. I didn't have anything at the moment. However, speaking of the festival... [11:31] Ubit Umarov: and 256x256 regions is a simple DSG [11:32] Andrew Hellershanks: I hope those of you who attended the festival over the weekend enjoyed it. [11:32] Ubit Umarov: considering SL mainland for example [11:32] Ubit Umarov: just a fixed size dsg [11:32] Ubit Umarov: see what i mean ? [11:33] Ubit Umarov: ( the scene its the entire mainland ) [11:33] Vincent.Sylvester @hg.zetaworlds.com:8002: I looked at this before, my main concern is figure out at what point it does make sense to use and whether in terms of inventory and assets it would improve loading in any way, but just reading code as someone barely familiar with C I can't quite make up my mind [11:33] Selby.Evans @grid.kitely.com:8002: Festival was good promotion and I got some new contacts [11:34] Vincent.Sylvester @hg.zetaworlds.com:8002: I am mainly just looking at various things to see if there is anything that could help with loading times, not letting stones unturned as they say [11:35] Ubit Umarov: and DSG is not putting 100 regions on a single machine [11:35] Ubit Umarov: is putting part of a region on just one :p [11:36] Ubit Umarov: so a grid is a entire datacenter :) [11:36] Ubit Umarov: or one of those nice machines with tons of cpus [11:36] Vincent.Sylvester @hg.zetaworlds.com:8002: I also see issues with locking so it is questionable whether it helps [11:38] Ubit Umarov: sometimes looking also has issues :p [11:38] Andrew Hellershanks: Vincent, it never hurts to have another pair of eyes looking at code. The biggest problem is getting to know where to look for things. I found Mono Develop helped me to find things when I needed to dig in to the code. [11:39] Vincent.Sylvester @hg.zetaworlds.com:8002: Yeah my main aim really is just to explore concepts that could help improve performance seeing as there are more and more things that will bring that down to a crawl [11:40] Andrew Hellershanks: Vincent, you could always keep notes as you explore the code. The notes could become part of some code documentation. That is something we really still need. [11:40] Andrew Hellershanks: <<:) [11:40] Andrew Hellershanks: hm... how did those hats wind up there? [11:41] Ubit Umarov: hats? [11:41] Vincent.Sylvester @hg.zetaworlds.com:8002: Not sure my incoherent mumbling would be useful as code documentation, but if I find interesting things I already tend to bug Ubit about them at any given chance :D [11:42] Andrew Hellershanks: Ubit, the extra << symbols I didn't mean to type. [11:46] Andrew Hellershanks: On the subject of documentaiton I've been remiss in not having gotten around to adding pages to the wiki about the newer OSSL functions. I've been distracted by an electronics project of mine. [11:46] Andrew Hellershanks: I will raise the priority of that task on my ToDo list. [11:47] Ubit Umarov: ( ie making is area fire dep crazy ) [11:47] Andrew Hellershanks: ? [11:47] Vincent.Sylvester @hg.zetaworlds.com:8002: I been trying to get around to testing lsl implementations so an update to that page could be made, sadly been distracted with writing php code all week [11:47] Ubit Umarov: nothing :) [11:48] Andrew Hellershanks: Ubit, no soldering iron use yet. I haven't ordered the PCBs. I've been testing parts of my circuits on solderless breadboard. :) [11:49] Ubit Umarov: i did place back the assetvieewr cap code [11:49] Andrew Hellershanks: Vincent, you are talking about testing XEngine and YEngine? [11:49] Ubit Umarov: not only disabled on opensimdefaults [11:49] Vincent.Sylvester @hg.zetaworlds.com:8002: XEngine first actually, then YEngine [11:49] Ubit Umarov: so its easier for viewers to test [11:50] Ubit Umarov: FS did accept liru code fix on the bad issue they had on it for opensim [11:50] Ubit Umarov: so guess will be out on next release [11:51] Ubit Umarov: and then we can mk it standard feature [11:51] Gavin.Hird @grid.xmir.org:8002: they usually release 2 times per year [11:52] Ubit Umarov: for now we can't, since so many use fs [11:52] Ubit Umarov: and all love to dance :) [11:52] Gavin.Hird @grid.xmir.org:8002: hehe [11:52] Andrew Hellershanks: Those two nearby mesh objects have been dancing up a storm for the last few weeks. [11:53] Ubit Umarov: was very confusing when LBSA had it active [11:53] Gavin.Hird @grid.xmir.org:8002: There has just been a minor update to the viewer this last week, but it should yield 5-10 FPS on most hardware [11:53] Gavin.Hird @grid.xmir.org:8002: https://www.dayturn.com/viewer/index.php?resources/ [11:53] Vincent.Sylvester @hg.zetaworlds.com:8002: I am really liking the progress on Dayturn so far, problems being solved at rapid pace [11:54] Vincent.Sylvester @hg.zetaworlds.com:8002: Now if only I could figure out this darn shader thing, I been messing with those files yesterday and only managed to break stuff xD [11:54] Ubit Umarov: well some grids withou hg, all regions same version, can use it [11:54] Ubit Umarov: ie like sl [11:54] Gavin.Hird @grid.xmir.org:8002: updating the shaders is an art [11:54] Vincent.Sylvester @hg.zetaworlds.com:8002: It's sorcery far as I can tell [11:54] Ubit Umarov: :) [11:54] Gavin.Hird @grid.xmir.org:8002: it is [11:55] Vincent.Sylvester @hg.zetaworlds.com:8002: Seriously just changing a single number and it's all suddenly pink [11:55] Gavin.Hird @grid.xmir.org:8002: hehe [11:55] Ubit Umarov: ohh i do that on opensim also [11:55] Ubit Umarov: change a line and pofff region dies [11:56] Ubit Umarov: :) [11:56] Vincent.Sylvester @hg.zetaworlds.com:8002: Hopefully with enough trial and error I can figure out how to lower than limit to -100, will be so much fun to build underwater worlds and explore them with submarines [11:56] Andrew Hellershanks: Gavin, Is there a link you can share to pre-compiled versions of Dayturn? [11:56] Gavin.Hird @grid.xmir.org:8002: https://www.dayturn.com/viewer/index.php?resources/ [11:56] Bill Blight: was gonna say, he just did [11:56] Bill Blight: LOL [11:56] Bill Blight: :P [11:58] Vincent.Sylvester @hg.zetaworlds.com:8002: Feel free to report whatever problems you find on the forums, Gavin has been doing a stellar job responding to them and solving things as they come up :) [11:58] Andrew Hellershanks: I was looking at those pages. The "Go to download" link doesn't go to a page. It just starts download of an .exe file. [11:58] Gavin.Hird @grid.xmir.org:8002: which is the installer [11:58] Andrew Hellershanks: I can't use that as I'm not running Windows. [11:59] Gavin.Hird @grid.xmir.org:8002: there is no Linux version [11:59] Vincent.Sylvester @hg.zetaworlds.com:8002: There are mac versions too [11:59] Andrew Hellershanks: Still doesn't help me. [11:59] Gavin.Hird @grid.xmir.org:8002: and I am pretty sure the code will not build on Linux as it is [11:59] Vincent.Sylvester @hg.zetaworlds.com:8002: I shall try to compile it on my smart-fridge xD [12:00] Andrew Hellershanks: I would give it a try but I can't find a link to the source code. [12:00] Gavin.Hird @grid.xmir.org:8002: The repository has code for it https://bitbucket.org/dayturn/dayturn-windows [12:00] Gavin.Hird @grid.xmir.org:8002: that respository should still build Linux with small adjustments [12:00] Andrew Hellershanks: Thanks, Gavin. I'll give it a try and see what happens. [12:01] Vincent.Sylvester @hg.zetaworlds.com:8002: I wonder if it would be possible to get them to run under wine [12:01] Gavin.Hird @grid.xmir.org:8002: someone said he did [12:01] Vincent.Sylvester @hg.zetaworlds.com:8002: I don't have a linux box with a gpu unfortunately, I should fix that [12:01] Gavin.Hird @grid.xmir.org:8002: someone on LBSA plaza [12:02] Vincent.Sylvester @hg.zetaworlds.com:8002: Given the improvements to that sector lately I would not be surprised if it even runs halfway decent too [12:02] Andrew Hellershanks: I don't like doing that as I have found programs under Wine don't always quite work right or have stability issues. [12:03] Vincent.Sylvester @hg.zetaworlds.com:8002: Even native linux programs like crashing though, just look at Hexchat xD [12:03] Gavin.Hird @grid.xmir.org:8002: the parson said it was fast [12:03] Gavin.Hird @grid.xmir.org:8002: person* [12:03] Vincent.Sylvester @hg.zetaworlds.com:8002: Gaming support on linux is making leaps these days, really nice to see that [12:03] Andrew Hellershanks: We reached the top of the hour already. Any last minute items for todays meeting? [12:04] Gavin.Hird @grid.xmir.org:8002: well hop over and crash my Pi region [12:04] Andrew Hellershanks: Games? Who has time for games? I rarely do. :) [12:04] Vincent.Sylvester @hg.zetaworlds.com:8002: What was the hg again? [12:04] Gavin.Hird @grid.xmir.org:8002: grid.xmir.org:8002:Dayturn [12:05] Vincent.Sylvester @hg.zetaworlds.com:8002: *takes gloves off* let's do this [12:05] Gavin.Hird @grid.xmir.org:8002: hehe [12:05] Andrew Hellershanks: Hm... we could hold a meeting over there and see how many people it takes to crash OS running on a Pi. ;) [12:05] Gavin.Hird @grid.xmir.org:8002: right [12:06] Andrew Hellershanks: Thank you all for coming today. See you again next week.