Time Nick Message 00:04 paramat woohoo 00:04 paramat meetings are ok for me 00:06 rubenwardy cool 00:06 rubenwardy also 00:06 rubenwardy we include pr nums in the commit msg when merging right? 00:07 paramat yes for engine, no for mtg 00:07 paramat i suggest we have meetings when people are around. a fixed time usually failed 00:08 paramat but whatever 00:08 rubenwardy sure 00:08 paramat just not saturday eve. sunday is better 00:08 rubenwardy I'm free except when I'm not 00:23 paramat #8312 seems mergeable (?) 00:23 ShadowBot https://github.com/minetest/minetest/issues/8312 -- Remove minimal from release installations by rubenwardy 00:29 rubenwardy marked a load of things as Possible Close 00:29 rubenwardy mostly WIP or adoption needed 00:29 rubenwardy https://github.com/minetest/minetest/pulls?q=is%3Apr+is%3Aopen+sort%3Aupdated-desc+label%3A%22Possible+close%22 00:32 paramat yeah going through 00:33 rubenwardy #7621 is ready for review 00:33 ShadowBot https://github.com/minetest/minetest/issues/7621 -- Add secure require() function by rubenwardy 00:34 paramat is that useful for client-provided CSM too or only SSCSM? 00:35 rubenwardy only SSCSM for now 00:36 rubenwardy it's not as useful for CSM 00:36 rubenwardy the usecase is that it allows you to load pure Lua libraries embedded in a mod 00:41 paramat well, no objections. but generally i feel new CSM features should not go ahead until SSCSM looks like it will actually happen 00:41 rubenwardy this is not a new CSM feature 00:41 rubenwardy it's SSCSM only 00:42 * rubenwardy removes incorrect label 00:42 paramat yes i know 00:42 paramat my comment doesn't apply to this PR 00:43 rubenwardy ah 00:43 paramat so 'no objections' =) 00:47 paramat i know some people have said they wil work on it, but my feeling is it won't happen for a long time, if at all 00:49 paramat nerz can't do it alone, but who else has time for it? we're already overwhelmed 00:50 rubenwardy I am quite concerned about security 00:50 paramat i feel all CSM work should be put on hold until SSCSM happens, because it may be wasted work 00:51 rubenwardy oops 00:51 rubenwardy earlier, I got confused and though SSCSM was SSM 00:51 rubenwardy +t 00:51 paramat heh 01:00 paramat lol 01:01 rubenwardy argh 01:01 rubenwardy to many issues 01:01 rubenwardy have you looked in -hub? 01:01 rubenwardy I wish GH just had nicer tools 01:05 paramat will comment in -hub 01:31 rubenwardy there are 9 pathfinding issues :/ 01:32 paramat afk now, will go through updated stuff later 01:47 rubenwardy merging #8317 in 10 01:47 ShadowBot https://github.com/minetest/minetest/issues/8317 -- Add compatibility to vcpkg buildsystem by adrido 02:12 rubenwardy #7895 is now ready for review 02:12 ShadowBot https://github.com/minetest/minetest/issues/7895 -- Add luacheck to check builtin by rubenwardy 04:14 sofar oh neat 10:49 Krock will merge #7688 and #8348 in 10 minutes 10:49 ShadowBot https://github.com/minetest/minetest/issues/7688 -- Add disable_jump to liquids and ladders by SmallJoker 10:49 ShadowBot https://github.com/minetest/minetest/issues/8348 -- Formspecs: Close on metadata removal by SmallJoker 10:50 Krock thanks for the approvals :) 11:00 Krock merging 11:01 Krock done 16:22 Krock #8590 16:22 ShadowBot https://github.com/minetest/minetest/issues/8590 -- Settings: Disallow space characters entirely by SmallJoker 16:25 p_gimeno does that mean a mod could override e.g. a security setting? O.o 16:29 Krock yes 16:29 Krock rubenwardy, nerzhul: high priority. 16:30 Krock or whoever else is active right now. sorry for the ping :/ 16:31 sfan5 merged 16:31 Krock thanks 17:38 Krock will merge #8550 in 10' 17:38 ShadowBot https://github.com/minetest/minetest/issues/8550 -- Use CDB author and name in installed package's config by pauloue 17:48 Krock merging 18:24 paramat for #7987 this should be considered and tried https://github.com/minetest/minetest/pull/7987#issuecomment-447710228 18:24 ShadowBot https://github.com/minetest/minetest/issues/7987 -- Time-of-day-dependent face shading (#6165 revived) by Unarelith 18:26 paramat aplying the effect linearly causes weirdness in darker areas 18:26 paramat *applying 18:26 Krock well so, maybe passing it through a cosine transformation? 18:28 paramat not sure, but some sort of curve 18:28 Krock float outdoorsRatio = cos(gl_Color.a * M_PI) * 0.5 18:29 Krock that's what eased perlin noise uses 18:31 paramat essentially, only apply the effect in areas with near-direct-sunlight lighting 18:36 paramat but, that can be experimented with later 18:37 paramat i wouldn't let that delay merge 18:59 p_gimeno hm, * M_PI? that's not an easing function 18:59 Krock p_gimeno: key is cosine 19:00 Krock you take half of the period to make a wavy, non-linear interpolation 19:01 Krock there should be a +0.5 after the calculation to lift it up 19:02 Krock I can see our engine uses some fast math tricks to act like cosine: https://github.com/minetest/minetest/blob/master/src/noise.h#L229 19:03 p_gimeno now it starts to make sense, I think you mean (1 - cos(gl_Color.a * M_PI)) * 0.5 19:03 p_gimeno that is an easing function 19:04 p_gimeno that one is the 5th degree smoothstep-like function 19:04 Krock yes, but the "1 -" part is omitted since you'd need to subtract it from 1 there anyway 19:04 Krock I just entered it in wolframalpha and it looks the same as you'd get with cosine and the stuff around 19:07 p_gimeno mm ok, it seems a bit obscured to me that way 19:08 p_gimeno https://en.wikipedia.org/wiki/Smoothstep#Variations - Wikipedia calls that function smootherstep 19:08 p_gimeno (the 5th degree polynomial) 19:11 Krock til 19:52 p_gimeno thinking about it, probably a sigmoid-shaped kind of function won't be much help, because it leaves the centre unaltered (it outputs 0.5 for an input of 0.5) and it seems to me that the midrange is what needs to be altered 19:55 p_gimeno the pow() function with a constant exponent is a simple easing function to alter the centre 20:00 p_gimeno I'm unclear on whether the exponent should be > 1 or < 1 though. Exponent values > 1 give output values < 0.5 for an input of 0.5, and vice versa. 20:15 paramat yes, i'm suggesting midrange applies zero effect, then some kind of smoothstep above midrange