Time Nick Message 15:57 sfan5 5.9.1 soon? 16:02 [MatrxMT] imo we should revert https://github.com/minetest/minetest/commit/5280863300e88c6e42c320115a53f0957e113589 for 5.9.1 16:04 [MatrxMT] the commit apparently introduced a very relevant regression and we should fix that regression since we can. it's not very relevant whether the commit the root cause or not. 16:05 [MatrxMT] (#14999) 16:05 ShadowBot https://github.com/minetest/minetest/issues/14999 -- minetest 5.9 mapblock flickering depending on camera view 16:07 sfan5 feel free to 16:09 [MatrxMT] 👍 16:10 MTDiscord Regarding 5.9.1: Are there specific testcases on the remaining issues we should look at? As in "this issue is considered solved if under circumstance A result B is achieved" ? 16:10 [MatrxMT] ("not very relevant" because we have don't know the root cause and have no fix for it) 16:10 sfan5 grorp: it doesn't revert cleanly however, you might have to check where the calls moved in a follow-up commit 16:11 sfan5 @bastrabun: #15087 is still an unconfirmed fix. that's it. 16:11 ShadowBot https://github.com/minetest/minetest/issues/15087 -- [no squash] Fix network window size logic by sfan5 16:16 MTDiscord That's what I tried to repro but my setup is insufficient or I'm not smart enough to understand what's expected. What I tested is not what was expected (or there is a general threashold of bandwidth necessary - at some point we may need to acknowledge Minetest requires more than pidgeon carrier ping - at least smoke signal and above) 16:19 sfan5 ah, right. thanks for testing that anyway. 16:20 MTDiscord I just want to prevent people waiting on me. And I if I can I want to test whatever you throw at me. Bring it on 😛 16:31 MTDiscord grorp: reverting the use_texture_alpha commit also fixes it. might be worth having some people other than me test the ci builds ive posted links to 17:02 sfan5 stars with and without shaders look different: https://0x0.st/Xxms.png vs. https://0x0.st/Xxmz.png 17:03 sfan5 which one is correct? 17:03 sfan5 (I know the technical answer but want to fix know in which way this should be fixed) 17:06 sfan5 s/to fix know/to know/ 17:13 MTDiscord Is there some kind of brightness/intensity control for stars? 17:14 MTDiscord If gamedevs can choose how bright they want them to be, then the bottom one should represent maximum brightness. If not, and they only come in one default brightness, then the top one looks saner. 17:15 sfan5 api allows picking color 17:16 MTDiscord i would expect to see exactly the color specified in the API 17:18 sfan5 if there's an uncontroversial result I can prepare a fix 17:18 sfan5 if not i'll open an issue 17:20 MTDiscord "I would expect to see exactly the color" ... doesn't really ever happen anywhere in MT, though, tbh. When I once tried to make a solid-color skybox and then make the fog match it, I had to take a screenshot of the skybox and sample the color from there. 17:55 sfan5 #15143 17:55 ShadowBot https://github.com/minetest/minetest/issues/15143 -- Star color is incorrect 17:56 sfan5 I was wrong actually (thought the brighter color is correct) and with context it simply turns out to be a regression 18:26 [MatrxMT] Is it really a regression if it's 5 versions old? 18:28 sfan5 yes 18:45 MTDiscord being 5 versions old just puts you in the unenviable position where the bug is a regression and then the fix is also a regression for other people. 18:47 MTDiscord in this case i'm saying fix is fix 18:48 MTDiscord this was never a consistent regression, and the API docs are clear about what should happen 18:51 sfan5 are they? 19:00 Krock I'd honestly be surprised if the documentation turns out to be 100% clear 19:01 MTDiscord well okay 19:01 MTDiscord it says this: "star_color: ColorSpec, sets the colors of the stars, alpha channel is used to set overall star brightness. (default: #ebebff69)" 19:01 sfan5 that is indeed quite clear 19:02 Krock I assume that fog must be turned off so that this holds true 19:03 sfan5 would be weird to render the sky with fog applied 19:05 Krock yes, somewhat.