Time Nick Message 06:03 Swift110-mobile Hey a 06:03 Swift110-mobile Hey all 08:49 erlehmann why are there no discussions about cheating in minetest forums 08:49 erlehmann i want to improve anticheat 08:49 erlehmann but i was told discussing cheating is banned? 08:49 erlehmann in particular, i want to forbid flying 08:51 erlehmann sfan5 what do you think of discussing ways how to cheat and how to react to that 08:51 erlehmann i only found it on issue tracker somehow 08:51 erlehmann funniest thing i found was a suggestion to generate ores only at surface 08:51 rubenwardy Discussing cheating is fine, promoting ready made tools isn't 08:51 erlehmann to frustrate xray hackers 08:51 rubenwardy I don't necessarily agree with this rule 08:52 erlehmann lol so that is why there are no results for "dragonfire" or "waspsaliva" on the forums 08:52 erlehmann i mean no useful results 08:52 rubenwardy I think it's not very useful as anyone who wants to cheat will just Google it anyway 08:52 erlehmann yeah i think too 08:52 erlehmann or just read the source 08:52 erlehmann i mean 08:53 erlehmann bool checkPrivilege(const std::string &priv) const 08:53 erlehmann - { return (m_privileges.count(priv) != 0); } 08:53 erlehmann + { 08:53 erlehmann + return true; 08:53 erlehmann + // (m_privileges.count(priv) != 0); 08:53 erlehmann + } 08:53 erlehmann :D 08:53 erlehmann that is kinda easy to work out 08:53 erlehmann but the question is how to deal with it 08:53 erlehmann and that is complex 08:54 erlehmann rubenwardy i recently found a bug where i could drop a specific item and cheat client waspsaliva would crash, that was funny 08:54 erlehmann i mean i reported it and it was fixed 08:54 rubenwardy Hack clients are often also hacky 09:24 MTDiscord ? 09:40 erlehmann rubenwardy so if i post that example, do you think i would be banned from the forums? 10:37 whosit is having "map saving" 100-200ms long every 5 seconds considered normal? 10:43 entuland I guess you could change those settings, but somewhere you have to store the map changes not to lose them if something goes south 10:44 entuland the point is whether it's freezes the game or not, regardless of how much the save takes 10:44 entuland s/it's/it 10:44 whosit why it's not async from the game loop? 10:44 entuland that I don't know 10:46 whosit the 100-200ms is for server running postgres backend 10:46 whosit and it does the timestep lag / freeze thing 11:08 erlehmann rubenwardy to whom can i talk about improving anticheat? there is no way i can discuss anticheat without explaining cheats 11:09 erlehmann rubenwardy if i can't do it on the forums, i can only … talk to the authors of cheat clients, which i am doing lol 11:09 erlehmann they have little interest in better anticheat though 11:09 erlehmann though i have 11:09 erlehmann bc anticheat is stupid enough to break trampolins and mesecon piston teleport 11:10 rubenwardy You're not going to get banned without warning 11:11 rubenwardy sfan and krock are probably the people to talk to 11:11 erlehmann ok thx 11:12 erlehmann rubenwardy i am pretty sure if i link to dragonfire or waspsaliva code i risk a ban 11:12 erlehmann or even show a snippet as the one above 11:13 erlehmann rationale: the rules say intentions do not count 11:14 erlehmann btw, "beware the dark" is a pretty funny way to discourage nocip and flying 11:20 erlehmann (bc ppl who do that also have fullbright) 11:20 erlehmann (and bc the sky is dark) 19:40 AntumDeluge I don't see an option to adjust the label on an image button ( http://minetest.gitlab.io/minetest/formspec/#image_buttonxywhtexture-namenamelabel ). Is this not possible? 19:42 AntumDeluge I mean obviously I can just leave the image label empty & add another label over it. Was just curious if it were possible to do it from the image_button options. 19:44 entuland yep, I don't think you have control over it 19:47 AntumDeluge k, thanks. 19:47 MTDiscord yw 19:48 entuland dang I should avoid posting from discord in this channel 19:48 sfan5 why 19:49 entuland because ping & startle myself :P 19:50 entuland it's like an I/O conflict, I don't expect to get the ping immediately after I send a message :P 19:50 erlehmann sfan5 hi. if i want to discuss how cheat clients work and possible anticheat improvements on a more persistent platform than here, where to do it? i do not want to get banned on minetest forums. 19:50 erlehmann sfan5 i ask you because you are definitely not a cheat client advocate 19:50 MTDiscord The bridge is supposed to prevent self-pinging, but I wouldnt be surprised if it doesnt work 19:51 entuland oh the ping happened on the HexChat client 19:51 entuland not on Discord 19:51 MTDiscord I know 19:51 erlehmann entuland highlight? 19:51 entuland yes 19:51 MTDiscord Its supposed to add a special invisible character to the nick on the IRC side to prevent self-pinging 19:52 entuland can't tell, the message above is marked as if someone else sent a message pinging me 19:52 sfan5 erlehmann: idk I guess github? 19:52 entuland the "entuland" word above is also treated exactly as my handle, I can right click it, get a context menu and so forth 19:53 erlehmann sfan5 yeah ok then not i guess. github hellbanned me once accidentally, i may have made too many commits or something. 19:53 erlehmann not using it except if i get paid 19:54 erlehmann sfan5 can you explain the history behind the "no talking about cheats" thing though? 19:54 erlehmann it sounds incredibly weird 19:54 erlehmann like something from a steam forum 19:54 erlehmann or 1999 boomer shooter enthusiasts 19:55 erlehmann the minecraft community talks about cheats all the time, so i wonder, is it to keep those ppl away? 19:55 sfan5 I wouldn't know what to say about the history spontanously 19:56 erlehmann ah i thought maybe there was some big bad event or something 19:56 erlehmann i heard that on clamity pistons were once made unknown by an admin to prevent duping 19:57 erlehmann but unknown items can be stacked higher than known items 19:57 erlehmann so yeah, that backfired spectacularly 19:57 erlehmann when they were made known 19:59 sfan5 anyway within limits code-focused cheating discussion should also be fine on the forums 20:00 erlehmann people keep saying that but there exist no discussions like that 20:00 erlehmann so i guess whoever tried that ate a ban 20:01 erlehmann meanwhile, can anyone make sense of this commit? https://git.minetest.land/Mineclonia/Mineclonia/commit/8f30e223e2d45c4ead2237196c13446d6b9cb1e3 20:02 erlehmann i have the cursed task of backporting features with commit messages like "I hate reverting something xD" 20:54 entuland erlehmann: well, the only change in that seems to be replacing "ipairs" with "pairs" 20:54 entuland which may make sense if the result of connected_players() changed format 20:55 entuland as ipairs would ignore any non-numeric key 20:56 entuland connected_players() or any other source of data there 20:57 erlehmann entuland well, i can see that, but i don't know why because the commit message is crap 20:57 entuland oh that I can't tell 20:58 erlehmann i thought maybe someone knew 20:58 entuland maybe someone else replaced pairs with ipairs with an earlier commit and they had to fix the "bad idea"? 20:58 erlehmann i was told that ipairs is faster 20:58 erlehmann yeah but also this is on a feature branch that kay27 made for nether portals in mineclone2. it does not belong there. 20:59 erlehmann i keep finding unrelated small fixes with questionable commit messages on feature branches and i hate it 20:59 erlehmann bc then the feature branch depends on that thing 21:03 entuland I guess the best way to know is to ask the committer :P 21:03 entuland also, I may be wrong and that could be a bad fix - maybe all of those cases really only use numeric keys and maybe ipairs is really faster and maybe there was really no need to change those 21:05 entuland I also found this hit searching for which one is really faster: https://devforum.roblox.com/t/speed-comparison-ipairs-vs-pairs-vs-numerical-loop-benchmark/537134 21:05 entuland Quote: "I recently learnt that ipairs is only this fast with our custom Lua interpreter, Luau; in Vanilla Lua it is slower than pairs." 21:13 MTDiscord Ipairs is slower than pairs, regular loop is fastest 21:25 MTDiscord ipairs is faster than pairs 21:26 MTDiscord At least, I think. Might only be in JIT 21:36 MTDiscord https://medium.com/@sleitnick/battle-of-the-loops-c001bcb4961c 22:03 MTDiscord I'd be curious to see how newer versions of Lua (specifically, 5.4) compare 22:10 MTDiscord can't tell what they're benchmarking exactly but here is something in general: https://www.reddit.com/r/lua/comments/j9x33f/arewefastyet_benchmark_suite_applied_to_different/ 22:23 MTDiscord pokes @exe_/irus for tests he said that he would provide 22:50 MTDiscord Sorry, I only got through basic two client testing, still have to run my 5-10 client test here in a bit. Sadly I only get 30 minutes of free time a day during the week 22:51 MTDiscord tis fine