<<Prev Next>> Scroll to Bottom
Stuff goes here
L1[00:22:30] <Spider ​EveryOS> %tonkout
L2[00:22:31] <MichiBot> Wow! Spider ​EveryOS! You beat Va​ur's previous record of 3 hours, 12 minutes and 25 seconds (By 2 minutes and 43 seconds)! I hope you're happy!
L3[00:22:32] <MichiBot> Spider ​EveryOS has stolen the tonkout! Tonk has been reset! They gained 0.003 tonk points! plus 0.002 bonus points for consecutive hours! (Reduced to 50% because stealing) Current score: 3.03175908. Position #2 Need 0.0416892 more points to pass Forec​aster!
L4[00:28:02] <kim​apr> is there an OC1 emulator that works on mobile?
L5[00:34:12] <kim​apr> i suppose OCEmu could work in termux-x11..
L6[02:08:03] ⇦ Quits: jackie (~jackie@banana-new.kilobyte22.de) (Ping timeout: 207 seconds)
L7[02:08:29] ⇦ Quits: nadja (~dequbed@banana-new.kilobyte22.de) (Ping timeout: 183 seconds)
L8[02:09:21] ⇨ Joins: nadja (~dequbed@banana-new.kilobyte22.de)
L9[02:09:31] ⇨ Joins: jackie (~jackie@banana-new.kilobyte22.de)
L10[02:10:35] <Amanda> Ocvm should work under termux as well
L11[02:10:49] <Amanda> If you compile it anyways
L12[02:11:06] <Amanda> Not sure if Izzy's ocvmaas is still live or not
L13[02:11:26] <Izzy> I think it technically is, but it's not port forwarded.
L14[02:11:40] <Izzy> ie, no.
L15[02:30:52] <Amanda> Hrm
L16[02:31:14] <Amanda> I guess now that I've been up for something like 18h I should slero
L17[02:31:30] * Amanda lays her head on Elfi, does a heccen zzzpurr
L18[02:34:32] <Amanda> Night girls
L19[03:46:12] <Forec​aster> I have finished my slero!
L20[03:46:53] <Forec​aster> %tonk
L21[03:46:53] <MichiBot> Darn! Forec​aster! You beat Spider ​EveryOS's previous record of <0 (By 3 hours, 24 minutes and 22 seconds)! I hope you're happy!
L22[03:46:55] <MichiBot> Forecaster's new record is 3 hours, 24 minutes and 22 seconds! Forecaster also gained 0.00341 tonk points for stealing the tonk. Position #1.
L23[05:37:41] ⇨ Joins: Hawk777 (~Hawk777@2001:569:7ca4:2a00:67e6:3825:3e66:cf27)
L24[06:42:50] ⇦ Quits: Hawk777 (~Hawk777@2001:569:7ca4:2a00:67e6:3825:3e66:cf27) (Quit: Leaving.)
L25[06:58:53] ⇦ Parts: anon (~anon@2401:c080:3000:26e9:5400:4ff:fe90:3cd5) (Leaving))
L26[07:25:14] <Forec​aster> %tonk
L27[07:25:15] <MichiBot> Hooray! Forec​aster! You beat your own previous record of 3 hours, 24 minutes and 22 seconds (By 13 minutes and 59 seconds)! I hope you're happy!
L28[07:25:16] <MichiBot> Forecaster's new record is 3 hours, 38 minutes and 21 seconds! No points gained for stealing from yourself. (Lost out on 0.00023 x 4 = 0.00092)
L29[09:26:56] ⇨ Joins: Vexatos (~Vexatos@p200300eaef1cda0061f71971ffc8081a.dip0.t-ipconnect.de)
L30[09:26:56] zsh sets mode: +v on Vexatos
L31[10:36:07] <MichiBot> Amanda REMINDER: Make borgmatic on whitter not run when the laptop is locked
L32[10:54:00] ⇨ Joins: ben_mkiv (~ben_mkiv@200116b80041530067d6c040cc09978d.dip.versatel-1u1.de)
L33[11:04:26] <Spider ​EveryOS> %tonkout
L34[11:04:26] <MichiBot> Sard! Spider ​EveryOS! You beat Forec​aster's previous record of 3 hours, 38 minutes and 21 seconds (By 50 seconds)! I hope you're happy!
L35[11:04:27] <MichiBot> Spider ​EveryOS has stolen the tonkout! Tonk has been reset! They gained 0.003 tonk points! plus 0.002 bonus points for consecutive hours! (Reduced to 50% because stealing) Current score: 3.03775908. Position #2 Need 0.0390992 more points to pass Forec​aster!
L36[11:09:56] ⇦ Quits: simon816 (~simon816@secondary.machine.simon816.com) (Quit: ZNC 1.9.1 - https://znc.in)
L37[11:11:32] ⇨ Joins: simon816 (~simon816@secondary.machine.simon816.com)
L38[11:15:13] * Amanda meows and looks around
L39[11:32:45] * Amanda beams a couple breakfast bars next to herself for herself and Elfi
L40[11:40:35] * stephanj looks confused at the counter top... didn't I just.... hrm. but where is the bread? weird. i must still be sleepy
L41[11:40:51] * stephanj redoes his breakfast
L42[11:41:12] <stephanj> i just looked away for a second!
L43[11:42:38] * Amanda eats faster
L44[12:08:20] <Forec​aster> %sip
L45[12:08:21] <MichiBot> You drink a dull purple potion (New!). Forecaster gains a negligible amount of luck.
L46[12:11:00] <Amanda> Meow meow meow! %splash @Forecaster with mutable purple potion
L47[12:11:00] <MichiBot> You fling a mutable purple potion (New!) that splashes onto @Forecaster. @Forecaster turns into a cat boy until hell has a mild winter.
L48[12:11:11] <Forec​aster> not again D:<
L49[12:11:15] <Amanda> !
L50[12:11:38] * Amanda enrolls fore back into the training program
L51[12:11:47] <Forec​aster> nooo
L52[12:11:51] <Forec​aster> I don't like training D:
L53[12:13:04] <Amanda> But last time you dropped it just before Finding Sunbeams for Information Gathering 101
L54[12:27:50] <Forec​aster> I can find sunbeams fine already >:
L55[12:30:00] <Amanda> See, that's the neat part. With a cat's good hearing, _any_ sunbeam is great for information gathering, so it's just a Sunbeam nap block!
L56[13:01:46] <Forec​aster> I don't need a class for that > - >
L57[13:23:29] * Elfi eepy moth noises
L58[13:33:34] ⇦ Quits: ben_mkiv (~ben_mkiv@200116b80041530067d6c040cc09978d.dip.versatel-1u1.de) (Quit: Leaving)
L59[13:59:04] <Forec​aster> With some luck I should be able to escape class, and fortunately I have at least a handful of that
L60[14:08:59] <kim​apr> making a little text editor https://discordembeds.pc-logix.com/live/1725113339161.png
L61[14:14:40] <ff​_66> you know there's already `edit` ?
L62[14:14:54] <Corded> > <ff​_66> you know there's already `edit` ?
L63[14:14:54] <kim​apr> It sucks
L64[14:15:32] <ff​_66> then `nano` sucks since the UI is clearly inspired from it
L65[14:16:11] <kim​apr> Yes nano sucks too
L66[14:16:35] <kim​apr> my editor supports tabs and wide characters, `edit` doesn't https://discordembeds.pc-logix.com/live/1725113795204.png
L67[14:34:52] <kim​apr> one problem is that it's very slow to open files. like it takes about 2 seconds to load a 62k character file
L68[14:41:53] <Redston​eParkour> try making it not read the entire file
L69[14:49:37] <Corded> > <Redston​eParkour> try making it not read the entire file
L70[14:49:38] <kim​apr> nah i don't think i wanna do that
L71[14:55:39] <kim​apr> the reason its so slow is that i store text in a complicated Rope structure and i do shittons of logic on every character when building the string
L72[15:02:19] <Redston​eParkour> hm
L73[15:02:56] <Redston​eParkour> thing is you're saying opening the file takes 'around 2s'
L74[15:03:40] <Redston​eParkour> and calculating with a 2048 bytes/gt read speed (the default) you already get 1.5s of just reading the file
L75[15:03:57] <kim​apr> Wait
L76[15:04:18] <kim​apr> maybe that's why it's slow
L77[15:05:35] <kim​apr> Does OCEmu enforce read speed limits though?
L78[15:05:51] <Redston​eParkour> probably
L79[15:06:09] <kim​apr> catting the file to dev null is instant
L80[15:06:15] <Redston​eParkour> huh
L81[15:06:19] <Redston​eParkour> nvm then
L82[15:06:34] <kim​apr> it takes 3 seconds on real OC
L83[15:09:14] <kim​apr> though that one runs Lua 5.3, my OCEmu is on 5.2
L84[15:09:30] <Redston​eParkour> shouldn't matter
L85[15:11:54] <Corded> > <kim​apr> it takes 3 seconds on real OC
L86[15:11:54] <kim​apr> Actually that file is of different size
L87[15:12:18] <Corded> > <Redston​eParkour> shouldn't matter
L88[15:12:18] <kim​apr> Nope, lua 5.3 is considerably faster than 5.2, i think
L89[15:15:29] <kim​apr> This is really weird
L90[15:15:40] <kim​apr> i thought it would crash on invalid UTF-8
L91[15:15:51] <kim​apr> but i put 500kb of /dev/random into it and it opens it
L92[15:22:14] <Corded> > <kim​apr> i thought it would crash on invalid UTF-8
L93[15:22:14] <kim​apr> Oooh i see. OCEmu throws an error, but real OC doesn't
L94[15:25:36] <kim​apr> this means i will need another way to check for broken off utf bytes than just seeing if unicode.charWidth errors out
L95[15:29:21] <Corded> > <kim​apr> making a little text editor
L96[15:29:21] <Ocawes​ome101> Is that a Vi clone I spy?
L97[15:38:02] <Corded> > <Ocawes​ome101> Is that a Vi clone I spy?
L98[15:38:02] <kim​apr> https://discordembeds.pc-logix.com/live/1725118681903.png
L99[15:39:03] <Ocawes​ome101> Ah, something adjacent to that
L100[15:41:10] <Corded> > <Ocawes​ome101> Ah, something adjacent to that
L101[15:41:10] <kim​apr> cloning anything in the Vi family isn't the goal, i take whatever i like from there and mercilessly delete all the stuff i find annoying or useless
L102[15:41:25] <Ocawes​ome101> Respectable
L103[15:42:00] <kim​apr> the file is named "emacs.lua" because i thought it would be funny
L104[15:42:53] <kim​apr> it's nothing like emacs
L105[15:42:53] <kim​apr> unless you're in Evil Mode at all times, though i don't know for sure because i've never used evil mode
L106[15:44:50] <kim​apr> Welp https://discordembeds.pc-logix.com/live/1725119090120.png
L107[15:45:13] <kim​apr> this file isn't actually broken
L108[15:45:59] <kim​apr> the unicode api is quite limited
L109[15:46:30] <kim​apr> doesn't have a :byte equivalent or anything, nor any way to detect invalid encoding
L110[15:48:02] <Ocawes​ome101> 5.3 provides `utf8` also, might be worth looking at that
L111[15:50:03] <kim​apr> relying on that will make my editor 5.3-only
L112[15:50:06] <kim​apr> which is undesirable
L113[15:51:22] <Va​ur> %tonk
L114[15:51:23] <MichiBot> Fopdoodle! Va​ur! You beat Spider ​EveryOS's previous record of <0 (By 4 hours, 46 minutes and 55 seconds)! I hope you're happy!
L115[15:51:24] <MichiBot> Vaur's new record is 4 hours, 46 minutes and 55 seconds! Vaur also gained 0.00478 tonk points for stealing the tonk. Position #3. Need 0.05365644 more points to pass Spider ​EveryOS!
L116[15:53:02] <Corded> > <kim​apr> which is undesirable
L117[15:53:02] <ff​_66> you could make it 5.4-only, if you wanted to
L118[15:53:30] <Corded> > <ff​_66> you could make it 5.4-only, if you wanted to
L119[15:53:31] <kim​apr> that means i can't even use it in OC 🥴
L120[15:53:51] <Corded> > <kim​apr> that means i can't even use it in OC 🥴
L121[15:53:52] <ff​_66> you can since OC 1.8
L122[15:54:29] <ff​_66> just need to change a small part of the config and there you go
L123[15:55:04] <kim​apr> surely there has to be a reason it's not on by default
L124[15:58:35] <kim​apr> Okay i think i have another idea
L125[15:59:04] <Corded> > <kim​apr> surely there has to be a reason it's not on by default
L126[15:59:04] <ff​_66> it is only because it is experimental
L127[16:02:10] <Corded> > <kim​apr> Okay i think i have another idea
L128[16:02:10] <kim​apr> I'll just keep track of the last and first characters in a rope and when concatenating them i'll check if concatenating those edge chars will result in less unicode codepoints than expected
L129[16:06:00] <kim​apr> Oh my god https://discordembeds.pc-logix.com/live/1725120359937.png https://discordembeds.pc-logix.com/live/1725120360096.png
L130[16:10:06] <kim​apr> is OCEmu still maintained?
L131[16:11:25] <Ocawes​ome101> Zenith391's fork is more so
L132[16:11:55] <kim​apr> yeah that's where i'm looking
L133[16:14:22] <kim​apr> this is funny
L134[16:14:52] <kim​apr> `unicode.wlen(unicode.sub(s,1,1))` seems to work the exact same way as `unicode.charWidth(s)` except it doesn't crash on OCEmu
L135[16:23:18] <ff​_66> you can also enhance your program by making things like file tabs, a bit like ComputerCraft's shell tabs
L136[16:26:00] <kim​apr> how does `unicode.sub` manage to be really slow when you're taking a small substring from the very beginning of a very big string? the obvious way to implement it (which also seems to be what OCEmu does (as it has the performance trends i expect), but not OC) would be to iterate over the string and decode characters and put them in the
L137[16:26:01] <Corded> new string, which would make slicing slower the further away from the beginning of a string you slice
L138[16:28:38] <Ocawes​ome101> OCEmu probably does it in Lua where OC does it in C
L139[16:28:53] <Ocawes​ome101> That would be where the performance discrepancy comes from
L140[16:29:32] <kim​apr> if doing it in C is slower than Lua it should be done in Lua instead
L141[16:29:50] <Ocawes​ome101> Doing it in C is faster which is why OC does it in C
L142[16:31:59] <Corded> > <Ocawes​ome101> Doing it in C is faster which is why OC does it in C
L143[16:32:00] <kim​apr> It isn't, it seems. first image is OCEmu, second is OC https://discordembeds.pc-logix.com/live/1725121919262.png https://discordembeds.pc-logix.com/live/1725121919451.png
L144[16:32:06] <kim​apr> C won't save you from bad algorithms
L145[16:32:56] <Ocawes​ome101> String access in C is a _lot_ faster
L146[16:33:05] <Ocawes​ome101> It is possible OC does something different
L147[16:33:53] <kim​apr> OC does something *bad*
L148[16:36:53] <ff​_66> language is only a small factor in the algorithm speed, the most important one being the efficiency. If OCEmu has implemented better the functions, it is the fastest.
L149[16:38:21] <Corded> > <kim​apr> It isn't, it seems. first image is OCEmu, second is OC
L150[16:38:21] <Ocawes​ome101> The way I'm reading this OC takes like 1/5 the time of OCEmu?
L151[16:38:45] <Ocawes​ome101> https://github.com/MightyPirates/OpenComputers/blob/master-MC1.7.10/src/main/scala/li/cil/oc/server/machine/luac/UnicodeAPI.scala here's OC's Unicode implementation; it's actually Scala
L152[16:38:55] <Corded> > <Ocawes​ome101> The way I'm reading this OC takes like 1/5 the time of OCEm…
L153[16:38:55] <kim​apr> OC takes many orders of magnitude more time
L154[16:39:33] <Ocawes​ome101> Oh I missed the `e`
L155[16:39:43] <Ocawes​ome101> Huh
L156[16:40:31] <kim​apr> I think the issue here is the Lua - C(/jvm) boundary, across which the entire huge string has to be cloned
L157[16:41:15] <Ocawes​ome101> OCEmu wraps utf8.sub as unicode.sub
L158[16:41:46] <Ocawes​ome101> https://github.com/zenith391/OCEmu/blob/master/src/apis/unicode.lua#L42-L49
L159[16:41:58] <Ocawes​ome101> So OCEmu is indeed the C implementation
L160[16:43:07] <Corded> > <ff​_66> language is only a small factor in the algorithm speed, the…
L161[16:43:07] <kim​apr> This btw
L162[16:45:04] <Corded> > <kim​apr> This btw
L163[16:45:05] <ff​_66> what did i say wrong here ?
L164[16:45:43] <Corded> > <ff​_66> what did i say wrong here ?
L165[16:45:44] <kim​apr> you didn't, i'm agreeing with you
L166[16:45:56] <kim​apr> though language *is* a factor
L167[16:46:46] <Ocawes​ome101> ^^ If getting one subcharacter of a string for comparison requires a memory allocation versus just a memory index, that could make up a considerable distance - though perhaps not quite one so big
L168[16:49:07] <Ocawes​ome101> Doing lots of repeated memory allocations, like repeatedly indexing a string in Lua, can get really slow. Most of the time you don't notice it but for iterating over really long strings like this I think it conceivably could.
L169[16:49:31] <Redston​eParkour> ff_66 i believe the way ocemu goes faster is by (when fast mode is on) not restricting component calls
L170[16:50:09] <Ocawes​ome101> Not relevant for Unicode
L171[16:50:16] <kim​apr> yeah, this isn't relevant here
L172[16:50:39] <kim​apr> no component calls are made while building and slicing a huge string
L173[16:50:54] <Redston​eParkour> correct
L174[16:51:07] <Redston​eParkour> only matters when reading the file from disk
L175[16:52:34] <kim​apr> also, OCEmu seems to crash completely when a "too long without yielding" error happens rather than just killing the coroutine that caused it
L176[16:53:16] <Redston​eParkour> oh yeah
L177[16:53:25] <Redston​eParkour> you know when i said it doesn't restrict component calls?
L178[16:53:56] <Redston​eParkour> it also doesn't seem to consider you to have yielded when a component call should yield
L179[16:54:23] <Redston​eParkour> and yeah that crashes it
L180[16:55:34] <kim​apr> huh. doesn't always happen https://discordembeds.pc-logix.com/live/1725123333707.png
L181[16:55:54] <Redston​eParkour> hm
L182[16:56:28] <Redston​eParkour> try doing a `while true do component.gpu.set(1,1," ") end`
L183[17:00:30] <Corded> > <Redston​eParkour> try doing a `while true do component.gpu.set(1,1," ") end`
L184[17:00:30] <kim​apr> `while true do print() end` crashes OCEmu, locks up OC
L185[17:00:41] <Redston​eParkour> right
L186[17:00:58] <kim​apr> if you wrap it in a pcall, then sleep, and then display the error, OCEmu doesn't crash
L187[17:01:22] <Redston​eParkour> ah
L188[17:02:23] <kim​apr> if you do the same without the sleep OCEmu will crash
L189[17:25:23] <Kristo​pher38> Have you considered using ocelot
L190[17:26:56] <Corded> > <Kristo​pher38> Have you considered using ocelot
L191[17:26:57] <kim​apr> what is this?
L192[17:27:54] <kim​apr> okay it seems great, i think i would try it
L193[17:29:10] <kim​apr> is ocelot more resource-intensive than OCEmu?
L194[17:30:10] <Kristo​pher38> It's certainly more accurate
L195[17:31:25] <kim​apr> Ooh there's a web version
L196[17:35:00] <kim​apr> i pressed Ctrl + W and it just closed the tab 😭
L197[17:35:09] <kim​apr> instead of closing edit
L198[17:39:48] <fingercomp> use Ctrl + E for that
L199[17:39:57] <fingercomp> we're remapped the key precisely for this reason
L200[17:41:06] <fingercomp> ...or you could try Ocelot Desktop, which is much more powerful than the web demo
L201[17:45:55] <ff​_66> because it runs on your CPU
L202[17:46:16] <Corded> > <kim​apr> is ocelot more resource-intensive than OCEmu?
L203[17:46:17] <Ocawes​ome101> Perhaps slightly but it makes up for that by being way more accurate
L204[17:46:23] <Ocawes​ome101> Also i3 user spotted
L205[17:49:26] <Corded> > <ff​_66> because it runs on your CPU
L206[17:49:26] <The P​atmann> I'm pretty sure that Ocelot Web doesn't utilize cloud computing, it just ends up having to be a webapp which historically has more fuss between it and the baremetal.
L207[17:49:38] <The P​atmann> I've not investigated it personally, so I may be wrong
L208[17:51:01] <fingercomp> ocelot online runs on our server; the browser just renders whatever's on the screen
L209[17:51:19] <fingercomp> the instance is shared among all users, btw
L210[17:51:38] <The P​atmann> Oh, neat!
L211[17:53:21] * Amanda looks around, wonders how it became 1400 already
L212[17:53:26] <fingercomp> just in case: we've had some folks repeatedly breaking it by overwriting the eeprom with some junk, after which the computer instance won't boot
L213[17:53:35] <Amanda> It was just 0800 and I started reading more of a story
L214[17:53:49] <fingercomp> so if you see the blank screen and pressing the "turn on" button does nothing, ping me here on IRC, and I'll reset the instance
L215[17:59:06] <Forec​aster> That might be because I've been pawing at this clock I found, but probably not
L216[18:02:25] <Corded> > <The P​atmann> Oh, neat!
L217[18:02:26] <ff​_66> I knew it
L218[18:03:15] <Amanda> I mean, it's not outside the realm of possibility to be running entirely in the browser, just probably way more work than it's worth to run a funny MC computer
L219[18:03:32] <kim​apr> fingercomp: small issue found: invalid utf-8 doesn't render as replacement characters
L220[18:05:05] <Corded> > <kim​apr> fingercomp: small issue found: invalid utf-8 doesn't render…
L221[18:05:06] <ff​_66> if invalid utf-8 codes are found it should load it as an ASCII, OEM or ANSI encoding
L222[18:07:42] <Corded> > <ff​_66> if invalid utf-8 codes are found it should load it as an AS…
L223[18:07:42] <kim​apr> that's not what OC does
L224[18:07:47] <fingercomp> @kimapr I presume you're talking about ocelot-online? as it's just a demo, it has quite a few bugs related to screen rendering, so I wouldn't be surprised
L225[18:08:00] <kim​apr> yeah, i'm talking about ocelot-online
L226[18:08:37] <kim​apr> the invalid characters seem to occupy the appropriate amount of space, but display as empty
L227[18:09:27] <kim​apr> seems like `unicode.sub` is broken in OCEmu as well
L228[18:10:26] <kim​apr> it *skips* over any invalid characters
L229[18:11:23] <fingercomp> I'd suggest using Ocelot Desktop if you care about emulation accuracy; we haven't really been working on the online version very much
L230[18:31:40] ⇨ Joins: Keroe (~Keroe@lim4.joinserver.xyz)
L231[18:32:08] ⇦ Quits: Keroe (~Keroe@lim4.joinserver.xyz) (Client Quit)
L232[18:32:56] <Corded> > <kim​apr> it *skips* over any invalid characters
L233[18:32:56] <kim​apr> weird, i can't repro that anymore
L234[18:39:13] <Forec​aster> %sip
L235[18:39:14] <MichiBot> You drink a powdery pearlpeas potion (New!). Forecaster turns into a bear spaghetti boy until they sneeze.
L236[18:39:41] <Forec​aster> well that's easier to fix than the last one
L237[18:39:53] <Amanda> D:
L238[18:40:01] <Amanda> How did FSM's kid get in the training hall!?
L239[18:44:12] <Amanda> %choose crackers or nilla wafers
L240[18:44:12] <MichiBot> Ama​nda: The proof is in the pudding. Definitely "nilla wafers". Now please get it out of my pudding.
L241[18:44:26] * Amanda adds more nilla wafers to the pudding
L242[18:45:27] <kim​apr> this is so bad https://discordembeds.pc-logix.com/live/1725129927527.png
L243[18:45:35] <kim​apr> wlen literally lying
L244[18:50:19] <Corded> > <kim​apr> this is so bad
L245[18:50:19] <Na​dja> Really, this is your fault for daring to speak a language that can't be pressed into 7-bit ASCII! (/s in case it isn't obvious)
L246[18:51:17] * Amanda meows hello to nadja, quietly standing between her and the minor tear in space-time she made while using her scratching planet
L247[18:52:29] <nadja> Amanda, keep this up and I will press you into a 7-bit plane. Now go close that rift.
L248[18:52:37] <nadja> And give me that planet, planets aren't toys.
L249[18:53:00] * Amanda pushes a napkin over the tear, quietly slinks out of the room
L250[18:59:56] <Corded> > <na​dja> Really, this is your fault for daring to speak a language t…
L251[18:59:56] <kim​apr> that on its own works fine, the problems occur when you feed OCEmu invalid UTF-8
L252[19:02:42] <Forec​aster> well stop feeding it junk!
L253[19:02:48] <Forec​aster> it's unhealthy
L254[19:05:51] <Amanda> OCEMu doens't use the same unicode libraries that OC does, it's literally a thin wrapper around Love2D AIUI
L255[19:06:29] <kim​apr> Love2D? it seems to use a lua SDL wrapper
L256[19:07:36] <Amanda> I might be thinking about another one then
L257[19:09:21] <Amanda> Either way, it's still using different code than oc itself does
L258[19:10:03] <Amanda> If you want the most accurate emulation, use ocelot. They literally take oc and rip a lot of the Minecraft stuff out and fake it
L259[19:10:42] <Amanda> So it's using the same lua binaries and libraries as oc itself
L260[19:10:55] <Amanda> AIUI anyways
L261[19:24:13] <Corded> > <na​dja> Really, this is your fault for daring to speak a language t…
L262[19:24:14] <ff​_66> and my language ? ASCII has been designed mainly for english !!!
L263[19:30:46] <Corded> > <ff​_66> and my language ? ASCII has been designed mainly for englis…
L264[19:30:46] <Na​dja> Same, איך אתה מעז?!
L265[19:31:38] <ff​_66> oh no, a language that is written right to left and not left to right !
L266[19:37:12] <Corded> > <kim​apr> this is so bad
L267[19:37:13] <Redston​eParkour> you're taking the last byte of a wide character then yeeting a `[` after it and then taking the first character
L268[19:37:29] <Redston​eParkour> i'm actually suprised that didn't result in an invalid unicode character
L269[19:47:47] <Amanda> I'm pretty sure it did, and that's what they're trying to say, that OCEmu's unicode.wlen isn't processing invalid UTF-8 the same as OC would
L270[19:53:43] <Forec​aster> Saying things on the internet?! What is the world coming to...
L271[19:57:20] <Amanda> %choose wake the laptop for stories?
L272[19:57:20] <MichiBot> Ama​nda: A faraway lamp turns pink
L273[19:57:40] * Amanda stares at the lamp, wondering why that lamp's suddenly a different colour
L274[19:57:44] <Corded> > <Redston​eParkour> i'm actually suprised that didn't result in an invalid unic…
L275[19:57:44] <kim​apr> It did result in an invalid unicode character, but OCEmu displays it as zero-width, then claims it would actually be displayed 1-width
L276[19:58:47] <kim​apr> apparently OC's unicode.sub turns invalid UTF-8 into replacement characters
L277[19:59:18] <kim​apr> .. this is actually not what i want
L278[20:02:27] <Corded> > <Ama​nda> stares at the lamp, wondering why that lamp's suddenly a di…
L279[20:02:27] <Forec​aster> I stepped on the remote
L280[20:31:25] <Amanda> Stories: Updated, laptop: Back to naptime
L281[20:35:18] <Diz​hka> https://tenor.com/view/skull-issues-gif-13031152103567454559
L282[21:16:14] <Izzy> krill issue
L283[21:17:23] <Spider ​EveryOS> %tonk
L284[21:17:23] <MichiBot> Consarn it! Spider ​EveryOS! You beat Va​ur's previous record of 4 hours, 46 minutes and 55 seconds (By 39 minutes and 5 seconds)! I hope you're happy!
L285[21:17:24] <MichiBot> Spider EveryOS's new record is 5 hours, 26 minutes and 1 second! Spider EveryOS also gained 0.00325 (0.00065 x 5) tonk points for stealing the tonk. Position #2. Need 0.0358492 more points to pass Forec​aster!
L286[21:31:38] ⇦ Quits: Vexatos (~Vexatos@p200300eaef1cda0061f71971ffc8081a.dip0.t-ipconnect.de) (Quit: Insert quantum chemistry joke here)
L287[21:41:37] ⇨ Joins: Gay (~Gay@91.144.161.160)
L288[21:43:59] ⇦ Quits: Gay (~Gay@91.144.161.160) (Client Quit)
L289[22:26:11] <kim​apr> so i pulled out the computronics Magical Memory from my computer and.. https://discordembeds.pc-logix.com/live/1725143171333.png
L290[22:26:53] <kim​apr> this thing managed to load 65.7k characters from the file before it died
L291[22:27:16] <kim​apr> there's 2MB ram installed on the computer..
L292[22:27:40] <kim​apr> does OC force garbage collection when it runs out of memory?
L293[22:28:45] <kim​apr> even after my program is gone the memory is still not being freed, and everything that needs more just crashes
L294[22:35:09] <Corded> > <kim​apr> does OC force garbage collection when it runs out of memory?
L295[22:35:10] <Kristo​pher38> From my experience - yes; if there's not enough memory for new allocation and it can garbage collect and the new allocation will fit in the freed region it will still continue to work
L296[22:37:18] <kim​apr> Wait
L297[22:37:26] <kim​apr> the global environment, is actually global?
L298[22:37:46] <Elfi> Frankly impressed the computer doesn't just halt like most do when you rip out their memory
L299[22:37:46] <kim​apr> https://discordembeds.pc-logix.com/live/1725143865954.png
L300[22:38:23] <Kristo​pher38> More accurate to say "per computer"
L301[22:38:25] <Corded> > <kim​apr> the global environment, is actually global?
L302[22:38:25] <Jas​onS> Uh yeah, the sandboxing isn’t exactly the greatest lol
L303[22:38:26] <Elfi> ...there might be a bug in computronics :'D
L304[22:39:13] <kim​apr> what bug?
L305[22:39:36] <Elfi> Idk, last time I played there wasn't even magical memory
L306[22:39:41] <kim​apr> Magical Memory is just like regular OC memory but bigger
L307[22:40:14] <Elfi> But causing the entire memory space to go higgledy-piggledy certainly points to something going wrong
L308[22:40:26] <kim​apr> whag?
L309[22:40:57] <Elfi> tl;dr glhf
L310[22:41:05] <kim​apr> removing magical memory only revealed that my program is VERY hungry for memory
L311[22:41:19] <kim​apr> like google chromium
L312[22:41:54] <kim​apr> and also i just found out that the global environment is shared with the entire system
L313[22:42:57] <Corded> > <Jas​onS> Uh yeah, the sandboxing isn’t exactly the greatest lol
L314[22:42:57] <kim​apr> per-process global env feels like the obvious thing to do 🥴 pretty sure the OpenOS shell is even capable of this
L315[22:44:56] <Elfi> Well, there's the joy of OC over CC
<<Prev Next>> Scroll to Top