<<Prev Next>> Scroll to Bottom
Stuff goes here
L1[00:14:58] <Kristo​pher38> I wanna connect it to something relevant but don't know what could it be, I don't have anything that is capable of outputting serial but not graphics by design
L2[00:22:45] <Kristo​pher38> If I had a breadboard computer on the other hand...
L3[00:33:25] ⇨ Joins: Base (~Base@168.195.5.187)
L4[00:37:41] ⇦ Quits: Base (~Base@168.195.5.187) (Ping timeout: 189 seconds)
L5[00:41:32] <Va​ur> %tonk
L6[00:41:33] <MichiBot> Consarn it! Va​ur! You beat Mic​hiyo's previous record of 2 hours, 2 minutes and 58 seconds (By 26 minutes and 20 seconds)! I hope you're happy!
L7[00:41:34] <MichiBot> Vaur's new record is 2 hours, 29 minutes and 19 seconds! Vaur also gained 0.00132 (0.00044 x 3) tonk points for stealing the tonk. Position #1.
L8[00:44:44] <Amanda> @Kristopher38 time to write a tui frontend for something!
L9[00:49:33] <Ar​iri> @lunar_sam thought you might like this https://cdn.discordapp.com/attachments/464810688553025546/929996090340683846/cars.mov
L10[00:56:35] <Kristo​pher38> If only I could pipe OC output to the serial port 🤔
L11[01:12:43] <luna​r_sam> >Ariri: <@!175686996461617162> thought you might like this https://cdn.discordapp.co
L12[01:12:44] <luna​r_sam> i love it
L13[01:23:51] <Amanda> @kristopher38 with some light patching you could probably get ocvm working fine under a serial terminal
L14[01:24:21] <Amanda> Namely, replacing Unicode glyphs with their equivalent in whatever codepage it uses
L15[01:24:53] <Amanda> And rerenderimg the rest as ?
L16[01:36:20] ⇦ Quits: Vexatos (~Vexatos@port-92-192-6-121.dynamic.as20676.net) (Quit: Insert quantum chemistry joke here)
L17[01:36:36] ⇨ Joins: Base (~Base@168.195.5.187)
L18[01:36:51] <Base> hello
L19[01:36:58] <Base> hola
L20[01:37:01] <Base> eae
L21[01:37:07] <Base> conitiua
L22[01:37:13] <Amanda> How'd you get in here! I told Forecaster to lock the door!
L23[01:40:15] <Amanda> %choose laptop nap time or comfort
L24[01:40:16] <MichiBot> Ama​nda: I received a telegram from a long lost relative that only read "comfort". Weird.
L25[01:40:53] <Amanda> MichiBot I think you're getting spam from "relatives"
L26[01:44:50] <Amanda> If only it were this easy:
L27[01:45:22] <Amanda> https://matrix.camnet.site/_matrix/media/r0/download/camnet.site/zFwZQwkwlEXYgqkXGjkUOufV/screenshot-1641865483.png
L28[01:46:33] <Base> See you later I'm going to sleep or I don't know
L29[01:47:03] <Amanda> ta-ta
L30[01:47:56] ⇦ Quits: Base (~Base@168.195.5.187) (Quit: Base)
L31[02:19:49] <Ar​iri> Amanda: what game?
L32[02:21:07] <Amanda> @ariri bitburner, free on steam
L33[02:22:23] <Ar​iri> might check it out, thanks
L34[02:22:27] <Ar​iri> %pet Amanda
L35[02:22:27] <MichiBot> Ariri is patting Amanda with a Magic furai furai furai! (25%). Amanda regains 1d4 => 2 (Magic +2) => 4 hit points!
L36[02:22:54] <Amanda> It's like an idle hacking game
L37[02:23:57] <Ar​iri> humu
L38[02:25:51] ⇦ Parts: lunar_sam (c44a7f2987@jabberfr.org) ())
L39[02:26:20] ⇨ Joins: lunar_sam (c44a7f2987@jabberfr.org)
L40[03:01:49] ⇨ Joins: DBotThePony (~Thunderbi@31.220.170.28)
L41[03:06:56] <Comput​erCoco> So how does power in OC work? I know on the wiki it say 14.3RF is 1"OpenComputersPowerUnit" but how would I calculate how much power a computer uses per tick?
L42[03:06:56] <Comput​erCoco> I know of the `computer.energy()` function, and thought a good idea would be running that on loop (on a tablet) and seeing how the power changes, but the fact of running the function costs power itself and actually drains the tablet waaayy faster than idling...
L43[03:07:35] <Comput​erCoco> I also tried running `computer.energy()-computer.energy()` to see what the differnece in energy would be after running it to find how much one check consumed, but that outputs 0
L44[03:10:28] <Comput​erCoco> Like... how much energy does one command take? or how much energy does a computer use per tick? how does it calculate that?
L45[03:24:17] <Amanda> It depends an a very large number of variables
L46[03:25:27] <Comput​erCoco> is there anywhere to see what they are besides the source code?
L47[03:25:53] <Amanda> If you're changing the screen a lot, or have a very large number of bright pixels, or send a lot of network packets, etc etc all changes it
L48[03:27:21] <Comput​erCoco> Wait it even formulates in the bright number of pixels?
L49[03:27:21] <Comput​erCoco> Is there like... a list anywhere of what it all changes it?
L50[03:30:43] <Amanda> I'm pretty sure someone found out that an all black screen used less. In any case, turning off the screen uses less power
L51[03:33:35] <Amanda> Anyway, I don't know if any such compiled list exists, and it's not really calculated per tick exactly, rather only the base rate is, making component calls and running the CPU costs more
L52[03:34:02] <Amanda> That said if you wanted to make one, in sure it'd be welcome on the ocdoc wiki
L53[03:34:48] <Comput​erCoco> Hmm
L54[03:34:48] <Comput​erCoco> Ill let you know if I do :3
L55[03:35:22] <Comput​erCoco> Running tests rn and defently seems like events and processing them take up a lot of the power
L56[03:35:44] <Comput​erCoco> But probably would have to go through the source code to get the exact values
L57[04:02:59] <Comput​erCoco> Does OpenComputers delete unused filesystem from your actual computer to save space?
L58[04:03:00] <Comput​erCoco> I just deleted a hard drive in game with the address "dda...", but in my opencomputers folder, its still there?
L59[04:03:01] <Amanda> I'm pretty sure it's configurable
L60[04:03:20] <Amanda> The power usage that is
L61[04:04:03] <Amanda> And how did you delete the disk? There's some ways that OC can't track deletion and clean up
L62[04:04:32] <Amanda> Like throwing it into the creative inventory iirc
L63[04:04:58] <Comput​erCoco> Just in my inventory?
L64[04:04:58] <Comput​erCoco> Like with the creative trashcan thing
L65[04:04:58] <Comput​erCoco> I saved and quit as well just to see if that owuld do anything
L66[04:05:37] <Comput​erCoco> >Amanda: Like throwing it into the creative inventory iirc
L67[04:05:38] <Comput​erCoco> is there not some backup that scans for all filesystem items in the game and deletes all the unused ones?
L68[04:06:19] <Amanda> How would you check if the item still exists? It could be in a chest in an unloaded chunk
L69[04:06:51] <Amanda> The clean up is best effort, so usual survival methods are handled
L70[04:06:54] <Michiyo> %tonk
L71[04:06:55] <MichiBot> Holy pentagram pendant Batman! Mic​hiyo! You beat Va​ur's previous record of 2 hours, 29 minutes and 19 seconds (By 56 minutes and 2 seconds)! I hope you're happy!
L72[04:06:56] <MichiBot> Michiyo's new record is 3 hours, 25 minutes and 21 seconds! Michiyo also gained 0.00279 (0.00093 x 3) tonk points for stealing the tonk. Position #4. Need 0.02371 more points to pass Forec​aster!
L73[04:07:10] <Michiyo> damn it 4 minutes
L74[04:07:14] <Amanda> Like chucking it into fire or leave
L75[04:07:21] <Amanda> Lava*
L76[04:07:34] <Comput​erCoco> Hmm
L77[04:07:35] <Amanda> %remindme 9h update your stories
L78[04:07:36] <MichiBot> I'll tell you "update your stories" in 9h at 01/11/2022 01:07:35 PM
L79[04:08:00] * Amanda collapses into a pile of floof containing 1 (one) Elfi fairy fren
L80[04:08:08] <Amanda> Night nerds
L81[04:08:14] <Michiyo> Night
L82[04:48:48] ⇨ Joins: s_a_m (~sam@172.58.236.226)
L83[04:50:56] <Comput​erCoco> made a program that logs energy usage per second over a period of time then averages it after
L84[04:50:56] <Comput​erCoco> I've got mostly averages of 1, but then got 2 a few times and then 5 once?
L85[04:50:56] <Comput​erCoco> Weird
L86[05:07:21] ⇨ Joins: flappy (~flappy@91-154-0-54.elisa-laajakaista.fi)
L87[05:50:29] <Comput​erCoco> Message contained 4 or more newlines and was pastebined https://paste.pc-logix.com/ececepuqej
L88[06:22:16] ⇦ Quits: s_a_m (~sam@172.58.236.226) (Ping timeout: 195 seconds)
L89[06:45:22] ⇨ Joins: ThePiGuy24 (~ThePiGuy2@host-92-17-122-58.as13285.net)
L90[06:45:57] ⇦ Quits: TPG24 (~ThePiGuy2@host-92-17-122-58.as13285.net) (Ping timeout: 189 seconds)
L91[07:15:52] ⇦ Quits: lordpipe (~ba7888b72@66.109.211.50) (Remote host closed the connection)
L92[07:19:37] ⇨ Joins: lordpipe (~ba7888b72@66.109.211.50)
L93[08:04:54] <Va​ur> %tonk
L94[08:04:55] <MichiBot> Boo-yah! Va​ur! You beat Mic​hiyo's previous record of 3 hours, 25 minutes and 21 seconds (By 32 minutes and 38 seconds)! I hope you're happy!
L95[08:04:56] <MichiBot> Vaur's new record is 3 hours and 58 minutes! Vaur also gained 0.00216 (0.00054 x 4) tonk points for stealing the tonk. Position #1.
L96[08:17:34] ⇨ Joins: ben_mkiv (~ben_mkiv@2001:16b8:1494:b000:fe34:97ff:fea9:75f2)
L97[08:39:25] ⇦ Quits: Hawk777 (~chead@2607:c000:8277:2e00:2383:fed9:7bdb:14a1) (Quit: Leaving.)
L98[09:34:17] <KOR_A​PUcard> Is this mod is still will not ported or updated to 1.16.5?
L99[09:35:19] <Mic​hiyo> Correct, several people have started doing so, then quickly gave up
L100[09:36:18] <KOR_A​PUcard> No way, it seems like a Bermuda Triangle!
L101[09:37:33] <Mic​hiyo> It's a fairly complex mod, written in a language that is no longer natively supported by forge.
L102[09:37:38] <Mic​hiyo> it's basically a total rewrite
L103[10:37:06] ⇨ Joins: Vexatos (~Vexatos@port-92-192-6-121.dynamic.as20676.net)
L104[10:37:06] zsh sets mode: +v on Vexatos
L105[11:27:48] ⇦ Parts: lunar_sam (c44a7f2987@jabberfr.org) ())
L106[11:28:33] ⇨ Joins: lunar_sam (c44a7f2987@jabberfr.org)
L107[11:33:27] <Va​ur> %sip
L108[11:33:28] <MichiBot> You drink a tiny pear potion (New!). When Vaur brings the bottle down they see A Silver colored plastic flamingo. It stares into their soul.
L109[12:14:39] <Izaya> @Ariri, CompanionCube: cashed out my doge, sitting on an extra 250 doge that I'm going to turn into more money
L110[12:14:51] <Izaya> (or try to, anyway)
L111[12:15:52] <Va​ur> %tonkout
L112[12:15:52] <MichiBot> Awesome! Va​ur! You beat your own previous record of 3 hours and 58 minutes (By 12 minutes and 57 seconds)! I hope you're happy!
L113[12:15:53] <MichiBot> Va​ur has tonked out! Tonk has been reset! They gained 0.004 tonk points! plus 0.006 bonus points for consecutive hours! Current score: 0.08563, Position #1
L114[12:36:11] ⇦ Quits: ThePiGuy24 (~ThePiGuy2@host-92-17-122-58.as13285.net) (Ping timeout: 195 seconds)
L115[12:43:54] ⇨ Joins: ThePiGuy24 (~ThePiGuy2@host-92-17-122-58.as13285.net)
L116[13:00:54] <kapo​w_12> Hello I was wondering why curseforge hasn't been updated since 2019?
L117[13:01:52] <kapo​w_12> Also would it make sense to switch the default branch on GitHub to track the latest version instead of 1.7.10?
L118[13:07:36] <MichiBot> Amanda REMINDER: update your stories
L119[13:09:05] <kapo​w_12> >Michiyo: It's a fairly complex mod, written in a language that is no longer natively…
L120[13:09:06] <kapo​w_12> On the MinecraftForge GitHub page there is a repo called Scorge https://github.com/MinecraftForge/Scorge which appears to be maintained by the Forge team. It seems like they just no longer ship it by default
L121[13:18:30] <kapo​w_12> just an idea lmk if you think it's feasible
L122[13:30:10] ⇨ Joins: serega (~serega@178.155.6.16)
L123[13:30:19] ⇦ Quits: serega (~serega@178.155.6.16) (Client Quit)
L124[14:30:43] ⇨ Joins: s_a_m (~sam@172.58.236.226)
L125[14:34:43] ⇦ Quits: s_a_m (~sam@172.58.236.226) (Ping timeout: 189 seconds)
L126[14:41:22] ⇨ Joins: s_a_m (~sam@172.58.236.226)
L127[14:53:16] ⇦ Quits: s_a_m (~sam@172.58.236.226) (Ping timeout: 195 seconds)
L128[14:54:09] ⇨ Joins: TPG24 (~ThePiGuy2@host-92-17-122-58.as13285.net)
L129[14:55:32] ⇦ Quits: ThePiGuy24 (~ThePiGuy2@host-92-17-122-58.as13285.net) (Ping timeout: 189 seconds)
L130[15:25:42] <Amanda> If it was as feasible/ simple as you seem to think, One of the at least three attempts so far to port it to 1.16 would have panned out, @kapow_12
L131[15:27:05] <Ko​dos> %TONK
L132[15:27:05] <MichiBot> Boom! Ko​dos! You beat Va​ur's previous record of <0 (By 3 hours, 11 minutes and 13 seconds)! I hope you're happy!
L133[15:27:06] <MichiBot> Kodos's new record is 3 hours, 11 minutes and 13 seconds! Kodos also gained 0.00319 tonk points for stealing the tonk. Position #2. Need 0.01995 more points to pass Va​ur!
L134[15:27:08] <Amanda> AIUI the original plan for 1.13+ anyway was a clean break from OC, into an OC2 with some structural changes, but probably still based on lua. @Sangar just happen to take that a different direction with the OC2 (Is is still a working title?) he ended up putting up
L135[15:30:27] <Amanda> And the 1.7.10 branch is the default on github because it's still the primary branch, every other version gets merged ontop of that
L136[15:30:38] <luna​r_sam> ~~i can't be the only person who finds scala hard to read, right?~~
L137[15:31:41] <Amanda> No, which is why the clean break was going to be written in Java anyway.
L138[15:32:20] <Amanda> they found that the scala part was seriously hindering people from submitting stuff. so a clean break in a new codebase for 1.13+ was planned, then Life Happen™
L139[15:34:39] <luna​r_sam> ic
L140[15:41:32] <Amanda> OC1 for 1.13+ was never going to happen, it was going to be a clean break which didn't even nessarially suppoet backward compat
L141[15:42:38] ⇨ Joins: s_a_m (~sam@172.58.236.226)
L142[15:45:28] <Va​ur> I'd be happy with oc2 even without any backward compatibility tbh
L143[15:46:17] <Amanda> OC1 accumulated so much technical debt across the MC versions it supported, it was inevitable that a clean break would be nessary eventually
L144[15:47:36] <Amanda> Life Happening to @payonel was probably just the final nail in the coffin
L145[15:49:26] <Ocawes​ome101> if i knew java i might try my hand writing an OC-like - but i don't know java really at all
L146[15:49:43] <Amanda> I appologise @payonel and @Sangar if I divulged anything that I was mean to keep in cofidence, I'm just so sick of kids coming in and bitching and moaning that it turned out this way
L147[15:49:54] <Ocawes​ome101> hmm, in that vein, does OC2 have a way to do custom architectures?
L148[15:50:27] <Ocawes​ome101> if it does, one could just make a Lua arch for it
L149[15:51:07] <Amanda> judging from some of Sangar's github activity, he may be working on a eeprom that runs lua, but AFAIK no it doesn't support additional arches
L150[15:51:14] <Ocawes​ome101> ah
L151[15:51:30] <Ocawes​ome101> custom architecture support would be very cool
L152[15:51:59] <Amanda> It's also a large part of what added to OC1's technical debt. Technically anything that can run on risc-v embedded could be an "Arch" though
L153[15:52:03] <Z0id​berg> Isn't OC2 RISC based?
L154[15:52:07] <Ocawes​ome101> it is, yes
L155[15:52:11] <Z0id​berg> I don't think OC2 needs custom architectures then
L156[15:52:20] <Z0id​berg> It should be easy enough to emulate any architecture from it
L157[15:52:24] <Ocawes​ome101> fair enough
L158[15:53:11] <Z0id​berg> I'm still thinking if I sued use my custom arch for a fabric kotlin mod
L159[15:53:14] <Z0id​berg> but that's word
L160[15:53:17] <Z0id​berg> work*
L161[15:53:22] <Z0id​berg> I have too many projects 😦
L162[15:53:28] <Ocawes​ome101> of course you should :P
L163[15:53:43] <Z0id​berg> Well I don't see many people being interested in stack machines for one
L164[15:53:51] <Z0id​berg> it would be a very very simple system though
L165[15:54:12] <Ocawes​ome101> i'm working(ish) on a 24-bit arch which could be just the right amount of complexity for minecraft
L166[15:54:28] <Ocawes​ome101> still gotta get the C emulator put together though and make sure it's feasible for use
L167[15:54:35] <Z0id​berg> The current MMU/DMA architecture for my cpu arch is 16 bit local 24 bit external.
L168[15:54:44] <Z0id​berg> so it maps 16 MB of space in 64 KB chunks
L169[15:55:03] <Z0id​berg> You know @Ocawesome101 ...
L170[15:55:08] <Z0id​berg> I've been thinking about writing a C compiler
L171[15:55:27] <Z0id​berg> and the idea is that it will compile C into a "C" AST.
L172[15:55:36] <Ocawes​ome101> give it an easily customizable code generation backend and i might use it! :P
L173[15:56:11] <Z0id​berg> It'd be in Elixir, so it could just compile C into an AST that can then be recursively fed to generate data for whatever linker format you wanted etc
L174[15:56:27] <Z0id​berg> Elixir has amazing binary data packing ability
L175[15:56:34] <Z0id​berg> and it works to the bit
L176[15:56:45] <Z0id​berg> so I can pack a data in say a 19 bit value.
L177[15:56:50] * Amanda flops down around Elfi, contemplates the number 42
L178[15:56:53] <Ocawes​ome101> my arch supports MMIO, has 6 16-bit ports, and it uses what i'm calling a "segment descriptor table" for sort-of memory protection; the CPU can fire an interrupt when entering a page marked as "protected" and let software do the rest
L179[15:57:13] <Ocawes​ome101> >Z0idberg: so I can pack a data in say a 19 bit value.
L180[15:57:13] <Ocawes​ome101> that's something i sometimes wish i had in e.g. Lua
L181[15:58:10] <Ocawes​ome101> >Ocawesome101: my arch supports MMIO, has 6 16-bit ports, and it uses what i'm calling a "segm…
L182[15:58:11] <Ocawes​ome101> (also 10 general purpose registers and 6 "attribute registers" that store specific information about what's going on in the processor)
L183[15:59:30] <Z0id​berg> User programs with my arch doesn't really make use of the external MMIO space. It is used by the DMA system. You can thing of it analogy wise as, you have your local MMIO address space in a 16 bit, 64KB area. That's your "solar system" And when you're talking to the MMU part of the MMU/DMA (it's a combo chip) you're creating wormholes to other devices that are negotiated and coordinated using the DMA system.
L184[15:59:58] <Z0id​berg> Additionally as I mentioned once before you can use memory mapped IO with the MMU/DMA to execute external instructions that the CPU doesn't internally support
L185[16:00:05] <Ocawes​ome101> neat
L186[16:00:06] <Z0id​berg> The CPU has instruction space reserved for that
L187[16:00:26] <Z0id​berg> This means that in say the world of OC, you could map components to their own instructions.
L188[16:00:46] <Z0id​berg> DMA takes care of the rest
L189[16:01:05] <kapo​w_12> >Amanda: If it was as feasible/ simple as you seem to think, One of the at least thr…
L190[16:01:06] <kapo​w_12> My intent was to attempt a port, but it seems as if you don’t really want one
L191[16:01:53] <Z0id​berg> @Ocawesome101 What kind of architecture is yours? Mine is stack machine as I said
L192[16:01:59] <Z0id​berg> so it's all about stack flow
L193[16:02:20] <Ocawes​ome101> @kapow_12 if you pull off a port that'd be great - and by all means, go ahead - it's just that there have been at least three failed attempts to do so, which doesn't really inspire much confidence
L194[16:03:45] <Z0id​berg> maybe we should port Minecraft off the JVM
L195[16:03:49] <Z0id​berg> 🙂
L196[16:04:01] <Z0id​berg> Let's rebuild Minecraft in Piet
L197[16:04:18] <Ocawes​ome101> @Z0idberg http://0x0.st/oiDg.txt
L198[16:04:50] <Z0id​berg> Oh gawd
L199[16:04:56] <Z0id​berg> you have a multiplier / divider
L200[16:05:04] <Z0id​berg> so evil 😄
L201[16:05:29] <Z0id​berg> I decided to move the multiplier & divider off-chip
L202[16:05:53] <Z0id​berg> OH!
L203[16:06:02] <Z0id​berg> I don't know if you knew this ocawesome
L204[16:06:54] <Z0id​berg> but you can determine greater than and less than in most architectures without having flags for them at all just by looking at the overflow bits etc.
L205[16:07:05] <Ocawes​ome101> interesting
L206[16:07:19] <Z0id​berg> most architectures compare by subtraction
L207[16:07:24] <Ocawes​ome101> i just had 24 bits of flag space and went "what completely unnecessary stuff can i chuck in here"
L208[16:07:31] <Z0id​berg> haha
L209[16:07:49] <Z0id​berg> also don't feel that just because you have a 24 bit chip you also have to have 24 bit everything
L210[16:09:00] <Ocawes​ome101> A5 as a 24-bit register does indeed seem a little overkill
L211[16:09:03] <Z0id​berg> Consider the hardware you will run it on too if you ever do; Most RAM chips aren't 24 bit; You could use a 16 bit chip and an 8 bit one next to eachother though
L212[16:09:12] <Z0id​berg> this is why I never made my 12 bit arch on hardware
L213[16:09:29] <Z0id​berg> there were always 4 bits of unused space everywhere
L214[16:09:57] <Ocawes​ome101> i don't plan to make physical hardware for this arch, mostly due to the fact that when it comes to hardware design i know next to nothing :P
L215[16:10:05] <Z0id​berg> My arch uses a 24 bit address space but is 16 bit
L216[16:10:17] <Z0id​berg> ok. if you don't plan to make any hardware it isn't an issue
L217[16:10:37] <Ar​iri> Izaya: naisuuu
L218[16:11:14] <Ocawes​ome101> it's called the FC-1 for "Fantasy Console-1" (more or less)
L219[16:11:24] <Z0id​berg> there are ways you could work around this with hardware; For example you could have a 24 bit address register where the first 8 bits is the page and it accesses 16 bit information, giving you 8 extra bits for rotations and shifts and such to do whatever you want with for playing around
L220[16:11:35] <Z0id​berg> that way you'd only ever need 16 bit memory hardware
L221[16:11:46] <Z0id​berg> but you'd lose 24 bit storing of numbers
L222[16:11:56] <Z0id​berg> (without splitting them)
L223[16:12:16] <Z0id​berg> you could also do hardware multiplexing of data
L224[16:12:33] <Ocawes​ome101> ~~just use 3x 8-bit chips everywhere~~
L225[16:12:34] <Z0id​berg> where it takes more than one memory read / write to transfer
L226[16:12:38] <Z0id​berg> lol
L227[16:13:00] <Z0id​berg> My architecture's DMA transfers 1 bit at a time
L228[16:13:25] <Z0id​berg> it has its own clock separate from the CPU, in fact the CPU gets its clock cycle from demultiplexing and dividing the DMA clock
L229[16:13:29] <Z0id​berg> several times
L230[16:14:05] <Ocawes​ome101> interesting
L231[16:14:38] <Z0id​berg> basically, every time the CPU clock gets a full cycle the DMA reads in another 16 bits to store / load on either the master or slave depending which direction we're handling. Then, the actual clock input which is WAY faster than the CPU transmits each of the 16 bits at super high speed down a fiber cable.
L232[16:14:50] <Z0id​berg> the fiber cables in mine are just SPDIF / TOSLINK
L233[16:15:37] <Ocawes​ome101> have you implemented this in hardware?
L234[16:15:43] <Z0id​berg> all external memory read and writes, whether you're copying lots of data or just sending commands etc are just sent down fiber.
L235[16:15:51] <Z0id​berg> Not yet! I've been building this in Kicad
L236[16:15:57] <Ocawes​ome101> ah
L237[16:15:58] <Z0id​berg> I have a CPLD thoug hfor teh core
L238[16:16:04] <Z0id​berg> I'm using a Lattice Mach
L239[16:16:34] <Ocawes​ome101> neat
L240[16:16:35] <Z0id​berg> but hey you wont have any large parallel MMIO edge connectors
L241[16:16:58] <Z0id​berg> you just use TOSLINk cables you can buy at the store to wire em up. There's a device I'm working on to MUX/DEMUX this signal
L242[16:17:10] <Z0id​berg> kind of like a network switch
L243[16:17:26] <Z0id​berg> the idea is that the MMU maps memory in local 64KB space
L244[16:17:36] <Z0id​berg> and when you write / read it uses the DMA to copy it over like RDMA
L245[16:17:48] <Z0id​berg> the data will all be there before the next CPU instruction
L246[16:18:08] <Z0id​berg> I also want to allow the DMA to be programmed to copy large sections of memory for you
L247[16:18:21] <Z0id​berg> or handle events from devices like say a serial port UART
L248[16:18:28] <Z0id​berg> and automatically transfer to local memory
L249[16:18:42] <Z0id​berg> Additionally, my CPU's stack controller will have a "queue mode"
L250[16:18:56] <Z0id​berg> so you cna turn a user stack into a queue instead
L251[16:19:01] <Z0id​berg> so you can do FIFO stuff
L252[16:19:52] <Z0id​berg> "Push some data from memory onto stack, do some math, transfer (pop and push) from stack onto serial UART data FIFO queue stack
L253[16:19:55] <Z0id​berg> repeat
L254[16:20:41] <Z0id​berg> I'm still working on the full implementattion of the stack controller
L255[16:21:34] <Z0id​berg> But anyways, as a Minecraft mod this would be set up like "every component device in MC would be its own DMA slave"
L256[16:21:43] <Z0id​berg> it'd be its own IO controller per say
L257[16:22:19] <Ocawes​ome101> i'd probably have an external I/O chip for it
L258[16:22:36] <Ocawes​ome101> do all your component IO over bus 2 or w/e
L259[16:22:52] <Ocawes​ome101> MMIO works too and might be better though
L260[16:24:09] <Z0id​berg> There is nothing wrong with MMIO
L261[16:24:32] <Z0id​berg> since my memory system is more of an actor model with devices on serial fiber links, it HAS to have an IO controller on every device.
L262[16:24:42] <Z0id​berg> which.. is just another MMU/DMA 🙂
L263[16:25:19] <Z0id​berg> I have -not- determined if I want full duplex or not. The TOSLINK connectors are like $7 a piece
L264[16:25:24] <Z0id​berg> at leats the onesI found
L265[16:25:53] <Z0id​berg> that would mean like $14 just for connectors on each IO board minimum not counting the other end
L266[16:26:14] <Z0id​berg> full duplex is kinda cool but tbh the CPU only does half duplex memory IO
L267[16:26:25] <Z0id​berg> so having a single wire setup is clean and not that unusual
L268[16:27:24] <Z0id​berg> Buuut if I did full duplex it means that I could be copying memory out while serial data or something is buffered in to some middle cache chip or something. That makes everything more complicated.
L269[16:27:38] <Z0id​berg> Maybe an arch 2.0 thing
L270[16:28:21] <Z0id​berg> Another option I have other than fiber btw is SATA cables.
L271[16:29:29] <Ocawes​ome101> ah yes, since those are serial aren't they
L272[16:30:24] <Z0id​berg> Yeah but they aren't very flimsy
L273[16:30:33] <Z0id​berg> nothing like a argular wire is
L274[16:30:46] <Z0id​berg> https://www.mouser.com/ProductDetail/Amphenol-Commercial-Products/SATA01120271?qs=St27ATL478vaFeWJF81RJw%3D%3D
L275[16:30:52] <Z0id​berg> Looks like these are $1 a piece
L276[16:31:05] <Z0id​berg> and I could use the other wires for things like hardware flow control?
L277[16:31:13] <Z0id​berg> ready to send, clear to send, etc
L278[16:31:23] <Ocawes​ome101> could be, yes
L279[16:31:36] <Ocawes​ome101> at the cost of very marginally less speed
L280[16:31:45] <Ocawes​ome101> because copper vs. light
L281[16:31:52] <Z0id​berg> I could also power it
L282[16:32:00] <Z0id​berg> well
L283[16:32:05] <Z0id​berg> TOSLINK is fast but it's not insane
L284[16:32:12] <Z0id​berg> many connectors only allow up to say 1Mhz
L285[16:32:17] <Ocawes​ome101> ah
L286[16:32:19] <Z0id​berg> but my instruction speed goal is only 40Khz
L287[16:32:28] <Z0id​berg> If I can get 40,000 instructions per second I will be happy
L288[16:32:44] <Z0id​berg> You can do a lot with that.
L289[16:33:06] <Z0id​berg> Since my design is based on memory movement instead of peak performance
L290[16:33:31] <Ocawes​ome101> rather amusing how we need billions of instructions per second to compete with what could be done on millions of instructions per second ten years ago
L291[16:33:34] <Z0id​berg> in a gpu like thing, copying a ton of sprite data would be fairly fast
L292[16:33:48] <Z0id​berg> yeah.
L293[16:34:04] <Z0id​berg> I think people sometimes forget that we don't usually actually need that kind of performance
L294[16:34:19] <Ocawes​ome101> bUt mUh wEb bRoWseR
L295[16:34:30] <Z0id​berg> but also computers now do more than one thing at a time. ALSO. a 1 Ghz CPU is still taking 30+ cycles to read and write to and from memory you know
L296[16:34:48] <Z0id​berg> so a lot of those cycles are wasted in terms of a per instruction performance, but the CPU does other things
L297[16:34:53] <Z0id​berg> and pipelining helps
L298[16:35:49] <Z0id​berg> I'm only using SRAM in my system so
L299[16:36:09] <Ocawes​ome101> so, ignoring all else, assuming no pipelining, that's only like 30MHz of actually executing instructions? unless i've done the math wrong
L300[16:36:19] <Z0id​berg> every memory operation only takes half a cycle
L301[16:36:27] <Ocawes​ome101> and that's assuming every instruction does memory IO, which it doesn't
L302[16:37:39] <Z0id​berg> I think a majority of the work done on a CPU these days is for things like DMA and such. DMA for the audio. DMA for the video card, DMA for the disk IO.. you don't need super fast instructions for that
L303[16:37:55] <Z0id​berg> plus we have cache
L304[16:38:00] <Z0id​berg> and SDRAM
L305[16:38:11] <Ocawes​ome101> true
L306[16:38:12] <Z0id​berg> which helps speed that up and reduce the number of cycles wasted
L307[16:39:16] <Z0id​berg> But you are correct ocawesome that, a 2Mhz 6502 is faster and way more efficient than a 5Mhz Intel 8080
L308[16:39:20] <Z0id​berg> er, 8086
L309[16:39:48] <Ocawes​ome101> 6502 is a simpler arch, isn't it?
L310[16:39:54] <Z0id​berg> yes.
L311[16:40:01] <Z0id​berg> and it uses SRAM
L312[16:40:07] <Z0id​berg> or depends on it anyways
L313[16:40:32] <Z0id​berg> so memory has no latency other than the small ammount of latency on the datasheet for everything to set
L314[16:40:53] <Z0id​berg> there's no "waiting for capacitor charge time" or anything like that.
L315[16:41:17] <Z0id​berg> the 6502 also doesn't have any internal RAM
L316[16:41:24] <Z0id​berg> other than the registers it has
L317[16:44:38] <Z0id​berg> wtf Ocawesome
L318[16:44:52] <Z0id​berg> now you have me working with details on my 24 bit address space
L319[16:45:26] <Z0id​berg> You made me come up with an idea for an unsolved problem I had: "Device detection and initialization"
L320[16:45:56] <Z0id​berg> the question I've had for a while is, if I have a bunch of devices in memory space all over the place and I don't know where they all are, how do I detect all of the devices I have efficiently?!
L321[16:46:27] <Z0id​berg> and the idea I just came up with is a "recursive tree structure like token ring"
L322[16:46:38] <Z0id​berg> when the system boots a token works its way down the IO tree
L323[16:46:53] <Z0id​berg> and informs every device when its time to register itself
L324[16:47:03] <Z0id​berg> that way they don't step all onto eachother.
L325[16:54:53] <ThePi​Guy24> this conversation somewhat reminds me of what a lot of people are doing in a game called logic world, which is sorta like minecraft redstone, but you have premade gates and wires and such
L326[16:55:02] ⇨ Joins: testuser (webchat@p54a20e9f.dip0.t-ipconnect.de)
L327[16:55:09] <ThePi​Guy24> is pretty neat, a few peeps from here have already found i t;
L328[16:55:12] <ThePi​Guy24> is pretty neat, a few peeps from here have already found it ; [Edited]
L329[16:55:13] <ThePi​Guy24> is pretty neat, a few peeps from here have already found it ;p [Edited]
L330[16:55:58] ⇦ Quits: testuser (webchat@p54a20e9f.dip0.t-ipconnect.de) (Client Quit)
L331[17:06:03] *** TPG24 is now known as ThePiGuy24
L332[17:06:27] *** ThePiGuy24 is now known as TPG24
L333[17:06:59] ⇦ Quits: s_a_m (~sam@172.58.236.226) (Ping timeout: 189 seconds)
L334[17:08:08] *** TPG24 is now known as ThePiGuy24
L335[17:11:01] ⇨ Joins: s_a_m (~sam@172.58.236.226)
L336[17:28:17] <Z0id​berg> https://tinyurl.com/yxrmlae8
L337[17:28:17] ⇦ Quits: s_a_m (~sam@172.58.236.226) (Read error: Connection reset by peer)
L338[17:28:28] <Z0id​berg> Lol. apparently Alabama's #1 dating site is ancestry.com
L339[17:28:42] <Ocawes​ome101> lmao
L340[17:29:11] <Z0id​berg> >ThePiGuy24: this conversation somewhat reminds me of what a lot of people are doing in …
L341[17:29:11] <Z0id​berg> linky?
L342[17:30:15] <ThePi​Guy24> to the game or the community discord?
L343[17:32:23] <Z0id​berg> Holy shit
L344[17:32:31] <Z0id​berg> this health care provider is looking for an "assembly developer?!"
L345[17:32:39] <Z0id​berg> (looking for jobs)
L346[17:32:55] ⇨ Joins: s_a_m (~sam@172.58.236.226)
L347[17:33:07] <Z0id​berg> Lol watch me get to that interview and they go "we want an assembly programmer because we need a programmer to "assemble things together for us"
L348[17:35:47] <Va​ur> damn it, why doesn't industrial foregoing block placer not show up as inventory in storage scanner 😢
L349[17:36:20] <Kristo​pher38> damn, I missed computer architecture conversation
L350[17:37:44] <Z0id​berg> I need to write a cover letter
L351[17:37:51] <Z0id​berg> @Kristopher38 Not really
L352[17:37:56] <Z0id​berg> you missed the highlight of it
L353[17:38:45] <Z0id​berg> >ThePiGuy24: to the game or the community discord?
L354[17:38:45] <Z0id​berg> this game you're talking about!
L355[17:39:34] <ThePi​Guy24> https://store.steampowered.com/app/1054340/Logic_World/
L356[17:40:51] <Kristo​pher38> i've seen it
L357[17:59:52] <luna​r_sam> i wanna make a CPU arch
L358[17:59:53] <luna​r_sam> i have
L359[17:59:54] <luna​r_sam> Ideas
L360[18:00:00] <luna​r_sam> but nothing working yet
L361[18:00:01] <luna​r_sam> besides
L362[18:00:04] <luna​r_sam> i wanna make a plane game too
L363[18:07:36] <ThePi​Guy24> mood
L364[18:07:45] <ThePi​Guy24> those are both things that i want to do
L365[18:10:36] <Elfi> Could always port uxn to OpenComputers, though I guess that'd require a specialized screen
L366[18:13:58] <Elfi> Or just no screen device, I suppose
L367[18:23:24] * Amanda offers Elfi some snacks to distract her while she steals something from the lämp room
L368[18:36:19] ⇦ Quits: ThePiGuy24 (~ThePiGuy2@host-92-17-122-58.as13285.net) (Ping timeout: 192 seconds)
L369[18:37:51] ⇦ Quits: s_a_m (~sam@172.58.236.226) (Ping timeout: 195 seconds)
L370[18:38:21] ⇨ Joins: ThePiGuy24 (~ThePiGuy2@host-92-17-122-58.as13285.net)
L371[18:38:53] ⇨ Joins: s_a_m (~sam@172.58.236.226)
L372[18:42:33] ⇦ Quits: s_a_m (~sam@172.58.236.226) (Ping timeout: 192 seconds)
L373[19:01:35] <Va​ur> %tonkout
L374[19:01:36] <MichiBot> Crud! Va​ur! You beat Ko​dos's previous record of 3 hours, 11 minutes and 13 seconds (By 23 minutes and 17 seconds)! I hope you're happy!
L375[19:01:37] <MichiBot> Va​ur 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: 0.09163. Position #1
L376[19:21:11] ⇨ Joins: TPG24 (~ThePiGuy2@host-92-17-122-58.as13285.net)
L377[19:22:59] ⇦ Quits: ThePiGuy24 (~ThePiGuy2@host-92-17-122-58.as13285.net) (Ping timeout: 189 seconds)
L378[19:52:57] <Va​ur> Conway's game of life powering my mana needs 😄 https://tinyurl.com/yytfcw94
L379[19:57:34] <Z0id​berg> why
L380[19:57:46] <Z0id​berg> are you going to make logic gates with it?
L381[19:59:37] <Z0id​berg> >ThePiGuy24: https://store.steampowered.com/app/1054340/Logic_World/
L382[19:59:37] <Z0id​berg> this better be Linux compatible.
L383[19:59:56] <ThePi​Guy24> it has a native version :D
L384[20:00:21] <Forec​aster> %sip
L385[20:00:21] <MichiBot> You drink a hairy automato potion (New!). Forecaster sees the sky briefly flash solid dark blue then go back to normal.
L386[20:00:24] <Z0id​berg> yay
L387[20:01:12] <Z0id​berg> @ThePiGuy24 One problem I have I didnt see yet was that project red gave me ICs
L388[20:01:18] <Z0id​berg> and I could fit 96 bits of RAM on those fuckers
L389[20:01:26] <Z0id​berg> or use them for address multiplexers, etc
L390[20:01:36] <Z0id​berg> it allowed me to make machines much smaller
L391[20:01:37] <Michiyo> I miss PR ICs :(
L392[20:02:08] <Z0id​berg> Do you remember the other onelike it that let you make miniature redstone contraptions on a 1x1 grid?
L393[20:02:13] <Z0id​berg> I forget the name of that mod
L394[20:03:17] <Z0id​berg> @ThePiGuy24 Thanks for showing me this. I know a ways back @Enthalpy and I wanted to do a tutorial series on youtube on computer engineering.
L395[20:03:28] <Z0id​berg> This would have been a better suitor for it
L396[20:03:35] <Z0id​berg> Perhaps
L397[20:03:37] <ThePi​Guy24> >Z0idberg: <@!258769251547086849> Thanks for showing me this. I know a ways back <@!1239…
L398[20:03:38] <ThePi​Guy24> np ;p
L399[20:16:17] <Enth​alpy> hmm, $12. Maybe @Z0idberg.
L400[20:17:15] <Va​ur> >Z0idberg: are you going to make logic gates with it?
L401[20:17:15] <Va​ur> no, it generate mana
L402[20:30:16] <lunar_sam> mana is stored in the [blank]
L403[20:31:43] <Va​ur> >lunar_sam: mana is stored in the [blank]
L404[20:31:43] <Va​ur> mana pool
L405[21:42:28] <Amanda> %choose spoon sink $; l or n
L406[21:42:28] <MichiBot> Ama​nda: spoon sink n
L407[21:42:35] <Amanda> sounds good
L408[21:58:59] ⇨ Joins: dequbed (~dean@2001:16b8:49e6:2c00:9527:f255:d3ce:9345)
L409[21:59:38] <dequbed> the server my bouncer on died and I completely forgot for like the last 11 days that I can like ... connect to IRC directly? Well anyway.
L410[22:00:08] <Amanda> haha
L411[22:03:47] <deq​ubed> Sadly, of all the things that died that day, Discord was not amongst them <.<
L412[22:04:27] <Michiyo> %tonk
L413[22:04:28] <MichiBot> Dagnabbit! Mic​hiyo! You beat Va​ur's previous record of <0 (By 3 hours, 2 minutes and 51 seconds)! I hope you're happy!
L414[22:04:29] <MichiBot> Michiyo's new record is 3 hours, 2 minutes and 51 seconds! Michiyo also gained 0.00305 tonk points for stealing the tonk. Position #4. Need 0.02066 more points to pass Forec​aster!
L415[22:16:10] ⇦ Quits: DBotThePony (~Thunderbi@31.220.170.28) (Ping timeout: 189 seconds)
L416[22:17:04] ⇨ Joins: DBotThePony (~Thunderbi@31.220.170.28)
L417[22:17:19] ⇦ Quits: ben_mkiv (~ben_mkiv@2001:16b8:1494:b000:fe34:97ff:fea9:75f2) (Ping timeout: 192 seconds)
L418[22:29:36] <Amanda> %choose rain box time or recompile the universe again
L419[22:29:36] <MichiBot> Ama​nda: I have a pamphlet that says never to engage in "recompile the universe again", so you should definitely do it!
L420[22:29:48] <Amanda> hrm, rain box time
L421[22:34:50] <Amanda> %choose while waiting for the mutex to unlock; space management or waves
L422[22:34:51] <MichiBot> Ama​nda: waves while waiting for the mutex to unlock
L423[22:35:42] <Ocawes​ome101> %oclogs
L424[22:35:42] <MichiBot> https://irclogs.pc-logix.com/?dir=logs/%23oc
L425[23:11:49] ⇦ Quits: Vampyre (~spy@87.214.188.202) (Read error: Connection reset by peer)
L426[23:11:54] ⇨ Joins: Vampyre (~spy@87.214.188.202)
L427[23:14:21] <Amanda> You know, while I appreciate that they group them together, I find it somewhat amusing that there's a possibility that we're using uranium for circuit boards in Survivng Mars
L428[23:14:26] <Ocawes​ome101> @Kleadron you ported CC to Beta 1.7 or something like that, right?
L429[23:14:35] <Klea​dron> yeah
L430[23:14:52] <Ocawes​ome101> do you have the code up somewhere?
L431[23:15:00] <Klea​dron> it should be on my github
L432[23:15:06] <Klea​dron> I don't maintain it anymore though
L433[23:15:58] <Ocawes​ome101> thanks :)
L434[23:34:04] <Amanda> %choose space management or more radiation
L435[23:34:05] <MichiBot> Ama​nda: Eeny, meeny, miny, space management.
L436[23:34:19] <Amanda> First: dinner
L437[23:36:18] <Amanda> Then: More radiation, I don't feel like pressing buttons
L438[23:37:01] <Amanda> %choose pizza or cheese and mac
L439[23:37:01] <MichiBot> Ama​nda: I received a telegram from a long lost relative that only read "pizza". Weird.
L440[23:37:22] <Amanda> I think you should go to the poilce, MichiBot, these telegrams are getting a bit frequent
L441[23:45:30] <Izaya> you know how you're not supposed to parse XML with regex?
L442[23:45:34] <Izaya> how about Lua patterns? B)
L443[23:45:41] <Izaya> https://shadowkat.net/tmp/7wKo.png
L444[23:46:37] <Amanda> XML is somewhat fine-er to parse with regex, HTML is a right nightmare
L445[23:47:32] <luna​r_sam> https://tinyurl.com/y3wq23qa
L446[23:47:59] <Amanda> Stop teabagging jet fighters, sam
L447[23:48:02] <dequbed> %bap Izaya
L448[23:48:02] <MichiBot> dequbed baps Izaya with the Magic outer-endian x86! (25%)!
L449[23:48:13] <dequbed> %pet Ariri
L450[23:48:14] <MichiBot> dequbed is petting Ariri with 25-man orgy. Ariri regains 1d4 => 4 hit points! 25-man orgy rides off into the sunset on a horse with no name.
L451[23:48:16] <Izaya> B)
L452[23:48:26] <dequbed> Why is it *always* the 25-man orgy?! <.<
L453[23:48:34] <dequbed> %blame Forecaster
L454[23:48:34] * MichiBot blames Forecaster for E.T for Atari being terrible!
L455[23:48:36] <Izaya> https://shadowkat.net/tmp/RPc2.png
L456[23:48:40] <Amanda> %blame CompanionCube
L457[23:48:41] * MichiBot blames CompanionCube for adding a spinal surgeon to the inventory!
L458[23:48:52] <Amanda> no, that was me, he's to blame for the orgy
L459[23:48:56] <dequbed> I call shenanigans!
L460[23:49:07] ⇦ Quits: flappy (~flappy@91-154-0-54.elisa-laajakaista.fi) (Ping timeout: 192 seconds)
L461[23:49:39] <dequbed> Izaya: is warm
L462[23:49:50] <Izaya> yeah
L463[23:50:04] <Izaya> but I wanted to know how warm so now I'm scraping the BOM once an hour to get the forecast for the next week
<<Prev Next>> Scroll to Top