| --- Log opened Fri Jun 09 00:00:50 2017 |
| 00:36 | | Reiv [NSkiwiirc@Nightstar-ih0uis.global-gateway.net.nz] has joined #code |
| 00:36 | | mode/#code [+o Reiv] by ChanServ |
| 00:39 | | RchrdB [RchrdB@Nightstar-qe9.aug.187.81.IP] has quit [[NS] Quit: Leaving] |
| 01:37 | | celmin|sleep is now known as celticminstrel |
| 06:06 | | celticminstrel is now known as celmin|sleep |
| 06:12 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [Ping timeout: 121 seconds] |
| 06:16 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has joined #code |
| 06:21 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code |
| 06:21 | | mode/#code [+ao VirusJTG VirusJTG] by ChanServ |
| 06:22 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has quit [Operation timed out] |
| 07:10 | | * McMartin feeds results from a BASIC program into Wolfram Alpha to make better sense of them. |
| 07:10 | <&McMartin> | This has been kind of a weird night. |
| 07:11 | <~Vornicus> | what've you got |
| 07:11 | <&McMartin> | A system time expressed in centiseconds since midnight Jan 1 1900 UTC |
| 07:11 | <&McMartin> | Transformed into a usable time and date. =P |
| 07:12 | <~Vornicus> | ah |
| 07:27 | | himi [sjjf@Nightstar-dm0.2ni.203.150.IP] has quit [Ping timeout: 121 seconds] |
| 07:33 | <&McMartin> | https://bumbershootsoft.wordpress.com/2017/06/09/project-mehitabel-the-bbc-basic s/ |
| 08:15 | <@abudhabi> | The project I mentioned last night seems pretty hopeful! It compiles and runs on the first try! |
| 08:34 | | Kindamoody[zZz] is now known as Kindamoody |
| 08:35 | <&McMartin> | Good stuff |
| 09:48 | | * TheWatcher eyes that blog post, notes he wasted entirely too much time playing Lander on the archimedes, wonders if it works on the pi |
| 09:51 | <@TheWatcher> | (almost certainly not, of course; I doubt anyone has ported it) |
| 10:10 | <&McMartin> | Did it run on the RISC PC? |
| 10:12 | <&McMartin> | There's a limited emulator that's designed to make 26-bit stuff run on modern ARM chips |
| 10:15 | | * McMartin goes digging, finds http://ajax3d.sourceforge.net/ |
| 10:15 | <@TheWatcher> | Hrm, near as I can tell, Lander only shipped as a demo with the Acorn Archimedes, the real game was Zarch on the Archimedes, with ports to the AtariST, Amiga, IBM PC, and ZX speccy |
| 10:15 | <~Vornicus> | 26-bit always makes me go fizba wizh |
| 10:16 | <@TheWatcher> | McMartin: ... |
| 10:16 | <@TheWatcher> | WHY? WHY WOULD YOU DO THIS TO ME?! |
| 10:16 | | * TheWatcher closes it before he gets absolutely no work done today |
| 10:17 | <~Vornicus> | this game seems difficult to control |
| 10:19 | | * McMartin digs through the Archimedes Software Preservation Project, does not in fact find Zarch or Lander. |
| 10:20 | | himi [sjjf@Nightstar-v37cpe.internode.on.net] has joined #code |
| 10:20 | | mode/#code [+o himi] by ChanServ |
| 10:21 | | * McMartin did confirm that Zool works as well as it did |
| 10:21 | <&McMartin> | I swear though |
| 10:21 | <&McMartin> | Britain's UX designers in the 20th century were imported directly from the Hell Dimension |
| 10:22 | <&McMartin> | QAOP is not an acceptable set of directional controls! |
| 10:22 | <~Vornicus> | quick, someone tell Foddy that |
| 10:23 | <&McMartin> | Speaking of Fizba Wizh, man, somebody who actually understands the game really really needs to do a dissection LP of ISotMAT. |
| 10:23 | <&McMartin> | TheWatcher: The list of Games With Redistro Rights that got the 26-to-32-bit treatment are at http://forums.jaspp.org.uk:9000/forum/viewforum.php?f=25 |
| 10:23 | <&McMartin> | I recognize a lot of these games from the DOS/VGA world, of course |
| 10:25 | <@TheWatcher> | All this vaguely reminds me, I keep thinking I should try faffing with Unity, see if I can make any progress remaking Tower of Babel. |
| 10:26 | | * McMartin looks at some of the earlier games |
| 10:26 | <@TheWatcher> | ... why have they put that on port 9000, that seems a bit random |
| 10:26 | <@TheWatcher> | Oh, wait |
| 10:26 | <&McMartin> | Man, some of these are only one step up from GORILLAS.BAS |
| 10:26 | <~Vornicus> | they didn't think they were powerful enough to go over |
| 10:27 | <@TheWatcher> | It's for the RISC 9000, isn't it~ |
| 10:31 | <&McMartin> | Looking at the sample videos, they do seem to be running them on a Pi through a custom filesystem module |
| 10:32 | <&McMartin> | U tgubj nist if tgese were |
| 10:32 | <&McMartin> | Blargh |
| 10:32 | <&McMartin> | I think most of these were, like, ARM710s and that generation |
| 10:32 | <&McMartin> | Pre-GBA~ |
| 10:32 | <&McMartin> | My understanding of pre-GBA ARM chips is extremely limited >_> |
| 10:33 | <&McMartin> | PARISC 9000 doesn't seem to have anything to do with Acorn. |
| 10:34 | <@TheWatcher> | Yeah, seems like, I just dredged it out of the dark recesses of my headbones wihtout checking it |
| 10:34 | <&McMartin> | The A7000 used ARM8s. |
| 10:34 | <&McMartin> | Not to be confused with ARMv8, which is what I'm using over here~ |
| 10:35 | <&McMartin> | (Specifically, the RPi3 uses the Cortex-A53) |
| 10:38 | | * McMartin watches the gameplay video for Redshift, completely fails to make sense of it |
| 10:38 | <&McMartin> | I can't tell if this is supposed to be Kobo or not. |
| 11:26 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds] |
| 11:48 | | Kindamoody is now known as Kindamoody|afk |
| 13:55 | <@himi> | I like me some ARMv8 niceness |
| 13:57 | <@himi> | I have to say, the Raspberry Pi 3 is a solid machine - outside a few use cases it'd make a decent desktop |
| 13:59 | <@TheWatcher> | I just wish that it had a sata interface |
| 14:04 | <@TheWatcher> | (my experience is that using an SD card for the OS works fine for a while, until it inevitably shits itself and you need to rebuild the filesystem from scratch, and the USB interface has some odd instabilities with every raspbian kernal I've used so that it's not safe to use with an exernal HD over nontrivial periods) |
| 14:06 | <@himi> | Oh god yes - it sucks for anything disk intensive /at all/, and it's finicky at best |
| 14:07 | <@himi> | Though I've had very little issues with running external HDs - possibly due to filesystem conservativeness |
| 14:07 | <@himi> | ext3 with conservative settings |
| 14:08 | <@himi> | SATA support is the biggest thing that makes me keep looking at alternatives, but it's hard to find anything that isn't a Via-based x86 which /does/ have SATA |
| 14:09 | <@himi> | Or, of course, and Atom - the price tends to be non-competitive in any x86 boards, though |
| 14:10 | <@TheWatcher> | I kept having problems where I'd hook up the external HD, mount it, and for some random time it'd be fine, then the USB interface would suddenly lose the drive entirely and then add it back on a different /dev node |
| 14:11 | <@himi> | Huh |
| 14:11 | <@himi> | With more than one drive? |
| 14:12 | | * himi has had . . . four or five different drives, with different manufacturers, all quite reliable |
| 14:12 | <@TheWatcher> | Yep, tried it with a couple of enclosures |
| 14:12 | <@himi> | This is on Pi 1, b and b+ - I haven't compared any Pi 2 or 3 |
| 14:13 | | * himi has a Pi 3 but it's currently on media player duty rather than server duty |
| 14:14 | <@TheWatcher> | I generally just chalked it up as "Eh, probably Broadcom funkiness" and found a different setup |
| 14:14 | <@himi> | What's that? Broadcom, /funky/? |
| 14:14 | <@himi> | How could you suggest such a thing! |
| 14:15 | <@TheWatcher> | I know, perish the thought, right? |
| 14:15 | <@himi> | I guess it's possible I just got lucky, but with USB it could be something as simple as accidentally sacrificing my firstborn to the right deity |
| 14:15 | <@himi> | (though that's arguably "getting lucky" |
| 16:41 | | Jessikat [Jessikat@Nightstar-e0t2ae.dab.02.net] has joined #code |
| 16:41 | | Jessikat [Jessikat@Nightstar-e0t2ae.dab.02.net] has quit [[NS] Quit: Bye] |
| 17:41 | <@ErikMesoy> | Python: why is it that creating an instance of object and trying to give it attributes is an error, but creating an instance of "class SecondaryObject(object)" and giving it attributes works fine? |
| 17:51 | <@gnolam> | Only the latter has a __dict__. |
| 18:00 | <@gnolam> | Remember: all datatypes are objects. |
| 18:11 | <@ErikMesoy> | But why do not all objects have dict? How does my custom object type get a dict? |
| 18:25 | <&ToxicFrog> | If I had to guess, I'd guess a leaky implementation detail where objects implemented in python get __dict__ and objects implemented in C don't, but that's just a guess |
| 19:37 | | RchrdB [RchrdB@Nightstar-qe9.aug.187.81.IP] has joined #code |
| 19:52 | | * ToxicFrog contemplates C++ operator overload resolution rules |
| 19:52 | <@gnolam> | Again: all datatypes are objects. Classes are objects. Objects are not classes. |
| 19:52 | <@gnolam> | Anyway, the details are in https://docs.python.org/3/reference/datamodel.html if you want to do some reading |
| 19:53 | <&ToxicFrog> | We appear to have a thing here where operator== is overloaded for (say) Date in the global namespace and Info in the ::util namespace (names changed to protect the guilty) |
| 19:54 | <&ToxicFrog> | And we have function in ::util that equality-tests two Dates, where lookup resolution fails to find ==(Date,Date) and only finds ==(::util::Info,::util::Info) |
| 19:55 | <&ToxicFrog> | If we move that function to an anonymous namespace at the top level, it works. |
| 19:55 | <&ToxicFrog> | I was under the impression that resolution would work outwards from the namespace the function was defined in to the top level and thus find both, and then match types and settle on the Date version. |
| 19:56 | <&ToxicFrog> | gnolam: your answer clarifies nothing |
| 19:59 | <&ToxicFrog> | Like, given a class A |
| 19:59 | <&ToxicFrog> | and: |
| 19:59 | <&ToxicFrog> | class B(A): pass |
| 20:00 | <&ToxicFrog> | It makes intuitive sense that A and B should be functionally equivalent |
| 20:01 | <@gnolam> | Sure. If both are in fact /classes/. |
| 20:04 | <&ToxicFrog> | gnolam: so object is a data constructor that can be inherited from but is not a class? |
| 20:05 | <&ToxicFrog> | I mean, I know that's not right, but it's the only way I can interpret what you're saying |
| 20:13 | | Netsplit Deepthought.Nightstar.Net <-> Krikkit.Nightstar.Net quits: @Syloq, @ToxicFrog, @PinkFreud |
| 20:15 | | Netsplit over, joins: @PinkFreud, @Syloq, &ToxicFrog |
| 20:16 | | Syloq [Syloq@NetworkAdministrator.Nightstar.Net] has quit [Ping timeout: 122 seconds] |
| 20:17 | | Syloq [Syloq@NetworkAdministrator.Nightstar.Net] has joined #code |
| 20:17 | | mode/#code [+o Syloq] by ChanServ |
| 20:23 | | * ToxicFrog gnaws on the notforth lexer |
| 20:24 | <&ToxicFrog> | This is a pretty straightforward redesign until I start thinking about how to AOT compile notforth into C and then it starts getting ugly. |
| 20:32 | <&ToxicFrog> | Like, calls to C words are fine, you just emit the name of the C word. Calls to notforth words are also find because if you're AOT compiling you know the C version of the notforth word is available and can call that directly. |
| 20:33 | <&ToxicFrog> | But anything that straight up pushes a value onto the stack is grody, because that value could be: |
| 20:33 | <&ToxicFrog> | - a constant in source code, in which it's fine |
| 20:33 | <&ToxicFrog> | - a constant declared earlier using `const` and being resolved at compile time |
| 20:34 | <&ToxicFrog> | - a pointer to a dynamically allocated null terminated string |
| 20:34 | <&ToxicFrog> | - a pointer to a Word, suitable for use with `exec` |
| 20:34 | <&ToxicFrog> | and all of these look the same in bytecode but need to compile to different things in C. |
| 20:35 | <&ToxicFrog> | I'm probably going to end up having a bunch of compile_<type>() functions that on the host, in c/file mode, all do different things and then forward to compile_literal, and on the target are just #defines for compile_literal, which just emits OP_PUSHLITERAL,<value> |
| 20:37 | <&ToxicFrog> | And there are some fun land mines, e.g. things you can do when defining builtin words that will work, and compile, but are not actually safe to use at runtime, like |
| 20:37 | <&ToxicFrog> | :buffer 80 alloc const |
| 20:38 | | RchrdB [RchrdB@Nightstar-qe9.aug.187.81.IP] has quit [Ping timeout: 121 seconds] |
| 20:38 | <&ToxicFrog> | Which will embalm in the builtin dictionary a constant named `buffer` that points to some memory that was allocated during the bootstrapping process, in a different binary and quite possibly a different processor architecture. |
| 20:39 | <&ToxicFrog> | I'm not sure how much I should worry about guarding about this. |
| 20:50 | | Kindamoody|afk is now known as Kindamoody |
| 20:55 | <@gnolam> | More that a class is a specific kind of object that, among other things, accepts arbitrary attributes. |
| 20:55 | <@gnolam> | https://www.theverge.com/tldr/2017/6/9/15768800/reddit-worst-volume-sliders-ui-d esign |
| 20:59 | <@ErikMesoy> | Reading the datamodel did not help me understand. Playing around with type() has excavated new realms of Object Is A Special Case. |
| 21:08 | <@ErikMesoy> | Instances of class "object" do not seem to behave like instances of any other class. |
| 21:36 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 21:39 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has joined #code |
| 21:39 | | mode/#code [+o Alek] by ChanServ |
| 22:02 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 22:02 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 22:07 | | Kindamoody is now known as Kindamoody[zZz] |
| 22:07 | | Netsplit Deepthought.Nightstar.Net <-> Krikkit.Nightstar.Net quits: @PinkFreud, @ToxicFrog, @Syloq |
| 22:24 | | Netsplit over, joins: ToxicFrog |
| 22:34 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds] |
| 22:36 | | Syloq [Syloq@NetworkAdministrator.Nightstar.Net] has joined #code |
| 22:39 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 23:17 | | PinkFreud [WhyNot@NetworkAdministrator.Nightstar.Net] has joined #code |
| 23:17 | | ServerMode/#code [+o PinkFreud] by *.Nightstar.Net |
| 23:17 | | mode/#code [+o Syloq] by ChanServ |
| 23:28 | < ToxicFrog> | gnolam: so, first of all, we aren't talking about classes, we're talking about class instances here |
| 23:28 | < ToxicFrog> | Although this behaviour also seems to apply to classes, i.e. `object` doesn't have a __dict__ but other classes do |
| 23:42 | | RchrdB [RchrdB@Nightstar-qe9.aug.187.81.IP] has joined #code |
| 23:48 | | * TheWatcher fingertappity, starts downloading unity |
| --- Log closed Sat Jun 10 00:00:52 2017 |