#jsmess 2020-01-24,Fri

↑back Search

Time Nickname Message
00:43 🔗 SketchCow Use the good one
00:43 🔗 SketchCow Use the one we can compile with.
01:51 🔗 azakai has quit IRC (Read error: Operation timed out)
02:17 🔗 SketchCow DFJustin: Can you compile the apple driver to current?
02:43 🔗 arkiver has quit IRC (Remote host closed the connection)
02:43 🔗 arkiver has joined #jsmess
03:07 🔗 kripken__ has joined #jsmess
03:24 🔗 db48x we can't compile with either, as far as I can tell
03:28 🔗 arkiver has quit IRC (Remote host closed the connection)
03:29 🔗 arkiver has joined #jsmess
04:03 🔗 SketchCow DFJustin thought he could
04:16 🔗 arkiver has quit IRC (Remote host closed the connection)
04:18 🔗 arkiver has joined #jsmess
05:18 🔗 arkiver has quit IRC (Remote host closed the connection)
05:19 🔗 arkiver has joined #jsmess
05:20 🔗 arkiver has quit IRC (Read error: Connection reset by peer)
05:20 🔗 arkiver has joined #jsmess
05:47 🔗 DFJustin http://fos.textfiles.com/dfjustin/apple2e/apple2ee/
06:26 🔗 db48x meanwhile, I deleted emscripten and reinstalled latest-fastcomp, and it seems to be working
06:27 🔗 arkiver has quit IRC (Remote host closed the connection)
06:27 🔗 arkiver has joined #jsmess
06:28 🔗 SketchCow Want me to hold off an experimental swap-in?
06:28 🔗 db48x no, go for it
06:30 🔗 SketchCow Interesting ponderable:
06:30 🔗 SketchCow -rw-r--r-- 1 root root 11885117 Jan 19 05:57 mameapple2e.js.gz
06:30 🔗 SketchCow -rw-r--r-- 1 root root 603441 Jan 19 05:57 mameapple2e_wasm.js.gz
06:30 🔗 SketchCow -rw-r--r-- 1 root root 5333330 Jan 19 05:57 mameapple2e_wasm.wasm.gz
06:31 🔗 SketchCow In the directory related to this:
06:31 🔗 SketchCow -rw-r--r-- 1 root root 3492310 Jan 24 06:30 mameapple2e.js
06:31 🔗 SketchCow -rw-r--r-- 1 root root 26815221 Jan 24 06:30 mameapple2e.wasm
06:31 🔗 SketchCow I mean, obviously I gzip them.
06:31 🔗 SketchCow -rw-r--r-- 1 root root 262620 Jan 24 06:30 mameapple2e.js.gz
06:31 🔗 SketchCow -rw-r--r-- 1 root root 5256865 Jan 24 06:30 mameapple2e.wasm.gz
06:31 🔗 SketchCow But still, that seems... incomplete.
06:32 🔗 SketchCow Anyway, let's break apple.
06:33 🔗 SketchCow Well, before I do, what's your opinion.
06:33 🔗 db48x those are the files you got from DFJustin?
06:33 🔗 SketchCow Yes.
06:33 🔗 db48x they need to be renamed, but they look fine
06:33 🔗 SketchCow OK, what name
06:33 🔗 db48x mameapple2e_wasm.js.gz and mameapple2e_wasm.wasm.gz
06:34 🔗 arkiver has quit IRC (Read error: Connection reset by peer)
06:34 🔗 SketchCow Snapped in, let's see what explodes.
06:34 🔗 SketchCow (Obviously it takes a little)
06:34 🔗 balrog has quit IRC (Read error: Operation timed out)
06:34 🔗 arkiver has joined #jsmess
06:36 🔗 godane has quit IRC (Ping timeout: 247 seconds)
06:37 🔗 SketchCow Well, the apple is definitely booting.
06:37 🔗 SketchCow And self-identifies as 217.
06:37 🔗 SketchCow (0.217)
06:38 🔗 balrog has joined #jsmess
06:38 🔗 db48x excellent
06:40 🔗 SketchCow Now for the card test
06:40 🔗 arkiver has quit IRC (Remote host closed the connection)
06:41 🔗 dboard420 has quit IRC (Ping timeout: 240 seconds)
06:41 🔗 dboard420 has joined #jsmess
06:42 🔗 SketchCow Want to take odds it all "just" works
06:44 🔗 SketchCow https://archive.org/details/TotalReplayArcade
06:44 🔗 SketchCow So, of course it explodes.
06:44 🔗 SketchCow BUT
06:44 🔗 SketchCow That could be some oversight by me
06:45 🔗 SketchCow The official command is: mame64 apple2ee -sl7 cffa2 -hard1 00playable.2mg
06:45 🔗 SketchCow https://ia601906.us.archive.org/7/items/emularity_engine_v1/apple2cffa2.json is the json I set
06:50 🔗 SketchCow exception thrown: 11893560 loader.js:1001:53
06:50 🔗 SketchCow RuntimeError: abort(63). Build with -s ASSERTIONS=1 for more info. mameapple2e_wasm.js.gz:1:3190873
06:50 🔗 SketchCow uncaught exception: 11893560
06:52 🔗 SketchCow JSON looks right, maybe DFJustin sees something wrong in it.
06:52 🔗 SketchCow I mean, current status is "It is compiled and it works for everything but this new feature they just added."
06:52 🔗 SketchCow So maybe there's a setting I don't know
06:55 🔗 godane has joined #jsmess
07:02 🔗 SketchCow Intereestingly, Mame wiki says: mame64 apple2ee -sl6 diskii -sl7 cffa2 -hard1 TotalReplay.2mg
07:08 🔗 SketchCow I'm going to put that in
07:08 🔗 SketchCow The theory being that 4am lives in mame, but mame guys were born in it
07:08 🔗 SketchCow molded by it, shaped
07:13 🔗 SketchCow Still blowing up
07:15 🔗 SketchCow Trying some more hail marys
07:23 🔗 SketchCow Hail marys not working.
07:25 🔗 db48x the arguments we're using for that item are "apple2ee -verbose -rompath emulator -window -nokeepaspect -resolution 560x384 -samplerate 44100 -hard1 /emulator/00playable.2mg -sl6 diskii -sl7 cffa2"
07:26 🔗 db48x which looks correct, unless the order matters
07:28 🔗 SketchCow Well, as we both note, it is absolutely not starting even the console log for MAME
07:28 🔗 SketchCow I switched to Total Replay 2 (not the one that came out two days ago) JUST in case it introduced a stupid
07:29 🔗 SketchCow I'm including diskii because MAME suggests it and my concern is always diskii does a "little more" than what you'd think
07:29 🔗 SketchCow you could try making a super weird version of the loader for external_js=1 just in case.
07:30 🔗 SketchCow I mean, it blows up very quick, before it gets anywhere
07:31 🔗 SketchCow Maybe order matters? That seems very unlike mame
07:38 🔗 DFJustin order does matter, -sl7 cffa2 creates the -hard1 slot
07:41 🔗 SketchCow Well, that explains that
07:42 🔗 SketchCow is there an easy solution to this
07:42 🔗 SketchCow I could make a custom one-off emulator called totalreplay
07:42 🔗 SketchCow Since this and pitchdark are the only two things that do this
07:42 🔗 DFJustin also you can put the 2mg in a zip to save a lot of bandwidth
07:43 🔗 SketchCow I'm inclined to super-jank this because there's two use cases
07:43 🔗 DFJustin slots adding media types is a thing on other systems though
07:46 🔗 db48x is there ever a case where we would have to put arguments after the disk image?
07:46 🔗 db48x because it's easy enough just to swap the order
07:50 🔗 SketchCow Can we just test and snap loader.js?
07:51 🔗 SketchCow I mean, if it in fact only matters for this, what situation murders the machine if disk image is last
07:51 🔗 SketchCow Because yeah, just test it with an external_js
07:55 🔗 DFJustin slots first, media last should be ok in all situations I think
07:57 🔗 SketchCow Well, let's do it then.
08:04 🔗 SketchCow Switched it back to v3.0.
08:04 🔗 SketchCow Fixing the metadata
08:05 🔗 db48x switched what back?
08:05 🔗 SketchCow Sorry
08:05 🔗 SketchCow Part of the thing where I was straw-grasping was I uploaded in Total Replay 2.0 instead of 3.0
08:05 🔗 db48x oh, ok
08:05 🔗 SketchCow JUST in case they malformed or broke something
08:11 🔗 db48x https://archive.org/details/TotalReplayArcade&external_js=1
08:12 🔗 db48x seems to work
08:14 🔗 SketchCow Going to do some quick checks
08:15 🔗 SketchCow DFJustin: When you mention compressing it - do you have to set emulator_ext to zip?
08:17 🔗 SketchCow db48x: So, fullscreen seems to not be working anymore
08:20 🔗 SketchCow (Doing some tests)
08:20 🔗 SketchCow This is not related to your change
08:20 🔗 db48x it's related to the new build of mame
08:20 🔗 SketchCow Yeah
08:23 🔗 db48x heh
08:23 🔗 db48x they changed the name of the method
08:23 🔗 db48x it used to be requestFullScreen
08:23 🔗 db48x now it's requestFullscreen
08:24 🔗 SketchCow So, in theory, your changes all relate to mame
08:25 🔗 SketchCow Checked, DOS loads fine under new loader
08:25 🔗 db48x thanks for checking
08:25 🔗 db48x I hope that requestFullscreen takes the same arguments as requestFullScreen did
08:26 🔗 SketchCow Arcade loads fine
08:26 🔗 db48x no, I get an error when I call it: TypeError: The expression cannot be converted to return the specified type. mameapple2e_wasm.js.gz:1
08:27 🔗 SketchCow Atari 2600 loads fine
08:28 🔗 SketchCow Atari 800 loads fine. This is likely going to be the case all down the line then
08:28 🔗 SketchCow So that's good, the reversed new loader works fine
08:33 🔗 SketchCow I am doing the zip experiment with total replay, so it'll either be fine or blow up in a weird way, do not be bothered
08:34 🔗 SketchCow It works.
08:34 🔗 SketchCow OK, so thanks, DFJustin - people will load in 12mb instead of 32
08:42 🔗 SketchCow So, db48x - what do you think the fullscreen solution is?
08:42 🔗 SketchCow Bang at it and try and make it go both ways?
08:43 🔗 SketchCow Or something else
08:56 🔗 db48x banging on it didn't work, and the documentation isn't very helpful
09:16 🔗 SketchCow is emscrioten doing this?
09:18 🔗 db48x yea
09:25 🔗 SketchCow https://twitter.com/textfiles/status/1220638696841654273?s=19
09:35 🔗 db48x RequestFullscreen :)
09:36 🔗 db48x https://git.sprintf.io/emscripten-ports/emscripten/commit/ff58311baa59001f4e7286546d2f5bdd40f08ef3?style=split
10:19 🔗 ted__ has quit IRC ()
10:19 🔗 ted__ has joined #jsmess
10:28 🔗 db48x ok, I fixed it
10:46 🔗 db48x it required a change to archive.js though
10:46 🔗 mattl has joined #jsmess
12:53 🔗 Ctrl-S_ has quit IRC ()
12:54 🔗 Ctrl-S_ has joined #jsmess
15:45 🔗 kripken__ has quit IRC (Ping timeout: 255 seconds)
16:27 🔗 godane has quit IRC (Read error: Operation timed out)
16:37 🔗 godane has joined #jsmess
17:11 🔗 azakai has joined #jsmess
17:25 🔗 SketchCow Alon said they did it to stay consistent with comventions
17:25 🔗 SketchCow I want to tell him about referer_log
18:19 🔗 godane has quit IRC (Read error: Connection reset by peer)
18:47 🔗 godane has joined #jsmess
23:13 🔗 Lord_Nigh has quit IRC (se.hub efnet.portlane.se)
23:13 🔗 pengan has quit IRC (se.hub efnet.portlane.se)
23:25 🔗 godane has quit IRC (Read error: Connection reset by peer)
23:37 🔗 db48x yes :)
23:42 🔗 godane has joined #jsmess
23:42 🔗 SketchCow Anyway, you have the answer, they swapped it due to things.
23:43 🔗 SketchCow They might have done it some time ago.
23:43 🔗 db48x 3 years ago, apparently
23:43 🔗 db48x they added a shim under the old name so that it would keep working
23:44 🔗 db48x perhaps a more recent change removed the shim
23:46 🔗 SketchCow Yeah, probably someone went "what's THIS old thing"
23:47 🔗 SketchCow A lot of software engineering is people going "Why the fuck is THIS still here" followed by a quick and fervent demonstration of why the fuck it is still ehre
23:48 🔗 bai "wait, so that zip tie was holding up the ENTIRE SKYSCRAPER?"
23:52 🔗 SketchCow https://en.wikipedia.org/wiki/Citigroup_Center#Engineering_crisis_of_1978
irclogger-viewer