[02:51] *** Lord_Nigh has quit IRC (Read error: Operation timed out) [02:53] *** Lord_Nigh has joined #jsmess [03:49] *** balrog has quit IRC (Read error: Operation timed out) [03:49] *** balrog has joined #jsmess [07:30] *** balrog has quit IRC (Read error: Operation timed out) [07:47] *** logchfoo0 starts logging #jsmess at Sun Apr 26 07:47:09 2020 [07:47] *** logchfoo0 has joined #jsmess [07:47] *** irc.Prison.NET sets mode: +nt [09:19] *** irc.Prison.NET sets mode: -o logchfoo0 [09:19] *** hub.efnet.us sets mode: +s [09:19] *** balrog has joined #jsmess [09:19] *** Lord_Nigh has joined #jsmess [09:19] *** Sgeo_ has joined #jsmess [09:19] *** DFJustin has joined #jsmess [09:19] *** mattl has joined #jsmess [09:19] *** defacto2 has joined #jsmess [09:19] *** datajerk has joined #jsmess [09:19] *** bai has joined #jsmess [09:19] *** dboard420 has joined #jsmess [09:19] *** drk||razi has joined #jsmess [09:19] *** Ctrl-S___ has joined #jsmess [09:19] *** Vito` has joined #jsmess [09:19] *** SketchCow has joined #jsmess [09:19] *** arkiver has joined #jsmess [09:19] *** hub.efnet.us changes topic to: http://www.emularity.com | Next goals: Media Swap Ability and MAME Refactoring [09:31] *** godane has joined #jsmess [19:47] *** balrog has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** Sgeo_ has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** defacto2 has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** datajerk has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** mattl has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** drk||razi has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** Ctrl-S___ has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** Vito` has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** godane has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** dboard420 has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** Lord_Nigh has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** DFJustin has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** SketchCow has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** arkiver has quit IRC (irc.Prison.NET hub.efnet.us) [19:47] *** bai has quit IRC (irc.Prison.NET hub.efnet.us) [20:01] *** godane has joined #jsmess [20:01] *** balrog has joined #jsmess [20:01] *** Lord_Nigh has joined #jsmess [20:01] *** Sgeo_ has joined #jsmess [20:01] *** DFJustin has joined #jsmess [20:01] *** mattl has joined #jsmess [20:01] *** defacto2 has joined #jsmess [20:01] *** datajerk has joined #jsmess [20:01] *** bai has joined #jsmess [20:01] *** dboard420 has joined #jsmess [20:01] *** drk||razi has joined #jsmess [20:01] *** Ctrl-S___ has joined #jsmess [20:01] *** Vito` has joined #jsmess [20:01] *** SketchCow has joined #jsmess [20:01] *** arkiver has joined #jsmess [20:09] *** LordNigh2 has joined #jsmess [20:09] *** Lord_Nigh has quit IRC (Ping timeout: 272 seconds) [20:09] *** LordNigh2 is now known as Lord_Nigh [20:10] *** Sgeo__ has joined #jsmess [20:11] *** DopefishJ has joined #jsmess [20:11] *** DopefishJ has quit IRC (Read error: Connection reset by peer) [20:14] *** DopefishJ has joined #jsmess [20:16] *** SketchCo1 has joined #jsmess [20:16] *** DFJustin has quit IRC (se.hub irc.underworld.no) [20:16] *** SketchCow has quit IRC (se.hub irc.underworld.no) [20:16] *** arkiver has quit IRC (se.hub irc.underworld.no) [20:20] *** defacto29 has joined #jsmess [20:20] *** Sgeo_ has quit IRC (Read error: Operation timed out) [20:21] *** defacto2 has quit IRC (Ping timeout: 372 seconds) [20:21] *** defacto29 is now known as defacto2 [20:21] *** godane has quit IRC (Ping timeout: 864 seconds) [20:27] *** datajerk_ has joined #jsmess [20:39] *** datajerk has quit IRC (Read error: Operation timed out) [20:44] *** Sgeo has joined #jsmess [20:45] *** datajerk_ has quit IRC (Max SendQ exceeded) [20:46] *** defacto2 has quit IRC (Ping timeout: 372 seconds) [20:47] *** datajerk has joined #jsmess [20:47] *** defacto2 has joined #jsmess [20:52] *** godane has joined #jsmess [20:53] *** Sgeo__ has quit IRC (Read error: Operation timed out) [21:47] *** godane has quit IRC (Read error: Operation timed out) [22:02] *** godane has joined #jsmess [22:16] bai, what happened with DOSBox-x on IA? Was that supposed to solve the emulation stopping for Win95? [22:17] I'm experimenting with loading an item's files into Win95 in DOSBox, and that part has gone more smoothly than DOSBox emulating Win95 itself [22:18] The copy.sh/v86 suggestion is not.. very pleasing to me because it doesn't seem to like command prompts very much. Also DOSBox's fake DOS is convenient for the aforementioned loading into the .img file dynamically [22:25] if that's the project with all the game-specific dosbox.conf files, I think we did roll that out to all the supported games that were in the archive, SketchCow would know more [22:26] I don't think we've specifically tried win95 to see if the dosbox-x configs for that improve things, the bigger problem with win95 stuff I think is just that most software moved onto CDROMs, so the 600mb preload kind of kills a lot of the turn-on-and-go aspect of the emulator [22:27] but yeah, to get win95 working you need to boot off of a disk image configured with an official version of DOS, 5.0 or 5.5 I forget [22:28] I stole win95.ajf.me's image [22:30] but yeah, it should work and I think the way it works is, if you want your item to use a custom dosbox.conf like one from the dosbox-x project, you can put it into the zip file for the item, I can't remember if you need special handling if you're using the .img approach [22:31] I seem to remember something about mapping dosbox.conf into the underlying virtual browserfs filesystem using some config settings [22:31] https://archive.org/details/some_dosbox_test_95 the main problem I'm having is the nested emulation timeout [22:31] but I can't remember if I did that with emularity (eg, if it's config driven) or with my own custom loader (code-driven) [22:32] I'm just imgmounting the .img and copying files into it before loading with boot [22:32] hmm, it's failing to load because of a 404 on the dosbox-x.json [22:35] ...it was working before, maybe switching to dosbox-x didn't actually take effect when I went to test [22:36] What's dosbox-alt? [22:37] uhhh not sure...for a while we had dosbox and dosbox-sync (or dosbox-nosync? I forget) and one of those handled the nested emulation timeout problem better [22:43] Bloop [22:44] *** SketchCo1 is now known as SketchCow [22:48] The dosbox-alt version crashed trying to get gamepad data [23:13] My understanding is that Em-DOSBox avoids emterpreter for certain paths and when it runs through those paths and they take a long time, that's the cause of the problem [23:14] What was the new Emscripten thing for sync functions? [23:15] yeah, the sync stuff had to do with emterpretify, you could whitelist certain function names which would be run in a way which allowed them to be paused and resumed, which helps with the emulation timeout problem, but it's possible that win95 is triggering some function calls that we don't normally see for the rest of the dos stuff [23:15] and yeah, I think there is a new way of handling that but I haven't read up on it yet [23:21] https://github.com/dreamlayers/em-dosbox/search?q=DOSBOX_RunMachineNoSleep&unscoped_q=DOSBOX_RunMachineNoSleep [23:22] https://github.com/dreamlayers/em-dosbox/blob/4526ed7fa56fc70a5de7b666f97a097c27b9ee9f/src/dosbox.cpp#L173 [23:22] Not being allowed to sleep because of wanting to avoid the emterpreter is the reason afaict [23:22] So if we threw everything into emterpreter we might be able to avoid the error at the cost of poor performance [23:23] https://github.com/dreamlayers/em-dosbox/issues/11 mentions jDOSBox doing something relevant [23:24] yeah there are three different options for enabling emterpreter - you provide a list of whitelisted functions, a list of blacklisted functions, or you say "screw it, just do 'em all" [23:24] the nuclear option is good for knowing if the problem is fixable [23:24] then you can extend the whitelist to get performance usable, by whitelisting paths that you know won't cause problems [23:25] or, more accurately, your whitelist contains a list of functions that WOULD cause problems, if if they were allowed to run without being asyncified [23:34] https://kripken.github.io/blog/wasm/2019/07/16/asyncify.html suggests that Asyncify is the new thing, for some reason I thought it was the old thing [23:35] old as in pre-emterpreter [23:36] oh, I thought asyncifying was just a term they made up for what emterpreter does to your code :D