[00:20] Loader locked in [00:25] Screenshotting not quite working [00:30] So, something happens [00:31] I think it crashes my screenshotter firefox [00:34] what ff version does the screenshotter use? [00:37] Oh god who knows [00:37] I think I can't update it [00:37] http://teamarchive1.fnf.archive.org/AMIGA/SCREENBIN/00_coverscreenshot.jpg [00:37] It's pretty though [00:39] Oh man, the Amiga thing's been uploading for 2 hours [00:39] how many amiga items are we slamming in? [00:40] First pass is 888 [00:40] that's a nice incredibly-round number [00:41] Yeah. So, it seems to be working. [00:41] The mass of ADF images I have, many of them are dupes with minor variances (cracker groups, etc) [00:49] Will deal in a bit [00:50] Could try a different machine for screenshots [00:50] Geez, do I have any lying around. [00:51] SketchCow: got a query, what drive are you using for these rips? (different drives have different read offsets) [00:51] What, the amiga? [00:51] no, sorry, should be more specific [00:51] the Mac cue/bin images, like the Panic one [00:52] Isbuster [00:52] Isobuster [00:52] Rip bin/cue [00:52] yeah I see the software, wondering what drive [00:52] I use random drives [00:52] They're all generic CDRs [00:54] If I can't say to ISObuster "read these discs" and have it produce a useful bin/cue [00:54] I might as well quietly snap every CD in two [00:58] when audio is involved, make/model of CD drive (or more specifically the read offset) becomes a valuable parameter since it indicates where the audio starts; most of these seem to be single or multisession data though [00:59] mmph [00:59] I'll do them and if someone hates them they can do them again. [01:00] I wonder what the crash/cost is on the firefox with sae [01:00] physical media is annoying, bleh. [01:24] all i can think is that the late 80s / early 90s was great for getting bbs servers on CDs [01:25] as much as physical media is annoying at least most of the software and stuff was backup onto shareware discs [01:48] *** Coderjoe has quit IRC (Read error: Operation timed out) [01:54] *** Coderjoe has joined #jsmess [03:24] *** GLaDOS has joined #jsmess [03:32] Now to think about the screenshotting [03:58] 800 PD disks going [04:09] Not sure of a way to see the console of this dying web browser [04:37] https://archive.org/details/3D_Game_Series_Mk1_The_1992_Shad-Art_PD baaaarf [04:38] Something's weird with firefox [04:39] you ever figure out which version it's running? [04:39] nope [04:39] this is second box [04:40] It fails for some reason [04:40] the mouse situation probably needs some love but that's unrelated to this problem. hmm. [04:41] trying to think what the easiest way of capturing console output is with firefox [04:43] maybe easiest to just configure the browser to start with it open, and make the screenshotter shoot the whole window without cropping [04:43] but even then we'll only see the most recent logs [04:45] I know when I run chrome and ff through karma they output logs to stdout, I wonder how they accomplish that [04:48] The problem in THIS case (the new one) is that (and I've seen this before) - firefox won't run the emulator. [04:49] oh, does it not run any of them now? [04:50] it might be a problem with webgl, you might have to install some additional mesa packages to get accelerated gl for headless browsers [04:50] are you running this via xvfb? [05:19] Yes [05:19] No, no. [05:19] This is a different problem. [05:19] My own firefox on my own machine doesn't work either, by the way [05:19] I assume it's some dead cookie or something. [05:22] oh, weird [05:24] Trying to figure out how to wipe my entire firefox cookies and cache [05:24] thought I did, it didn't. [05:32] Bummer. [05:35] Just don't know what to do. It blows it up. [05:37] this._sae = new ScriptedAmigaEmulator(); [05:37] Makes firefox (my firefox) blow up) [06:14] https://ia601505.us.archive.org/18/items/Beasties_II_The_Ultimate_Conflict_demo-playable_1994_Firstrate_Effex_5D_Licencew/00_coverscreenshot.jpg [06:14] Also Chrome has same issue [06:33] http://fos.textfiles.com/AMIGA/screenshot_00.jpg [06:44] Verified. this._sae = new ScriptedAmigaEmulator(); blows it up [06:44] In loader.js [06:57] hmm [06:58] what does the dev console look like? [08:09] *** DFJustin has quit IRC (Ping timeout: 260 seconds) [08:13] *** DFJustin has joined #jsmess [08:18] *** joepie91 has quit IRC (Ping timeout: 244 seconds) [08:22] *** arkiver has quit IRC (ny.us.hub irc.servercentral.net) [08:22] *** SketchCow has quit IRC (ny.us.hub irc.servercentral.net) [08:22] *** decay has quit IRC (ny.us.hub irc.servercentral.net) [08:23] *** joepie91 has joined #jsmess [08:38] *** balrog has quit IRC (Read error: Operation timed out) [08:38] *** balrog has joined #jsmess [08:40] *** Coderjoe has quit IRC (Read error: Operation timed out) [08:47] *** Coderjoe has joined #jsmess [09:10] *** arkiver has joined #jsmess [09:10] *** SketchCow has joined #jsmess [09:10] *** decay has joined #jsmess [09:10] *** irc.servercentral.net sets mode: +o SketchCow [11:18] *** Coderjoe has quit IRC (Read error: Operation timed out) [11:47] *** Coderjoe has joined #jsmess [14:39] Sorry, fell alseep. [14:48] So, Mike got back to me. [14:48] (Owns or licenses some aspect of Amiga.) [14:48] I intend to talk to him [14:49] morning [14:56] Morning. So as you might have seen, behavior is such that in some cases, the stuff doesn't quite load. [14:56] I thought it was something stupid with my personal setup, but I suspect it's in other locations too. [15:00] hmm [15:01] anything interesting show up in the js console when it happens? [15:01] Let me go do it again. [15:01] (I'm assuming the problems are similar) [15:02] So the error on your loader is "Failed to download game data!" [15:03] true [15:04] that error will come up long before it runs SAE though [15:04] Correct [15:04] No, I should explain, there's two things at work here. [15:04] One is that on my screenshotter machine, it freezes on running. [15:05] mmm [15:05] That's one thing. Likely due to such an old firefox. [15:05] firefox freezes, or the machine hangs, or? [15:05] So I installed both Chrome on screenshotter and chrome on other machine (with firefox already installed) [15:05] So the screenshotter chrome and other machine chrome/firefox all get loader errors. [15:06] So, the "firefox stops" issue is something separate [15:06] ok [15:06] It could be a range of things and at this juncture it's difficult to diagnose. [15:07] But the fact that all my other installations give out that error, that's a problem. [15:07] oh, I see a small bug on the loader [15:07] if it gets an error early in the process, it'll still try to launch SAE [15:07] skipping all of the steps in between :) [15:08] fixed it and pushed it to the loaderlab [15:09] that might make it easier to find the real problem [15:10] So, it gives an error of "UnknownError" in browserfs.js [15:13] And of course browserfs is huge, so it's not overly crazily helpful since it's closured. [15:31] hmm [15:31] I've written to Colanto, basically indicating that we are doing this, using AMOS, and there you go. [15:31] He already says his "hands are tied" due to "third parties" but this makes me wonder what sort of deal he has. [15:32] https://archive.org/details/3D_Game_Series_Mk1_The_1992_Shad-Art_PD shows the chrome dying on boot [15:42] could you install the browserfs source and sourcemap like you did for sae? [15:44] We need to think about that - won't it then go after everything on the archive? [15:45] what do you mean by "go after"? [15:45] Replace the functionality of. [15:45] (I might be off) [15:46] ah [15:46] I'm probably being unclear. When we mess with scriptedamigaemulator.js, it affects the amiga, but this could affect all the items on the archive being accessed. [15:47] I'm fine with that, but it has to be measured against blowing up [15:47] the sourcemap just allows the browser to run the compressed/compiled script as normal, while displaying the original sources in the debugger [15:47] OK. [15:47] If you hand me things to put in, I'll happily put them in. [15:48] actually, can reproduce this with external_js=1? if so, we can put the sourcemap into the loaderlab [15:53] Let me try. [15:53] Yes. [15:54] great [15:54] I installed the source map into the loaderlab [15:56] Should I reload? It seems to fail the same way with the same obtuse block of code [15:56] yes [15:57] Same, browserfs:5:7614 [15:57] "Unknown error" [15:57] ok, go to the debugger tab [15:58] look at the list of sources [15:58] do you have browserfs.js? [15:58] or do you have a src/main.ts? [15:59] I do not see browserfs.js [15:59] I do have a src/main.ts [15:59] excellent [16:00] ok, also in the debugger, click on the gear icon on the right-hand side and choose Pause on Exceptions [16:01] Done [16:01] deltaFS = new AsyncMirrorFS(deltaFS, [16:01] now it should stop when the exception is created, pausing execution so that you can see what's going on [16:02] hmm [16:02] IndexedDB = BrowserFS.FileSystem.IndexedDB; is the line before [16:02] (Maybe it pauses the line after) [16:02] that's actually in loader.js... [16:02] Yes [16:03] I have the exception [16:03] Can't cut and paste! Fie! [16:03] TypeError [16:03] AsyncMirrorFS is not a constructor [16:06] ok, I think that's a red herring [16:06] looks like jvilk changed the name from AsyncMirrorFS to AsyncMirror :P [16:07] I've uploaded a loader.js that uses the new name [16:07] do you get any interesting errors now? [16:08] throw new ApiError(ErrorCode.EPERM, "OverlayFS is not initialized. Please initialize OverlayFS using its initialize() method before using it."); [16:08] In backend/OverlayFS.ts [16:08] bah [16:08] ok, so that's another red herring [16:08] I forgot that jvilk told me he had changed it so that was now required [16:09] I think there was something else too, something more complicated [16:10] or maybe that was the complicated thing that I didn't want to do at the time [16:23] *** naTmeg has joined #jsmess [16:23] evening [16:23] Hey, naTmeg [16:23] I opened the conversation with Mike [16:23] SketchCow [16:23] He sounds pretty convinced they can't even help us. [16:23] We'll see if they want to hurt us. :) [16:24] I assume conversations already happened with him and you, hence use of the AMOS [16:24] ahm, he not willing to give to roms now? [16:24] he's -to [16:25] but he's the copyright-owner or i'm wrong? [16:25] He claims third-party entanglements prevent him. But we'll see what he does. [16:28] well, it does not really suprise me. i guess they will never release it :/ [16:28] Well, I'd say he claims to have it as part of his will [16:29] But it also seems that he gets to be the punching bag (owner) while also having obligations that strip him of the capabilities of an owner. [16:31] db48x: Let me know if you need me to do anything [16:50] SketchCow: the whole rom-situation is fubar. those roms are 30 years old and they still think they can make some significant bucks out of it, instead of supporting the community. i don't remember the conversation exactly, but it was the same song.. [16:51] I know. That's why I'm bringing light to the situation. [16:52] and how does this continue? [16:52] Which [16:53] about the rom-usage [16:53] There's several points of reference for this one. [16:53] First, this use on the IA is going to bring many more people back to Amiga. [16:54] true [16:54] That's going to increase awareness of it beyond the relatively small amount of players right now. (less than 1000) [16:54] Then I will be sure, based on the discussions, to bring to light the unique situation of the ROMs [16:54] And the Amiga trademark, etc [16:55] ok ic. and you'r going to use AROS meanwhile? [16:56] Yes [16:56] ok [16:56] And this is also going to give you feedback on developing this thing, and maybe even some contributors [16:56] yea :) [16:57] After we nail out this screenshotter thing [16:57] i saw you've already filled some stuff in [16:57] Not really. It's the shotter failing [16:58] what's the prob? [16:58] That's what db48x is on right now. [16:58] We'll figure it out. Once it works, it'll screenshot the 1,600 items I have up already [16:58] yes, i noticed the missing SS [16:59] And we'll see which ones explode, put them into a trashbin collection, and hide them off until later. [16:59] ahh ok [16:59] https://archive.org/details/softwarelibrary_apple [16:59] It'll end up like that. Collections and screenshots ahoy. [17:00] Much easier to find brokes [17:00] https://archive.org/details/softwarelibrary_win3 Windows 3 [17:02] many systems, i like that. [17:03] Oh, dude. [17:04] http://archive.org/details/consolelivingroom [17:04] many, many, many systems. [17:05] heh yeah. how many TB total? [17:05] TB is not a good measurement for pre-1995 systems [17:06] i mean IA [17:06] Oh. [17:06] Then TB is sort of inadequate. But let's do it. [17:06] 30,000. [17:06] nice :) [17:08] Oh, fun new thing for bai and db48x [17:08] Let me double-check [17:10] (Ignore for now.) [17:12] SketchCow: i there anything i can/should do? [17:12] Two things. [17:12] is [17:12] 1. Continue to improve the work on SAE; all improvements will be immediately reflected in the playing of thousands of programs. [17:12] 2. Be prepared for a world where MAME overtakes you [17:13] what does mean 2? [17:13] MAME is a juggernaut [17:14] *** SketchCow sets mode: +oooo arkiver balrog Coderjoe db48x [17:14] *** SketchCow sets mode: +oooo DFJustin GLaDOS joepie91 jvilk [17:14] *** SketchCow sets mode: +oo Lord_Nigh naTmeg [17:14] Functionally, it continues to improve across a mass of platforms and methodologies. [17:15] It's not great at Amiga compared to SAE in the Javascript world. SAE is a very strong leader in this. [17:16] But they have a lot of energy aimed in a lot of directions. [17:16] Hey, you asked what you can do. :) [17:16] I don't know your motivations for writing SAE, any more than you might know my reasons for doing what I'm doing [17:16] yea, was just unsure what you meant [17:16] (Although I can talk at length) [17:16] MAKE is way bigger that SAE imho [17:16] (As all these poor saps in this channel know) [17:19] to 1: i will continue to improve, but i just worked 6 months (4 days/week) on it. i need a break [17:19] Oh, agreed [17:20] have to do some other (important) things the next weeks. i will come back to work on it [17:20] I'm just saying we're about to give you a LOT of data on its use, now. [17:20] Once that screenshot hits, I'm going viral [17:21] i know and thats great. have work tomorrow and then a have 4 days free [17:21] :) [17:22] i have a 3day/week, so some time left for things. like coding :) [17:35] Well, I for one really thank you for the amount of speed and effort you put in from "hey, naTmeg" through to this booting. [17:35] db48x and bai and everyone are incredibly hard workers, but at least they have had months of warning and effort in this [17:36] heh thx, i like doing that :) [17:36] Still, it's like waking up and finding out you've been asked to help put up a circus tent [17:36] And then you meet the lions [17:36] and the creepy clowns [17:36] (I'm the creepy clown) [17:36] *** joepie91 has left Konversation terminated! [17:37] :) [17:41] it's also a suprise for me. i never thought that such a big site as IA would use my simul. so also thanks form my side! i'm curious about the feedback i'll get. [17:48] oh hey naTmeg, question for you [17:48] on Firefox Nightly the Z and Y keys seem to be swapped; I was poking at input.js and found https://github.com/naTmeg/ScriptedAmigaEmulator/blob/master/sae/input.js#L448-L449 [17:48] is that intentional? [17:49] Best... bug ever [17:49] well given the way web APIs work I had no idea if it was intentional or not :P [17:50] it defaults to the us-keyboard. i plan to rework the whole input section. [17:50] yeah, that's fine; I was using US keyboard layout when I came across that behavior [17:50] intput.js is the only old part i took from 0.8.3 [17:51] https://twitter.com/TheStrangeLog by the way [17:51] The best thing [17:53] yipdw: ahh i swapped the keys at line 448. didn't even remembered that.. [17:54] that whole file does need a rework/report [17:54] it was pretty easy to read at least [17:55] the original input-file from winuae is >10k lines. hadn't the time yet [17:55] Oh, bringing that in, too.. [17:55] i will [17:59] i think with 0.9.0 are major feature are implemented. for the future i plan to code fpu-support and maybe port picasso96 (gfx-card), but thats huge. [18:00] all [18:00] good ol QWERTZ [18:01] and i'm sure there are some minor fixes to be done the next weeks.. [18:01] Mostly, I'm fascinated at the "doppler effect" [18:15] it would be nice if mame could do that too instead of popping [18:15] agreed [20:12] i noticed that the mute-button on IA did not work. maybe you just not linked it yet? anyway, as a small contribution for today, iv'e added sae.mute() for later use.. [20:12] nice. works at runtime I assume? [20:13] yes [20:13] muting in mame is a huge hack right now [20:14] maybe we can also get the fullscreen-option to work, buts thats a little more tricky [20:14] there's also something I wanted to ask about the mouse pointer. do you implement pointerlock anywhere in SAE? [20:15] right now you get both your regular pointer and the emulated pointer, and they don't necessarily move at the same speed or occupy the same position [20:15] no, the coords are just passed if mouse is over canvas [20:16] maybe we're scaling the canvas which is causing the speed mismatch [20:16] thats very tricky to sync. i planned that for the future [20:17] on most of our other systems we call canvas.requestPointerLock() when the emulator is focused [20:17] this captures the system mouse pointer, and you get relative delta movements instead [20:17] then the user can release the pointer by hitting esc [20:18] thx, i need to look at this [20:18] that belongs to the input-stuff i want to redo [20:19] works like this (of course it doesn't have to wrap at the edges, that's just their demo) - http://mdn.github.io/pointer-lock-demo/ [20:20] very nice. i'll take a closer look tomorrow [20:21] cool, let me know if you need any help as well [20:22] great thx [20:24] i've already played with the fullscreen-api. i gave up on browser-incompatibilities, don't have a nerve for such things :) [20:24] basic support for it is already implemented, see video.js [20:25] how long ago was that? it was a bit of a pain at first but now I think all browsers implement it in the standard way [20:25] 4-5 months [20:25] not so long [20:29] hmm, not sure I'm following it, I see the toggle_fullscreen function but it seems to just set some config values...where do the changes get applied to the canvas element? [20:34] line 555: getbestmode() and later in create_windows_2() [20:34] fullscreen-support is commented out [20:36] support is basic/alpha. iv'e not sayed it worked, need some work and testing [20:36] oh yeah, just trying to wrap my brain around what you've got so far so I can offer some surgical-strike suggestions rather than just vague recommendations :) [20:37] hehe [20:39] the video-file is a little complex. there are serval winuae files in it. will take some time for me too to get into it again.. [20:43] heh, that file was about 2-3x the size it has now. because i was too lazy to write my own cmdset, i've also (partially) ported DirectDraw to JS, but due to speed reasons i dropped that again.. [20:46] so this project is a manual JS port of a native app? [20:47] yes [20:47] http://scriptedamigaemulator.net/_video_full_new.js something for the trashcan :) [20:55] the first public version i'd released (0.5) was almost written from scratch. i did some experiments with html5/js and so i ended up in sae. some part were ported from various uae-sources. [20:55] cool. it's all very impressive [20:55] starting with v0.6 i ported more and more from winuae, because it coast many time to find all those little infos required that some tiles work [20:57] today, sae is 85% winuae and the rest is own code [20:59] credit also goes to Toni Wilen, the maintainer of winuae. he's a working monster and that since years. without him, sae would not exist in his current state. [20:59] he did a great contribution to the community with winuae [21:00] he just added ppc support, so you could run nextgen-amigas like AmigaOS4 :p [21:01] *** azakai has joined #jsmess [21:29] I wikl probably nedd to screeenshot via amazon [21:33] SketchCow: does it work now? [21:33] the screenshoots [21:34] -o [21:37] no hurry btw, for me it could take 20 more hours :) [21:55] bed-time for me, n8 guys [21:55] *** naTmeg has quit IRC (Leaving) [22:11] night