[01:38] so looking at the first one, 1944, mame says it's 384x224 but in the json it's set to 384x448 [01:40] so you have the choice of doing 384x224 which is pixel exact but too wide or 384x288 which is 4:3 but an uneven vertical stretch which will look kinda crappy in a vertical scroller [01:50] alternatively you can run every horizontal game at a higher res like 640x480 or 800x600, we weren't doing that because of potential performance hit but maybe it doesn't matter anymore [02:19] *** azakai_ has joined #jsmess [03:31] *** pengan has quit IRC (hub.efnet.us irc.efnet.nl) [03:31] *** BnARobin has quit IRC (hub.efnet.us irc.efnet.nl) [03:31] *** dboard420 has quit IRC (hub.efnet.us irc.efnet.nl) [03:33] *** pengan has joined #jsmess [03:33] *** BnARobin has joined #jsmess [03:33] *** dboard420 has joined #jsmess [03:39] So I'll try a few things. [03:40] All the Aspect Ratios in this will be handled. I'm now going down through all of them and we are going to end up with a list of "used to boot, no longer boots". [03:41] Could use your help with ideas [03:41] I suspect there's trends (like it's all of one driver type) [03:44] I'll be done with the list tonight. [03:44] Do you want them in here or want to look at the google doc [03:45] Turns out a lot of them were that I had special merged .zip files because our loader handled one, but then db48x fixed that, and so now qsound_hle.zip and sente6vb.zip needed to be added to others. [03:47] I should say it's absolutely the Sega 16b driver that's causing trouble here, for one. [04:03] Invalid function pointer '0' called with signature 'viiii' [04:13] Trying to get samples working on Zaxxon [04:13] Out of ideas [04:14] Can't make a .zip with "samples/*.wav" and tried putting "*.wav" in a samples.zip, both no dice [04:39] Aspect ratios are fixing fast [05:48] I thought we were just putting the samples into the same zip with the rom [05:51] I don't think anything else will work with our setup [05:52] the sega 16b thing sounds like the same as outrun, I know what it's coming from but I have no idea how to fix it [05:56] looking through the others, crater and niyanpai are probably fixed already by the fix I did for tapper [05:56] the other different errors are probably fixable but will need investigating [06:14] Ok, need to bed [06:14] Zaxxon will get the suggested fix. [06:15] (But of course one ROM update and it's poof again) [06:15] And I got aspect ratio done on almost everything [13:25] *** Sgeo__ has joined #jsmess [13:28] *** Sgeo_ has quit IRC (Read error: Operation timed out) [14:43] https://archive.org/details/arcaderepair [14:43] DFJustin and bai and maybe balrog and others - this is everything that doesn't work. [14:43] and Lord_Nigh [14:43] Lord_Nigh probably can help the most here... [15:16] *** Sgeo__ has quit IRC (Read error: Operation timed out) [15:18] *** Sgeo has joined #jsmess [15:22] Well, it'd be good to know. All of those are dead in the water. [15:26] *** azakai_ has quit IRC (Read error: Operation timed out) [15:48] *** kripken has joined #jsmess [16:08] *** Sgeo has quit IRC (Read error: Operation timed out) [17:10] Most seem to be the sega item [17:10] So, I can't get Zaxxon to do samples, no matter what [17:11] I suspect the loader isn't loading subdirectories [17:22] hmm. looks like many things if not everything from sega super-hang-on/outrun hardware (segaorun.cpp) is broken in jsmess, though i'm not sure why... also stuff from sega system 18 (segas18.cpp) is broken, i.e. moonwalker and a few games from sega system 16b (segas16b.cpp) like dynamite dux [17:23] a lot of the games which aren't working have an fd1094 encrypted 68k cpu, so that could be the common cause [17:23] but not all of them do [17:36] I already tracked outrun down to commit 7285b74fc63b7359f317cca210e9760f191db211 which is the devcb3 rewrite [17:37] stack trace https://gist.github.com/DopefishJustin/3a8e444639875af876db11c81a4ef7a0 [17:38] it's deep in template crap [17:38] the specific error is that it's trying to call a null function pointer [19:11] Well, to framework this [19:11] - I don't know how to fix it [19:12] - db48x won't spend time fixing it, he's on loader and oh yes all of Internet Archive Infrastructure [19:12] So if you guys come up with something, great. [19:12] I WILL say there's a floating thing, worth bringing in azakai or kripken in, which is that the WASM bundles on IA are fuckin' HUGE now [19:12] Easily twice the size [19:19] that could be the memory rewrite [19:20] Yeah, I don't know. Obviously, that's a concern. It's just wasted bandwidth [19:21] Not as big a concern as "35 machines stopped working" [19:21] Meanwhile, there's a page with "all the other emulators that are not arcade" [19:22] I'm going to focus on getting those working, or getting them removed so they don't confuse people. [19:22] After that, we can call out anything that concerns us. [19:22] I'm going to then work with db48x on an upgrade path going forward [19:22] Some way we can do this, audit the updates, and knock out the ones that died. [19:22] And fix [19:23] this thing where MAME makes things ever so slightly worse for accuracy - I get it, but it really messes up this framework [19:38] if wasm sizes increase due to an emscripten change, let me know, that would be a pretty bad bug we need to fix [19:51] https://archive.org/details/cpc_Fuck_Double_A_1987_Speedy doesn't work because it's a floppy disk and cpc464 doesn't have a floppy drive [19:51] should work with cpc6128 but you have to type the load command [19:52] https://archive.org/details/demo_shymmer_megadriv_2013 doesn't work in native either but the 32x driver is unfinished and it's a demoscene prod so there's probably no quick fix [20:12] Right [20:38] I suspect 32x should be backed out and killed [20:38] The others just got experiments and were never looked at for any real amount [20:53] DFJustin: I believe there's some in there that qualify as "Justin wanted them very much" [20:53] Let me know which those are, you can also add them to the spreadsheet as notes [22:23] *** Sgeo has joined #jsmess [22:25] *** kripken has quit IRC (Ping timeout: 268 seconds) [22:39] *** Sgeo_ has joined #jsmess [22:41] *** Sgeo has quit IRC (Read error: Operation timed out) [23:43] *** Sgeo has joined #jsmess [23:44] *** Sgeo_ has quit IRC (Read error: Operation timed out)