| --- Log opened Sun Aug 11 00:00:40 2019 |
| 00:57 | <@celticminstrel> | I still have my 10.7 if that helps (but not homebrew). |
| 00:59 | <&McMartin> | Probably eventually |
| 01:00 | <&McMartin> | I've started development on UQM 0.7.1 |
| 01:00 | <&McMartin> | Step 1 was "get it to run with brew dependencies instead of having a 40 step build process" |
| 01:00 | <&McMartin> | So that works |
| 01:00 | <&McMartin> | And by then end I'll probably still need the 40-step process to get a proper binary out, but I'd *also* like a working brew formula too |
| 01:11 | | Alek [Alek@Nightstar-o723m2.cicril.sbcglobal.net] has quit [[NS] Quit: bbl] |
| 01:21 | | Vorntastic [uid293981@Nightstar-2dc.p8m.184.192.IP] has joined #code |
| 01:21 | | mode/#code [+qo Vorntastic Vorntastic] by ChanServ |
| 01:25 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 01:25 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 02:42 | | Alek [Alek@Nightstar-o723m2.cicril.sbcglobal.net] has joined #code |
| 02:42 | | mode/#code [+o Alek] by ChanServ |
| 03:26 | | Derakon[AFK] is now known as Derakon |
| 03:30 | | Vorntastic [uid293981@Nightstar-2dc.p8m.184.192.IP] has quit [[NS] Quit: Connection closed for inactivity] |
| 03:56 | | Vorntastic [uid293981@Nightstar-2dc.p8m.184.192.IP] has joined #code |
| 03:57 | | mode/#code [+qo Vorntastic Vorntastic] by ChanServ |
| 04:42 | | Derakon is now known as Derakon[AFK] |
| 04:49 | <@celticminstrel> | Well, I do plan to get a new computer soon, but I'll keep this one around indefinitely running 10.6. |
| 04:49 | <@celticminstrel> | So I can run PPC programs sometimes. |
| 04:50 | <@celticminstrel> | (I say programs but I'm mostly thinking of games.) |
| 05:05 | <&McMartin> | Games are programs! |
| 05:05 | <&McMartin> | It looks like I'm probably going to have to drop support for i386 completely for 0.7.1 |
| 05:05 | <&McMartin> | We dropped PPC awhile ago~ |
| 05:06 | <~Vornicus> | 0.7.1 of what |
| 05:07 | <~Vornicus> | ah, uqm |
| 05:11 | <@celticminstrel> | Well, this machine is x64 FTR |
| 05:11 | <@celticminstrel> | Though it doesn't support the 64-bit kernel... |
| 05:15 | <&McMartin> | I guess I don't understand the "so I can run PPC games" - wasn't Rosetta long gone by then? |
| 05:16 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Connection closed] |
| 05:16 | <@celticminstrel> | Rosetta was removed in 10.7. |
| 05:16 | <@celticminstrel> | I'm running 10.7, but I didn't overwrite the previous 10.6 install on this computer, so I can boot into 10.6 and run PPC games. |
| 05:17 | <@celticminstrel> | Or PPC apps in general. |
| 05:17 | <@celticminstrel> | Like GameMaker. |
| 05:19 | <@celticminstrel> | Though even if I had overwritten the previous 10.6 install, I still have the install disks... in fact I still have the 10.4 install disks. So I could in theory boot back to Tiger. |
| 05:20 | <@celticminstrel> | Sadly, that won't allow me to run classic apps though. :( |
| 05:20 | <&McMartin> | Sheepshaver ftw, I guess |
| 05:20 | <@celticminstrel> | Yeah, I guess, tho I never had much luck with Sheepshaver IIRC. |
| 05:20 | <@celticminstrel> | I've managed to get a working Basilisk II, I think. |
| 05:20 | <@celticminstrel> | It's a pain mind you... |
| 05:45 | | * McMartin continues poking about the Swift ecosystem |
| 05:45 | <&McMartin> | ... this library is called Starscream |
| 05:46 | <&McMartin> | Presumably for when you want a fully conforming WebSockets implementation that betrays you at every possible opportunity |
| 05:46 | < Yossarian> | https://www.anandtech.com/show/14694/amd-rome-epyc-2nd-gen |
| 05:46 | <@celticminstrel> | XD |
| 05:47 | < Yossarian> | benchmarks fresh |
| 05:50 | <&McMartin> | Hahahaha |
| 05:50 | | * McMartin also checks the TFWiki to make sure |
| 05:51 | <&McMartin> | Yep |
| 05:51 | <&McMartin> | The Objective-C equivalent of it is called Jetfire |
| 05:51 | <&McMartin> | For when, I suppose, you want the betrayal to actually work and possibly do some good |
| 05:58 | <&McMartin> | Also re UQM 0.7.1, the flagship upgrade is going to be "SDL2 backend" and that may punch up the minimum sysreq |
| 07:12 | <@celticminstrel> | Battle for Wesnoth did end up dropping 10.7 support, though I don't recall if SDL2 was related to the reason for it... in any case, I can try UQM on 10.7 or even 10.6 if you'd like. |
| 07:21 | | Kindamoody[zZz] is now known as Kindamoody |
| 07:24 | | Kindamoody [Kindamoody@Nightstar-eubaqc.tbcn.telia.com] has quit [[NS] Quit: Rebooting] |
| 07:27 | | Kimo|autojoin [Kindamoody@Nightstar-eubaqc.tbcn.telia.com] has joined #code |
| 07:27 | | mode/#code [+o Kimo|autojoin] by ChanServ |
| 07:29 | | Kimo|autojoin is now known as Kindamoody |
| 08:24 | | Alek [Alek@Nightstar-o723m2.cicril.sbcglobal.net] has quit [[NS] Quit: ] |
| 08:38 | | Alek [Alek@Nightstar-o723m2.cicril.sbcglobal.net] has joined #code |
| 08:38 | | mode/#code [+o Alek] by ChanServ |
| 08:56 | | * McMartin installs Starscream, is promptly betrayed by it, Xcode, and the App Sandbox |
| 08:56 | <&McMartin> | checks out |
| 09:00 | <~Vorntastic> | Did it really |
| 09:00 | <&McMartin> | Well, none of the test code worked, and the test code also started out acting like it was a Mac application while secretly being for iOS and confusing Xcode terribly |
| 09:01 | <&McMartin> | And then it turns out that it didn't work because my actual Mac code defaulted to "not allowed to be a network client", which is kind of a bummer when, like Starscream, your whole reason for existence is to implement a networking protocol |
| 09:01 | <&McMartin> | So now I've gotten that sorted out and the test programs run. |
| 09:08 | <&McMartin> | I suppose this *also* checks out, in that its feeble treachery was swiftly dealt with |
| 09:12 | <&McMartin> | celticminstrel: Now that I check it, apparently SDL2 will *run* on anything back to 10.5, but it can only be *built* on the 10.7 or later SDKs. |
| 09:13 | | * McMartin is pretty firm on 10.8 being the oldest acceptable SDK at this point, since that was when Objective-C took its final form, and also its final form was the first where ObjC becomes a minimally acceptable application development language. |
| 09:22 | <&McMartin> | Also I have decided. The Consummate Vs will be all the various third-party packages I'm experimenting with |
| 09:29 | <~Vorntastic> | And the chiaroscuro? |
| 09:30 | <&McMartin> | Burninatino |
| 09:30 | <&McMartin> | Also burnination |
| 09:30 | <~Vorntastic> | Trogdor strikes again! |
| 09:30 | <&McMartin> | It also occurs to me that we have a lot of dragon imagery in our codebase anyway |
| 09:31 | <&McMartin> | So this is "in character" for a throwaway project |
| 09:31 | <&McMartin> | Also Swift starts with an S |
| 09:31 | <&McMartin> | And then has a more different S |
| 09:31 | | Kindamoody is now known as Kindamoody|afk |
| 10:41 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has joined #code |
| 14:25 | | Derakon[AFK] is now known as Derakon |
| 14:40 | | gnolam [lenin@Nightstar-hfrbpd.cust.bahnhof.se] has quit [Ping timeout: 121 seconds] |
| 15:18 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 15:18 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 16:20 | | Vash [Vash@Nightstar-sjaki9.res.rr.com] has joined #code |
| 16:34 | | * Vornicus eyes this bit of code |
| 16:36 | <~Vornicus> | it's in the init routine, but it's skipped over: for each of the 50 initialized elements of this array, it compares against 0xff and if it's different it jumps further on... but the elements are in the range [0..3] |
| 16:50 | | Kindamoody|afk is now known as Kindamoody |
| 16:52 | <~Vornicus> | cmp $#18 bcs (somewhere) --- that's ...carry is set on a subtraction A - B if B > A, approximately, and in this case it's accumulator - listed value, which means this is if A < 0x18 |
| 18:50 | <&McMartin> | Might just be dead code |
| 18:51 | <~Vornicus> | might be, but it does the check 50 times |
| 18:54 | <&McMartin> | "but it's skipped over"? |
| 18:54 | <&McMartin> | Also yes, BCS/BCC is unsigned compare, BMI/BPL is signed compare |
| 18:59 | <~Vornicus> | it goes: ldx #$31; loopstart: lda $2500, X; cmp $#$ff; bne later; ... later ... dex; bpl loopstart; |
| 19:00 | <~Vornicus> | Which should mean that this index's page-25 byte should be -1 for the code between the jump and later to run |
| 19:01 | <&McMartin> | Page 37~ |
| 19:01 | <~Vornicus> | yes yes |
| 19:02 | <~Vornicus> | But this is in the init code and page 37 is filled with nonnegative numbers |
| 19:02 | <&McMartin> | I'd say "OK, skip as no op for now, move on |
| 19:02 | <&McMartin> | Maybe it will make sense later |
| 19:03 | <&McMartin> | Maybe it once made sense and then got skipped |
| 19:03 | <&McMartin> | Maybe it's useless defensive coding |
| 19:03 | <~Vornicus> | Maybe. |
| 19:04 | <~Vornicus> | Yeah I've just translated it into non-assembler in my notes and moved on figuring if i end up back there it will be more obvious what it's doing |
| 19:04 | <&McMartin> | Zanac has a *lot* of dead code |
| 19:12 | | Kindamoody is now known as Kindamoody|afk |
| 19:22 | <&McMartin> | To the extent that I suspect they licensed or prewrote an engine and then just bunged it in |
| 19:22 | <&McMartin> | Oh yeah, that's the other thing that might be relevant |
| 19:23 | <&McMartin> | If this was written with an older tool he would basically be assembling code directly into memory, where it would then sit |
| 19:23 | <&McMartin> | If he had to modify it later, it wouldn |
| 19:23 | <&McMartin> | 't "back up" everything else to cover the gap |
| 19:23 | <&McMartin> | So then it would be simpler to skip/delete or cheat a test |
| 19:49 | <~Vornicus> | Truth |
| 19:49 | <~Vornicus> | mmm. do jump back to just before this segment at the end of this hunk |
| 19:50 | <&McMartin> | Ooh maybe it's a periodic map scan |
| 19:50 | <~Vornicus> | Perhaps. |
| 19:50 | <&McMartin> | would definitely recommend "first identify puzzle pieces, then knit them together" |
| 19:51 | <&McMartin> | Sadly no C64 emulator has the kind of interactive monitor that's actually worth using to watch stuff evolve. |
| 19:51 | <~Vornicus> | Yeah, I'm doing that. So far it's only used three subroutines, which I've called "wild page writer" and "wild page reader" and "get two noise bits" |
| 19:54 | <~Vornicus> | why are they called ASL and LSR |
| 19:54 | <~Vornicus> | (shift left and right respectively) |
| 19:56 | <~Vornicus> | I've gotten what I need so far from vice's monitor |
| 20:09 | <~Vornicus> | Aha |
| 20:09 | <~Vornicus> | dec $2500,x |
| 20:11 | <~Vornicus> | I think I figured it out: this is an updater loop of some sort, 2500 is the countdown between updating individual entities; 2400 is the size of that countdown (some things update faster, some slower?); 2300 and 2200 are direction vectors; 2100 and 2000 are locations |
| 20:29 | <~Vornicus> | While it's doing that it has a thing where it... okay - it... it's doing something to the wildcard space repeatedly |
| 20:36 | <~Vornicus> | Okay . lda $0b4b the wildcard low byte, compare to ---hahaha! Okay! Now we are in *business*! I think i know what this whole thing does now. |
| 20:38 | <~Vornicus> | So, mcm, you know how in the game, when you kill something, it explodes in a shower of particles? |
| 20:38 | <~Vornicus> | *the particle sprite is generated randomly on game init* |
| 21:29 | <~Vornicus> | Yeah, here it goes: it generates 50 particles in a 4x4 area at the center of the sprite; each particle has a random position, direction, and inverse velocity. It then runs 22 frames of animation, updating each particle and placing it on the sprite. |
| 21:31 | <~Vornicus> | ...it does this 4 times. |
| 21:31 | <~Vornicus> | so there are ...apparently 4 distinct explosion sprites. |
| 21:31 | <~Vornicus> | sprite sequences. |
| 22:24 | | Kindamoody|afk is now known as Kindamoody |
| 22:52 | | Vash [Vash@Nightstar-sjaki9.res.rr.com] has quit [[NS] Quit: Leaving] |
| 22:58 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has quit [Ping timeout: 121 seconds] |
| 23:21 | <&McMartin> | Arithmetic Shift Left, Logical Shift Right |
| 23:22 | <~Vornicus> | weirdly there isn't an ASR |
| 23:25 | <&McMartin> | Yeah, generally you end up wanting ROR instead |
| 23:33 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Connection closed] |
| 23:50 | | Kindamoody is now known as Kindamoody[zZz] |
| --- Log closed Mon Aug 12 00:00:42 2019 |