ScummVM IRC Log - 26 Oct 2014

[Back to Index]

[00:06] <-- lukaslw left irc: Quit: Page closed
[00:10] <-- SugarCube left irc: Ping timeout: 272 seconds
[00:10] <-- WooShell left irc: Quit: Walking upside down in the sky, between the satellites passing by. Gliding along the black rainbow, I fly away with my shadow. Scratching the moon like a DJ, the night follows its odyssey.
[00:18] <-- dreammaster left irc:
[00:38] --> Manul1206 joined #scummvm.

[00:40] <clone2727> Maybe giving some more info about why it doesn't run might help us
[00:40] <clone2727> And then file a bug report about it
[00:41] <Manul1206> I don't know -it just loads a progress bar then exits to desktop
[00:41] <clone2727> http://developer.android.com/tools/help/logcat.html
[00:41] <Manul1206> It stays in running apps list btw, but switching to it doors only brief flash of black screen and then guess to desktop again
[00:42] <Manul1206> Ok, I'll try this tool
[00:42] <clone2727> or https://play.google.com/store/apps/details?id=org.jtb.alogcat&hl=en
[00:48] <-- waltervn left irc: Quit: Ik ga weg
[00:51] <Manul1206> I am not an Android expert, this program (alogcat) shows only menu and blank space instead of logs. Do I need root?
[00:52] <clone2727> You don't need root
[00:52] <Manul1206> Regular logcat in terminal shows some info but i dont know the parameters i need
[00:53] <Manul1206> It does not display anything about scummvm if run without parameters
[00:55] --> ignalina joined #scummvm.
[00:56] <clone2727> You might need to turn Android debugging on
[00:56] <clone2727> http://developer.android.com/tools/help/adb.html#Enabling
[01:08] <-- ignalina left irc: Quit: ignalina
[01:09] --> SugarCube joined #scummvm.
[01:17] --> Manul012 joined #scummvm.
[01:21] <Manul012> Hello, that's me again from my PC now. I used
[01:21] <Manul012> Usb debugging
[01:21] <Manul012> here are the results in pastebin
[01:21] <Manul012> http://pastebin.com/LmenjhpV
[01:21] <-- Manul1206 left irc: Quit: AndroIRC - Android IRC Client ( http://www.androirc.com )
[01:22] <-- SugarCube left irc: Ping timeout: 272 seconds
[01:26] <clone2727> He clearly missed the "And then file a bug report about it" part
[01:27] <Manul012> ok
[01:29] <Lightkey> clone2727: you clearly missed he's still here :-D
[01:30] <clone2727> Granted, it looks like an ART bug
[01:30] <-- D0SFreak left irc: Ping timeout: 244 seconds
[01:30] <clone2727> But we might as well track it anyway
[01:33] <Lightkey> <Beavis> F/art, hrhr
[01:35] --> SugarCube joined #scummvm.
[01:38] <Manul012> Ok, I created the ticket
[01:38] <Manul012> https://sourceforge.net/p/scummvm/bugs/6741/
[01:38] <clone2727> Thanks!
[01:43] <-- Manul012 left irc: Quit: Page closed
[01:53] <Lightkey> https://groupees.com/bm15 SPY Fox in "Dry Cereal" (plus another game of your choosing) for $1 the next 11 days
[01:53] <Lightkey> also Rex Nebular and the Cosmic Gender Bender now on Steam
[02:16] --> Vampire0_ joined #scummvm.
[02:20] <-- Vampire0 left irc: Ping timeout: 255 seconds
[02:21] <-- clone2727 left irc: Remote host closed the connection
[02:38] --> clone2727 joined #scummvm.
[02:38] #scummvm: mode change '+o clone2727' by ChanServ!ChanServ@services.
[02:50] --> DominusExult joined #scummvm.
[02:50] <-- Dominus left irc: Ping timeout: 244 seconds
[02:50] Nick change: DominusExult -> Dominus
[04:00] <-- SylvainTV left irc: Ping timeout: 255 seconds
[04:03] --> Polynomial-C joined #scummvm.
[04:05] <-- Poly-C left irc: Ping timeout: 250 seconds
[04:06] <-- Cheeseness left irc: Quit: Leaving.
[04:18] <-- Ramal left irc: Ping timeout: 258 seconds
[04:22] <-- ScummBot left irc: Ping timeout: 245 seconds
[04:23] --> ScummBot joined #scummvm.
[05:01] --> Poly-C joined #scummvm.
[05:04] <-- Polynomial-C left irc: Ping timeout: 265 seconds
[05:12] <-- Schnabeltierchen left irc: Ping timeout: 265 seconds
[05:13] --> Schnabeltierchen joined #scummvm.
[05:42] <-- demonimin left irc: Ping timeout: 260 seconds
[05:45] --> LordHoto joined #scummvm.
[05:45] #scummvm: mode change '+o LordHoto' by ChanServ!ChanServ@services.
[05:48] --> demonimin joined #scummvm.
[05:48] <-- demonimin left irc: Changing host
[05:48] --> demonimin joined #scummvm.
[05:56] --> Jon_God joined #scummvm.
[06:22] --> Cheeseness joined #scummvm.
[06:33] <-- Lightkey left irc: Ping timeout: 265 seconds
[06:46] --> Lightkey joined #scummvm.
[06:47] --> frankyboy_ joined #scummvm.
[07:30] <-- chkr left irc: Ping timeout: 256 seconds
[07:30] <-- broosky left irc: Ping timeout: 255 seconds
[07:30] --> broosky joined #scummvm.
[07:31] --> chkr joined #scummvm.
[07:35] --> borosky joined #scummvm.
[07:35] <-- broosky left irc: Ping timeout: 245 seconds
[07:40] --> Raziel^ joined #scummvm.
[07:40] #scummvm: mode change '+v Raziel^' by ChanServ!ChanServ@services.
[07:40] <-- LordHoto left irc: Quit: leaving
[07:41] <-- clone2727 left irc: Quit: later
[07:46] <-- borosky left irc: Ping timeout: 260 seconds
[07:47] --> broosky joined #scummvm.
[08:23] --> t0by joined #scummvm.
[08:23] #scummvm: mode change '+v t0by' by ChanServ!ChanServ@services.
[08:25] <-- Jon_God left irc: Remote host closed the connection
[08:33] --> GitHub139 joined #scummvm.
[08:33] <GitHub139> [scummvm] sev- pushed 2 new commits to master: http://git.io/D8cbkg
[08:33] <GitHub139> scummvm/master 522cd1c Eugene Sandulenko: PRINCE: Fix warning
[08:33] <GitHub139> scummvm/master 3b9ac03 Eugene Sandulenko: PRINCE: Fix variable sign
[08:33] GitHub139 (GitHub139@192.30.252.38) left #scummvm.
[08:35] --> GitHub109 joined #scummvm.
[08:35] <GitHub109> [scummvm] sev- pushed 1 new commit to master: http://git.io/sYthjQ
[08:35] <GitHub109> scummvm/master 123e52f Eugene Sandulenko: PRINCE: Fix another variable sign
[08:35] GitHub109 (GitHub109@192.30.252.39) left #scummvm.
[08:53] <fuzzie> hm, that accessed stale local reference thing looks like our bug
[08:55] <fuzzie> omg, yes, really stupid mistake on my part :(
[08:55] <fuzzie> I am amazed that ever worked
[08:55] <fuzzie> but I guess this would be an appropriate moment to take pr#433 instead
[09:01] <Raziel^> fuzzie: i vote for #512
[09:01] <Raziel^> ;-)
[09:05] <fuzzie> so I have time as of the last few days
[09:05] <fuzzie> but I'm really sick :(
[09:12] --> _marc` joined #scummvm.
[09:19] <Raziel^> fuzzie: chicken soup, hot tea and lots of blankets
[09:24] <-- _marc` left irc: Ping timeout: 260 seconds
[09:32] --> _sev|work joined #scummvm.
[09:32] #scummvm: mode change '+o _sev|work' by ChanServ!ChanServ@services.
[09:33] <-- _sev|work left irc: Client Quit
[09:34] --> TMM joined #scummvm.
[09:36] --> _marc` joined #scummvm.
[09:36] <-- Mellified_Man left irc: Ping timeout: 244 seconds
[09:41] --> Mellified_Man joined #scummvm.
[09:47] --> _sev|work joined #scummvm.
[09:47] #scummvm: mode change '+o _sev|work' by ChanServ!ChanServ@services.
[09:48] <-- _sev|work left irc: Client Quit
[10:06] <-- _marc` left irc: Quit: _marc`
[10:11] --> edheldil_ joined #scummvm.
[10:13] --> waltervn joined #scummvm.
[10:17] --> D0SFreak joined #scummvm.
[10:24] <Raziel^> anyone using ubuntu here?
[10:26] <fuzzie> what do you need?
[10:27] <Raziel^> i have a dual boot with win 7
[10:27] <Raziel^> upgraded to 14.04
[10:27] <Raziel^> and now win 7 is stuck on the grub screen, all purple
[10:27] <Raziel^> never goes any further
[10:30] <Raziel^> how can i fix that? i'm not good with linux
[10:32] <fuzzie> you can get to grub, and win7 is in the list?
[10:32] <Raziel^> yes
[10:33] <Raziel^> choosing win 7 gives a short beep from the machine and stalls with an all purple screen
[10:34] <fuzzie> does it actually stall?
[10:34] <fuzzie> or do you just not get video?
[10:34] <Raziel^> well, i have sitting it there for 20 minutes now, the purple screen is still here
[10:35] <Raziel^> i don't get any welcome sounds and such
[10:35] <Raziel^> and no hdd activity
[10:37] --> _marc` joined #scummvm.
[10:48] --> ajax16384 joined #scummvm.
[10:48] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[10:51] --> uruk-hai joined #scummvm.
[10:51] #scummvm: mode change '+o uruk-hai' by ChanServ!ChanServ@services.
[10:55] --> pcercuei joined #scummvm.
[10:58] <pcercuei> _sev: hi, I see you pushed the new gcw zero port to github. So is it working already, or are there more changes needed?
[10:59] <_sev> more changes are needed
[10:59] <_sev> working on them
[10:59] <pcercuei> hmm
[10:59] <pcercuei> like what?
[10:59] <_sev> let me commit my intermediate results
[11:00] <pcercuei> one thing to do that would be very useful, is enable the mouse on joystick by default
[11:01] <_sev> yes, will try to do it
[11:01] --> GitHub56 joined #scummvm.
[11:01] <GitHub56> [scummvm] sev- pushed 1 new commit to master: http://git.io/d8BGqQ
[11:01] <GitHub56> scummvm/master d4c687d Eugene Sandulenko: GCW0: Further work on the port
[11:01] GitHub56 (GitHub56@192.30.252.34) left #scummvm.
[11:01] <_sev> you never told me what configure parameters you used
[11:02] <_sev> and why do you have SamsungTV stuff in your port directory
[11:02] <_sev> ...currently for some reason .scummvmrc doesn't get copied
[11:02] <_sev> looking into this problem right now
[11:03] <pcercuei> because that was a long time ago :) I don't remember the parameters
[11:04] <-- Cheeseness left irc: Quit: Leaving.
[11:04] <-- ajax16384 left irc: Read error: Connection reset by peer
[11:05] --> uruk_hai joined #scummvm.
[11:05] #scummvm: mode change '+o uruk_hai' by ChanServ!ChanServ@services.
[11:05] <_sev> that's why you need to stick to GPL, and publish your sources right away
[11:06] --> ajax16384 joined #scummvm.
[11:06] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[11:06] --> zear joined #scummvm.
[11:07] Action: pcercuei slaps zear
[11:07] <zear> hey guys
[11:08] <-- uruk-hai left irc: Ping timeout: 265 seconds
[11:11] <_sev> pcercuei: so how to enable the mouse joystick by default?
[11:11] --> WooShell joined #scummvm.
[11:11] <pcercuei> there is a config option for it, so you could just enable it in the default config
[11:13] <WooShell> good meowning =^.^=
[11:16] <_sev> pcercuei: could you please be just a bit more specific?
[11:17] <-- ajax16384 left irc: Read error: Connection reset by peer
[11:18] <_sev> pcercuei: do you mean --joystick?
[11:18] --> ajax16384 joined #scummvm.
[11:18] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[11:19] <_sev> yep, that's it
[11:20] <pcercuei> yes
[11:20] <_sev> there
[11:20] <pcercuei> but that's for ./configure, right?
[11:20] <_sev> no
[11:20] --> GitHub60 joined #scummvm.
[11:20] <GitHub60> [scummvm] sev- pushed 2 new commits to master: http://git.io/KoJzXg
[11:20] <GitHub60> scummvm/master 2699259 Eugene Sandulenko: GCW0: Correct launch options
[11:20] <GitHub60> scummvm/master 0ee20ce Eugene Sandulenko: GCW0: Enable analog joystick by default
[11:20] GitHub60 (GitHub60@192.30.252.36) left #scummvm.
[11:21] <_sev> just committed
[11:21] <_sev> one question
[11:21] <_sev> I see that the mouse button is 'B' instead of 'A". Is it OK or better to change it?
[11:21] <pcercuei> ah ok, I thought it was a config option (in scummvmrc)
[11:21] <_sev> I just don't know yet GCW0 guidelines
[11:21] <pcercuei> yes, please do
[11:21] <_sev> please suggest the key map
[11:22] <_sev> A -> left click
[11:22] <_sev> B -> right click
[11:22] <_sev> START -> F5 menu?
[11:22] <_sev> Y -> middle click???
[11:22] <_sev> anything which you consider viable
[11:22] <_sev> and I'll work on it in the evening
[11:22] <pcercuei> is middle click used in games?
[11:22] <_sev> yes, sometimes, but very rarely
[11:23] <_sev> also probalby left shoulder -> Shift??
[11:23] <pcercuei> one should be Esc
[11:24] <pcercuei> the standard way to open the main menu is the start button, or the power switch (which delivers the "home" key event when raised then lowered without any combo entered)
[11:25] <pcercuei> maybe make X the middle button
[11:26] <pcercuei> Y as the on-screen keyboard, Start as the menu, Select as Esc?
[11:34] <_sev> ok, will do that
[11:34] <_sev> and shift?
[11:34] <_sev> shoulder button?
[11:34] <_sev> A -> left click
[11:35] <_sev> B-> right click
[11:35] <_sev> Start -> F5 menu
[11:35] <_sev> X -> middle button
[11:35] <_sev> Y -> keyboard
[11:35] <_sev> Select -> ESC
[11:35] <_sev> I'm off
[11:36] <pcercuei> what is shift for?
[11:36] <pcercuei> ok, see you
[11:44] --> SylvainTV joined #scummvm.
[11:44] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[12:01] <-- frankyboy_ left irc: Remote host closed the connection
[12:08] <-- sact left irc: Quit: Leaving.
[12:09] <-- Poly-C left irc: Quit: GNU/Linux, because I'd rather own a free OS than steal one that's not worth paying for.
[12:14] <-- uruk_hai left irc: Quit: Leaving
[12:20] --> Polynomial-C joined #scummvm.
[12:25] <zear> hey _sev
[12:25] <zear> so you're the guy who's working on an official GCW0 port of ScummVM?
[12:26] <Raziel^> _sev: too late, they found you...
[12:26] <Raziel^> :-D
[12:26] <zear> oh I knew about it for months from his comments on the KickStarter of the project ;P
[12:26] <zear> I just noticed GCW0 related commits in the master today
[12:27] <zear> and thus came to check if sev received his unit already :)
[12:27] --> uruk-hai joined #scummvm.
[12:27] #scummvm: mode change '+o uruk-hai' by ChanServ!ChanServ@services.
[12:27] <zear> <_sev> I see that the mouse button is 'B' instead of 'A". Is it OK or better to change it?
[12:27] <zear> I'd strongly prefer the left mouse click to be mapped to 'A'
[12:28] <zear> since this is what the majority of software uses as their "accept" button
[12:28] <zear> while 'B' should be the right click, as it's "go back" button in games
[12:28] <zear> all the software I ported follows this order
[12:30] <-- _marc` left irc: Read error: Network is unreachable
[12:53] --> Endy_ joined #scummvm.
[12:53] --> Huerbine joined #scummvm.
[12:54] --> zear_ joined #scummvm.
[12:54] --> Harekiet2 joined #scummvm.
[12:55] <-- WooShell left irc: Disconnected by services
[12:55] Nick change: Huerbine -> WooShell
[12:56] <-- Harekiet left irc: Ping timeout: 260 seconds
[12:56] <-- Endy left irc: Ping timeout: 260 seconds
[12:56] <-- mramberg left irc: Ping timeout: 260 seconds
[12:56] <-- zear left irc: Ping timeout: 260 seconds
[12:56] <-- KingGuppy left irc: Ping timeout: 260 seconds
[12:56] Nick change: zear_ -> zear
[12:57] --> KingGuppy joined #scummvm.
[13:31] --> Robin_Watts_ joined #scummvm.
[13:34] <wjp> does anybody know anything about this Hugo email on the list?
[13:34] --> fuzzie_ joined #scummvm.
[13:34] --> sirlemonhead joined #scummvm.
[13:34] <wjp> I do see this "Workaround: SCRIPT.FON doesn't load properly at the moment" comment in the hugo engine
[13:34] --> heroux__ joined #scummvm.
[13:35] --> Styrbjorn joined #scummvm.
[13:35] --> heroux___ joined #scummvm.
[13:39] Schnabeltierchen (~Schnabelt@xdsl-87-79-167-141.netcologne.de) got netsplit.
[13:39] heroux_ (sandroco@50708355.static.ziggozakelijk.nl) got netsplit.
[13:39] heroux (sandroco@50708355.static.ziggozakelijk.nl) got netsplit.
[13:39] Robin_Watts (~chatzilla@91.84.110.196) got netsplit.
[13:39] fuzzie (~fuzzie@li141-249.members.linode.com) got netsplit.
[13:39] ^Styr_ (~foobar@c83-249-213-144.bredband.comhem.se) got netsplit.
[13:39] Nick change: heroux___ -> heroux
[13:39] Possible future nick collision: heroux
[13:43] <-- uruk-hai left irc: Quit: Leaving
[13:45] Schnabeltierchen (~Schnabelt@xdsl-87-79-167-141.netcologne.de) returned to #scummvm.
[13:46] <Raziel^> Best Vaseline ad ever...
[13:46] <Raziel^> http://cdn.themetapicture.com/media/funny-vaseline-ad-girl-sitting.jpg
[13:50] fuzzie (~fuzzie@li141-249.members.linode.com) got lost in the net-split.
[13:50] ^Styr_ (~foobar@c83-249-213-144.bredband.comhem.se) got lost in the net-split.
[13:50] Robin_Watts (~chatzilla@91.84.110.196) got lost in the net-split.
[13:50] heroux_ (sandroco@50708355.static.ziggozakelijk.nl) got lost in the net-split.
[13:55] <-- heroux left irc: Ping timeout: 255 seconds
[13:55] Nick change: heroux__ -> heroux
[13:55] --> heroux_ joined #scummvm.
[13:56] --> Ramal joined #scummvm.
[13:59] <-- D0SFreak left irc: Read error: Connection reset by peer
[14:01] --> _marc` joined #scummvm.
[14:30] --> Sylvain joined #scummvm.
[14:33] <-- SylvainTV left irc: Ping timeout: 255 seconds
[14:59] <-- _marc` left irc: Quit: _marc`
[15:20] --> SimSaw joined #scummvm.
[15:20] #scummvm: mode change '+v SimSaw' by ChanServ!ChanServ@services.
[15:21] --> _marc` joined #scummvm.
[15:30] <t0by> <Raziel^> http://cdn.themetapicture.com/media/funny-vaseline-ad-girl-sitting.jpg <=== I swear to god I'm not getting this. Isn't this an ad for a hand cream anyway.
[15:31] <Raziel^> t0by: :-D
[15:31] <Raziel^> take a closer look at the metal things that ships are secured with...
[15:32] <t0by> I never said I /wanted/ to get it.
[15:32] <Raziel^> heh, kk :-P
[15:33] Nick change: fuzzie_ -> fuzzie
[15:37] --> uruk-hai joined #scummvm.
[15:37] #scummvm: mode change '+o uruk-hai' by ChanServ!ChanServ@services.
[15:37] <-- uruk-hai left irc: Client Quit
[15:39] --> lukaslw joined #scummvm.
[15:39] #scummvm: mode change '+o lukaslw' by ChanServ!ChanServ@services.
[15:40] --> clone2727 joined #scummvm.
[15:40] #scummvm: mode change '+o clone2727' by ChanServ!ChanServ@services.
[15:41] --> D0SFreak joined #scummvm.
[15:53] <-- Ramal left irc: Ping timeout: 265 seconds
[15:58] <clone2727> wjp: That one font has the version field set to 0x100. According to Microsoft documentation, it doesn't exist. According to freetype2, it doesn't exist. I never wanted to load Windows into IDA to find out the truth
[16:01] --> jdxjet joined #scummvm.
[16:10] <_sev> zear: yes, I am
[16:10] <-- jdxjet left irc:
[16:10] <_sev> zear: I am thinking about following what PSP port does
[16:11] <zear> _sev, if I were to guess I'd say the previous GCW0 port was recompiled based on GP2X config
[16:11] <zear> because it uses the same keys GP2X
[16:11] <zear> but the button layout is very different on GCW0
[16:11] <pcercuei> no, dingoo A320
[16:11] <_sev> I saw Dingux inside
[16:11] <_sev> yep
[16:11] <zear> pcercuei, which was based on GP2X config
[16:12] <zear> here: http://origin.arstechnica.com/reviews/hardware/gp2x-review.media/003.jpg
[16:12] <zear> _sev, what does the PSP port do?
[16:12] <zear> there are unwritten keymapping standards for GCW0, I'd prefer if ScummVM followed them (maybe it's time I write them down)
[16:13] <_sev> zear: https://github.com/scummvm/scummvm/blob/master/backends/platform/psp/README.PSP
[16:13] <wjp> clone2727: How mysterious. I hope the PR will be useful
[16:13] <_sev> zear: just tell me those standards, or rather how PSP differs from it
[16:15] <zear> _sev, http://pastebin.com/L8hwJfEF
[16:16] <zear> that's what can apply to any kind of game/program
[16:16] <_sev> ok, not really different except the start button
[16:16] <zear> specific game genres will follow more advanced button standards
[16:16] <zear> ie. first person shooters
[16:17] <_sev> what I'll do
[16:17] <_sev> I'll try to use keymapper, and then let people voice their opinions on our forums if they want default to be different
[16:18] <zear> I guess that is a way to solve this, however remember that people might not be familiar with these standards
[16:18] <zear> and they might want to make a "homer simpson's car" of keymapping
[16:18] <_sev> what needs to be decided.
[16:18] <_sev> ESC, F5 (menu), Virtual Keyboard, Mouse
[16:18] <zear> the most essencial part is that A is left mouse click and B is right mouse click
[16:18] <zear> the rest shouldn't be as important
[16:18] <_sev> where to ut those
[16:19] <_sev> Mouse=Pause
[16:19] <zear> we don't need all of them, I believe
[16:19] <zear> ESC doubles pause
[16:19] <zear> unless pause can still let you reposition the cursor
[16:19] <zear> for some quick typing actions
[16:19] <_sev> and we have only 3 buttons
[16:20] <_sev> or...
[16:20] <zear> oh, ESC as in skip cutscene?
[16:20] <_sev> ESC=X
[16:20] <zear> not as in F5 menu
[16:20] <_sev> Virtual Keyboard=Y
[16:20] <_sev> F5=SELECT
[16:20] <_sev> Pause=Start
[16:20] <zear> virtual keyboard is usually done with start/select or R in other emulators on GCW0
[16:20] <_sev> aha, R
[16:20] <_sev> ok, I could use it
[16:20] <zear> try to keep the face buttons (A/B/X/Y) to most common actions
[16:21] <_sev> mouse left/right, esc and menu
[16:21] <zear> A/B are taken for mouse buttons
[16:22] <zear> i'd say put F5 (menu) under START
[16:22] <zear> ESC under SELECT
[16:22] <zear> virtual keyboard under R
[16:22] <zear> since these are all less commonly used actions
[16:23] <zear> so they're offloaded to buttons of minor importance
[16:23] <_sev> ok
[16:23] <zear> is there any use of Pause when you have F5 Menu?
[16:23] <_sev> yes, we can drop it
[16:23] <zear> sweet
[16:23] <_sev> and L fir shift
[16:24] <_sev> fir=for
[16:24] <zear> what does shift do?
[16:24] <zear> uppercase letters in keyboard mode?
[16:24] <_sev> shift click, plus special options in main menu
[16:24] <_sev> I can do it GUI-specific, though
[16:24] <zear> ah, like mass game add?
[16:24] <_sev> (if key mapper works for me)
[16:24] <_sev> yes
[16:24] <zear> yes, I like that mapped to L
[16:25] <zear> how does scummvm handle extra buttons required by certain games? Ie. arcade sections where you need to use keyboard instead of mouse?
[16:26] <zear> if there are common keys for them, I'd map them to X/Y
[16:26] <-- TMM left irc: Ping timeout: 265 seconds
[16:26] <_sev> if key mapper works (and i'll know it in couple of minutes), then you can have per-game mappings, and even switch them within same engine, though that feature is not widely used yet
[16:27] <zear> oh sweet
[16:27] <zear> oh yea, one thing that would be nice - custom key label names
[16:27] <zear> so not "LCTRL" but "A"
[16:28] <_sev> sure
[16:28] <zear> since people not necessarily know what keyboard buttons are mapped to which physical GCW0 buttons
[16:28] <_sev> that's not hard to do
[16:28] <zear> yep
[16:29] <zear> and I think both d-pad and joystick can work as cursor movement?
[16:29] <zear> at the same time
[16:30] <_sev> yes, it does already
[16:30] <zear> awesome
[16:32] <zear> also, when the port is complete, I'd like to add it to our official repository: repo.gcw-zero.com
[16:33] <pcercuei> what icon will you use in the download page of scummvm.org? :)
[16:33] <zear> oh yea, we should have some "official" GCW0 icons
[16:34] <zear> this one: http://www.gianas-return.de/images/web/ico_gcwzero.gif
[16:35] <zear> pcercuei, you're prefer OD logo instead? :P
[16:35] <pcercuei> I don't really care TBH
[16:35] <zear> ah, of course :)
[16:38] --> sact joined #scummvm.
[16:39] <-- t0by left irc: Ping timeout: 250 seconds
[16:40] <_sev> well, I see that current port is a bit problematic. But that could be fixed a bit later
[16:41] <_sev> it crashes on aspect ratio change, curses leaves trails when it should not, palette rotation doesn't work...
[16:44] <zear> you mean cursor?
[16:44] <zear> or is there ncurses involved?
[16:44] <-- SimSaw left irc:
[16:44] <zear> also, does it use SDL 1.2?
[16:45] --> GitHub120 joined #scummvm.
[16:45] <GitHub120> [scummvm] digitall closed pull request #524: WII: missing ASCII codes for some keys (master...WII-1.7.0-Beta-2) http://git.io/n6s_Lg
[16:45] GitHub120 (GitHub120@192.30.252.34) left #scummvm.
[16:45] --> GitHub95 joined #scummvm.
[16:45] <GitHub95> [scummvm] digitall pushed 2 new commits to master: http://git.io/IzF0DA
[16:45] <GitHub95> scummvm/master e738de7 Alexander Reim: WII: missing ASCII codes for some keys
[16:45] <GitHub95> scummvm/master d5d2b39 David Turner: Merge pull request #524 from AReim1982/WII-1.7.0-Beta-2...
[16:45] GitHub95 (GitHub95@192.30.252.44) left #scummvm.
[16:49] --> ny00123 joined #scummvm.
[16:50] --> GitHub173 joined #scummvm.
[16:50] <GitHub173> [scummvm] digitall closed pull request #507: Install the theme-able SVG icon (master...patch-1) http://git.io/kgkCPA
[16:50] GitHub173 (GitHub173@192.30.252.42) left #scummvm.
[16:50] --> GitHub94 joined #scummvm.
[16:50] <GitHub94> [scummvm] digitall pushed 2 new commits to master: http://git.io/r1mOlA

[16:50] <GitHub94> scummvm/master 2208eda David Turner: Merge pull request #507 from Mailaender/patch-1...
[16:50] GitHub94 (GitHub94@192.30.252.46) left #scummvm.
[16:50] <_sev> zear: it uses SDL 1.2, and I mean cursor.
[16:50] <clone2727> digitall: Since I know you're reading this, at least fix the damned commit message before merging
[16:50] <zear> ok, then crashing on aspect ratio change shouldn't be SDL backend related
[16:51] <zear> since GCW Zero should handle aspects different than 4:3, by adding black bars
[16:51] <zear> what is the resolution these games request though?
[16:51] <pcercuei> by "change of aspect ratio" you mean power+A?
[17:01] <_sev> interesting
[17:01] --> RUBICN64 joined #scummvm.
[17:01] <_sev> it dies in drawEllipse()
[17:03] <zear> pcercuei, based on the code where drawEllipse() is - it's not IPUs aspect ratio
[17:04] <pcercuei> yes, scummVM doesn't change its resolution
[17:05] --> t0by joined #scummvm.
[17:05] #scummvm: mode change '+v t0by' by ChanServ!ChanServ@services.
[17:09] <-- lukaslw left irc: Quit: Page closed
[17:17] --> digitall joined #scummvm.
[17:17] #scummvm: mode change '+o digitall' by ChanServ!ChanServ@services.
[17:17] --> GitHub175 joined #scummvm.
[17:17] <GitHub175> [scummvm] digitall force-pushed master from 2208eda to 5b5a2bc: http://git.io/xtS8kg

[17:17] GitHub175 (GitHub175@192.30.252.44) left #scummvm.
[17:17] <digitall> Fixed dud commit message in the PR merge by force push...
[17:18] <digitall> Anyone who has pulled in the last 10 minutes may need to do some fixup... Hopefully that should not affect anyone.
[17:19] <clone2727> It's not worth force pushing to fix a commit message....
[17:19] <digitall> Sigh.
[17:19] <digitall> Damned if I do, damned if I don't
[17:20] <digitall> Right... dinner.
[17:20] digitall (digitall@cpc3-hitc7-2-0-cust674.9-2.cable.virginm.net) left #scummvm ("Food").
[17:20] <-- sirlemonhead left irc: Ping timeout: 244 seconds
[17:24] <wjp> oh
[17:24] <wjp> :-(
[17:25] <clone2727> We should really document on the commit guidelines "don't ever ever ever force push"
[17:26] <wjp> yes
[17:26] <fuzzie> but also "don't complain at poor digitall", who takes on a disproportionate amount of the janitorial stuff which we don't
[17:28] <clone2727> fuzzie: Someone doing something wrong while attempting to do something right doesn't make them above complaints
[17:29] <fuzzie> my point is that I think a force-push there in response was pretty predictable
[17:29] <clone2727> wjp: Actually, it is there: http://wiki.scummvm.org/index.php/Git_tips#Don.27ts
[17:30] <fuzzie> and complaining at anyone from both directions is not really a positive thing, so I want to jump up and down and object to the negativity
[17:31] <clone2727> fuzzie: So, saying nothing and opening the potential for it happening again is the right thing to do?
[17:31] <fuzzie> we all want technical perfection
[17:32] <fuzzie> clone2727: not sure there is a good answer
[17:33] <fuzzie> always something which someone could do beter :-/
[17:33] <wjp> sorry, was definitely not intending that to be the second comment to that merge
[17:35] <wjp> anyway, speaking of janitorial work, let's have a look at that CONTRIBUTING.md
[17:36] <DrMcCoy> Dunno how applicable it is to ScummVM, but this is what I put into xoreos' repository: https://github.com/xoreos/xoreos/blob/master/CONTRIBUTING.md
[17:36] <DrMcCoy> Basically linking to the xoreos wiki and saying "Yes, I do complain a lot about style"
[17:37] <clone2727> DrMcCoy: Yeah, that's more of what I was hoping that our CONTRIBUTING.md would be like
[17:38] <DrMcCoy> Sure, the actual meat is a click away, but I don't think duplicating information is a good idea either
[17:38] <wjp> hrm, pull requests are kind of annoying for collaboratively writing a piece of text
[17:38] <t0by> <wjp> anyway, speaking of janitorial work, let's have a look at that CONTRIBUTING.md <==== Oh, btw, I was wondering: at which point one does get credited in the actual credits? I look forward to the day I will be able to brag with my friends who run ScummVM on their PC or smartphone. :P
[17:39] <wjp> what does that have to with CONTRIBUTING.md? :-)
[17:39] <wjp> but you're not in there yet?
[17:39] <fuzzie> easy to fix
[17:39] <clone2727> t0by: Just add yourself. Make sure you edit credits.pl and update everything properly
[17:39] <t0by> <wjp> what does that have to with CONTRIBUTING.md? :-) ===> they're two pieces of ASCII text :P
[17:40] <wjp> the credits are actually a perl script
[17:40] <wjp> but, yes, feel free to add yourself
[17:40] <wjp> or we can do it
[17:40] <clone2727> wjp: Yeah, that's why I explicitly stated credits.pl ;)
[17:41] <t0by> <clone2727> t0by: Just add yourself. Make sure you edit credits.pl and update everything properly ===> Really? I will proceed to acquire bragging rights when I've finished fighting with game descriptors. To motivate me.
[17:41] <DrMcCoy> t0by: Keep in mind that you have to run that perl script after editing and then check in the changed files. Also into the web repository
[17:41] <DrMcCoy> t0by: I think I sent a mail to -devel about that a few months ago how to do that
[17:42] <wjp> so, what are the points we want to make in CONTRIBUTING.md?
[17:42] <fuzzie> commit messages plus link to wiki pages?
[17:42] <t0by> Oh, yes, in repsonse to Fedor. Sorry for not paying attention.
[17:43] <wjp> It looks like it can be pretty short
[17:44] <clone2727> wjp: I'm really thinking something along the lines of what DrMcCoy has
[17:44] [md5] --> (~md5@unaffiliated/md5/x-729473) joined #scummvm.
[17:44] #scummvm: mode change '+o [md5]' by ChanServ!ChanServ@services.
[17:44] <[md5]> good evening
[17:49] <wjp> How about something like http://www.usecode.org/scummvm/c.txt ?
[17:50] <DrMcCoy> I like it
[17:51] <clone2727> Agreed
[17:52] <clone2727> wjp: Only nitpick would be to use the serial comma before "procedure" (though, not strictly necessary)
[17:54] <wjp> serial comma?
[17:54] <-- Schnabeltierchen left irc: Ping timeout: 255 seconds
[17:54] <clone2727> having the comma before "and" or "or" in a list
[17:55] <wjp> oh, you mean after procedure
[17:55] <wjp> sure
[17:55] <clone2727> Ah, yes, after :)
[17:56] <DrMcCoy> We are decent people! In our house, we do not use the Oxford comma!
[17:56] <DrMcCoy> :P
[17:56] <wjp> ok, you two fight it out and let me know the result afterwards :-)
[17:57] Action: wjp will try to find out how markdown works in the meantime
[17:57] <DrMcCoy> http://daringfireball.net/projects/markdown/syntax
[17:57] --> Schnabeltierchen joined #scummvm.
[17:57] <DrMcCoy> Or https://help.github.com/articles/github-flavored-markdown/ for GitHub specifics, but we don't need those here
[18:00] <wjp> thanks
[18:01] <wjp> so something like this?
[18:01] <-- Sylvain left irc: Ping timeout: 250 seconds
[18:02] <DrMcCoy> Looks good to me
[18:02] <DrMcCoy> Except, of course, that Oxford comma ;)
[18:02] Action: clone2727 slaps DrMcCoy
[18:02] <DrMcCoy> :(
[18:04] <eriktorbjorn> DrMcCoy: Commas are fun: http://en.wikipedia.org/wiki/Serial_comma#Ambiguity
[18:05] --> SylvainTV joined #scummvm.
[18:05] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[18:05] <clone2727> eriktorbjorn: The "unresolved" section is best
[18:05] <DrMcCoy> Yes, there's arguments for and against that thing
[18:05] <wjp> I kind of dislike putting github-specific files in the repo
[18:06] <wjp> but I suppose there's no other way of getting this?
[18:06] <clone2727> wjp: I agree with that as well
[18:06] <DrMcCoy> wjp: You can of course still do plaintext CONTRIBUTING, without the md extension. But no clickable links, then
[18:06] <-- RUBICN64 left irc: Quit: AndroidIrc Disconnecting
[18:07] <DrMcCoy> wjp: Or do you mean the concept of a CONTRIBUTING file in general?
[18:09] Action: wjp nods
[18:11] <wjp> but it's fine
[18:11] <DrMcCoy> Well, the idea is that it's not solely a GitHub-specific file
[18:12] <DrMcCoy> It can still be read and understood by people extracting the usual tarball
[18:13] <clone2727> DrMcCoy: Not if it's markdown
[18:14] <DrMcCoy> Markdown is very readable in raw. Which was one of its initial design decisions
[18:14] <DrMcCoy> That the reason for markdown existing in the first place
[18:15] <DrMcCoy> Can be read and written with a normal text editor, but also rendered to something nice/useful in HTML
[18:16] <clone2727> DrMcCoy: But at that point I'd rather have actual plaintext if I'm going to read it by hand
[18:16] <clone2727> (just IMHO)
[18:18] <DrMcCoy> You can also use "Please see [our wiki][0]" and then at the bottom "[0]: https://wiki.scummvm.org/blahblah"
[18:18] <DrMcCoy> Isn
[18:18] <DrMcCoy> 't that "actual plaintext"?
[18:21] <DrMcCoy> I for one at least really like Markdown
[18:21] <clone2727> For [0], it's fine, but the [our wiki] wouldn't. An unfamiliar mind wouldn't know what the [] is.
[18:22] <DrMcCoy> You mean a mind completely devoid of any transfer of similar concepts
[18:22] <clone2727> This isn't a criticism of markdown, just that I'd rather not read markdown text as if it weren't
[18:22] <DrMcCoy> That hypothetical mind would have problems reading in the first place
[18:23] <clone2727> DrMcCoy: No, not that they wouldn't be able to read it. That they'd be unable to figure out what the [] is for
[18:23] <DrMcCoy> That's what I'm saying. A mind that would be unable to figure that out would be unable to figure out reading
[18:23] <clone2727> ...
[18:24] <DrMcCoy> Because that mind would be of a 2 years old
[18:25] --> [md5]_ joined #scummvm.
[18:25] #scummvm: mode change '+o [md5]_' by ChanServ!ChanServ@services.
[18:26] [md5] <-- (~md5@unaffiliated/md5/x-729473) left irc: Ping timeout: 245 seconds
[18:33] --> GitHub58 joined #scummvm.
[18:33] <GitHub58> [scummvm] sev- pushed 4 new commits to master: http://git.io/FmJKqg
[18:33] <GitHub58> scummvm/master f5e273d Eugene Sandulenko: GCW0: Create log file at game launch
[18:33] <GitHub58> scummvm/master 293a0d0 Eugene Sandulenko: GCW0: Create target with unstripped binaries
[18:33] <GitHub58> scummvm/master 6046863 Eugene Sandulenko: BASE: Display virtual keyboard and keymapper in game options
[18:33] GitHub58 (GitHub58@192.30.252.40) left #scummvm.
[18:42] <pcercuei> _sev: what's with the log feature?
[18:42] <_sev> what do you mean?
[18:42] <pcercuei> gmenu2x has a built-in log feature, if you didn't notice it
[18:42] <pcercuei> "Output logs" in the menu
[18:45] <_sev> where is it?
[18:45] <zear> yep, plus if the file is being written to during game run-time you'll generate a lot of disk writes and: 1) waste CPU cycles, 2) wear out the SD card
[18:46] <zear> _sev, while in gmenu2x, press START
[18:46] <zear> it will bring you to gmenu2x options
[18:46] <_sev> so where the logs get written then?
[18:47] <zear> After program runtime, you can find logs in settings->Log Viewer
[18:47] <zear> (assuming the program outputted to stdout/stderr and not a file ;D)
[18:48] <_sev> let me see
[18:48] <zear> keep in mind this also logs to SD card, so CPU usage during game run-time is increased, especially if the game outputs a lot of logs
[18:49] <zear> so keep gmenu2x logs ON only for debugging sessions and not your regular gaming
[18:50] <zear> also, don't do this: https://github.com/scummvm/scummvm/blob/f5e273ded942d2de09be4673cc6b6f136d6ed1e3/dists/gcw0/scummvm.sh#L6
[18:50] <pcercuei> _sev: it's written in /var/log/gmenu2x.log
[18:50] <zear> this is assuming too much
[18:50] <zear> you assume that the mount location of the opk is in /mnt/ScummVM/
[18:50] <pcercuei> and zear, for the Nth time - logs are not saved on the SD card
[18:51] <zear> pcercuei, at least at some point - they were
[18:51] <zear> since I remember debugging and seeing a lot of disk io
[18:51] <pcercuei> at the beginning of *2013*, yes
[18:51] <zear> pcercuei, ok, so that's fixed now
[18:52] <zear> _sev, is scummvmrc something ScummVM is able to generate?
[18:52] <zear> or do you need to provide one with the OPK?
[18:53] --> sirlemonhead joined #scummvm.
[18:53] <zear> also, please don't output directly to $HOME, make your program create a subdirectory in $HOME, like $HOME/.scummvm/ and output there
[18:53] <zear> I understand the port is still in progress, I am just listing things I noticed :)
[18:56] <_sev> ok, thanks, guys.
[18:57] <_sev> I am removing the redirect
[18:57] <zear> so, if ScummVM is able to generate a default scummvmrc, I'd generate it directly in $HOME/.scummvm/ (at the same time creating that dir if doesn't exist)
[18:57] <_sev> zear: how to get the mount point then? Why it could change?
[18:57] <zear> instead of using shell scripts
[18:58] <_sev> you mean generate it in C code?
[18:58] <zear> yes
[18:58] <_sev> ugly
[18:58] --> dreammaster joined #scummvm.
[18:58] <zear> no, this is ugly :D
[18:58] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[18:58] <zear> but I believe the location of the mount point is $PWD
[18:58] <_sev> hardcoding config values is a no-no
[18:59] <zear> but you are technically hardcoding them inside a read-only OPK
[18:59] <_sev> I could use sed(1) after all
[18:59] <zear> so yea, I believe ./scummvmrc $HOME/.scummvmrc
[18:59] <zear> should work
[18:59] <_sev> to substitute it with `pwd`
[19:00] <_sev> then thing is
[19:00] <_sev> that I have those paths is scummvmrc as well
[19:00] <_sev> which I need to substitute then
[19:00] <zear> then they can't be there
[19:00] <zear> it is absolutely not guaranteed that the mount point won't change
[19:00] <zear> and you shouldn't depend on that
[19:00] <pcercuei> _sev: yes, using the mount point is a problem. It will probably change soon
[19:00] <_sev> well, how can I refer to files from .opk?
[19:01] <zear> during the program runtime - with $PWD
[19:01] <_sev> that is hardcoding once again
[19:01] <pcercuei> you can always assume that the current directory when the OPK is executed is /mnt/scummVM
[19:01] <_sev> e.g. currently they're free form, you set them up in your computer
[19:01] <zear> pcercuei, he shouldn't assume /mnt/ is a mountpoint of OPKs and that /mnt/scummVM/ is the mountpoint of that particular OPK
[19:02] <_sev> there is no such concept in the world, you know. What if I run it with absolute path from different directory
[19:02] <pcercuei> zear: that's what I said
[19:02] <zear> pcercuei, you wrote:
[19:02] <zear> <pcercuei> you can always assume that the current directory when the OPK is executed is /mnt/scummVM
[19:02] <zear> so you wrote he can assume /mnt/scummVM/ will always be valid
[19:02] <zear> which isn't true
[19:03] <pcercuei> I mean that the current directory will always be the mountpoint when executing the OPK
[19:03] <zear> ah, that's different then
[19:03] <zear> I agree with that
[19:03] <_sev> I would rather leave it hardcoded right now, and when/if it will change in the future, I'd adapt
[19:03] <pcercuei> _sev: you can hardcode for now, but it will change
[19:03] <pcercuei> yes
[19:03] <zear> _sev, users won't adapt
[19:03] <pcercuei> I plan to switch the mountpoint to /mnt/opk so that hardcoding becomes possible
[19:04] <zear> by hardcoding it, you are forcing the users to update ScummVM OPK to a new version when they upgrade the firmware and mountpoint path changes
[19:04] <zear> and you should never depend on the users
[19:04] <zear> since it's almost guaranteed they won't do that
[19:04] <zear> and instead will come here to this channel, complaining that ScummVM isn't working
[19:04] <_sev> look
[19:04] <pcercuei> don't listen to zear :)
[19:04] <_sev> when they upgrade their firmware, .scummvmrc will get obsolete instantly
[19:05] <zear> pcercuei, you mean /mnt/opk/[app_name]/ or just /mnt/opk?
[19:05] <_sev> as it will contain absolute path this way or another
[19:05] <pcercuei> I mean just /mnt/opk
[19:05] <zear> pcercuei, then what if I mount two OPKs at once?
[19:05] <pcercuei> you can't
[19:05] <zear> _sev, which I argue is a bad habit and shouldn't be present
[19:05] <pcercuei> unless you SSH/telnet into the device
[19:05] <_sev> zear: where?
[19:06] <zear> _sev, hardcoding paths
[19:06] <_sev> zear: where it should not be present?
[19:06] <zear> so paths should never be absolute, unless you can retreive the current path and amend
[19:06] <pcercuei> hardcoding is fine, as long as the mountpoint is fixed, and that's one thing I have to take care of
[19:07] --> Surkow joined #scummvm.
[19:07] <_sev> zear: ScummVM runs on over 30 platforms. You're doing GCW firmware wrong
[19:07] <pcercuei> actually, I can just symlink /mnt/opk to /mnt/$MOUNTPOINT
[19:07] <pcercuei> then it will still work
[19:07] <-- Raziel^ left irc: Ping timeout: 265 seconds
[19:08] <zear> _sev, what sort of stuff is kept in scummvmrc?
[19:08] <_sev> paths to games, to themes, savageness etc
[19:09] <_sev> to plugins
[19:09] <zear> by default, so that you copy an original scummvmrc from OPK to $HOME
[19:09] <pcercuei> _sev: the problem is that most Unix apps require their data to be at specific places (/usr/lib etc), but that collides with the OPK format, which has been designed to contain everything, binary+data
[19:09] <zear> because paths to games, etc. can and should indeed be absolute
[19:09] <pcercuei> it's not really that the firmware has been done wrong
[19:09] <zear> I am just wondering what stuff you're keeping there by default that you need to copy a template scummvmrc
[19:09] <_sev> why it should change in the future?
[19:09] <zear> instead of just creating one on the first run
[19:10] <_sev> zear: take a look?
[19:10] <pcercuei> the only thing I don't want you to do, is hardcode /usr/local/home instead of using $HOME
[19:10] <_sev> zear: https://github.com/scummvm/scummvm/blob/master/dists/gcw0/scummvmrc
[19:10] <zear> ok, so all these things could be set by default in the C code
[19:10] <pcercuei> but scummVM doesn't so it's fine with me
[19:10] <zear> oh, and joystick=0 is bad :D
[19:11] <pcercuei> huh?
[19:11] <zear> _sev, I see it sets the theme, engine-data and plugins to $PWD
[19:11] <zear> so why not make it default in C code?
[19:12] <zear> "no path given, so it defaults to ./"
[19:12] --> LordHoto joined #scummvm.
[19:12] #scummvm: mode change '+o LordHoto' by ChanServ!ChanServ@services.
[19:12] <pcercuei> or better, if the path is foo=path/to/data without leading slash, consider it's relative to the current directory
[19:13] Action: zear nods to that
[19:13] <_sev> zear: and how can you guarantee that you run it from the very directory you're at?
[19:13] <pcercuei> then that would be a valid improvement for all platforms
[19:13] <zear> _sev, because OPK format guarantees that
[19:13] <_sev> you're telling me that mount point of OPK may change in the future, and now you're saying that I should rely on another behaviour o_O
[19:13] <zear> _sev, precisely in this line: https://github.com/scummvm/scummvm/blob/f5e273ded942d2de09be4673cc6b6f136d6ed1e3/dists/gcw0/default.gcw0.desktop#L10
[19:14] <zear> you are launching "scummvm.sh", or in other words: $PWD/scummvm.sh aka ./scummvm.sh
[19:14] <_sev> ok, let me see
[19:14] <zear> _sev, absolute mount point will change
[19:14] <zear> because you're hardcoding pathes to /mnt/scummVM/
[19:14] <zear> that *will* change
[19:15] <zear> but the fact that OPK changes the local directory to whatever the mountpoint is - won't change
[19:15] <zear> so even if the mountpoint changes to /foo/bar/svmmVM_blah/
[19:15] <zear> you can still expect the OPK to cd to that directory
[19:15] <zear> and the relative directories in ./ to still be there
[19:18] <_sev> nice, it works
[19:18] <zear> sweet
[19:18] <_sev> though now joystick stopped to work for me...
[19:18] <zear> isn't that because it's set to =0 in scummvmrc ?
[19:18] <pcercuei> _sev, I just tested - relative paths (from the top of the OPK) work just fine
[19:19] <_sev> zear: no, that is joystick #0, exactly what I want
[19:19] <zear> ah! :)
[19:19] <_sev> maybe it is because of key mapper
[19:19] <zear> pcercuei, do we guarantee that internal analog will always be #0?
[19:20] <pcercuei> I think, yes
[19:21] <_sev> question
[19:21] <_sev> where to put saves?
[19:22] <pcercuei> it saves to the right path by default
[19:22] <pcercuei> ~/.scummvm/
[19:22] <Surkow> most linux programs would throw them in $HOME/.programname
[19:23] <pcercuei> one thing you will want to enable, is SDL_HWSURFACE | SDL_DOUBLEBUF
[19:23] <pcercuei> on my old build it was using SDL_SWSURFACE
[19:25] <Surkow> although assets should be stored in $HOME/.local/share/programname as seen in http://standards.freedesktop.org/basedir-spec/basedir-spec-latest.html
[19:25] <Surkow> for example, emulators like mupen64plus on my pc stores savegames there
[19:25] <pcercuei> you can't change it now
[19:25] <LordHoto> I think we're still too old fashioned to switch to that new stuff ;-)
[19:26] <Surkow> oh well
[19:26] <Surkow> I suppose $HOME/.config/programname for config files is too modern as well :P
[19:26] <LordHoto> But, speaking about SDL_DOUBLEBUF, I am pretty sure we don't want that, since we're pretty much having a rendering pipeline based on partial updates without any double buffering in mind
[19:26] <LordHoto> Surkow: what I mean is: we didn't switch yet
[19:26] <pcercuei> ouch x.x
[19:27] <pcercuei> ok, that's too bad. The UI flickers a little bit because of SDL_SWSURFACE
[19:27] <LordHoto> Surkow: We have some FR about using XDG_foo stuff
[19:27] <pcercuei> but it's not very disturbing
[19:27] <Surkow> so no triple buffering?
[19:27] <Surkow> :(
[19:27] <pcercuei> no point, when all games run fullspeed with 4% CPU ...
[19:28] <LordHoto> If you use some GLES based output, it will probably flicker lss.
[19:28] <LordHoto> *less
[19:28] <LordHoto> which reminds me, I need to push my GLESv2 support at some point
[19:28] <pcercuei> GLES for composition of the image, yes, maybe
[19:28] <clone2727> LordHoto: scalers.
[19:28] <zear> Surkow, I dislike $HOME/.local/ stuff
[19:28] <zear> because it confuses GCW Zero users
[19:29] <Surkow> it's a nice separation between assets and configuration files
[19:29] <zear> they already learned that config can be found in /media/data/home/
[19:29] <LordHoto> I'll let you guys argue that out on your own ;-)
[19:29] <zear> and suddenly there's a new location
[19:29] <zear> LordHoto, ;D
[19:29] <Surkow> just backup your whole home directory
[19:29] <Surkow> problem solved
[19:29] <zear> SWSURFACE is a *bad* idea
[19:29] <zear> because it doesn't provide vsync
[19:29] <zear> and you'll have screen tearing
[19:30] <zear> now, HWSURFACE with DOUBLEBUF eliminate that and provide crystal clear visuals
[19:30] <zear> with an even faster buffer because it's hardware
[19:30] <LordHoto> won't HWSURFACE fallback to SWSURFACE when no native output mode is supported?
[19:30] <pcercuei> as LordHoto said, the rendering pipeline is based on partial updates
[19:30] <zear> correct
[19:30] <zear> LordHoto, so it will fallback to SWSURFACE on X11 and Win32
[19:30] <pcercuei> so double buffering is not possible
[19:30] <LordHoto> so, it will be SWSURFACE anyway?
[19:30] <zear> yes, except for platforms providing hardware buffers
[19:31] <zear> like GCW Zero
[19:31] <LordHoto> you have 16bit graphics hardware buffers?
[19:31] <zear> 16 and 32
[19:31] <pcercuei> nah
[19:31] <zear> but I'll let pcercuei explain it in more detail
[19:31] <zear> yea, since I don't really understand this stuff :)
[19:31] <pcercuei> basically, SDL_HWSURFACE makes the main SDL surface correspond to the framebuffer. So you can write to the framebuffer directly
[19:32] <pcercuei> SDL_SWSURFACE will make the main SDL surface backed by a separated buffer, whose content will be copied to the framebuffer when SDL_Flip is called
[19:32] <LordHoto> please, forget about SDL_Flip, no double buffering for us :-P
[19:33] <pcercuei> you still call it
[19:33] <LordHoto> I'm sure we don't
[19:33] <LordHoto> at least not directly
[19:33] <pcercuei> otherwise nothing is copied to the framebuffer and you see no image
[19:33] <wjp> SDL_UpdateRect does that
[19:33] <zear> yes, but it sucks :)
[19:33] <wjp> SDL_Flip is equivalent to SDL_UpdateRect on the full screen for SWSURFACE
[19:33] <pcercuei> yes, ok
[19:33] <wjp> "sucks"? Please be a bit more constructive
[19:34] <zear> wjp, doesn't provide double buffering :)
[19:34] <LordHoto> as I said, no double buffering for us
[19:34] <LordHoto> :-P
[19:34] <zear> but that makes the GCW0 port pointless
[19:34] <pcercuei> no double buffering is fine
[19:34] <zear> since you'll have tearing
[19:34] <pcercuei> but do you have VSYNC?
[19:35] <fuzzie> I'm going to regret asking, but doesn't it do OpenGL ES?
[19:35] <pcercuei> yes it does
[19:35] <LordHoto> so, just use that
[19:35] <zear> that's overengineering
[19:35] <LordHoto> you guys only have a tiny resolution of 320x240 anyway
[19:35] <LordHoto> so all the SDL pipeline scalers etc. are useless
[19:36] <LordHoto> so, just use our GLES output
[19:36] <pcercuei> we don't use scalers
[19:36] <LordHoto> what would be overenginerring is to change our whole SDL based output for GCW0, when there's already something in our codebase which does VSYNC etc. stuff
[19:36] <pcercuei> and GLES output is much slower
[19:36] <fuzzie> hm
[19:36] <fuzzie> you have no hw accel API which doesn't suck?
[19:37] <LordHoto> I am pretty positive, that for 320x240 games GLES should be just fine for you guys.
[19:38] <pcercuei> there is nothing to be changed
[19:38] <zear> you just need to flip twice
[19:38] <pcercuei> ScummVM already works with the SDL 1.2 backend
[19:38] <zear> that's it if i'm not mistaken
[19:38] <zear> update both buffers in one iteration of the rendering loop, problem solved
[19:38] <pcercuei> so there's no point to use GLES, especially since it would be slower
[19:39] <LordHoto> but you want double buffering?
[19:39] <zear> it is necessary for v-sync
[19:39] <pcercuei> double buffering is always better, but it's not possible for scummVM
[19:40] <pcercuei> so there's no point in even discussing it
[19:40] <LordHoto> well, you guys are discussing it?
[19:40] --> mramberg joined #scummvm.
[19:40] <LordHoto> but maybe now I'm just completely confused what this discussion is about
[19:40] <pcercuei> I mentionned it, before you told me about partial updates
[19:41] <pcercuei> partial updates requires the rendering to happen in one single buffer
[19:41] <LordHoto> anyway, as I said: if you want vsync just output via GLES
[19:41] <LordHoto> +,
[19:41] <pcercuei> yes
[19:41] <LordHoto> there's other low end devices using that
[19:41] <fuzzie> it's just strange if directfb is faster than opengl
[19:42] <LordHoto> so, shouldn't be too slow for you
[19:42] <pcercuei> for ScummVM the performance penalty probably doesn't matter much, ScummVM is very light
[19:42] <fuzzie> but you know best :-p
[19:42] <pcercuei> fuzzie: yes, it is strange
[19:42] <LordHoto> yes, we don't do fancy 3d rendering :-P
[19:42] <fuzzie> obviously the driver situation on this kind of hw is difficult
[19:43] <pcercuei> fuzzie: 320x240 pixels aren't that much, it's faster to render in software than going through the whole GLES stack
[19:43] <LordHoto> anyway, GLES output also features true 32bit colorspace
[19:44] <LordHoto> probably hard to notice on those tiny screens, but I'm just trying to sell it to you anyway
[19:44] <LordHoto> :-P
[19:44] <Surkow> what is the native scummvm resolution?
[19:44] <pcercuei> I will let _sev try it :)
[19:44] <Surkow> it varies between games?
[19:44] <fuzzie> Surkow: yes
[19:44] <LordHoto> everything old is 320x240 anyway
[19:44] <fuzzie> 320x200 up to and including 1080p, at least
[19:44] <Surkow> 320x240 is only the restriction of the lcd
[19:44] <Surkow> there is hdmi/tv-out
[19:45] <LordHoto> well, for 1080p GLES seems even more attractive than SDL surface, right? :-P
[19:45] <Surkow> not really
[19:45] <LordHoto> oh, it is
[19:45] <pcercuei> if you render natively at 1080p, of course
[19:45] <Surkow> if you scale up to 1080p you need to use the IPU of the device
[19:45] <pcercuei> if you render at 320x240 and use the GPU to upscale, then SDL is faster :)
[19:46] <pcercuei> because we have hardware acceleration there
[19:46] <LordHoto> so, upscaling by GLES is slower than software rendering+GPU?
[19:46] <_sev> for some stupid reason joystick stopped working for me.
[19:46] <pcercuei> LordHoto: yes
[19:46] <_sev> even if I rollback to previous version of code
[19:46] <LordHoto> pcercuei: what?
[19:46] <pcercuei> <@LordHoto> so, upscaling by GLES is slower than software rendering+GPU?
[19:46] <pcercuei> ^ yes
[19:46] <LordHoto> pcercuei: I know
[19:47] <LordHoto> pcercuei: but that doesn't make any sense
[19:47] <LordHoto> pcercuei: so, rendering a 320x240 texture is slower than software rendering+upscaling?
[19:47] <pcercuei> yes
[19:47] <LordHoto> pcercuei: why do you even use GLES support as a feature? :-P
[19:47] <pcercuei> we have a sucky GPU :(
[19:48] <LordHoto> appareantly it's really good at upscaling though
[19:48] <LordHoto> *apparently
[19:48] <pcercuei> it's a separate, dedicated chip that handles upscaling
[19:48] <LordHoto> I see
[19:48] <LordHoto> oh well
[19:49] <LordHoto> let's say: I'm 100% sure you don't want to use SDL output from us for all the really highres games
[19:49] <LordHoto> because they use 32bpp anyway, and we convert it to 16bpp and push it to the output and then it might convert it back
[19:49] <LordHoto> but just my two cents on that matter
[19:49] <pcercuei> but there ScummVM composes the image in GLES? Or renders in software then use GLES to upload to the screen?
[19:49] <LordHoto> basically, all the 'rendering' is done in software anyway
[19:50] <t0by> somaen, I am starting to fear that one does not simply /extend/ the C-style struct (not class) ADGameDescription, especially if one still has plans to have a static const list-initialized array thereof somewhere.
[19:50] <pcercuei> ok, then using SDL output will always be faster
[19:50] <Surkow> pcercuei, the gpu is only restricted by the limited bandwidth, no?
[19:50] <pcercuei> see it that way: it's faster to render to a 320x240 buffer, than render to a 320x240 buffer and upload it to the GPU
[19:50] <LordHoto> SDL has an additional internal scaling step (even for 1x), so it's doing it twice
[19:50] <Surkow> Exophase said it was faster than the one from the pandora
[19:51] <fuzzie> pcercuei: for partial updates we usually only change small bits
[19:51] <pcercuei> no, SDL doesn't scale anything
[19:51] <t0by> The least that has been happening to me is a number of warnings re:not being very ISO C++
[19:51] <LordHoto> pcercuei: we do internally
[19:51] <pcercuei> it doesn't even have scalers
[19:51] <_sev> pcercuei: any idea why joystck could stop working?
[19:51] <pcercuei> ah
[19:51] <LordHoto> pcercuei: it's all our code doing that
[19:51] <LordHoto> pcercuei: trust me :-P
[19:51] <t0by> Would you or anybody complain much if I added a separate struct for WME-specific details?
[19:51] <zear> _sev, which firmware version do you have installed?
[19:51] <t0by> (wjp, maybe you care as well)
[19:51] <pcercuei> LordHoto: ok, but is it done unconditionally, or when the required resolution is not accepted by SDL?
[19:51] <_sev> zear: 8.20
[19:52] <LordHoto> pcercuei: unconditionally
[19:52] <zear> _sev, 2014?
[19:52] <_sev> zear: it was working couple hours ago
[19:52] <pcercuei> that sucks :p
[19:52] <_sev> zear: yes
[19:52] <pcercuei> _sev: rebooting doesn't fix it?
[19:52] <zear> _sev, does it work in the input testing app?
[19:52] <_sev> pcercuei: nope
[19:52] <pcercuei> try with a different app
[19:52] <_sev> it works in the launcher, so hardware works
[19:52] <LordHoto> pcercuei: basically our rendering piple for SDL is: game updates screen into temp buffer -> scale with scaler onto SDL output surface
[19:53] <LordHoto> pcercuei: even for normal1x last time I checked
[19:53] <zear> _sev, something else is joystick #0?
[19:53] <_sev> works in hwtest too
[19:53] <zear> _sev, cat /dev/input/js0
[19:54] <LordHoto> pcercuei: it's really just uniform handling :-P
[19:54] <zear> if that works when you move around the analog, then it is joystick #0 and all should be fine
[19:54] <_sev> zear: catting works
[19:54] <pcercuei> _sev: do you have joystick_num=0 in the scummvmrc?
[19:54] <LordHoto> pcercuei: GLES is actually also keeping it in a buffer around (don't ask) and pushing it through to GLES
[19:55] <pcercuei> if ScummVM converts and scales in software, then GLES might be faster
[19:55] <fuzzie> isn't normal1x no-op?
[19:55] <fuzzie> or memcpy?
[19:56] <_sev> pcercuei: heh, you're right
[19:56] <LordHoto> fuzzie: memcpy
[19:57] <LordHoto> pcercuei: yeah... for 32bpp games it probably is
[19:57] <LordHoto> pcercuei: even on desktops... GL output is faster there
[19:58] --> GitHub12 joined #scummvm.
[19:58] <GitHub12> [scummvm] sev- pushed 2 new commits to master: http://git.io/HpVPwA
[19:58] <GitHub12> scummvm/master 732f958 Eugene Sandulenko: GCW0: Remove absolute paths
[19:58] <GitHub12> scummvm/master f160646 Eugene Sandulenko: GCW0: Fix default joystick number
[19:58] GitHub12 (GitHub12@192.30.252.35) left #scummvm.
[19:59] <pcercuei> it's just sub-optimized for us, but a properly optimized implementation would require a lot of changes
[19:59] <pcercuei> anyway, if GLES brings VSYNC, that's already a valid reason to use it
[20:10] <t0by> Please, general question: you know ADGameDescription, right? Suppose you need an extra field of information (in my case, the executable version of the original interpreter) which basically only depends on the files hash. How do you do it sane? Is there anything like that in existing engines?
[20:11] <clone2727> t0by: As long as it's POD, you're fine
[20:12] <t0by> Payable On Death?
[20:12] <clone2727> plain old data
[20:12] <clone2727> t0by: http://en.cppreference.com/w/cpp/concept/PODType
[20:12] <clone2727> t0by: Here's how AGI does it: https://github.com/scummvm/scummvm/blob/master/engines/agi/detection.cpp#L46
[20:13] <t0by> Thanks for the AGI pointer
[20:21] <wjp> there's one small gotcha that's easy to miss: you need to pass the size of the new struct to the AdvancedMetaEngine constructor
[20:21] <wjp> (see line 157 in that AGI file)
[20:26] <pcercuei> _sev: btw, we have Fluidsynth in the firmware
[20:27] <_sev> I know, the problem is soundfont
[20:27] <pcercuei> the GeneralUser one sounds very good, but is fat and eats a lot of resources
[20:27] <pcercuei> I didn't find a better one
[20:27] --> hidnplayr joined #scummvm.
[20:28] <-- dreammaster left irc: Ping timeout: 240 seconds
[20:28] Nick change: hidnplayr -> KolibriOS|hidnpl
[20:28] --> dreammaster joined #scummvm.
[20:28] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[20:29] Nick change: KolibriOS|hidnpl -> KolibriOS|jeff
[20:34] <-- ajax16384 left irc: Read error: Connection reset by peer
[20:39] KolibriOS|jeff (hidnplayr@12.104.145.5) left #scummvm.
[21:06] <-- dreammaster left irc: Ping timeout: 256 seconds
[21:06] --> dreammaster joined #scummvm.
[21:06] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[21:08] <-- ny00123 left irc: Quit: Leaving
[21:10] <-- dreammaster left irc: Client Quit
[21:21] --> GitHub189 joined #scummvm.
[21:21] <GitHub189> [scummvm] sev- pushed 1 new commit to master: http://git.io/j5gR_w
[21:21] <GitHub189> scummvm/master 9b8b737 Eugene Sandulenko: KEYMAPPER: Fix off-by-one error with keymapper action names
[21:21] GitHub189 (GitHub189@192.30.252.43) left #scummvm.
[21:21] --> TMM joined #scummvm.
[21:23] <_sev> heh, it was lurking there for a while
[21:35] <wjp> hm
[21:35] <wjp> strange code
[21:37] <wjp> does the use of ACTION_ID_SIZE in Keymap::saveMappings make sense?
[21:43] <_sev> it parses keymap_<keymapname>_<action> from .scummvmrc
[21:43] <_sev> and it uses this size for strcpy
[21:43] <_sev> if you switch to String, it will be way too slow
[21:44] <clone2727> I find that hard to believe
[21:44] <wjp> yes, but there seems to be some confusion about length
[21:44] <clone2727> Why would it be so slow with Common::String?
[21:44] <wjp> saveMappings seems to assume ID_SIZE is _excluding_ terminating 0
[21:44] <wjp> while the rest _including_
[21:49] <wjp> I'm guessing somebody originally intended this to be non-terminated, but this was lost when moving from strncpy to strlcpy. It's all bit weird.
[21:50] <wjp> right, it used to be integer comparison
[21:51] <wjp> but then it was poorly rewritten into char arrays
[21:52] <_sev> clone2727: err, because it was slow?
[21:52] <_sev> clone2727: probably you forgot, but something like 5 years ago that piece of code was rewritten to char instead of Strings
[21:52] <_sev> because it took like 3 secs to load on SDL backend
[21:54] <-- Harekiet2 left irc: Quit: I brought my pogo stick just to show her a trick
[21:54] --> Harekiet joined #scummvm.
[21:55] <wjp> strange
[21:55] <wjp> you mean at startup?
[21:58] <wjp> This code was from the keymapper gsoc project, right? Was that a rewrite of earlier code using String?
[21:58] <wjp> I'm having trouble tracking it down in history
[22:02] <clone2727> _sev: But there's nothing in there that would really slow it down. Especially not with compiler optimizations
[22:04] <_sev> it generated about 500 key variants
[22:04] <_sev> and thus was slow
[22:04] <clone2727> 500 isn't a lot
[22:04] <_sev> then somebody (eriktorbjorn?) rewrote it
[22:04] <_sev> exactly, but it was slow
[22:04] <_sev> the code written by student was very slow
[22:05] <wjp> the char[ACTION_ID_SIZE] stuff is from the student
[22:06] <wjp> see 0b8203b0e5, d92909203b5
[22:15] --> Jon_God joined #scummvm.
[22:16] --> Ramal joined #scummvm.
[22:51] <-- _marc` left irc: Quit: _marc`
[23:04] --> GitHub19 joined #scummvm.
[23:04] <GitHub19> [scummvm] sev- pushed 1 new commit to master: http://git.io/2GhOcA
[23:04] <GitHub19> scummvm/master 238b3e1 Eugene Sandulenko: GCW0: Hardcoded (for now) GCW0 controls
[23:04] GitHub19 (GitHub19@192.30.252.44) left #scummvm.
[23:05] --> GitHub174 joined #scummvm.
[23:05] <GitHub174> [scummvm] sev- pushed 1 new commit to master: http://git.io/OEjHKw
[23:05] <GitHub174> scummvm/master 8e3b6df Eugene Sandulenko: GCW0: Switch to HW floats and introduce the platform define
[23:05] GitHub174 (GitHub174@192.30.252.45) left #scummvm.
[23:08] <-- waltervn left irc: Quit: Ik ga weg
[23:21] --> sloof joined #scummvm.
[23:25] --> RUBICN64 joined #scummvm.
[23:30] <-- clone2727 left irc: Quit: brb
[23:34] <-- edheldil_ left irc: Ping timeout: 260 seconds
[23:35] --> clone2727 joined #scummvm.
[23:35] #scummvm: mode change '+o clone2727' by ChanServ!ChanServ@services.
[23:38] --> edheldil_ joined #scummvm.
[23:44] <-- pcercuei left irc: Ping timeout: 258 seconds
[23:46] --> GitHub179 joined #scummvm.
[23:46] <GitHub179> [scummvm] bluegr closed pull request #503: GUI: Fix for bug 6719 (last line of help missing) (master...fixforbug6719) http://git.io/Sylu_g
[23:46] GitHub179 (GitHub179@192.30.252.42) left #scummvm.
[23:46] --> GitHub71 joined #scummvm.
[23:46] <GitHub71> [scummvm] bluegr pushed 2 new commits to master: http://git.io/Mg7R_g
[23:47] <GitHub71> scummvm/master dd3e877 Ben Castricum: GUI: Fix for bug 6719 (last line of help missing)...
[23:47] <GitHub71> scummvm/master 17d7f3b Filippos Karapetis: Merge pull request #503 from BenCastricum/fixforbug6719...
[23:47] GitHub71 (GitHub71@192.30.252.33) left #scummvm.
[23:56] <-- t0by left irc: Remote host closed the connection
[00:00] --- Mon Oct 27 2014