Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2008-12-08

00:00:48bluebrotherpixelma: just committed my rename. Hopefully didn't broke anything else ...
00:02:00 Quit robin0800 (Remote closed the connection)
00:06:35pixelmanot very important probably but is a combined log of the 5th and 6th (so the first part is doubled)
00:09:56Zagorhow odd
00:12:39pixelmafixed now?
00:13:28Zagorthe 6th was a tough day for logbot
00:13:37pixelmaI already thought I was imagining it... :)
00:14:09Zagoramazing that it got anything logged at all
00:16:27kugelUnhelpful: you didn't change pictureflow to use your scaling. Is there a reason for that (such as the pluginlib scaler looks better)?
00:16:58Unhelpfulkugel: only that the album art and sliding_puzzle changes were already done
00:17:21kugelah ok :)
00:17:38Unhelpfuli only had to make minor changes to sliding_puzzle vs the previous resize-on-load patch. AA loading needed nothing new done at all, except to handle the rebuffering
00:19:24 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
00:21:01 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]")
00:23:54 Join Lynx_ [0] (
00:24:04 Quit bmbl ("Woah!")
00:24:16 Quit bertrik ("Leaving")
00:24:28 Quit tyfoo ("Carpe diem")
00:26:20 Quit moos ("Rockbox rules the DAP world")
00:28:22 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
00:28:33 Quit Zagor ("Client exiting")
00:28:36 Join OvermindDL1 [0] (
00:29:44OvermindDL1Greetings, does anyone who manages the website here right now?
00:30:25advcomp2019OvermindDL1, there is people.. why do you ask?
00:31:30OvermindDL1Just tried to sign up with the forum, it apparently skipped a few character in my username, so I deleted the account, and I apparently cannot sign-up with the correct username until an admin finishes validating the deletion; it does not let me sign up again due to being from the same IP
00:32:39 Quit RETiRED`ShorTLiF ()
00:33:20OvermindDL1Perhaps I could ask my question here though, I currently have an old Sandisk e250v1, been running rockbox on it since day one, but the thing is getting a bit beaten up, so was thinking about getting a Fuze since they are rather cheap on NewEgg right now, was wondering about the current status of the Fuze port, yes I found that long Forum Topic, but it does not really say how far they have to go for it, I think I gathe
00:34:19advcomp2019i do not know what is going on with that tho.. as far as i know, the fuze still has a few issues
00:35:36 Join unstable [0] (
00:36:29 Quit n1s ()
00:36:45unstableI want to buy a micro sdhc card, for the Sansa e280 .. can someone just verify what I'm buying before I check out...
00:37:08unstableWill that card work? It says microsdhc, but the shape in the picture doesn't seem right to me. Is that the right card for the e280?
00:37:34OvermindDL1I did not know SDHC cards ran on the e2?0's, do they run on the e250 as well? If so that would be quite nice, what is the max size supported?
00:38:30_Auron_I hope I'll be able to rockbox the sansa I ordered from woot
00:38:44woosterunstable that shape looks wrong but they say it sells with an SD I dunno
00:38:51unstableOvermindDL1: I think someone said in theory, it can go up to 32GB
00:39:03unstableOvermindDL1: But the biggest micro sdhc card you can buy is 16GB I think.
00:39:08 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
00:39:14bluebrotherOvermindDL1: SDHC are supported on Rockbox. They are not supported by the OF, so you need a card reader to transfer files
00:39:16OvermindDL1And the e250 will support it with rockbox?
00:39:16unstablewooster: ok, I'm not crazy. That shape does look wrong.
00:39:19_Auron_I thought microSDHC was limited to 16GB and SDHC is limited to 64GB
00:39:31OvermindDL1I have an sdhc card reader, just never tried sticking one in the e250...
00:39:58bluebrotherOvermindDL1: why shouldn't it? There's no difference in regard of memory cards between the e200 models ...
00:40:37bluebrotherthe SDHC spec is currently limited at 32GB, but as far as I understood that's quite arbitrary and can easily be changed
00:40:38OvermindDL1just between v1 and v2 then?
00:41:25bluebrothere200 v1 and v2 are completely different hardware. But in the same line there is no real difference except the size of the internal memory
00:41:34unstableWell 32GB is hard to find, I think there is a prototype card, and it costs a lot of money.
00:41:44unstable16GB is the practical limit at this point in time right now.
00:42:07bluebrothertrue, but that will change
00:42:20bluebrotheranyway, gtg now.
00:42:22 Quit bluebrother ("leaving")
00:44:51OvermindDL1So, should I just go ahead and get the fuze while it is very cheap on newegg and just let it sit until rockbox works, or should it still take a while (by a while I mean at least 6 months) till it gets to a full working status on par with the e2?0?
00:46:19***Saving seen data "./dancer.seen"
00:47:01 Quit ender` (" It's bad luck to be superstititious. -- Law of Superstition")
00:47:16 Quit Schmogel ("Miranda IM! Smaller, Faster, Easier.")
00:47:28OvermindDL1has anyone messed with the Sandisk View?
00:48:15scorcheyou can check on the forum thread for previous info...
00:48:51OvermindDL1not just for rockbox, I just noticed it on newegg and wondering how people like it compres to the e2?0 and fuze?
00:49:10unstableThat card will work right? with rockbox/e280
00:49:14unstableThat form factor, etc?
00:49:18scorcheOvermindDL1: being that this is #rockbox, rockbox discussion is the only thing that is on-topic here
00:49:58scorcheand which username were you having issues with on the forum?
00:50:33OvermindDL1it somehow did not catch the dDL1...
00:50:43alexbobpIs it, or will it ever be, possible to get rockbox on a zune?
00:50:50Nico_PBagder: any idea why the latest commit emails apparently didn't come through?
00:50:57scorchealexbobp: not if no one works on it..
00:51:05OvermindDL1And I did look on the rockbox site, I could not really find anything for the Sandisk View, kept coming up with the 'view' in the twiki links...
00:51:33alexbobpscorche: ah, heh
00:51:42alexbobpI don't own a zune or ever plan to... I was just curious.
00:51:54scorcheOvermindDL1: done
00:53:21unstablescorche: Will that card work on the e280 with rockbox? I'm always a little wearly when I order parts online.
00:53:49scorcheUnhelpful: *micro* SD
00:53:56mc2739OvermindDL1: Snadisk View thread on forum -
00:54:01scorcheunstable: ^^
00:54:18unstablescorche: o I see, that's why it's so inexpensive.
00:54:22unstableGood thing I didn't order it.
00:54:31 Quit herrwaldo ("Konversation terminated!")
00:57:44saratogaOvermindDL1: theres a status link on the front page
00:57:59saratogaas far as I know its up to date on the View port
00:58:36unstableWill that work? heh. What is with the adapter, I already have an sd card reader on my laptop.
00:59:31scorcheUnhelpful: any microSD card should work
01:01:58Unhelpfulyou did it again ;)
01:02:11Unhelpfulunstable: *micro*sd is smaller
01:02:27scorcheUnhelpful: your nick isnt being helpful with my tab-completion...
01:02:52unstableThat one says "microsdhc" will that work? heh
01:03:06Unhelpfulscorche: only one of the services i provide.
01:03:41scorcheunstable: microsd/microsdhc ...happy? ;)
01:04:22unstablescorche: So that last amazon link i pasted will work right?
01:05:00scorchei dont see any reason why it shouldnt..........
01:05:44unstableok, heh. Just making sure.
01:05:51unstableI made a few mistakes before trying to find the card.
01:05:54unstableI'll buy that one.
01:07:50 Quit Nico_P (Remote closed the connection)
01:10:15 Quit xnyhps ("Zzzzz..")
01:11:16jhMikeSAre the ipod mini and mini 2g supposed to be able to record?
01:12:50pixelmaI believe they don't
01:13:30pixelmaor aren't
01:14:03jhMikeSI'm getting on google that they have wm8731 codecs, not wm8721 like 1g/2g which is curious.
01:16:03OvermindDL1If you all had your pick of any MP3 player that rockbox fully supports, what would be chosen?
01:16:03OvermindDL1As stated, I am looking to replace my Sansa e250, rotary thing is not working well anymore....
01:19:18pixelmajhMikeS: Rockbox wiki says WM8711 and has scans/photos
01:21:28jhMikeSpixelma: for both minis? I tried looking for HW info and failed.
01:22:08 Join Llorean [0] (n=DarkkOne@
01:22:10pixelmaI only found pages for the 2nd gen Mini
01:23:09jhMikeSpixelma: thanks for digging that up. :)
01:25:33grndslm??? forget that
01:36:46 Quit lasser (Read error: 110 (Connection timed out))
01:38:05 Quit sadmac2 (Read error: 60 (Operation timed out))
01:42:03pixelmajhMikeS: you're welcome :) maybe there would be some more info in the ipodlinix wiki but it seems it's not really usable at the moment (at least I can't search it)
01:43:13 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.4/2008111319]")
01:44:49 Quit BHSPitLappy (Read error: 110 (Connection timed out))
01:46:54 Quit MethoS- (Remote closed the connection)
01:47:01 Quit Lynx_ ("Konversation terminated!")
02:11:08 Part pixelma
02:11:20 Join pixelma2 [0] (n=marianne@rockbox/staff/pixelma)
02:14:00 Quit Zom (Read error: 131 (Connection reset by peer))
02:17:59 Join Zom [0] (
02:18:34 Part hashbangfoo
02:36:22Aurix_Lexicofor a plugin, what's a good example for creating a menu?
02:37:37JdGordonsearch for rb->do_menu
02:38:13Aurix_Lexicok, thank you
02:38:45 Quit Zarggg ()
02:41:42 Quit Bagder (Read error: 110 (Connection timed out))
02:46:14 Join Bagder [241] (n=daniel@rockbox/developer/bagder)
02:46:21***Saving seen data "./dancer.seen"
02:48:20 Join Davide-NYC [0] (n=Davide-N@
02:51:02 Part unstable
02:52:11 Join webguest [0] (n=454c83f0@gateway/web/cgi-irc/
02:54:23webguestHi, I have a Toshiba Gigabeat F40, and have recently installed Rockbox. Unfortuantley, I can't play any of the music, because of the .sat encryption. I understand there is a place for this in the wiki, but the instructions do not work. Is there another way to decrypt the .Sat file so I can simply play the .mp3's?
02:55:41LloreanYou can just use the original MP3s. There isn't another known method.
02:55:43 Quit webguest (Client Quit)
02:57:52saratogaLlorean: who has to sign off on the BMP resize patch before it can be commited?
02:58:08saratogaor rather who will get angry if its commited without consulting them first
02:58:28Lloreansaratoga: I don't think we need a signature.
02:58:43 Join webguest10 [0] (n=4b48c4e8@gateway/web/cgi-irc/
02:58:46LloreanIt's one of those "generally wanted" features, so as long as the patch seems to be pretty good it's safe to go with it.
02:58:57LloreanWe MAY want to wait until after the 3.1 freeze ends.
02:59:21saratogai want it in for 3.1 if possible
02:59:43LloreanWell, I don't see any real problem with it trying to go in.
03:00:11saratogai'm of the opinion that it could go in tonight, though I don't know that I want to be the one to commit it
03:00:15 Join CrippsFX [0] (
03:00:31 Quit webguest10 (Client Quit)
03:00:43 Quit mofux (Remote closed the connection)
03:00:43Lloreansaratoga: Does the "rebuffer on WPS change" stuff work?
03:03:34JdGordonI wouldnt be against not allowing it to rebuffer on wps change and just show "no aa" untill the next one is available
03:03:44*JdGordon would also like it in for 3.1 to show a nice new feature
03:03:55LloreanI kinda like the idea of having a new feature for 3.1 as well.
03:04:32LloreanAnd it's "small enough" that we can probably revert its changes if we can't get it working quite right by the time we branch.
03:04:34JdGordonalthough, without proper testing it could get nasty
03:10:11LloreanUnhelpful: How do you feel about the status of the patch itself?
03:17:27Davide-NYCHas the GigabeatS bootloader been updated recently?
03:17:29 Quit Bagder (Read error: 110 (Connection timed out))
03:22:23saratogaLlorean: the current patch was updated today to rebuffer on WPS change
03:26:13jhMikeSDavide-NYC: Not in any major way I'm aware of.
03:26:21Davide-NYCHello Mike
03:26:25Lloreansaratoga: Well, if it's gonna be in for 3.1 it's really gotta go in very soon.
03:26:41jhMikeSDavide-NYC: Hello
03:26:50Davide-NYCI'm trying to do some testing on the GigabeatS but am running into the "file not found" error
03:27:04Davide-NYCOf course I forget how to fix and can't find reference to it in the wiki
03:27:09jhMikeSUsing the .tar trick doesn't work? It does for me.
03:27:32Davide-NYCplease describe the .tar trick. I'm unfamiliar
03:28:02Davide-NYCI can get into the bootloader's USB mode to access both partitions
03:28:43jhMikeS'make tar' dump the .tar in /Content/0b00/
03:28:50Lloreansaratoga: If it's expected to be working, maybe it should go in now. If anyone gets upset, we can agree to remove it before we branch for the RC if problems aren't fixed by then? Gives us a fair chance at getting the feature in, but we don't let it hold up the release?
03:29:02 Quit amiconn (Nick collision from services.)
03:29:06 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
03:29:10 Quit pixelma2 (Nick collision from services.)
03:29:18 Join pixelma2_ [0] (n=marianne@rockbox/staff/pixelma)
03:29:20 Nick pixelma2_ is now known as pixelma2 (n=marianne@rockbox/staff/pixelma)
03:29:51Davide-NYCso just "make" as normal and instead of make zip I make tar?
03:30:12Davide-NYCshould I delete the .rockbox dir first?
03:30:27jhMikeSshouldn't matter
03:30:35Davide-NYCthanks, am attempting
03:32:44Davide-NYCwhat is the state of the charging code?
03:33:09Davide-NYCjhMikeS: worked like a charm. Thanks!
03:33:21 Quit crwl (Remote closed the connection)
03:33:25jhMikeSgetting there. I had to make some power management code more friendly first.
03:33:42Davide-NYCAnything I should be looking out for and making note of?
03:33:57jhMikeSwhat do you mean?
03:34:27UnhelpfulLlorean: i feel pretty good about it. i'd like to know that more than akio and i have messed with it before i consider it done, mostly because of how many bugs "appeared" as soon as it went on FS and he started looking at it
03:34:39Unhelpfuldifferent usage patterns, and the like
03:35:16*Unhelpful did not know of this "tar trick"... i've solved it by reformat
03:35:29 Quit perrikwp (" ajax IRC Client")
03:36:33LloreanUnhelpful: Well, it shouldn't be *too* usage pattern dependent, should it?
03:36:50Davide-NYCjhMikeS: I mean is there anything you need an uninformed enthusiast (like myself) to do?
03:36:58jhMikeSthere's something odd about the filesystem there where it can get inconsistent somehow between rockbox writes and USB writes
03:37:38jhMikeSDavide-NYC: not at the moment. n1s has a radio patch if you would like to mess around with that.
03:38:34Davide-NYCwell, don't hesitate to ask for whatever
03:38:42*Davide-NYC not promising anything
03:41:07UnhelpfulLlorean: i'd hope not, most of the trouble arose from me mostly testing with sliding_puzzle, and then just shaking out issues w/ the greyscale scaler once it was done
03:41:50 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
03:42:19LloreanUnhelpful: I think the main question is: Do you want it in 3.1? If it's buggy it's probably gonna depend on you fixing the issues people have over the next week.
03:42:23jhMikeSDavide-NYC: I never hesitate to ask :)
03:42:38 Join smartguyz_ [0] (
03:42:50 Part smartguyz_ ("Ex-Chat")
03:45:48 Join Bagder [241] (n=daniel@rockbox/developer/bagder)
03:47:21saratogaLlorean: maybe ask once tomorrow when the swedes are up and if they say yes commit?
03:48:00LloreanSounds like a plan to me.
03:51:44soapIf any interested developer wants a Sansa Clip - PM me and we can work out a very merry Christmas.
03:54:38_Auron_I'm still waiting for my sansa e260 to get here
03:58:26 Quit Aurix_Lexico (Read error: 110 (Connection timed out))
03:59:55 Quit Llorean ("Leaving.")
04:01:56JdGordonspeaking of which... soap whats the deal with iap? in for 3.1?
04:04:02 Quit BHSPitLappy (Remote closed the connection)
04:06:55 Quit _Auron_ ("Infinity repeatedly denies rumours of plotting with zero to bring down the Universe.")
04:07:29Unhelpful@llorean, i don't mind bug-fixing, but it can take me a day or two to have time to work on a bug, and a few have taken a day or two to identify
04:10:31 Join AndyIL [0] (i=AndyI@
04:11:28 Join _Auron_ [0] (
04:22:21 Quit AndyI (Read error: 110 (Connection timed out))
04:23:52 Quit XavierGr ()
04:25:10saratogaUnhelpful: fixing bugs in SVN would be the easiest way i think
04:25:24saratogaat least it was in WMA since then people actually sent me problem cases
04:25:42 Quit aarcane ("Leaving")
04:26:29saratogaJdGordon: i think the patch is out of sync and no one has resynced it
04:26:52saratogaotherwise it seems to be working on a lot of devices, but not all since the various versions of the protocol aren't fully understood i think
04:29:08 Quit saratoga ("CGI:IRC (EOF)")
04:29:54 Join Am0nXz [0] (n=normanz@
04:29:56 Part Am0nXz
04:32:22 Join massiveH [0] (
04:32:54 Join Llorean [0] (
04:34:26 Quit obo ("bye")
04:34:58 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
04:35:57Unhelpfulhuh. that's odd, it appears that wps_uses_albumart is *always* set to 0 when wps_data_load is called.
04:46:23***Saving seen data "./dancer.seen"
04:48:29 Join webguest08 [0] (n=63e3ca18@gateway/web/cgi-irc/
04:48:55 Nick webguest08 is now known as viperman03 (n=63e3ca18@gateway/web/cgi-irc/
04:49:45Unhelpfulneed help with something?
04:50:07 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
04:50:47 Quit OvermindDL1 (Read error: 110 (Connection timed out))
04:51:11viperman03been a while since I was on irc ... anyways, has there been any discussion of the multi-font patch that was closed a while back?
04:51:30 Quit jhulst (Connection timed out)
04:52:02Lloreanviperman03: I'm not sure why there should be discussion of a closed patch.
04:52:43Unhelpfulone more dead bug :D
04:54:13viperman03The patch was never made to be part of the svn build (at lease no attempt was being made to integrate it). I was merely curious is any discussions were made to attempt integration.
04:54:35 Join blkhawk- [0] (
04:54:47Lloreanviperman03: If it was never intended for inclusion, it shouldn't have been submitted in the first place.
04:55:03LloreanThe patch tracker is the place for people to put patches while they work toward getting them prepared for inclusion.
04:55:29LloreanIf an author wants to work on a version of multifont for inclusion, we'd welcome a task for it.
04:55:42Unhelpfulattempting integration would mean a developer working on it to make it ready
04:58:13viperman03ver y understandable ... I'm not even sure if the original author was keeping tabs on how the patch developed to was it was. At any rate, my coding skills are quite limited and I was simply cusious if there was any past discussion made on this type of patch.
04:58:34 Quit esthar ("KVIrc 3.4.0 Virgo")
04:58:35LloreanI think a well done multifont patch would have a fair shot at inclusion.
04:59:10LloreanThere's already a syntax for using multiple fonts in the WPS, though currently it's limited to the compiled-in system font and whichever one user font is loaded.
04:59:44Unhelpfulit looks like somebody synced it in september. how out of sync it is now depends on the amount of work done in the files it touches
05:00:45Unhelpfulwould maybe font= :-separated-list be a better idea, than specific named fonts?
05:01:43LloreanUnhelpful: I think the problem is more with the use of the existing glyphcache, etc, than concerns about the syntax.
05:01:45Unhelpfulwe could have a set limit to the list length, of course... but that seems more flexible in terms of having WPS that uses more than one font, unless people really do want a different font each for menu, browser, tuner, record, and wps.
05:03:24UnhelpfulLlorean: i know nothing about the font/text drawing code. stuff on the wiki suggests it's harder than what i've had to work with lately.
05:07:33viperman03"Font is a number - 0 is the built-in system font, 1 is the user-selected font" I assume this is what you were refering to as "syntax for using multiple fonts in the WPS" ?
05:08:22Davide-NYCLlorean: I'm wondering whatever happened (if anything at all) to the USB Charge settings patch? Particularly for the H1x0 targets? Has anyone expressed any interest as far as you know?
05:08:47Davide-NYCI'm talking about the "Ask, Always, Never" setting for USB cable connect.
05:09:00LloreanDavide-NYC: Well, the H100 isn't likely a particular concern for it anyway
05:09:12Davide-NYCWhy is that?
05:09:17LloreanWe don't have a habit of including specific features for hardware mods. That's why, for example, RTC support isn't compiled in for the H100...
05:09:47LloreanEven if the option goes in, you'll still have to enable it in a custom build for H100 since the hardware doesn't natively have USB charging and the option shouldn't be presented to users for a feature their hardware doesn't have.
05:10:03Davide-NYCOf course. But it would be fantastic if it was included in the same way the RTC is.
05:10:25LloreanWell, is there a patch for it?
05:10:31 Quit blkhawk (Read error: 113 (No route to host))
05:10:33 Nick blkhawk- is now known as blkhawk (
05:11:29Davide-NYCI do not think there is one for the H1x0 since it's hardware charging
05:11:47LloreanI meant, is there a patch to present the option in general that seems acceptable for commit?
05:12:21 Quit miepchen^schla (Connection timed out)
05:12:42Davide-NYCI cannot find one. Doubtful.
05:13:24LloreanWell, that would be the first step to actually seeing such an option. Write a patch for it, so that you can have people actually try the idea and comment on it in reality, rather than conceptually
05:13:58Davide-NYCIs USB_POWER defined for the H1x0? If so this patch might work: FS #6654
05:14:20 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
05:14:29LloreanThe H100 doesn't have USB power.
05:14:37LloreanI certainly wouldn't expect it to be defined.
05:14:45LloreanBut I thought we're on to the option in general, not the H100 specifically, now
05:15:07LloreanAlso, that patch is *described* as nasty
05:15:16LloreanIt's almost certainly not suitable for review yet.
05:15:37LloreanReally, if you want to see the feature at this point it's pretty clear nobody else is working on it, so your best bet is to start working on it yourself.
05:16:02Davide-NYCSure, but as a total neophite programmer I'm much better of modifying extant code
05:16:14LloreanRockbox is extant code.
05:17:04LloreanYou're probably going to need to improve your programming skills if you want something suitable for inclusion, anyway.
05:17:17Davide-NYCmaybe and maybe not
05:17:17LloreanAnd "after I've spent some time reading" is still sooner than "well, nobody's working on it so probably never"
05:17:34LloreanIf you don't want to work on it, that's fine.
05:17:38Lloreannobody's making you.
05:18:03LloreanI'm just saying, at this point it's unlikely to progress until someone puts in some real time on it.
05:18:08Davide-NYCif I get something cobbled together that works it'll show enough good will and effort that some good samaritan code guru will clean it up for submission
05:18:27LloreanThat's very, very unlikely.
05:18:30LloreanAnd a bad assumption to make.
05:18:46Davide-NYCI still suspect you are in fact a robot
05:18:50LloreanIf you cobble something together that requires work to clean up, you're just creating work for someone else.
05:19:06Davide-NYCthat's true
05:19:27LloreanLook at how many cobbled together patches have been closed after sitting for a while with only resyncs and nobody willing to actually work toward commit.
05:26:06UnhelpfulLlorean: would there be any objection, at this point, to dropping further work on the split-out patches, in favor of working on the combination of them together? these branches are starting to make me crazy :/
05:28:13Unhelpfulmaybe you've not seen the FS task, but i started work in essentially the same form that akio had last had it, with a patch that refactored bmp.c, a patch that added the resizer, a patch that added support for resized load to albumart, and another patch for sliding_puzzle
05:28:49LloreanI think a unified patch is probably acceptable.
05:29:09LloreanUsually it's "one patch per feature/fix"
05:29:21LloreanBut I don't think this is something we'd just include part of.
05:30:22Unhelpfuli ended up having to merge the first two, to be able to really work on them at all. the others are still reasonably separate, but my local branches to track changes to each of them separately are getting a bit messy
05:30:33 Quit massiveH ("Leaving")
05:31:28LloreanI think, in the end, they're all part of the same "feature".
05:31:57Davide-NYCnight all
05:32:02 Quit Davide-NYC ("ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]")
05:32:55 Join dberg918 [0] (
05:34:51Tannyit looks like the ipod video has the worst video-playback of all the players
05:35:10Lloreanof the color ones, yeah.
05:35:40 Quit Horscht (Read error: 104 (Connection reset by peer))
05:35:45UnhelpfulTanny: you pretty much have to use the apply firmware for video playback on ipod video. it uses a hardware video codec that we lack docs for.
05:35:53LloreanWe can't use the special processor in it, so it's limited to decoding video on the main CPU like other players, and it's got the worst screen size to CPU speed ratio.
05:37:07Tannyis there a comparably large capacity player that has better video playback?
05:37:13 Quit viperman03 ("CGI:IRC (EOF)")
05:37:17LloreanGigabeat F
05:37:43 Join webguest12 [0] (n=63e3ca18@gateway/web/cgi-irc/
05:37:56 Quit webguest12 (Client Quit)
05:40:40Tannyis playback unlikely to improve?
05:40:53LloreanUnlikely any time soon, certainly.
05:41:08LloreanThere's no guarantee it ever will, but there's very little interest shown from people who could actually do the work.
05:41:48 Quit HellDragon (Connection reset by peer)
05:42:14Tannydoes that mean that other features that display movement, like doom and pac-man suffer as well? or are they sufficiently simple to play smoothly?
05:42:24 Join HellDragon [0] (
05:42:38LloreanThey suffer much less.
05:42:54LloreanVideo playback and Gameboy games are, I believe, the worst.
05:43:59_Auron_how well does stuff run on sansa e20x's?
05:44:19LloreanA bit better on the e200v1
05:44:25Lloreanit's basically the same CPU with a smaller screen.
05:44:53LloreanVideo is much better, but gameboy games are more limited by the CPU itself than the screen size (larger screen size quickly increases the CPU use for video playback)
05:45:12_Auron_and emulation is CPU-intensive no matter what
05:45:47LloreanThe various players aren't really designed to be particularly fast. Just 'fast enough' usually.
05:45:56*Tanny watches the demo video on the website and decides its plenty good for him
05:46:04LloreanThe Gigabeat F seems to be a rather striking exception to this, since it's _much_ faster than necessary for what it's original firmware does
05:46:06 Nick fxb is now known as fxb__ (
05:46:12LloreanTanny: What demo video?
05:46:37LloreanThe Elephant's Dream video is for running on your player, for benchmarking and such.
05:47:29Tannythey seem tuned to the individual players
05:47:39LloreanThey're sized to various screen sizes.
05:47:43LloreanSince videos need to be pre-scaled.
05:47:54LloreanBut how they look in terms of speed is not at all indicative of actual performance on the player.
05:48:47Tannythe ipod accepts outside videos, though, right? that I could encode and load on the original firm-ware if I need to watch a tv episode or film?
05:48:57Tannyor is it all drm only stuff?
05:49:07LloreanYou can create your own videos, but that sort of question or discussion really doesn't belong here.
05:49:22LloreanWe don't provide support or information about the original firmware here.
05:49:36Tannywouldn't get that here, I understand, just wanted to ask if it was possible
05:49:56LloreanGoogle is your friend for any questions about the original firmware
06:01:21 Join Darksair [0] (n=user@
06:04:47 Join woodensoul [0] (
06:05:23woodensoulHello all. I'm looking for someone that's in-the-know on the Sansa Clip port.
06:06:01Lloreanwoodensoul: Did you have a question about something in the wiki or on the thread?
06:07:03woodensoulWell I noticed that there has been some progress made lately, especially sound working and I was curious how usable Rockbox is on the Clip since the wiki still has old info regarding the progress.
06:07:17LloreanIt's not supported yet.
06:07:36LloreanSo please don't ask for usability updates. If you'd like to contribute, we're always looking for more coders and you can help it finish up.
06:07:44LloreanOtherwise, please simply have patience until it's listed as a supported target.
06:08:59woodensoulYeah, I understand that. I was interested in using it as its current state and yes, I have considered contributing by coding (I've learned C in school). Is the USB the main area that needs work?
06:10:08LloreanAs I said, it's not supported. Even if you're interested in using it in it's current state, it's preferred that people not come asking for status updates.
06:10:21 Quit axionix (Read error: 110 (Connection timed out))
06:10:30LloreanAs to "the main area that needs work", there are many areas of Rockbox that need work.
06:10:48LloreanThere's plenty of general areas that aren't player specific. Optimizations and bugfixes. But yes, USB is one area that's weak more or less globally.
06:12:22woodensoulI gotcha on the status updates policy. Is it OK to ask about the bootloader installation for the clip?
06:12:30 Nick fxb__ is now known as fxb (
06:12:36LloreanBootloader installation is covered on its wiki page.
06:13:46woodensoulWhat do most developers use to write their code? I have Visual Studio 2008, I assume that would be fine as long as I compile with the tools mentioned on the 'for developers' wiki.
06:14:22LloreanAny editor that won't break line endings or convert spaces to tabs will work.
06:14:51Unhelpfulsaving with unix line endings might be an issue for VS2k8?
06:15:16woodensoulI assume that's an option within the program... line endings?
06:15:30scorchedont ask us...\
06:15:37LloreanUnhelpful: I think Notepad doesn't screw them up any more, so hopefully VS doesn't either.
06:16:01scorcheLlorean: really?..that is news to me
06:18:51Lloreanscorche: Maybe I'm thinking Wordpad.
06:19:20LloreanEither way, it's Microsoft stuff that doesn't mess up line endings.
06:19:35Unhelpfulwordpad has multiple text format options on save
06:19:53woodensoulI'm confused on this line in the wiki: "/rbutil/mkamsboot contains a tool that patches the Sandisk firmware with a compressed copy of a rockbox bootloader.
06:20:22woodensoulSo that isn't incorporated in the GUI of Rockbox Utility?
06:20:38LloreanIt's _not released_ on those players yet.
06:20:45LloreanIt's still primarily developer tools.
06:21:37woodensoulso how to use those tools are in the developers wiki?
06:23:02 Join saratoga [0] (n=41becb3b@gateway/web/cgi-irc/
06:23:09 Join Dhraakellian [0] (
06:23:12Lloreanwoodensoul: It tells you what the command does on that wiki page.
06:23:23Lloreanyou need to compile and run it on the appropriate files.
06:23:31LloreanAs it says.
06:24:11Unhelpfulsaratoga: was a little confused, by fixing it would be easier "in svn"... you mean after commit?
06:24:29saratogaUnhelpful: basically i meant let people report bugs and fix them as they turn up
06:24:53saratogabasically, unless we're going to make fundimental changes for how a feature works, I like to develop it in SVN when possible so that more people see it
06:25:40Unhelpfuloh, yes, i see what you mean. i was more trying to give a realistic impression of what my response time might be, in the worst case, if nobody else can/will debug a scaler problem.
06:27:24 Quit HellDragon (Read error: 104 (Connection reset by peer))
06:27:41 Join HellDragon [0] (
06:29:08woodensoulLlorean: I just read the wiki and am unsure what to do to "compile and run it on the appropriate files" as you said.
06:29:38woodensoulThe "appropriate files" is the sansa firmware file, yes?
06:30:07saratogawoodensoul: yes you'll need a copy of the clip firmware (from sandisk)
06:30:17saratogaand the rockbox bootloader (compiled by you)
06:30:24LloreanIt says this pretty clearly.
06:30:34Llorean"a copy of the Sandisk firmware bin file (download from Sandisk's website), a copy of rockbox's bootloader"
06:31:08saratogai intentionally didn't give the actual command to discourage people who have no idea what they're doing from trying this
06:31:11woodensoulOK, I understand now that the bootloader is what needs to be compiled before using it to patch the sansa firmware.
06:31:21saratogai assumed anyone who wouldn't brick their player would understand how to run a unix command
06:31:42Lloreanwoodensoul: You also need to compile the tool.
06:32:18woodensoulWell, we are making some progress now.
06:32:28LloreanIf you can compile Rockbox, this should provide no problem for you.
06:32:39LloreanIf you can't manage to compile Rockbox, you probably shouldn't be doing this at this time.
06:32:49scorchewoodensoul: if you are getting confused by these instructions, perhaps it is a sign that you should wait a bit before undertaking this task when the process is simpler and safer
06:32:51woodensoulI've come to get informed so I'm not "someone that has no idea what I'm doing".
06:33:09 Quit wooster ()
06:33:18saratogawe had a lot of people brick their players back in the v1 development days, so i'm hoping to avoid that
06:33:31woodensoulI have compiled before, but it's been awhile and I've never compiled a bootloader, just the SVN.
06:33:34scorchesaratoga: that would be nice..
06:33:56 Quit Tanny (Remote closed the connection)
06:34:01Lloreanwoodensoul: If you read what it says when you're being asked questions while compiling the SVN build, you can figure out how to build a bootloader very easily.
06:34:42woodensoulAaah, very nice Llorean. I had overlooked that before.
06:36:09 Join Tangent1 [0] (
06:36:27woodensoulSo what about compiling the tool to patch the firmware? Is that obvious when compiling as well?
06:36:39 Quit HellDragon (Remote closed the connection)
06:37:40 Join HellDragon [0] (
06:37:53Lloreanwoodensoul: You just compile it like you'd compile any normal program.
06:38:09 Quit HellDragon (Remote closed the connection)
06:41:18woodensoulthanks guys, i'll re-read the developers wiki page.
06:42:11 Join esthar [0] (
06:42:48 Quit dberg918 ("Ex-Chat")
06:43:46 Join HellDragon [0] (
06:45:19 Quit Darksair (Read error: 60 (Operation timed out))
06:46:25***Saving seen data "./dancer.seen"
06:48:54 Quit Dieterbe (Read error: 110 (Connection timed out))
06:53:27 Join n17ikh [0] (
06:54:34 Join bleh [0] (
06:54:58 Quit n17ikh|Lappy (Nick collision from services.)
06:55:08 Quit bleh (Client Quit)
06:55:36 Join n17ikh|Lappy [0] (
06:56:02 Quit n17ikh (Nick collision from services.)
06:58:20 Quit saratoga ("CGI:IRC (EOF)")
07:01:33 Join Dieterbe [0] (
07:11:57 Quit guyjones ("Leaving")
07:12:30 Nick fxb is now known as fxb__ (
07:12:56amiconn_dionoea: RED
07:13:00 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
07:14:11Lloreanamiconn: And the patch makes several existing tags redundant I think.
07:18:03kkurbjunjhMikeS: the gigabeat F/X bootloader should be able to use interrupts now
07:18:31kkurbjunso I don't think that a fake sleep is necessary anymore
07:19:58kkurbjuncrt0.s makes sure that the code is running from the right location and sets up the MMU now for both the normal build and the bootloader
07:21:28 Join stoffel_ [0] (
07:36:15 Quit stoffel_ ("leaving")
07:38:48jhMikeSkkurbjun: that was never altered and I didn't look deeper so far so I didn't bother.
07:39:09 Quit Dieterbe (Operation timed out)
07:39:19jhMikeSIf we can do that, it would be better. What about allowing other threads to execute?
07:39:19 Join Dieterbe [0] (
07:51:36jhMikeSkkurbjun: how can the vectors be overwritten while loading the firmware and the tick is running? is it loaded elsewhere and then copied over?
08:11:59 Join amiconn_ [0] (n=quassel@rockbox/developer/amiconn)
08:32:30DEBUGLost contact with server (snapshot: dancer.c line 124)
08:32:30***Saving seen data "./dancer.seen"
08:32:30***Started Dancer V4.16
08:32:30***Connected to on port 6667
08:32:30***Logfile for #rockbox started
08:34:47***No seen item changed, no save performed.
08:34:48***Started Dancer V4.16
08:34:48***Connected to on port 6667
08:34:48***Logfile for #rockbox started
08:35:41CtcpVersion from freenode-connect!freenode@freenode/bot/connect
08:35:41***Server message 501: 'logbot :Unknown MODE flag'
08:35:41Mode"logbot :+i" by logbot
08:35:42***Server message 477: 'logbot #rockbox :[freenode-info] why register and identify? your IRC nick is how people know you.'
08:35:43 Join logbot [0] (n=bjst@gateway/web/cgi-irc/
08:35:43 Join _lifeless [0] (n=lifeless@
08:35:43 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
08:35:43 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
08:35:43 Join kugel [0] (n=chatzill@unaffiliated/kugel)
08:35:43 Join amiconn_ [0] (n=quassel@rockbox/developer/amiconn)
08:35:43 Join Dieterbe [0] (
08:35:43 Join n17ikh|Lappy [0] (
08:35:43 Join HellDragon [0] (n=jd@Wikipedia/HellDragon)
08:35:43 Join esthar [0] (
08:35:43 Join Tangent1 [0] (
08:35:43 Join Dhraakellian [0] (
08:35:45 Join woodensoul [0] (
08:35:45 Join blkhawk [0] (
08:35:45 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
08:35:45 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
08:35:45 Join Llorean [0] (n=DarkkOne@rockbox/administrator/Llorean)
08:35:45 Join _Auron_ [0] (
08:35:45 Join AndyIL [0] (i=AndyI@
08:35:45 Join Bagder [241] (n=daniel@rockbox/developer/bagder)
08:35:45 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
08:35:45 Join CrippsFX [0] (
08:35:45 Join Zom [0] (n=zom@reactos/tester/Zom)
08:35:45 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
08:35:46 Join shadearg [0] (
08:35:46 Join nuonguy [0] (
08:35:46 Join rasher [50] (n=rasher@rockbox/developer/rasher)
08:35:46 Join bapdog [0] (
08:35:46 Join gromit` [0] (
08:35:46 Join kkurbjun [0] (
08:35:46 Join Rob2222 [0] (
08:35:46 Join mc2739 [0] (
08:35:46 Join reacocard [0] (
08:35:46 Join tvelocity [0] (
08:35:46 Join FOAD [0] (
08:35:46 Join kachna|lappy [0] (
08:35:46 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
08:35:46 Join GodEater [0] (i=c2cbc962@gateway/web/ajax/
08:35:46 Join Xerion [0] (
08:35:46 Join nplus [0] (n=nplus@141.25.Globcom.Net)
08:35:46 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
08:35:46 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
08:35:46 Join CaptainKewl [0] (
08:35:46 Join goffa [0] (n=goffa@
08:35:46 Join kharo [0] (
08:35:46 Join lastebil [0] (
08:35:46 Join andrew__ [0] (
08:35:46 Join HBK [0] (
08:35:48 Join Hostile [0] (
08:35:48 Join Tristan [0] (
08:35:48 Join grndslm [0] (
08:35:48 Join Seedy [0] (
08:35:48 Join preglow [0] (
08:35:48 Join DaCapn [0] (
08:35:48 Join shodanX [0] (
08:35:48 Join Zambezi [0] (
08:35:48 Join tim__b [0] (
08:35:48 Join SUSaiyan [0] (
08:35:48 Join blithe [0] (n=blithe@
08:35:48 Join flux [0] (
08:35:48 Join Acksaw [0] (
08:35:52 Join fred_2 [0] (
08:35:52 Join parafin [0] (
08:35:52 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
08:35:52 Join Shiny [0] (
08:35:52 Join agaffney [0] (n=agaffney@gentoo/developer/agaffney)
08:35:52 Join BlakeJohnson86 [0] (
08:35:52 Join J-23 [0] (
08:35:52 Join TMM [0] (
08:35:52 Join jfc [0] (
08:35:52 Join thegeek [0] (
08:35:52 Join sbhsu [0] (
08:35:52 Join havien [0] (
08:35:52 Join courtc [0] (n=court@unaffiliated/courtc)
08:35:52 Join denes_ [0] (
08:35:52 Join krazykit [0] (n=kkit@
08:35:52 Join scorche|sh [0] (
08:35:52 Join markun [50] (n=markun@rockbox/developer/markun)
08:35:52 Join balou [0] (
08:35:52 Join scorche [0] (i=Blah@rockbox/administrator/scorche)
08:35:52 Join tarbo [0] (n=me@unaffiliated/tarbo)
08:35:52 Join alexbobp [0] (n=alex@
08:35:52 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
08:35:52 Join Gareth [0] (
08:35:52 Join plus_M [0] (
08:35:52 Join neddy [0] (n=john@nat/sun/x-4ec565b8c4d76350)
08:35:54 Join martian67 [0] (i=lol3izer@about/linux/regular/martian67)
08:35:54 Join stu8ball [0] (
08:35:54 Join rvvs89 [0] (n=rvvs89@pdpc/supporter/base/rvvs89)
08:35:54 Join soap [50] (n=soap@rockbox/staff/soap)
08:35:54 Join desowin_ [0] (n=desowin@
08:35:54 Join midkay [0] (n=midkay@rockbox/developer/midkay)
08:35:54 Join synergist [0] (
08:35:54 Join pabs [0] (
08:35:54 Join @ChanServ [0] (ChanServ@services.)
08:35:54 Join ze [0] (i=ze@
08:35:54 Join idshark [0] (
08:35:54 Join Jabone [0] (
08:35:54 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
08:36:10 Join Unhelpful [0] (
08:36:10 Join Kopfgeldjaeger [0] (
08:36:10 Join jon-kha [0] (
08:36:10 Join Tuplanolla [0] (n=jani@unaffiliated/tuplanolla)
08:36:10 Join amigan [0] (i=dcp1990@unaffiliated/amigan)
08:36:10 Join dionoea [0] (n=dionoea@videolan/developer/dionoea)
08:36:10 Join freqmod_qu [0] (i=quassel@2001:700:300:1430:213:d3ff:fee9:5ed0)
08:36:10 Join feisar [0] (
08:36:10 Join suom1 [0] (
08:36:10 Join larstobi [0] (n=larstobi@
08:36:10 Join crashd [0] (
08:36:10 Join Galois [0] (
08:36:10 Join ch4os [0] (n=ch4os@gentoo/user/ch4os)
08:36:13 Join Slasheri [0] (i=miipekk@rockbox/developer/Slasheri)
08:36:13 Join Nimdae [0] (
08:36:13 Join fxb__ [0] (
08:36:13 Join lostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
08:36:13 Join maraz [0] (
08:36:13 Join liiwi [0] (
08:36:13 Join Hadaka [0] (
08:36:13 Join avacore [0] (
08:36:13 Join Neovanglist [0] (i=Neovangl@
08:50:32 Join Bagderr [0] (n=daniel@gateway/web/cgi-irc/
08:50:32 Nick Bagderr is now known as B4gder (n=daniel@gateway/web/cgi-irc/
08:50:32Lloreandionoea: This seems like one of those features that probably should've been brought up on the -dev mailing list. Doesn't it make some existing tags pointless?
08:50:32 Join robin0800 [0] (
08:50:32 Join Rob2223 [0] (
08:50:32 Part B4gder
08:50:32 Join Bagderr [0] (n=daniel@gateway/web/cgi-irc/
08:51:20 Join bertrik [0] (
08:52:35 Join axionix [0] (
08:55:01 Quit robin0800 (Remote closed the connection)
08:55:03 Nick Bagderr is now known as B4gder (n=daniel@gateway/web/cgi-irc/
08:55:20pondlifeSite seems ok again now, for the logs...
08:58:16 Quit woodensoul ()
09:02:16dionoeaLlorean: yeah ... except that existing tags are likely to be better in terms of performance (since you don't have to do a tag lookup in the global settings array)
09:03:26Lloreandionoea: Surely you don't need to do the lookup more than once so long as they don't leave the WPS. Settings don't change themselves.
09:06:05 Join ender` [0] (
09:06:24 Join axionix_ [0] (
09:06:55dionoeaLlorean: well stuff like volume can change
09:07:14dionoea(it would be quite stupid to use this tag for volume though)
09:07:19 Part Dhraakellian ("Meddle not in the affairs of ircops, for they are (not so) subtle, and quick to anger.")
09:07:21LloreanI wasn't really thinking of volume. But Shuffle, Repeat, probably some others.
09:07:24 Quit Rob2222 (Read error: 110 (Connection timed out))
09:07:39dionoeaMaybe i cache the lookup ... I don't remember really well (I coded that a few months ago)
09:08:28LloreanI'm just curious what this really adds. What tags for the WPS were missing?
09:09:09dionoeaSome people were asking for a few extra tags in the features request forum if i remember correctly
09:09:17LloreanIt seems there was a pretty decent binary size change for it.
09:09:31dionoeaoh, I didn't check that before commiting
09:10:08LloreanWhy'd you commit it now (right before the feature freeze, without having looked at it for months?)
09:10:23dionoeait's forum topic 18338
09:10:40 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
09:11:08dionoeaheh, I didn't even know that there was a feature freeze commign up. Feel free to revert if it's really not wanted :)
09:11:23dionoeaI was just cleaning up my tree and figured that I could commit this.
09:11:24amiconn_Llorean: Settings can change without (completely) leaving the wps, by using the quickscreen
09:11:35Lloreanamiconn_: You still leave the WPS to access the quickscreen.
09:12:48LloreanBut I'm just curious if we want this, or if we want explicit tags. It seems if we're going to use this, we should clean up as many redundant tags as possible if we can reclaim any size that way.
09:13:02amiconn_I agree
09:14:49Zagordionoea: have you seen that your commit broke player?
09:15:04dionoeayeah, and I have a patch for it but can't commit it since I'm at work
09:15:19dionoeabasically you have to move the function implementation out of an ifdef bitmap
09:15:35LloreanThe volume tag at least has some special behaviour (in regard to >0 volume levels) that probably mean it's impossible to replace it with this any-tag, but I'm sure at least a few others can be.
09:17:09dionoeaLlorean: that could be emulated with a simple option to this new tag (but i'm not sure that it'd be a good idea to replace everything with it though)
09:17:37 Join petur [50] (n=petur@rockbox/developer/petur)
09:18:33LloreanIt looks like it's only really suitable for replacing Repeat, Shuffle, and Crossfade mode tags
09:18:49 Quit axionix (Read error: 110 (Connection timed out))
09:22:39 Quit kachna|lappy (Read error: 113 (No route to host))
09:24:32pixelmawhat does it "give back" when not used as conditional? If I remember correctly, the usual repeat and shuffle tags (also some other, like virtual LED etc.) have some one letter defaults when not used as a condiotionsl (shuffle shows an "s" etc.), it is mentioned on the CustomWPS page. I doubt this functionality is used often, but it's there currently...
09:25:42 Quit bertrik ("Leaving")
09:25:51amiconn_Iiuc you can always emulate this by using it as a conditional, and placing the letters into the alternatives
09:27:56Zagordionoea: ok. I can do that if you like.
09:32:08dionoeaZagor: sure
09:42:46 Join Thundercloud [0] (
09:47:29 Join moos [0] (
09:54:12 Quit perrikwp (" ajax IRC Client")
09:58:16 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
10:01:31 Join LinusN [0] (n=linus@gateway/web/cgi-irc/
10:01:41 Part LinusN
10:01:47 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
10:06:29 Join Zagor_ [0] (n=bjorn@
10:07:03 Quit Zagor (Nick collision from services.)
10:07:11 Nick Zagor_ is now known as Zagor (n=bjorn@
10:12:38kugelCan anyone tell me where I find the discussion why FS #4733 (Multifont) was not suitable for SVN?
10:13:14pondlifeZagor: What's been going on with the site/DoS over the past week? Might be worth posting such news on the forum...
10:14:26B4gder is my summary from yday
10:14:37pondlifeAh, but was also down
10:14:52pondlifeLinusN might want to know that e-mails have been bounced too
10:14:55B4gdersome of yes
10:15:05pondlifedblog was :/
10:15:12B4gderyes, but it got back up yday
10:15:38pondlifeWas down earlier today when I first checked - fine now though
10:16:18Zagorposting in the forums would have been a good idea. I'm not there much, so I didn't think of it,
10:16:56pondlifeWith no IRC logs available, that's the next place I check.
10:17:13pixelmaI first searched for info - in the topic, then forums and only then looked closer at the frontpage
10:17:40Zagormea culpa
10:18:07pondlifeNot a biggie, but an idea if there's a next time :)
10:19:13 Join kachna|lappy [0] (
10:21:48 Join lasser [0] (
10:28:34 Quit Thundercloud (Remote closed the connection)
10:28:59scorcheanyone in need of a cloak?
10:30:31 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
10:32:31 Join blippe [0] (n=none_of_@
10:33:58 Quit tchan (Read error: 113 (No route to host))
10:34:49***Saving seen data "./dancer.seen"
10:37:52 Join miepchen^schlaf [0] (
10:45:49 Quit jhulst_ (Read error: 60 (Operation timed out))
10:49:27 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
10:50:58Zagordoes anyone have a clue to this warning:
10:51:46Zagorand why does it only occur on soap's machine?
10:53:32B4gdervery strange
10:53:57pondlifeA bad SVN pull?
10:54:28pondlifeDoes the entire source tree get cleanly checked out each build?
10:55:11pixelmaa different gcc version?
10:55:26JdGordonkugel: its not suitable because its done in a way which everyone agrees is wrong...
10:56:00pixelmahmm no
10:56:24B4gderpondlife: no, each server does a "svn update -r[rev]"
10:56:34B4gdera full checkout would be very time consuming
10:56:42B4gderand bandwidth
10:56:56 Join einhirn [0] (
10:57:02scorchemaybe have him look at his local copy?
10:57:24B4gderyes, that sounds like a good idea
10:58:53pondlifeMaybe pull his server from the list in the meantime?
10:59:43kugelJdGordon: a bit more precise if possible. "Everyone agrees it's bad" is hardly something you can base something upon
11:00:02scorcheit is just a warning..
11:00:09B4gderpondlife: it's only a warning at this point, let's see if he comes around soon and we can fix it
11:00:22kugelI've noticed that everyone agreed that it's bad by myself
11:00:58pondlifeI was more thinking that other code might be corrupted on his machine, but not generating warnings/errors - that would be unpleasant, if unlikely.
11:01:07pondlifee.g. a missing source line or 2
11:01:15B4gderah right
11:01:30pondlifeUntil we know for sure
11:02:19JdGordonkugel: the accepted solution is to rework the font cache so it can store more than 1 font in a single buffer...
11:02:39JdGordonthe rejection was based on that.. not on the apps level usage
11:06:20kugelI also was curious about dynamically alocating the memory (like tagcache etc), so multiple fonts are supported, but the memory wouldn't be wasted if you only use 1 font
11:07:17JdGordonwell yeah thats a possible stand is option
11:07:44JdGordonbut then causes problems when you want a theme which wants 3 fonts but only allocated pace for 2
11:08:06JdGordonpixelma: wanna have a look @ 2 manual patches?
11:08:29JdGordon9134 and 9524
11:08:33kugelI thought you allocate so many memory how you need to fit all requested fonts
11:08:41ZagorJdGordon: "this wps requires a reboot before functioning fully"
11:08:58Zagorwindows all over again
11:09:02Lloreankugel: We already don't allocate enough memory to fit every font.
11:09:05kugelthat would potentially involve stopping playback while changing font of course
11:09:24LloreanWe have a glypchache where we cache a certain amount of the most recently used characters.
11:10:21LloreanThis is why sometimes even with Database enabled you'll get disk spinups while browsing. The requirement for a character that isn't in the cache.
11:10:47LloreanI believe the idea with multifont is that it needs to effectively use the glyph cache for dealing with multiple fonts of varying sizes.
11:12:59kugelbut without using multiple glyph cache (one for each font) AND without raising the glyph cache size?
11:13:23Lloreankugel: The glyph cache size can be discussed separately.
11:13:26 Quit havien (Read error: 104 (Connection reset by peer))
11:13:32LloreanBut yes, without using separate glyph caches.
11:13:52LloreanJdGordon: Either one, really. =P
11:14:11JdGordonno... database will hit the disk anyway unless its loaded to ram
11:14:18 Join havien [0] (
11:14:48*JdGordon thinks mutlifont is pretty low down on peoples priorities
11:15:36Zagorat least the people who are candidates to implement it ;)
11:15:39LloreanJdGordon: I think though that it is one of the last GUI prettification features that everyone's sorta in favour of, as long as it's done well
11:15:43kugelwell, the demand is high (also between devs to be able to have a different font on the remote), but the requirements for svn seem to be pretty hig
11:15:52 Quit einhirn (Read error: 54 (Connection reset by peer))
11:16:34 Quit grndslm (Read error: 113 (No route to host))
11:17:15Zagorkugel: it is an important core function. the requirements must be high.
11:17:52kugelsure. I have not argued against it
11:17:56JdGordonoh, I dont tihnk anyone is against mutlifont.. but those with the inclination all have things theyd rather fiddle with
11:24:04 Quit kugel (Read error: 60 (Operation timed out))
11:28:17 Join Darksair [0] (n=user@
11:31:19 Quit Darksair (Client Quit)
11:32:28JdGordongoing back to the setting wps tag from a while ago... I have no problem keeping old (now "redundant") tags for simplicity
11:33:42LloreanJdGordon: I think we'd be better off choosing "settings get explicit tags" or "all pure setting tags are used through the generic tag'
11:34:19LloreanI think it's actually simpler if we have less special cases rather than mixing it. It means current WPS authors have to learn a few new things, but it won't take them long.
11:35:18 Join einhirn [0] (
11:36:51LloreanI don't say it's necessarily a rush that this be done (all the included WPSes would need to be updated, since I think they all use at least one redundant tag) but it may be one of those things to keep in mind when introducing the new theme site (or possibly mark the tags on the wiki page in red as 'old and scheduled for removal' or similar if we do decide we don't need them)
11:48:25 Join Darksair [0] (n=user@
11:48:26 Quit nuonguy ("This computer has gone to sleep")
11:55:48 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
12:04:50 Join robin0800 [0] (
12:06:27 Quit Acksaw (Connection timed out)
12:07:47 Quit jhulst_ (Read error: 60 (Operation timed out))
12:09:09 Join Acksaw [0] (
12:10:55 Join midkay_ [0] (
12:12:14 Quit petur ("bah, gone to customer site")
12:13:02 Join tyfoo [0] (
12:21:39 Join mofux [0] (
12:27:25 Quit midkay (Read error: 110 (Connection timed out))
12:34:51***Saving seen data "./dancer.seen"
12:35:56 Join TheSphinX^ [0] (
12:42:04 Join n1s [0] (n=nils@rockbox/developer/n1s)
12:45:05n1sjhMikeS: i'm not sure if i understand what you mean about audio_set_inout_source
12:45:20n1sgah s/inout/output/
12:45:53 Join funman [0] (
12:45:58n1sshould i not pipe the fm output directly to hp out? or is this something that's needed for recording only?
12:47:40 Join Aurix_Lexico [0] (n=comrade@
12:50:17soapB4gder, I'm awake now.
12:50:49soapLast time my build server had issues it was a bad stick of memory - let's hope that is not it again.
12:52:17 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
12:54:41 Quit bapdog ()
12:56:31funmann1s: if fm is piped to hp out, it's less work to do (if you are not recording), so I think it's a win
12:57:16n1sfunman: yes, it's dead simple wich is why i don't think i understood his comment correctly :)
12:58:24soapOh well, I'm off to work. PM me a list of what you want checked an I'll get on it 10 hours from now.
13:03:15B4gdersoap: I think we need to figure out why your host causes the weird build warnings, but I'm not sure exactly how to do that. We need to play by ear a bit I guess.
13:04:17soapshould I just log in, wipe SVN and redownload?
13:04:37soap(but later?) I really am off to work.
13:04:46soapor run memtest86 again?
13:04:56B4gderyes, start with a clean wipe
13:06:56 Join kugel [0] (n=chatzill@unaffiliated/kugel)
13:14:54n1samiconn: is there any reason that splash does a full screen update instead of a partial of only the are the splash covers?
13:15:25 Quit Darksair (Read error: 104 (Connection reset by peer))
13:17:36 Quit bmbl ("Woah!")
13:20:08 Join fredddy [0] (
13:22:39dionoeafunman: \o
13:22:51 Join daurn [0] (
13:23:08daurnhas any work been done on the sansa fuze
13:24:20B4gderfuze is an ams sansa
13:24:36funmandionoea: hi!
13:24:40B4gderdaurn: funman and the team has done a lot of great work on those
13:25:19funmandaurn: the major obstacle remaining on the Fuze is the access to the buttons missing, but luckily kugel is working on it
13:25:44daurnwhere do I get one
13:26:40dionoeaI don't think that's recomended ... since you wouldn't be able to use it without buttons :)
13:27:36 Quit Acksaw (Read error: 104 (Connection reset by peer))
13:27:56 Join Acksaw [0] (
13:28:05daurndionoea: no reasonI can't do buttons myself :P
13:28:41funmandaurn: if you want to help, all the code is in the svn repository
13:28:45gevaertssoap: I think these are a bit too consistent to be RAN issues
13:28:55daurnfunman: but I need the device first :P
13:29:01daurnwhere is generaly the cheapest place to get one
13:30:00funmanhmm it depends where you live; have a look at anythingbutipod forums, they track cheap resellers
13:32:10n1sfunman: re: fs#9609, do you mean to just initially fill write_bytes with an i2c read instead of hardcoding?
13:33:22 Quit kugel (Read error: 110 (Connection timed out))
13:33:42funmann1s: I think yes, but if the chip works that way, no need to bother I suppose
13:34:05 Join soap_ [0] (n=42c07542@gateway/web/cgi-irc/
13:34:17soap_gevaerts: that is what I was (hopefully) thinking as well.
13:34:36n1sfunman: shouldn't be hard to do, thanks for spotting the array iondex bug too :)
13:36:21funman;) I'll try to see what happens on the Clip this afternoon, for now we can't tune (I'll check if i2c writes are working, I don't know at which point bertrik went so far)
13:39:59 Quit Nico_P (Remote closed the connection)
13:44:31 Quit soap_ ("CGI:IRC (Ping timeout)")
13:50:19 Join Darksair [0] (n=user@
13:51:36 Quit mc2739 ()
13:54:37 Quit tyfoo ("Carpe diem")
13:56:00 Quit miepchen^schlaf ()
13:58:53n1sfunman: new patch is up, and the reception is better now that the spacing is correct :)
14:02:00 Quit Darksair (Remote closed the connection)
14:02:58 Quit Aurix_Lexico (Read error: 110 (Connection timed out))
14:03:31 Quit CrippsFX (Read error: 60 (Operation timed out))
14:03:41 Join kugel [0] (n=chatzill@unaffiliated/kugel)
14:07:03 Quit funman (Read error: 148 (No route to host))
14:11:07 Join massiveH [0] (
14:16:19 Quit robin0800 (Read error: 110 (Connection timed out))
14:16:25 Join Schmogel [0] (
14:20:58 Join LambdaCalculus37 [0] (i=44a04303@rockbox/staff/LambdaCalculus37)
14:23:06 Quit n1s ()
14:29:47 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
14:30:06 Join gregzx [0] (n=chatzill@
14:34:43 Quit Hostile (Read error: 60 (Operation timed out))
14:34:53***Saving seen data "./dancer.seen"
14:36:15 Quit _lifeless (Remote closed the connection)
14:43:50 Quit kachna|lappy (Read error: 110 (Connection timed out))
14:46:02 Quit kugel (Remote closed the connection)
14:46:15 Join kugel [0] (n=chatzill@unaffiliated/kugel)
14:51:13 Quit TheSphinX^ ("XChat@Linux")
14:53:07 Quit kugel (Remote closed the connection)
14:53:18 Join kugel [0] (n=chatzill@unaffiliated/kugel)
15:02:55 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]")
15:03:31 Quit kugel (Remote closed the connection)
15:17:47 Join kugel [0] (n=chatzill@unaffiliated/kugel)
15:28:19 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
15:31:13Unhelpfulkugel: yes?
15:32:36 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
15:36:20 Join robin0800 [0] (
15:37:30 Quit robin0800 (Remote closed the connection)
15:38:14 Join robin0800 [0] (
15:45:32 Quit robin0800 (Remote closed the connection)
15:47:31 Quit CaptainKewl (Read error: 110 (Connection timed out))
15:47:39 Part B4gder
15:48:03 Join Bagderr [241] (n=daniel@rockbox/developer/bagder)
15:48:21 Nick Bagderr is now known as B4gder (n=daniel@rockbox/developer/bagder)
15:51:37 Quit daurn ("ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]")
15:52:40 Join funman [0] (
15:54:29 Join robin0800 [0] (
16:01:09 Quit Zagor ("Client exiting")
16:01:49 Join {phoenix} [0] (
16:02:27 Join matsl [0] (
16:03:22 Join funman_ [0] (
16:03:31 Quit funman (Nick collision from services.)
16:03:35 Nick funman_ is now known as funman (
16:16:22 Join Mastrojan [0] (n=4dfdb48a@gateway/web/cgi-irc/
16:16:31 Quit Mastrojan (Client Quit)
16:16:44 Join Mastrojan [0] (n=4dfdb48a@gateway/web/cgi-irc/
16:17:03Mastrojanhow is current status for Sansa FUZE?
16:17:35funmanMastrojan: follow the "Status for current and work-in-progress targets" link on front page
16:17:53funmanthe status for fuze is described in the "Sansa V2 family" in new ports
16:20:22MastrojanFunman, when you finish Porting Sansa V2? ;) soory 4 my bad english :P
16:21:08MastrojanWhat You must finish? Buttons and?
16:21:21funmanMastrojan: no idea, we do a bit everyday, and when it's finish we'll write it on the front page
16:21:49funmaninternal storage, usb, fm, recording, write a manual, plus bugfixes, etc..
16:22:16 Join toffe82 [0] (
16:22:19MastrojanI understand
16:22:19 Join n1s [0] (n=nils@rockbox/developer/n1s)
16:23:05MastrojanBut for February you can finish?
16:23:18funmanFebruary 2066, perhaps
16:23:23Mastrojan:P :P
16:23:36Mastrojan2012 - end of world :P
16:24:04B4gderisn't that 2038?
16:24:19funmandepending the sect you are a member of :)
16:24:27B4gderthe 32bit time_t sect!
16:25:45MastrojanMatbe i can help for porting Sansa Fuze?
16:25:59MastrojanBut i can any skills for electronics, programming
16:26:34funmanMastrojan: programming skills are needed, but perhaps you can define a keymap for rockbox and its plugins, and test it in the fuze simulator
16:27:00 Part B4gder
16:27:05funmann1s: i added some comments to FS #9609
16:27:54Mastrojanthanks funman
16:28:08funmann1s: I wonder which text editor let you indent functions on 5 spaces :/
16:28:10n1sfunman: thanks
16:28:21n1sfunman: my spacebar does
16:28:49 Quit Mastrojan ("CGI:IRC (EOF)")
16:28:49funmann1s: ok, I just set tab to write 4 spaces, but if you need \t elsewhere that can be a problem
16:30:36n1sfunman: i need to think about the init stuff, because there might be another problem...
16:34:57***Saving seen data "./dancer.seen"
16:36:16n1syes, tuner_init gets called from radio_init which is called from init() which is the first thing called in apps on startup, that is before tuner_power is called so the fm chip is not even powered when i try to read :/
16:37:03funmanI believe radio_init() needs to be called after tuner_power() then
16:37:22n1salso you are right, the gpio interrupts ar enot needed for now
16:37:24*funman masters obviousness
16:38:28funmanalso write_bytes[] miss 2 registers (0x6 & 0x7)
16:39:54n1syes, i only made it contain the ones we actually change
16:40:12funmanhm but in register 0x7 there is a bit to enable the oscillator
16:40:51n1salso since this init is only called once i think itäs better to enable/disable the chip with tuner_set
16:41:21J-23did anybody remove Mastrojani's post in v2 thread?
16:41:54funmann1s: with a RADIO_INIT case?
16:42:24LambdaCalculus37J-23: It was a "port status update" request.
16:42:30n1sfunman: no in the RADIO_SLEEP case as it is now, i don't see the problem with that
16:42:55funmanI think RADIO_SLEEP isn't used prior to other commands, but I didn't check
16:43:47n1stuner_set(RADIO_SLEEP, 0); is the first command sent to the tuner in radio.c:radio_start
16:44:00funmanoh then it makes things simpler
16:47:28 Quit n1s ()
16:47:35Unhelpfulfunman: vim at least can do it, kate as well (5-space indents). and if emacs can't, i would be shocked.
16:48:08funmanUnhelpful: true, but 5 spaces are quite uncommon ;)
16:49:26Unhelpfuland off-topic here?
16:51:05kugelJ-23: !?
16:51:23kugelUnhelpful: Oh I was just curious about the upscaler.
16:51:41kugelIt wasn't enabled by default, unlike to what you said
16:52:05Unhelpfula bug. see the FS task for details :/
16:53:00kugelUnhelpful: well, I added the #define and it worked.
16:53:15kugelNice job, the image quality is definitely pleasing imho
16:53:41 Join grndslm [0] (
16:54:03Unhelpfulah, fine. the only other thing i fixed aspect ratio in the downscaling case, when upscaler was disabled, so you needn't bother with the new patch
16:54:23*Unhelpful can't believe he used "#ifdef HAVE_UPSCALER" and then "#define HAVE_SCALER" :/
16:54:39 Quit kugel (Remote closed the connection)
17:00:04 Join IudeX [0] (n=4e83d3b2@gateway/web/cgi-irc/
17:00:36IudeXin this one week there aren't any progress in sansa ams
17:00:51IudeXis it a lot of work?
17:01:29funmanIudeX: yes we are too busy telling people to not ask for status, so we can't work
17:02:19IudeXfunman. I'm not asking about status :P
17:02:30LambdaCalculus37Sounds to me like you are.
17:05:41 Quit IudeX ("CGI:IRC (EOF)")
17:05:44 Quit einhirn (Read error: 104 (Connection reset by peer))
17:18:22funmanwhat is the typical time for a tuner until it 'tunes' on a station ?
17:19:18funmanit looks like si4700_tune() always timeouts on the Clip, but if I check the tuned bit manually it is eventually set
17:21:53 Join n1s [0] (n=nils@rockbox/developer/n1s)
17:30:41funmann1s: hmm if i print out the registers while tuning I notice that the chip switches between power on and power off :o
17:31:15n1sthat does sound, umm, "wrong" :P
17:31:35 Join Strife89 [0] (n=michael@
17:32:05funmanhm i'm not reading the correct bytes ..
17:32:29funmanit toggles mute/non mute
17:33:18funmanrather, the tune bit isn't set
17:33:56funmanhmm no, again i'm reading wrong (this different read & write order is confusing)
17:34:17n1si'll change init to just zero out write bytes for now, do you need to touch register 7? that is the first one that needs its bits preserved
17:35:07funmani believe the XOSCEN bit is needed, but i'm not sure since we're not to the point to get sound already
17:37:52funmancan c200/e200 radio be recorded / does it use pcm driver to playback ?
17:38:03funmanor only a loopback
17:42:31LambdaCalculus37funman: The c200/e200 radio can be recorded as well.
17:43:58n1sfunman: iiuc XOSCEN is only needed when you do not have an external clock source connected?
17:44:37funmann1s: i'm not sure, but the description says it needs an external crystal anyway
17:47:37n1sfrom th "Initialization Sequence" it sounds like you can choose whether to use an external or internal oscillator
17:50:49 Join kugel [0] (n=chatzill@unaffiliated/kugel)
17:52:32*n1s spots another mistake, i should not do this stuff without tripple-checking thae day after apparently :)
17:52:32 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
17:52:59 Quit massiveH ("Leaving")
17:57:05 Nick fxb__ is now known as fxb (
17:57:19n1sthe detection would return 1 for present and -1 for not present. Both of which evaluate to true when interpreted as bools...
17:58:40jhMikeSn1s: it's basically recording related.
17:59:07n1sjhMikeS: ok, I'll leave it as a stub for now then
17:59:42jhMikeSwhat code is asking for it?
18:00:58 Join Nibbl [0] (
18:02:32 Quit XavierGr ()
18:02:54 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
18:03:02jhMikeSright, it's all over the place :)
18:08:32 Join pixelma_ [50] (i=pixelma@rockbox/staff/pixelma)
18:18:18 Join gregzx [0] (
18:18:23 Quit pixelma (Nick collision from services.)
18:18:24 Nick pixelma_ is now known as pixelma (i=pixelma@rockbox/staff/pixelma)
18:19:24n1sjhMikeS: iirc i needed to add fmradio to the input caps to make audio_patch.c build which then wanted that function
18:19:26 Join faemir [0] (
18:19:54 Join stoffel_ [0] (
18:20:23n1shmm, radio_hardware_present() is not playing nice
18:21:57jhMikeSn1s: how did you implement it? A one-time check?
18:22:57n1sjhMikeS: yes, just check for chip id, but i think radio_hardware_present() can be called before tuner_power
18:23:16 Join MethoS- [0] (
18:23:45jhMikeSit should expect to have to power up the chip to check. see the sanyo code.
18:24:01n1sbecause when i enter the root menu the first time it reads the chip and manufacturer id as 0 but when i exit the fm screen it reads it correctly
18:26:02jhMikeSit shouldn't be checked except once. the chip will get repowered repeatedly for that when I'm pretty sure the hardware won't change thereafter.
18:26:27n1sah, yes, that driver does indeed power it up before checking
18:27:19n1sI guess i could bring back the init function and power up and check once in there and then just cache the value
18:28:02 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:28:24jhMikeSall with radio power used to repower every time but I changed it to cache the result and avoid all that activity
18:28:39 Join tyfoo [0] (
18:29:05n1sthe init is only called once on start up so it should be ok
18:30:37kkurbjunjhMikeS: crt0.s masks the interupts and copies the firmware to a safe location (0x31000000) and reboots if there is not enough space starting from 0x30000000, the bootloader loads the main firmware at 0x31000000 and crt0.s takes care of copying it to 0x30000000 and enabling the mmu
18:31:14kkurbjunI set it up so that the boorloader runs under the same conditions as the main firmware
18:31:38kkurbjunso assuming that the thread stuff is not ifdefed you should be able to run threads too
18:32:06jhMikeSkkurbjun: so then irq_handler for the bootloader will definitely not be overwritten (affirmation please :)
18:32:18jhMikeSnot the function but the vector
18:32:34jhMikeSor both...hehe
18:34:52jhMikeSNo MMU for the bootloader? How are the vectors at 0x00000000 reached?
18:35:02***Saving seen data "./dancer.seen"
18:35:59kkurbjuneverything will be copied assuming that you are booting from the OF instead of flash and then it will be rebooted from 0x31000000 and then copied again to 0x30000000 and restarted once more, so the vectors for both the bootloader and main firmware will be in the proper location (0x30000000/0x0), the bootloader doesn't overwrite the vectors when it's loading the main firmware, the main firmware takes care of setting up it's vector loc
18:36:08kkurbjunthere is an mmu in the bootloader
18:36:19kkurbjunI was just using the physical locations rather than the virtual
18:36:40kkurbjunsince the MMU is restarted after the code reaches 0x30000000
18:36:49kkurbjunand then everything is mapped to 0x0
18:37:53 Join bertrik [0] (
18:38:16amiconn_That sounds like a lot of copying
18:38:20kkurbjunthe both the main firmware and the bootloader are linked to 0x0 so after crt0.s is all done it jumps to main running from some offset starting at 0x0
18:38:41kkurbjunamiconn it doesn't do most of the copying if everything is at 0x31000000
18:39:03kkurbjunit only does one copy then to 0x30000000
18:39:39kkurbjunthe other copies are checks for running from flash and for the OF bootcode loading our FWIMG01.DAT
18:40:25kkurbjunsince it loads it at I think 0x30004000 if I remember correct
18:40:37 Quit Nibbl (Read error: 60 (Operation timed out))
18:41:05kkurbjunits also setup this way so that you could use the main build in place of FWIMG01.DAT or you could run the main build from flash without the bootloader
18:42:09kkurbjunthe main build from flash I have tested and it works fine, I havn't tested the FWIMG01 replacement since I havn't had an incentive, but it should work
18:43:14kkurbjunworst case it is two copies in either situation, if you count the OF loader it would be 3 copies in that case
18:43:52kkurbjunbut it allows it to work without any defines to tell the code that you are building a flash build or a OF bootloader build
18:44:21 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
18:46:05 Quit GodEater (" ajax IRC Client")
18:46:23bertrikfunman, do we have udelay on AMS sansa targets?
18:46:43funmannot that i'm aware
18:46:56*jhMikeS didn't even notice a delay in bootup from any copying
18:48:29funmanbertrik: what delay do you need?
18:48:45jhMikeSkkurbjun: does flashing anything get rid of the annoying startup screen?
18:49:27kkurbjunjhMikeS: technically the bootloader is started twice when you use FWIMG01.DAT, but it only runs a few lines of code and then copies. Yeah running from flash is a bit faster too, it starts in about 5 seconds versus the 10 I counted before
18:49:38bertrikI was thinking of some actual delays in the fm i2c code, instead of using a for-loop with some random number of iterations
18:50:17saratogasomeone on ABI mentions getting a Fuze with a 2.xx firmware
18:50:23 Join evilnick [0] (i=0c140464@gateway/web/ajax/
18:50:27funmanbertrik: read/write seem to be fine
18:50:51funmanif needed we could use a busy loop which checks the kernel timer value
18:51:15bertrikhow fast does the kernel timer run?
18:51:41kkurbjunthe code I am working on allows you to change the startup screen since we have that code in the bootloader already to copy the previous location of the LCDADDR1 to the new location I point it to the location in flash and the flash_writer I am working on allows you to change the splash to any image you like
18:52:05LambdaCalculus37saratoga: I wonder if it may be a hardware revision, or just your typical firmware update with new features.
18:52:07 Quit stoffel_ ("leaving")
18:52:08kkurbjunyou could remove the copy from lcd_meg_fx too and that would completely remove any bootsplash
18:52:43kkurbjunthe bootloader build that's on the tracker doesn't have the code to show the splash, it is setup so that it's always verbose and shows the rockbox boot messages
18:52:48bertrikhm, for 400 kHz i2c operation, all delays are less than 1 us
18:52:59kugelsaratoga: apparently a new pcd version
18:52:59saratogaLambdaCalculus37: everytime sandisk has changed the first digit, its been because they cahnged the CPU
18:53:07jhMikeSkkurbjun: I was sort of hoping to get rid of those leafy things and treble clefs :)
18:53:25kkurbjunbooting from flash also give you a wakeup alarm :)
18:53:45kkurbjunyeah, the old screen wasn't one that I prefered
18:54:40jhMikeSthe gigabeat S is even more annoying. that one _has_ to get flashed at some point and just be formatted regular FAT32.
18:55:41LambdaCalculus37jhMikeS: At least the bootsplash on the Gigabeat X isn't so bad (just a band of colors).
18:55:50kkurbjun:), I havn't seen it, I can imagine that not running with fat32 would be a bit of a pain
18:56:01funmanbertrik: see the TIMER_FREQ define
18:56:26LambdaCalculus37kkurbjun: Do you think flashing the Gigabeat X would be safe as well?
18:56:29funmanit's 24MHz / 16 = 1.5MHz
18:56:45 Join nuonguy [0] (
18:57:10jhMikeSkkurbjun: the TFAT has some compatibility issues that can cause files written to not have their data visible and then theres' the occasional "File Not Found" when booting and the semi-random reformatting Toshiba never fixed.
18:57:13LambdaCalculus37saratoga: True, although the m200 series didn't have a CPU change until 4.XX.
18:57:52 Quit Strife89 ("Bye all.")
18:58:04amiconn_jhMikeS: The funny thing is that on my beast, only the firmware partition seems to be TFAT.
18:58:35jhMikeSamiconn_: the other is FAT32?
18:58:40kkurbjunLambdaCalculus37: I don't know, I haven't seen the firmware on the gigabeat X, I would be surprised if the LCD init routine was the same, but it could be, I think it was toshiba that did the screens in the F/X so it could be possible
18:58:44jhMikeSboth mine show up "TFAT"
18:59:00 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
18:59:44kkurbjunLambdaCalculus37: you could try running the flash writer code and check the md5sum of your flash, if it fails the check I wouldn't mind taking a quick look at the firmware to see how similar it is to the F
19:00:12kkurbjunactually I might try writing X firmware to my F and see if it runs
19:00:34 Part LinusN
19:00:50LambdaCalculus37kkurbjun: I know the F and X are largely the same internally. I think the only differences were in the LCD and the hard drive connector.
19:01:17jhMikeSamiconn_: so the data partition is actually FAT32 and the one with nk.bin is TFAT? Weird.
19:01:55n1sjhMikeS: i have never seen the "File Not Found" error but i sometimes get several dirs with the same name and content in the same subdir
19:02:36 Join PaulJam [0] (
19:02:52jhMikeSn1s: I get a similar one as well occasionally.
19:03:34n1sjhMikeS: isn't TFAT basically just FAT32 with an extra FAT? maybe the of sometimes mess up and reads the one the computer os didn't update?
19:03:37funmanbertrik: since we don't have recording yet on sansa ams, we should use the loopback feature of i2sin & i2sout to hear from FM radio (at least that's how I understand it)
19:03:48 Quit agaffney (Read error: 104 (Connection reset by peer))
19:04:00amiconn_Umm, I wanted to check, and I've got a strange message...
19:04:06bertrikoh, I thought we could just use some analog path from input to output
19:04:14jhMikeSn1s: I don't know the details. Sometimes I have dumped data to a file in rockbox (without error) only to find the file to be 0 bytes the first time its created.
19:04:17 Join massiveH [0] (
19:04:27funmani looked for that in the as3514 registers but didn't find anyhing like that
19:04:32amiconn_It wants me to restore it using some pc app...
19:05:07 Quit pondlife ("Leaving.")
19:05:08bertrikyuk, that's quite a detour ... :)
19:05:19jhMikeSamiconn_: It did the random reformat thing?
19:05:38amiconn_I don't know... never experienced anything like that before
19:05:42n1si have got to say that that bug is fantastic :)
19:05:54 Quit funman ("leaving")
19:05:59LambdaCalculus37n1s: Also bloody annoying. ;)
19:06:40toffe82if I remeber it was happening only when connected to usb
19:06:55toffe82after a transfer or something like this
19:06:59n1sLambdaCalculus37: and that they decided it was a good idea to reformat and leave the player in an unusable state is pretty amazing:)
19:08:02jhMikeStoffe82: Probably but it was something I remember many complaining about when I first search around for the player.
19:08:32LambdaCalculus37toffe82: It also happens if you mess with any of the OF files on the data partition.
19:08:35toffe82yes, it was never fixed by toshiba
19:08:46 Join agaffney [0] (n=agaffney@gentoo/developer/agaffney)
19:09:52 Quit robin0800 (Remote closed the connection)
19:09:56jhMikeSit's incredibly stupid to even do that automatically. what idiotic paranoia about copying will lead to... they could have just repaired an error rather than trashing the entire device.
19:12:22 Join karashata [0] (n=karashat@
19:12:47 Quit moos (Read error: 104 (Connection reset by peer))
19:16:35bertrikthe as3525 diagram on page 3 suggests that it should be possible to route e.g. LINE1 directly to the output mixer
19:17:08 Join herrwaldo [0] (
19:24:48 Join funman [0] (n=fun@rockbox/developer/funman)
19:25:13funmanbertrik: it seems we only have to set the volume for line1, and mute other channels
19:25:28funmansee page 145
19:26:15BigBambiLambdaCalculus37: If you mean then S, then not any file
19:26:32BigBambiI only have nk.bin on my firmware partition (I deleted all the others) and it works fine
19:26:36n1sfunman: i have a new patch that fixes all the things you mentioned and some others too :) but my beast has to charge first
19:27:15bertrikno wait, I think we just have to set LIN1_on or LIN2_on in the AUDIOSET_1 register, see page 147
19:29:09*gevaerts decides to try flashing his gigabeat
19:29:26funmanbertrik: makes sense
19:29:43LambdaCalculus37BigBambi: On the main storage partition on the beast are files like drmstore.dat, MediaLibrary_thumbs.edb, etc. I tried deleting those files once, and when I went to charge my beast via the OF, it flipped out and wiped the partition.
19:30:08funmann1s: I'll not be in the feet of bertrik while he codes ;)
19:31:49BigBambiLambdaCalculus37: I know them all no problem
19:32:03BigBambi*I deleted them all no problem
19:32:25 Join miepchen^schlaf [0] (
19:34:25LambdaCalculus37BigBambi: I wonder why my beast wiped the partition when I deleted them. :/
19:34:25 Quit tyfoo ("Carpe diem")
19:34:35bertrikI think I'll commit some of the parts that are now mostly stable and create a new patch with the latest si4700.c from n1s in an hour or so
19:35:13LambdaCalculus37bertrik: Cool. :)
19:35:22n1sbertrik: i will test my latest changes and post a new patch soon
19:37:03n1sbertrik: also maybe everything named si4700 should be si470x but that is easier to do once it's in svn of course
19:39:45saratogaanyone have a reason album art resize shouldn't go in today?
19:40:57gevaertskkurbjun: my gigabeat seems to have a non-recognized firmware
19:44:41 Quit nuonguy (Read error: 60 (Operation timed out))
19:45:17kkurbjungevaerts: it's not a modified image is it?
19:45:24kkurbjunif not could you post it to the tracker
19:45:39gevaertsI didn't modify it, but I got this player second-hand
19:45:54kkurbjunyou can also restore with the backup.bin that Kyle posted on the tracker
19:45:56funmanbertrik: i want to check other models to verify if they use the same i2c, so we can put the fmradio file one level down
19:46:24kkurbjunIf you could post it then I would be interested to check it out
19:46:25funmanunfortunately i can't access my disassembly right now, perhaps tomorrow evening or in the week
19:46:35gevaertsOK. I posted them
19:46:54kkurbjunif you still want to try flashing now, you can run restore with the backup.bin file on the root of your player
19:46:56 Quit matsl (Read error: 110 (Connection timed out))
19:47:07gevaertsI'll try
19:47:10funmanbertrik: oh n1s doesn't need to enable the crystal oscillator on the gigabeat, do you understand why it's not needed?
19:47:36kkurbjunthis is for an F right?
19:47:43 Join Thundercloud [0] (
19:47:50gevaertsAn F60, yes
19:48:10 Join jgarvey [0] (
19:48:20kkurbjunok, good, I wouldn't do a restore on an X yet, I was just making sure
19:49:23LambdaCalculus37kkurbjun: My girlfriend is the one with the Gigabeat X, and I don't think she would appreciate me breaking it on her. ;)
19:49:36LambdaCalculus37I'll ask her for the OF files so I can get an md5sum on them, though.
19:50:03 Nick amiconn is now known as amiconn__ (n=jens@rockbox/developer/amiconn)
19:50:09 Nick amiconn_ is now known as amiconn (n=quassel@rockbox/developer/amiconn)
19:50:17 Nick amiconn__ is now known as amiconn_ (n=jens@rockbox/developer/amiconn)
19:50:48bertrikfunman, I think both the radio and the RTC in the as3525 need a 32 kHz and they both have their own oscillator. AFAIK you can only enable oscillator to drive the crystal. I found some code in the clip OF that enables the radio oscillator.
19:50:58gevaertsAfter restoring from the other backup.bin it works. Nice fast boot :)
19:51:02bertrik*only enable _one_
19:51:09 Join robin0800 [0] (
19:52:46kkurbjungevaerts: great, if you build rockbox with #define HAVE_RTC_ALARM you will have the wakeup alarm too
19:52:51bertrikfunman, BTW I agree about the audio-as3525.c (or similar)
19:56:01 Join ameyer [0] (
19:56:45funmanafaiu the one for e200/c200 only uses as3514 (and one PP register) so we could share it
19:58:29 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.4/2008111319]")
19:58:56 Join kugel [0] (n=chatzill@unaffiliated/kugel)
20:01:04jhMikeSLambdaCalculus37: If that the case, you buy another one and tell her it's the same one. :) (sort of like the dead goldfish trick)
20:02:32pixelmado I see correctly on pictures that if you hold the Clip in front of you, the two volume are to the right and the upper one is "Volume Down" and the lower "Volume Up"?
20:03:00pixelmaand "Power" on the left side?
20:04:32 Join _Auron__ [0] (
20:05:39LambdaCalculus37pixelma: Power is on the left side, but the volume rocker on the right is "Volume Up" on the top and "Volume Down" on the bottom.
20:07:39*pixelma tries to define additional keys on the numpad for the sim
20:09:19pixelmathought I could use the + and - keys for volume but that would be wrong then, at least on my keyboard - is above +
20:10:53n1sfunman: when i tried reading the 06 and 07 registers in the init and copying them to write_bytes i get no sound anymore...
20:11:25funmanoh ..
20:11:50LambdaCalculus37pixelma: How about the * and - keys as VolUp and VolDown?
20:11:51bertrikactually on my clip, there is no indication as to which one is volup and which one is voldown. Both the OF and rockbox implement the upper button as volume up now.
20:12:22 Join mofux_ [0] (
20:12:26funmanregister 7 says 'do not write during powerup'
20:13:02n1sah, hmm, then this will get more complicated...
20:13:11pixelmaon the c200 there is a slight hint what's supposed to be vol up or down because one has a raised and the other a sunken triangly thing
20:14:03funmanwell there is 2 triangles (down on voldown, up on volup)
20:14:50pixelmaon the pictures of a Clip it looks like both are raised though
20:15:57LambdaCalculus37pixelma: Indeed, that is the case.
20:16:04 Join Zagor [242] (
20:18:29gevaertskkurbjun: I could, but then I'd have to set the clock correctly :)
20:22:35pixelmaLambdaCalculus37: but then it's on the side and the top one is volume up etc. probably makes sense
20:23:27 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
20:24:07 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
20:26:36LambdaCalculus37pixelma: I know it may throw people off, then, but using - and + for VolUp and VolDown will make more sense.
20:27:44LambdaCalculus37Perhaps the * key can be the home button.
20:28:45 Quit mofux (Read error: 110 (Connection timed out))
20:29:42 Quit mofux_ (Read error: 110 (Connection timed out))
20:35:03***Saving seen data "./dancer.seen"
20:35:26 Quit neddy ("Leaving.")
20:36:24*Bagder throws #FS9614 in Zagor's lap
20:36:37Bagderthat was letters in slightly wrong order
20:36:55*Zagor deciphers
20:37:51 Join neddy [0] (n=john@nat/sun/x-3e80c32e29d6c537)
20:37:56gevaertsI recently watched the credits on my clip (yes I was bored...), and I noticed that the default system font doesn't have all needed characters
20:38:23 Quit miepchen^schlaf ()
20:38:47 Join arohtar [0] (
20:39:03Zagorsince it's compiled-in, including everything would either take a lot of room or require some major cleverness to find out what to include
20:40:59pixelmaI think it was agreed the builtin system... what Zagor said... but: the Clip should be using a different default font which has some more characters (08-Rockfont, IIRC)
20:41:06funmanbertrik: i'm seeing random seeks (not skips) using buffering_flash.c since today .. does it also happen to you ?
20:41:31gevaertsMaybe an interim solution could be to transliterate the names that don't work? (but then, that's not easy either)
20:41:36funmanpixelma: more characters is better :)
20:41:46 Quit faemir (Read error: 54 (Connection reset by peer))
20:41:51bertrikfunman, haven't played anything yet today
20:45:04 Join botanical [0] (
20:45:13 Quit botanical (Client Quit)
20:46:48 Join kachna|lappy [0] (
20:47:09 Join aarcane [0] (
20:49:13pixelmadidn't see a missing character with 08-Rockfont but wonder who "Alastair S" is...
20:50:11 Quit robin0800 (Remote closed the connection)
20:52:10LambdaCalculus37pixelma: No idea, but I see a reference to his name here:
20:53:39*domonoky plays with fm on m200v4... but i am not sure how to test it. (i have the fm-screen, and can let the chip do scanning, it seems. But it doesnt find any stations, and i of course dont hear anything, as the ascodec is not correctly configured)
20:54:26gevaertsdomonoky: do you have earphones plugged in?
20:55:10funmandomonoky: try to enable line1 and/or line2
20:56:09 Quit arohtar (Client Quit)
20:56:22 Join faemir [0] (
20:58:49 Part kkurbjun
20:58:57domonokyfunman: will try. no ADC running needed ?
20:59:35funmanI don't think so, i believe the analog output of FM is directly sent to output
21:01:51 Join robin0800 [0] (
21:03:19kugelbertrik: I'm a bit confused about your i2c findings in the fuze firmware. Why did you even disassemble it for i2c?
21:03:22*domonoky enabled both lineIn but no sound... perhaps the fm-chip is not inited correctly. i dont even know if the communication works. i just used bertrik patch for generic_i2c.c and the existing tea5767 code :-)
21:04:01 Quit gevaerts (Nick collision from services.)
21:04:10 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
21:04:40funmanbertrik: why didn't you used the OF loop ? for(i=0;i<30;i++) asm volatile("nop"); I think it's both faster and smaller in code size
21:06:16 Quit Thundercloud (Remote closed the connection)
21:06:52bertrikkugel, why not? I found it when looking for button code
21:06:58domonokyfm-radio debug screen shows some reg-values, so i think i2c to the fm-chip should work communication..
21:07:10kugelbertrik: well, you have no fuze, do youß
21:07:22pixelmaLambdaCalculus37: I wonder about shortening the last name. We wouldn't allow "AlastairS" as wiki name, why is there one in the credts?
21:08:14 Join slitterell [0] (n=4b1b86cb@gateway/web/cgi-irc/
21:08:38slitterellI auto detect and I get the issue 'Your configuration is invalid. This is most likely due to a changed device path. The configuration dialog will now open to allow you to correct the problem.' but it never fixes no matter what I do
21:09:04domonokyslitterell: make sure the device and the path are correct
21:09:08slitterelland I have tried it on 2 computers, any ideas
21:09:20slitterellyeah, I have verfied both
21:09:49gevaertsslitterell: what device?
21:09:56domonokyrbutil checks if you selected a device, and if the path to the device is a valid location..
21:10:00funmanrunning as an administrator?
21:10:00slitterell4th gen ipod
21:10:22slitterelland on both computers I am admin
21:10:56domonokyslitterell: so autodetection finds the ipod, and fills in a path ?
21:11:13slitterellyes, autodetect finds ipod and path
21:11:43 Join crwl [0] (
21:12:02domonokyand you are absolutly sure the path is correct ?
21:12:26slitterellI am absolutely sure the path is correct. if I disconnect the device, it disappears, when I reconnect, it comes back
21:12:51gevaertsWhy on earth does compiling buffering_flash.c instead of buffering.c give me non-working USB on e200?
21:13:17slitterellonly thing I can think of is bad cable
21:13:29slitterellwhich is entirely possible
21:13:55LambdaCalculus37pixelma: You got me there.
21:13:56*domonoky checks the rbutil code.
21:14:49domonokyrbutil checks 3 things: if it has a player, a mountpoint, and if the mountpoint is writeable.
21:15:08 Join miepchen^schlaf [0] (
21:15:24slitterellit isn't writeable...
21:15:35domonokyif one of this 3 things are not correct, it will put you (back) into the config dialog
21:16:23slitterellwhen I click on the h: drive, in the situation, it says 'Please insert a disk into drive H:'
21:16:36slitterellbut I am DEFINTELY certain it is the right drive
21:17:15domonokycan you access the ipod drive normally via the explorer ?
21:17:43slitterellthat is what I am saying. Via explorer it says I need to insert disk
21:18:13bertrikdomonoky, are you using the right GPIO pins for i2c?
21:18:42domonokymaybe try a itunes restore first. before installing rockbox, the ipod should be in a working condition, and be visible as a drive in explorer.
21:18:50 Join xnyhps [0] (n=xnyhps@2001:470:1f14:da:219:e3ff:fed7:c57c)
21:19:26domonokybertrik: i am using the same gpio pins as you... (i thought maybe the engineers choose the same pins :-) )
21:19:54funmandomonoky: you can check GPIO usage in the wav_codec block
21:20:03 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
21:20:06funmanor in otg_functio (where the diagnosis mode is located)
21:20:28 Quit karashata ("G'bye everyone!")
21:20:37funmansorry I can't do that for you now :(
21:21:21*domonoky is not good with asm.. but i will find my way without it :-)
21:21:49 Join Jaykay [0] (
21:23:56bertrikLook for usage of 0xC80Y0400 where Y = B, C, D or E. This is the GPIO direction register.
21:26:08bertrikI read a slightly different device id for the si470x than n1s did. This id is used to determine if the radio chip is found, the debug menu my clip claims to have found the radio despite the different device id
21:26:41Jaykayi suggest closing of, because its actually not know what it does
21:27:18bertrikfunman, did you find a way to make tuning not time out on the clip?
21:27:52funmanbertrik: which ID do you read?
21:28:06gevaertsJaykay: well spotted. I think knowing what patches are supposed to do should be a requirement...
21:28:54bertrik0x1242, n1s expects 0x0a42
21:29:45funmanoh 0x0a42 is for si4700 and not si4702 perhaps
21:29:54 Join Aurix_Lexico [0] (n=comrade@
21:31:03gevaertsJaykay: closed
21:32:36bertrikfunman, but you did verify that writing worked?
21:32:53bertrikI think n1s is checking the wrong register index in his tuner code
21:33:09funmanbertrik: yes
21:33:35funmanbasically I read all the registers and printed them in the si4700_tune() loop
21:35:45bertrikI'm stll confused about which chip is driving the 32kHz crystal
21:35:55 Quit nplus (Remote closed the connection)
21:36:05funmandomonoky: I see references to gpiod in wav_codec
21:36:10 Quit Jaykay ("ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]")
21:37:07 Join einhirn [0] (
21:37:10bertrikor could the rectangular block be a self-contained oscillator+crystal?
21:37:37bertrikfunman, do you still have one of your clips opened to have a closer look at the rectangular thing next to the radio chip?
21:38:31funmanyes, moment please
21:39:14n1sbertrik: i fixed the detection in the latest patch, it now looks for 0x1242 too, i must have been very tired when i wrote that :)
21:40:01funmanit reads 'EAH3'
21:41:10funmani can't see the connections from the other side of the PCB
21:42:26funmanit seems objdump uses 16 bits ldr in thumb ..
21:43:09funmanoh no, it displays the addresses, not the value loaded
21:43:20funmanhow inconvenient :'(
21:43:58 Quit PaulJam (".")
21:44:39funmandomonoky: m200 i2c seems to use D6 & D7, now looking for which one is SCL
21:45:27funmandomonoky: for FM I2C on m200v4, D7 is SCL and D6 is SDA
21:45:59domonokyfunman: many thanks ! i will try it :-)
21:46:00jhMikeSyou're using bitbanging drivers when you have a hardware i2c module?
21:46:25*funman wonders if that 'you' is addressed to SanDisk
21:47:11*jhMikeS must be getting confused and should imbibe more often
21:47:38amiconnjhMikeS: There might be reasons to do that, e.g. forgotten pullups...
21:48:15funmanthere is 2 hardware i2c modules: one for audio/power and one unused
21:48:56 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.4/2008111319]")
21:50:08jhMikeSsometimes these hw designers have their heads up their *****. run one or two more traces or route them elsewhere and life could be superb.
21:50:56 Quit faemir ("Leaving")
21:52:10 Join fml [0] (n=4fd3c8ec@gateway/web/cgi-irc/
21:52:44slitterellfor the record, mine was formatted in mac mode and by restoring it in windows, it worked. Thank you for your help. :-)
21:53:41 Join faemir [0] (
21:54:11fmlRe r19357 (generic WPS tag): my opinion might not matter but I'm very in favour of keeping the specific tags. It's shorter, faster (if that matters) and is more from the 'business domain' IMO
21:54:28 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
21:54:50bertrikfunman, can't find much with EAH3 on google ... :(
21:55:35 Join kugel [0] (n=chatzill@unaffiliated/kugel)
21:57:08bertrikhmm, epson has some crystals starting with "EA"
21:57:55 Join matsl [0] (
21:58:33 Nick LambdaCalculus37 is now known as Trodgor (i=44a04303@rockbox/staff/LambdaCalculus37)
21:58:47 Nick Trodgor is now known as LambdaCalculus37 (i=44a04303@rockbox/staff/LambdaCalculus37)
21:59:11 Quit LambdaCalculus37 (" ajax IRC Client")
22:01:13domonokynow the fm-radio on m200v4 seems to commuicate correctly, and finds many stations. But sound is still missing :-)
22:03:12bertrikwow, do the station frequencies make sense?
22:04:37domonokythe look sensible, but im not sure, i dont hear radio often.. lets check a few .-)
22:06:34 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]")
22:08:29bertrikfunman, does it look like this :
22:10:20 Join Thundercloud [0] (
22:10:27bertrikdomonoky, did you have to change a lot of code to make it work so far?
22:12:05funmanbertrik: yes
22:12:06domonokybertrik: nope, i just copied your fm-i2c-code and used the tea5767 radio code.. with funmans gpio findings it seems to work. i find similar stations with my h100..
22:12:31funmansize and shape match
22:13:07funmandomonoky: you can move the file one level down and use ifdef for PINs
22:13:20funmanand work on audio for us :P
22:15:37 Quit fml ("CGI:IRC (EOF)")
22:17:23 Quit pondlife ("Leaving.")
22:20:48*domonoky hears some fm-radio feeping and the it shuts off again.. :-/
22:21:05domonokybut the is some fm-radio sound.. ! :-)
22:21:53domonokythe solution: enable line_in, un-mute it, and set the volume.. :-)
22:21:57*jhMikeS wonders why is whining about '/*' within comment
22:22:29jhMikeS* :p
22:23:55rasherdomonoky: When doing device autodetection in Rockbox, I think it'd make sense to pop up a dialog box saying "Found a Sandisk Sansa E200 at E: - You can now install Rockbox", and clicking OK will then close the Configuration dialog
22:24:03rasherdomonoky: Rockbox utility, that is.
22:25:18domonokyrasher: yes, that would make sense. there is even a not-ready patch in the tracker which does this (and also improved detection) but i dont have time for rbutil at moment :-)
22:26:02rasherIt's easy to miss that the autodetection actually worked
22:27:28 Join matsl_ [0] (
22:28:05 Quit matsl (Read error: 110 (Connection timed out))
22:28:44 Join {-phoenix-} [0] (
22:29:40 Quit {phoenix} (Read error: 110 (Connection timed out))
22:31:30*domonoky hears fm-radio !! :-)
22:34:20bertrikn1s, you seem to use only 9 bits to set the channel in your patch, but I think we can use 10 bits
22:34:41domonokyinteressting: i lowered the volume of the line_in channel, and dont get the shutdowns anymore. maybe it has something to do with "the input signals (in sum) should not be higher then 1Vp" ?
22:35:02n1sbertrik: ah, I'm off by one
22:35:08***Saving seen data "./dancer.seen"
22:37:45n1shmm, i wonder if it would make more sense to treat the registers as 16 bits instead of messing with the individual bytes
22:38:07funmanproblem is that they are transmitted as big endian
22:38:12 Quit {-phoenix-} (Remote closed the connection)
22:38:20 Quit matsl_ (Read error: 60 (Operation timed out))
22:39:14amiconnn1s: TEA5767 uses 5 byte transfer iirc. How would you handle that as 16-bit regs?
22:39:26n1sfunman: ah, right. My brain is big endian too :) Well, as long as you guys keep spotting all my mistakes i think we should be fine
22:39:46funmanamiconn: speaking about the SI4700
22:39:54n1samiconn: I meant for the si4700
22:39:59*n1s slow
22:40:38n1soften bitfields cross a byte boundary so it is a little fiddly to work with bytes
22:40:46 Quit amiconn_ (" HydraIRC -> <- s0 d4Mn l33t |t'z 5c4rY!")
22:41:06domonokybertrik, funman should i rename the fmradio-i2c-clip.c, if i move it one up ?
22:43:14funmanperhaps just remove the '-clip' , bertrik checked that the FuZe also use bit banging, perhaps i can have a look at e200v2
22:45:23 Quit ameyer ("leaving")
22:45:47bertrikbut does the fuze use the same GPIO bank for SDA and SCL?
22:46:21 Join MethoS-- [0] (
22:46:56funmane200v2 uses A7 for SDA and A6 for scl
22:46:58bertrikhmm, no, but we can probably work around that and still use a single fmradio-i2c-xxx.c
22:47:07n1sbertrik, funman: For scanning/seeking on the si4700 i was thinking about making an interface that just starts a seek and trigger an interrupt when it finishes which then tells the fm screen to update
22:48:05bertrikn1s, I don't know if any of the si4700 GPIOs is connected back to the as3525, so I'd rather stick with a software polled method
22:48:24 Join UzAt [0] (
22:48:54 Quit massiveH ("Leaving")
22:48:56n1sbertrik: ok, something like the current tuning loop could probably work, with a timeout then
22:49:00 Join matsl_ [0] (
22:49:12funmann1s: aren't GPIO1 & 2, SCL and SDA?
22:50:41bertrikfunman, no, SCL is mapped to SCLK and SDA is mapped to SDIO
22:50:45n1shmm, timeout should not even be needed, there's a helpful "seek failed" bit :)
22:51:38bertrikI just read that a tune is very quick, some the timeout can be drastically reduced. When we finally figure out what's wrong with the clip, we can probably remove the timeout
22:53:16n1stune never fails on the beast as it is now, but it thinks that most frequencies are stations...
22:53:46n1si'm not sure if the seek threshold etc would affect tune, that would be nice
22:58:05n1sanyway i'll save that for another day
22:58:33n1slargest problem is probably hacking radio.c
23:00:02 Quit MethoS- (Read error: 110 (Connection timed out))
23:02:40 Quit robin0800 (Read error: 104 (Connection reset by peer))
23:02:52 Quit Zagor ("Client exiting")
23:05:03 Quit funman ("leaving")
23:05:28 Quit domonoky (Read error: 104 (Connection reset by peer))
23:08:22jhMikeSkkurbjun: (for the logs) trying a real tick in the bootloader fails egregiously
23:08:30 Join brent113_ [0] (
23:08:31 Join blahrus [0] (n=blahrus@
23:08:50 Quit brent113_ (Client Quit)
23:09:01 Join massiveH [0] (
23:19:09 Quit aarcane ("Leaving")
23:19:36 Quit perrikwp (" ajax IRC Client")
23:22:16 Quit fredddy (Remote closed the connection)
23:23:45bertrikn1s, I'm running out of ideas why the si4700 won't work on the clip
23:24:33 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
23:26:09 Join Hostile [0] (
23:26:51 Quit jgarvey ("Leaving")
23:26:54 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
23:31:40Hostileso is the current status of rockbox on the clip kind of bootloader only right now?
23:31:48Hostilesandisk clip*
23:32:26LloreanHostile: It's in-development. Any updates are contained in the forum thread, the official Rockbox changelog, and the wiki pages associated with the player.
23:33:20Hostileokay. its just its the same forum thread for all sandisk v2 models, so its sifting through 30 pages of fuze and c200 talk to find a few tidbits about the clip
23:36:04 Quit Thundercloud (Remote closed the connection)
23:36:06LloreanWell, if you aren't going to be working on it, you can just wait until it's officially supported.
23:36:18LloreanIf you are going to be working on it, you need to sort through most of it anyway since there's a lot of overlap.
23:37:08jhMikeSkkurbjun: (more for logs) nvm, got to work.
23:38:03 Quit Schmogel (Read error: 104 (Connection reset by peer))
23:39:37 Quit matsl_ ("Leaving")
23:42:07n1sbertrik: can you seek and does RSSI vary in a sensible way? (IIRC this did not work for me before the 32kHz clock was supplied to the tuner)
23:42:24saratogaBagder: did you see my email to the SVN list?
23:42:52Bagderhe, I just responded to it too
23:43:27bertrikn1s, I'm now trying to "make it work" but not very systematically
23:43:43saratogaah good timing
23:43:55n1sbertrik: do you know what domonoky did to get it to work on the m200?
23:44:00Lloreansaratoga: I agree with Bagder regarding your email.
23:44:00saratogawell if he gets his account soon, we can make the 3.1 deadline
23:44:19bertrikn1s domonoky is working with a completely different radio chip
23:44:49n1sah, didn't know that
23:44:50bertrikRSSI is not communicated over the radio API, is it? how did you debug this?
23:45:15n1si just read all the regs over i2c and printed them in a loop
23:46:10bertrikn1s, how? do you just print them right over the other screen contents?
23:46:17LambdaCalculus37bertrik: Do you remember the FS# for funman's audio buffering patch?
23:46:44n1sbertrik: i just hacked it into one of the debug screens
23:46:58bertrikFS #9332 you mean? (that's nicop's patch by the way)
23:47:29n1son the beast's huge lcd printing them all was no problem but most are not so interesting
23:48:02LambdaCalculus37bertrik: That's the one; sorry I got confused. :)
23:49:20*LambdaCalculus37 tries out FS #9609
23:49:32bertrikn1s, what's a typical symptom of not having the 32 kHz clock?
23:50:41 Quit massiveH ("Leaving")
23:50:47n1sbertrik: i think seeking never found any stations and that RSSI didn't change much, it should be pretty obviously higher when on a station than between
23:52:40 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]")
23:54:02n1sbertrik: i tried to include register 6 and 7 in write_bytes earlier today but when i did that i got no sound on the beast, i see that you are doing that...
23:55:43n1soh and the init function is wrong too, i messed it up and forgot about the different starting points of read and write , fixed in my newest patch but it still has the problem when writing to those 2 registers
23:57:08n1sand you need to add si4700_init() to the TUNER_TYPE_CASE thing in tuner.c or the init function is never called
23:58:25 Quit n1s ()

Previous day | Next day