| 00:02.57 | Ian_Daniher_ | wow, that's absolutely minute! |
| 00:06.24 | g1powermac_PB | alright, its official, I'm voting for the green party if only I could actually vote for the green party in KY. . . |
| 00:13.16 | *** join/#edev dijenerate (n=dijenera@65.48.160.149) |
| 00:21.40 | MonMotha | mark me down for the WESD51 |
| 00:21.52 | MonMotha | awesome iron, reasonable price, readily available tips |
| 00:22.31 | MonMotha | it's no Metcal, but for the price, it's tough to beat |
| 00:24.06 | MonMotha | and really, you don't need a tiny tip for most things |
| 00:24.20 | MonMotha | it's not the size of your tip, it's how you use it ;-) |
| 00:24.28 | g1powermac_PB | heh |
| 00:24.48 | MonMotha | it does help for reworking QFNs |
| 00:24.57 | MonMotha | (well, QFNs and DFNs) |
| 00:24.58 | g1powermac_PB | yea |
| 00:25.07 | g1powermac_PB | and really small pitch qfp |
| 00:25.08 | MonMotha | but I find that for most things, the tiny tips just can't get the heat transfer needed |
| 00:25.21 | MonMotha | for QFP down to 0.5mm, I just use the stock tip |
| 00:27.14 | MonMotha | I've never managed to actually solder down a 0.5mm QFN/DFN, but I've never seriously tried, either |
| 00:27.23 | MonMotha | but I can do QFP down to 0.5mm |
| 00:27.30 | MonMotha | 0.65mm and larger is absolute cake |
| 00:27.45 | Ian_Daniher_ | MonMotha: yeah, I <3 my WESD51 |
| 00:27.54 | Ian_Daniher_ | I got it for $130 shipped |
| 00:27.58 | g1powermac_PB | MonMotha: the 0.5mm is tough |
| 00:28.09 | g1powermac_PB | I did it once with my el cheapo weller |
| 00:28.12 | MonMotha | 0.5mm takes me some extra time, but I can do it no problem |
| 00:28.24 | g1powermac_PB | I did the slather and remove method |
| 00:28.29 | MonMotha | I do have to clean the flux up afterwards |
| 00:28.36 | MonMotha | I just used a TON of flux :) |
| 00:28.46 | g1powermac_PB | slathered the solder and used wick to remove it all |
| 00:28.48 | MonMotha | only works if the board has a good soldermask |
| 00:28.55 | g1powermac_PB | yea |
| 00:29.08 | g1powermac_PB | luckily the board I was using was really well made |
| 00:29.10 | MonMotha | helps if the pads are oversized to "pull" away excess, but not required |
| 00:29.21 | MonMotha | all my protos are too, usually |
| 00:29.33 | MonMotha | I have a very small list of places I'm willing to buy protos from |
| 00:29.42 | MonMotha | I just don't want to deal with reworking crappy boards |
| 00:29.47 | g1powermac_PB | yea |
| 00:30.05 | MonMotha | not as big a deal at production where very little rework is required and they have people whose actual job is rework |
| 00:30.12 | g1powermac_PB | cheap perf boards are tough, even if you're dealing with through hole stuff |
| 00:30.24 | MonMotha | I'm not a solder tech, I'm an engineer. I just happen to be fairly good at soldering, so I rework my own protos. |
| 00:30.36 | MonMotha | you know, remove a trace here, add a connection here |
| 00:30.45 | g1powermac_PB | yea |
| 00:33.50 | g1powermac_PB | hmm, specialty heirloom berries. . .might be nice addition to the extreme green garden |
| 00:33.56 | g1powermac_PB | http://www.treesofantiquity.com/index.php?main_page=index&cPath=41_17 |
| 00:35.24 | Ian_Daniher_ | g1powermac: mmm |
| 00:35.33 | Ian_Daniher_ | g1powermac: looks tasty |
| 00:35.36 | g1powermac | yea |
| 00:36.23 | g1powermac | once I get all this work done at the store and house, I'll finally be able to work on the extreme green garden |
| 00:36.54 | g1powermac | its really been a mad dash to get all this work done before I head into surgery for my gallbladder |
| 00:44.25 | Ian_Daniher_ | :/ |
| 01:04.08 | GPSFan | g1powermac: our orchard is these: http://www.treesofantiquity.com/index.php?main_page=product_info&cPath=41_1&products_id=66 From: http://www.starkbros.com/index.jsp Planted in 1905... |
| 01:04.35 | g1powermac | nice |
| 01:04.57 | GPSFan | about 50 years past its prime ;>( but still producing... |
| 01:08.09 | g1powermac | thats cool |
| 01:09.21 | g1powermac | wishes we had more land |
| 01:09.44 | g1powermac | there's only so much you can do with a small lot |
| 01:09.57 | g1powermac | especially if you want stuff that only comes from big trees |
| 01:23.47 | g1powermac | gnight all |
| 02:32.15 | *** join/#edev Madsy (n=madsy@ti500710a342-0953.bb.online.no) |
| 03:38.56 | *** join/#edev Madsy2 (n=madsy@ti500710a342-0896.bb.online.no) |
| 03:59.29 | *** join/#edev DrZimmerman (n=theo@77-56-242-58.dclient.hispeed.ch) |
| 04:31.22 | *** join/#edev chouimat|away (n=dieu@r2351064.cidc.net) |
| 04:35.40 | *** join/#edev Chocobo (n=swinchen@strongbad.eece.maine.edu) |
| 04:44.55 | *** join/#edev KazaLite (n=kazaLite@116.58.8.130) |
| 05:02.30 | *** join/#edev mbuf (n=user@61.16.248.242) |
| 05:56.35 | *** join/#edev mbuf` (n=user@61.16.248.242) |
| 06:53.58 | *** join/#edev harpal (n=Harpal@122.169.108.195) |
| 07:06.26 | *** join/#edev [AD]Turbo (n=turbo@host45-243-dynamic.12-79-r.retail.telecomitalia.it) |
| 07:07.24 | [AD]Turbo | hola |
| 07:46.29 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 07:58.10 | *** join/#edev lyakh (n=lyakh@p57BD2324.dip0.t-ipconnect.de) |
| 09:03.46 | *** join/#edev roxfan2 (n=dunno@168.54-240-81.adsl-dyn.isp.belgacom.be) |
| 09:21.05 | *** join/#edev Chocobo (n=swinchen@strongbad.eece.maine.edu) |
| 10:39.41 | *** join/#edev g1powermac_PB (n=g1powerm@unaffiliated/g1powermac) |
| 10:52.09 | *** join/#edev mbuf`` (n=user@61.16.248.242) |
| 11:00.58 | *** join/#edev mbuf (n=user@61.16.248.242) |
| 11:07.23 | *** join/#edev g1powermac (n=g1powerm@unaffiliated/g1powermac) |
| 12:18.18 | wmat | morning |
| 13:06.26 | *** join/#edev mbuf (n=user@61.16.248.242) |
| 13:31.54 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 13:33.32 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 13:33.32 | *** mode/#edev [+o GPSFan] by ChanServ |
| 13:41.37 | prpplague | morning folks |
| 13:41.39 | prpplague | solar: ping |
| 13:45.48 | prpplague | MonMotha: ping |
| 13:50.40 | MothaWork | prpplague: yes? |
| 13:51.17 | prpplague | MonMotha: every had problems with intersil switchers? |
| 13:51.50 | prpplague | MothaWork: the softstart on some of these intersil switchers are exteremly long rampup time |
| 13:53.31 | MothaWork | I've not used any of the intersil switcher parts |
| 13:53.39 | MothaWork | but I could see that being a problem |
| 13:53.54 | MothaWork | may have to use a BOD or similar to hold everything in reset until power is stable |
| 13:56.05 | prpplague | MonMotha: we just moved over to a TI part |
| 13:56.12 | prpplague | MonMotha: the softstart is much better |
| 13:56.37 | MothaWork | that also works |
| 13:57.54 | MothaWork | I've been pretty satisfied with TI's power management lineup |
| 14:01.38 | prpplague | MonMotha: the intersil part has a 650uS softstart time |
| 14:05.00 | *** join/#edev dijenerate (n=dijenera@65.48.160.149) |
| 14:05.29 | *** join/#edev chouimat|work (n=dieu@209.217.106.98) |
| 14:07.44 | MothaWork | that's kinda excessive |
| 14:08.35 | MothaWork | does that part have some sort of special application? |
| 14:40.45 | prpplague | MonMotha: it says its for pda's and such |
| 14:40.53 | prpplague | MothaWork: but i don't think thats right |
| 14:41.38 | MothaWork | yeah, I can't think of any reason for a PDA to need "soft start" that is that log |
| 14:41.40 | MothaWork | *long |
| 14:44.56 | solar | rise and shine |
| 14:44.59 | solar | prpplague: pong |
| 14:47.30 | prpplague | solar: greetings |
| 14:47.42 | prpplague | solar: sorry about the lack of communication in the last couple of days |
| 14:47.53 | chouimat|work | prpplague: http://icanhascheezburger.files.wordpress.com/2008/05/funny-pictures-raccoon-subprime-mortgage.jpg |
| 14:47.54 | solar | figured you are busy |
| 14:47.55 | prpplague | solar: between debugging this hardware and "wtf meetings" |
| 14:48.10 | prpplague | solar: jsut been swamped, getting your tarballs together now |
| 14:51.00 | solar | it's ok. I started over on the toolchain. |
| 14:51.13 | solar | found a bug where eabi will default to arm5 |
| 14:51.26 | solar | arm10tdmi vs arm9tdmi |
| 14:57.50 | prpplague | solar: ahh lovely |
| 15:03.30 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 15:26.03 | ds2 | morning |
| 15:55.09 | prpplague | lordy lordy, i'll never deal with SFLC again |
| 15:55.49 | *** join/#edev CosmicPenguin (n=nobody@host-245-157.pubnet.pdx.edu) |
| 15:55.49 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 15:55.51 | solar | direct to the hall of shame next time? |
| 15:56.41 | chouimat|work | prpplague: ??? |
| 16:02.08 | ds2 | still no response from them? |
| 16:09.17 | *** join/#edev adye1 (n=adyer@adsl-69-223-13-78.dsl.chcgil.ameritech.net) |
| 16:18.30 | prpplague | solar: just post it on the net next time |
| 16:18.49 | prpplague | chouimat|work: i reported some GPL violations to SFLC a few weeks back |
| 16:18.57 | prpplague | chouimat|work: they've been total jerks |
| 16:19.16 | prpplague | ds2: leapfrog hasn't responded to me nor has SFLC |
| 16:19.19 | chouimat|work | prpplague: I knew the first part ... and the second part sound like lawyers to me :) |
| 16:19.24 | solar | <PROTECTED> |
| 16:20.00 | prpplague | solar: haven't got the time or the engery, at this stage, simply don't care, got bigger fish to fry |
| 16:24.45 | prpplague | just very disappointed |
| 16:25.28 | prpplague | chouimat|work: granted the lawyers are probably that way by default, but the SFLC is a non-profit organization, the rest of SFLC doesn't have to be jerks |
| 16:28.21 | wmat | prpplague: you should contact that kernel hacker that does all of the GPL violation stuff |
| 16:28.52 | prpplague | wmat: you talking about laforge aka harold welte? |
| 16:28.54 | wmat | also, there's a gpl-violations mailing list |
| 16:29.00 | wmat | yeah, harold |
| 16:29.11 | prpplague | yea, just as useless |
| 16:29.42 | prpplague | they did nothing about emsoft or mattel when i reported it to them back in 2003 |
| 16:30.13 | wmat | huh, well who the hell do that go after then? |
| 16:30.43 | wmat | i see gpl-violations.org is Harald and others |
| 16:31.17 | *** join/#edev lyakh (n=lyakh@p57BD2324.dip0.t-ipconnect.de) |
| 16:31.29 | prpplague | wmat: they only seem to go after people that are reported by the core group |
| 16:31.51 | wmat | they should have a big "Report Violation" button on their homepage |
| 16:35.41 | prpplague | wmat: yea well, then what? |
| 16:38.24 | wmat | yeah, good question |
| 16:38.26 | wmat | who knows |
| 16:39.27 | prpplague | which is the same issue with SFLC |
| 16:39.55 | prpplague | once you report it, they have no obligation to keep you informated or even communicate at all with you |
| 16:40.19 | prpplague | there is no procedure posted as to what to expect or who to talk to |
| 16:41.14 | CosmicPenguin | I think its like reporting a crime |
| 16:41.33 | CosmicPenguin | if you see a guy mug an old lady and call the cops, they aren't really obligated later to call you and tell you they collared the guy |
| 16:42.56 | prpplague | CosmicPenguin: slightly different, more like i'm a private investigator who comes across drug ring, you report it to the police, yet they don't ask for any of the photos you took, notes about comings and goings, or even the address where the house is |
| 16:44.19 | *** join/#edev flavioribeiro (n=flaviori@201.70.21.178) |
| 17:00.43 | *** join/#edev jkridner|work (n=a0321898@nat/ti/x-66ceb17e72f17ab4) |
| 17:01.00 | prpplague | jkridner|work: greetings |
| 17:01.22 | prpplague | jkridner|work: dev-n-hack channel, lots of professionals here |
| 17:01.25 | jkridner|work | hi prpplague. heard your mention of this channel and thought I'd lurk a bit. |
| 17:02.07 | prpplague | jkridner|work: np, lots of discussion ranging from heavy hardware to driver dev, and just about anywhere in between |
| 17:13.11 | *** join/#edev s (n=flaviori@201.70.21.178) |
| 17:23.12 | *** join/#edev bbradley (n=bbradley@78-105-167-16.zone3.bethere.co.uk) |
| 17:28.57 | *** join/#edev Ei5GTB_ (n=Paul@78.16.185.100) |
| 17:36.48 | *** join/#edev bbradley (n=bbradley@78-105-167-16.zone3.bethere.co.uk) |
| 17:36.58 | *** join/#edev prpplague^2 (n=dave@mail.americanmicrosystems.com) |
| 17:44.32 | prpplague | jkridner|work: don't suppose you know any wince developers in the are looking for work? |
| 17:53.16 | jkridner|work | I can check. Do they need to work in a certain place? |
| 17:54.56 | jkridner|work | At least I don't know that any of them are looking. |
| 17:57.57 | ds2 | ce is dead |
| 18:03.44 | MothaWork | ds2: if only |
| 18:06.38 | ds2 | there is no sane reason to ever use ce |
| 18:07.44 | ds2 | the maemo thumb pad is nicer then the osk |
| 18:13.13 | MothaWork | I won't argue with that, but the fact is that it's rather prevalent, and sometimes customers demand it |
| 18:14.01 | prpplague | ds2: people _still_ think they need it |
| 18:14.14 | ds2 | then they are noncustomers ;) |
| 18:14.19 | prpplague | jkridner|work: yea, we are looking for a full time person, euless area |
| 18:14.26 | chouimat|work | no they drive cars |
| 18:14.41 | prpplague | ds2: unfortunately, we need customers |
| 18:15.07 | prpplague | ds2: our sales department is having a hell of a time communicating with developers that use linux |
| 18:15.33 | prpplague | ds2: we had a turn over in staff for sales about 6 months ago, and now all the new sales people are just pissing the OSS/linux folks off |
| 18:15.44 | ds2 | get a nonlegacy sales team? |
| 18:15.48 | chouimat|work | prpplague: must be as hard with winblows developers :) |
| 18:16.16 | prpplague | ds2: ?? |
| 18:16.29 | ds2 | sales guys are disposable like a paper napkin |
| 18:16.32 | prpplague | chouimat|work: sorry, i did not understand that statement |
| 18:16.50 | prpplague | ds2: yea, well, i don't have any power over that |
| 18:17.03 | chouimat|work | prpplague: I just said the sales drones will have as much problem communicating with windows developers |
| 18:17.21 | prpplague | chouimat|work: they don't seem to |
| 18:17.32 | chouimat|work | now I'm scared .... |
| 18:18.05 | *** join/#edev Ian_Daniher (n=it@204.10.219.119) |
| 18:18.24 | ds2 | legacy/obsolete sales |
| 18:19.31 | MothaWork | the reality is that there are lots of people out there who are just plain used to "The Microsoft Way" |
| 18:19.48 | MothaWork | they complain about things like 3rd party commerical app support...on a fixed function device |
| 18:20.11 | MothaWork | They're used to Visual Studio and can't let it go |
| 18:20.43 | ds2 | and that is a dead end w/a 3000000000000meter drop |
| 18:21.30 | MothaWork | I'm not denying that, but the reality is that in some cases, if you want to sell the damn things, they have to be capable of running WiNCE |
| 18:21.40 | prpplague | exactly |
| 18:23.05 | ds2 | hange needs to start somewhere |
| 18:23.14 | chouimat|work | I think wince look a lot better than then non CE version of windows |
| 18:23.20 | prpplague | ds2: ?/ |
| 18:23.22 | prpplague | ds2: ?? |
| 18:23.28 | CosmicPenguin | CE != windows |
| 18:23.28 | Ian_Daniher | g1powermac: I'll ship your sht11 tonight |
| 18:23.43 | ds2 | change i mean |
| 18:23.46 | MothaWork | yeah, CE and NT have basically nothing to do with each other |
| 18:23.52 | chouimat|work | CosmicPenguin: it's why I said it look better :) |
| 18:23.54 | MothaWork | they share a name |
| 18:24.03 | MothaWork | like Outlook and Outlook Express |
| 18:24.08 | MothaWork | and IE for Windows vs. IE for Mac |
| 18:24.16 | Ian_Daniher | CosmicPenguin: I don't know whether you ever go tht update, but I got a forth script that works, now I just need to recompile the XO's kernel and I'm in business |
| 18:24.23 | Ian_Daniher | s/tht/the |
| 18:24.24 | CosmicPenguin | the core is completely different |
| 18:24.25 | ds2 | ce and nt are both abominations |
| 18:24.39 | CosmicPenguin | the CE core is rather smart - in some ways smarter then Linux in the same environment |
| 18:24.56 | MothaWork | ds2: well, here's what it comes down to. Company wants to sell hardware. Customers demand WinCE. Company finds that customers demanding CE support represent sales exceeding the cost of putting CE on the device. Ergo, the device will end up supporting CE. |
| 18:25.21 | chouimat|work | basic economy stuff |
| 18:26.08 | MothaWork | ds2: very rarely is a company willing to deprive itself of net income (or in some cases, even just plain old revenue) in the name of a philosophical statement |
| 18:26.27 | prpplague | MonMotha: exactly |
| 18:26.29 | ds2 | that's like saying roaches should be the gov't since they are numerous and we are a demoocracy |
| 18:26.45 | MothaWork | now, a problem many companies have is that they fail to include support costs in their analysis. Many CE houses will require more support than a Linux shop, but that is just another factor. |
| 18:27.00 | MothaWork | ds2: and if roaches were capable of rising up against the oppressive humans, you damn well bet they'd be in charge |
| 18:27.14 | prpplague | ds2: we don;'t have the luxory of being idealistic with sales |
| 18:27.15 | MothaWork | ds2: fortunately, roaches are relatively easily taken care of in the grand scheme of things |
| 18:27.28 | CosmicPenguin | Very shortly, there will be an excellent almost 1:1 comparsion between CE and Linux |
| 18:27.30 | MothaWork | ds2: complete lack of sales, however, is something that dooms a company |
| 18:27.44 | CosmicPenguin | when android comes out you can compare it to phones that run CE and draw your own conclusions |
| 18:27.58 | MothaWork | CosmicPenguin: I presume you are speaking of CE itself, not the entire "Windows Mobile" platform? |
| 18:28.02 | chouimat|work | android ewwwww |
| 18:28.15 | CosmicPenguin | in the cell phone market - its hard to compare |
| 18:28.18 | MothaWork | I'm told that, under the hood, CE isn't as terrible as the Windows Mobile environment makes it appear |
| 18:28.41 | CosmicPenguin | but I think even withteh whole windows mobile shell, stuff like java performance and power management still shines through |
| 18:28.45 | MothaWork | for example, there is the capability of actual process management. Windows Mobile just hides it too well. |
| 18:29.06 | prpplague | our problem boils down to the new sales people aren't understanding how to do business with OSS/Linux developers |
| 18:29.09 | CosmicPenguin | I have always been impressed with CE when I have been exposed to some of its features |
| 18:29.27 | MothaWork | Linux is doing better and better in the power management department, though it's still highly processor/platform dependent (seems to be in CE, at times, too). Now, Java acceleration is hit or miss it seems. |
| 18:29.27 | ds2 | android is just another userland like alp |
| 18:30.28 | ds2 | which java accel? |
| 18:31.44 | prpplague | MothaWork: well since we've been selling our devices with linux as far back as 2000, we had(past tense) a pretty large OSS/Linux customer base |
| 18:31.55 | MothaWork | prpplague: I see |
| 18:31.59 | chouimat|work | starts to enjoy tftpboot in uboot |
| 18:32.33 | MothaWork | what seems to have happened is that Microsoft has convinced all the chipmakers to provide CE BSPs, but they only provide minimal Linux support |
| 18:32.40 | prpplague | MonMotha: we've never had a wince device |
| 18:32.53 | MothaWork | that means that time to market (which is often THE driving factor in embedded) is often quicker with WinCE |
| 18:33.12 | prpplague | yea |
| 18:34.21 | chouimat|work | wow marvell has a freebsd 7 port for their kirkwood soc |
| 18:34.52 | MothaWork | that's kinda amazing |
| 18:38.27 | Ian_Daniher | chouimat|work: marvell doesn't... usually do anything like that, do they? |
| 18:39.48 | prpplague | MothaWork: yea the sales people apparent have never heard the phrase "source code" |
| 18:40.02 | MothaWork | Ian_Daniher: Marvell does a lot of things internally for customers that they do not publish |
| 18:40.29 | MothaWork | prpplague: well, and WinCE customers are often unaware of the flexibility that having all the source and the ability to modify it brings |
| 18:40.42 | MothaWork | you can get source for chunks of WinCE, but I believe parts of it are binary blobs |
| 18:40.44 | chouimat|work | <PROTECTED> |
| 18:40.58 | Ian_Daniher | chouimat|work: ah, gotcha :) |
| 18:41.07 | adye1 | prpplague: they are however, familiar with the phrase "first born", as in "it'll take your first born before we even acknowledge me make the chips you're talking about" |
| 18:41.23 | prpplague | adye1: ?? |
| 18:41.34 | adye1 | prpplague: Marvell :-( |
| 18:41.42 | chouimat|work | Ian_Daniher: but iirc they will eventually publish the freebsd port because they started to put it the perfarce repository |
| 18:42.18 | prpplague | adye1: hehe, ahh, no we are talking about AML's sales people |
| 18:42.23 | MothaWork | chouimat|work: "perfarce" sic I presume? |
| 18:43.00 | chouimat|work | MothaWork: ? it's just the name I give to perforce :) |
| 18:43.23 | prpplague | adye1: we got a new set of sales people, and they don't have the first clue about selling open source devices |
| 18:43.47 | MothaWork | chouimat|work: exactly |
| 18:44.01 | MothaWork | I was saying that I presume "perfarce" is an intentional "misspelling" |
| 18:45.33 | MothaWork | prpplague: have you ever actually used Jazelle under Linux? |
| 18:45.44 | MothaWork | I've never heard of anybody actually using it. I'm sure it's been done, but I've never heard of it. |
| 18:46.00 | prpplague | not i |
| 18:47.09 | chouimat|work | MothaWork: I read a lot about it (and also the java support of the avr32) but never tried it yet and since I hate java it's not on the top of my todo list |
| 18:47.38 | ds2 | i had it working before |
| 18:47.47 | MothaWork | chouimat|work: thing that came up here is that WxGTK is so slow, that we think we might have actually been able to use Java and have it run faster |
| 18:47.57 | chouimat|work | hehe |
| 18:48.11 | MothaWork | and of course, Java tends to be a bit easier to make large multi-developer GUI apps with than C++ |
| 18:48.29 | ds2 | problem is a lack of jvms supporting jazzelle |
| 18:48.39 | MothaWork | ds2: that's what I gathered from googling a bit |
| 18:49.09 | ds2 | afaik sun is the only one |
| 18:49.35 | prpplague | ds2: we strongly encourage our customers to stay away from java |
| 18:50.03 | MothaWork | hum, ARM926EJ-S has the DSP instruction set? |
| 18:50.27 | prpplague | MonMotha: yea, iirc |
| 18:50.31 | ds2 | that is a good idea |
| 18:50.32 | MothaWork | yeah, I just read that |
| 18:50.34 | prpplague | MonMotha: since that is what the davinci is |
| 18:50.35 | MothaWork | didn't realize it |
| 18:51.43 | MothaWork | and yet it still takes GTK several seconds to create a few buttons |
| 18:51.50 | MothaWork | (not that the DSP instruction would help any with that) |
| 18:55.44 | *** join/#edev s (n=flaviori@201.70.21.178) |
| 18:56.48 | ds2 | desktop folks are scary:( |
| 18:57.40 | chouimat|work | gtk is bad ... and wxgtk add a huge MFC like layer to it :( |
| 18:58.21 | MothaWork | chouimat|work: you won't get me denying that |
| 18:58.25 | MothaWork | I didn' t pick the toolkit |
| 18:58.36 | MothaWork | we originally wanted to use WxX11, but somehow we ended up with WxGTK |
| 18:58.58 | solar | that one takes forever to compile ;/ |
| 18:59.13 | chouimat|work | MothaWork: I had to work with for a project ... and went I went back to qt2.something I was happy |
| 18:59.24 | MothaWork | any time it's slow, you can look at the profiling and it's all in GTK doing lots of (unnecessary) locking while creating buttons |
| 18:59.35 | MothaWork | chouimat|work: well, we wanted to keep this a "no royalties" device |
| 18:59.55 | MothaWork | so that means LGPL, BSD, or similar licensed toolkits |
| 19:00.46 | MothaWork | though in hindsight it might have been cheaper to just use Qt and pay the royalties due to lessened development time |
| 19:00.55 | MothaWork | tough to tell, of course |
| 19:01.38 | prpplague | solar: what changes were you wanting to do for the bootloader? |
| 19:04.18 | solar | apex? none. |
| 19:05.27 | solar | prpplague: Does tslib/evtouch support the s3c2410 TS ? |
| 19:05.32 | prpplague | oh ok, i thought you had asked for the source for that and the openocd config |
| 19:05.36 | prpplague | solar: yep |
| 19:05.43 | *** part/#edev adye1 (n=adyer@adsl-69-223-13-78.dsl.chcgil.ameritech.net) |
| 19:05.44 | solar | prpplague: I did ask for it all. |
| 19:06.05 | solar | need that to be able to make firmware/jffs2/squashfs images etc. |
| 19:06.28 | CosmicPenguin | ubifs for the win! |
| 19:06.32 | CosmicPenguin | or so I am told |
| 19:06.35 | prpplague | solar: i'll provide the source, but i'm not sure why you need it for generating the i mages |
| 19:07.01 | solar | prpplague: so I'll be able to reflash the device? |
| 19:07.25 | prpplague | solar: you can do that through the uart and sd card |
| 19:12.11 | prpplague | solar: jtag will only be available during production |
| 19:22.16 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 19:22.30 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 19:23.01 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 19:28.55 | MothaWork | prpplague: is there any sort of API for OTP MTD devices? |
| 19:29.10 | MothaWork | like many NAND devices have a few OTP pages, and Atmel dataflash also has some OTP functions |
| 19:29.44 | prpplague | MothaWork: API for what aspects of OTP devices? |
| 19:30.06 | prpplague | MothaWork: normally the OTP area is just like the rest of the device, just can't be erased |
| 19:30.10 | MothaWork | prpplague: I just notice that none of the MTD drivers seem to support any OTP features, and there's no apparent good way to get it in there |
| 19:30.24 | MothaWork | prpplague: on NAND and Dataflash, they use special instructions to access those areas |
| 19:31.15 | prpplague | MothaWork: ahh, i was thinking more like OTP devices, not just areas |
| 19:31.23 | MothaWork | yeah |
| 19:31.32 | MothaWork | we're looking for a place to chuck a serial number |
| 19:32.24 | MothaWork | I do have an I2C ROM I can write protect in hardware, but then you can't program things entirely with software (have to program then break off the jumper) |
| 19:37.54 | solar | prpplague: eh? You saying the firmware image itself is going to live on the SD? |
| 19:38.22 | prpplague | solar: no, but you can upgrade the device using the sd card, |
| 19:38.40 | prpplague | solar: i.e. boot from to the sd card and flash the device image from there |
| 19:40.30 | solar | so you have the real bootloader already flashed into the device then? |
| 19:41.27 | prpplague | yea |
| 19:41.58 | prpplague | solar: the customer should never have to flash the bootloader |
| 19:42.59 | solar | prpplague: ok. Just not sure how you were sending this to me. I figured a blank flash by default. |
| 19:43.48 | prpplague | solar: yea, this will have the bootloader, a sample kernel and a sample rootfs |
| 19:44.02 | prpplague | solar: just trying to stabilize the hardware now |
| 19:48.31 | solar | oh that sure makes life easy then |
| 19:52.49 | *** join/#edev adye2 (n=adyer@adsl-69-223-13-78.dsl.chcgil.ameritech.net) |
| 19:54.52 | *** join/#edev like2wise (n=likewise@82-171-51-231.ip.telfort.nl) |
| 20:00.50 | *** join/#edev Ian_Daniher (n=it@204.10.219.119) |
| 20:09.40 | *** join/#edev CosmicPenguin (n=nobody@host-245-157.pubnet.pdx.edu) |
| 20:09.40 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 20:40.01 | MothaWork | prpplague: are you aware of any opensource media decoders that are capable of taking advantage of the ARM9E DSP instructions? I presume gcc isn't exactly great at emitting them. |
| 21:02.31 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 21:07.50 | roxfan | maybe this https://datatype.helixcommunity.org/2005/aacfixptdec |
| 21:09.02 | roxfan | hmm mp3 too it seems |
| 21:09.04 | MothaWork | that's just a fixed point decoder |
| 21:09.31 | MothaWork | there are special DSP oriented instructions on ARM9E that can combine several instructions into one in many cases |
| 21:10.14 | Madsy | What is the latest version of the ARM instruction set? Version 7? |
| 21:10.43 | Madsy | My hardcopy version of ARMARM only covers up to v5, so I guess it is pretty outdated |
| 21:10.55 | roxfan | https://helixcommunity.org/viewcvs/datatype/mp3/codec/fixpt/real/arm/ |
| 21:11.09 | roxfan | Madsy: they recently updated public spec to v6 |
| 21:11.15 | roxfan | v7 is still nda only |
| 21:11.24 | Madsy | Aha, thanks :-) |
| 21:11.32 | Madsy | I'm setting up my ARM forum atm |
| 21:12.05 | MothaWork | roxfan: doesn't look like it uses the DSP stuff, just has some optimized loops for ARM "in general" |
| 21:12.21 | roxfan | well, v7-A&R that is... v7-M is available |
| 21:12.41 | ds2 | s the v6 stuff downloadable now? |
| 21:13.36 | roxfan | yes |
| 21:15.01 | roxfan | http://www.arm.com/community/university/eulaarmarm.html |
| 21:16.08 | *** join/#edev dijenerate (n=dijenera@72.22.144.246) |
| 21:16.54 | roxfan | MothaWork: do you mean the saturating instructions? |
| 21:17.06 | MothaWork | roxfan: yeah, there are also some MAC instructions and such |
| 21:17.17 | roxfan | hmm |
| 21:17.42 | MothaWork | I don't have my quickref card handy |
| 21:17.56 | MothaWork | I don't actually deal with assembly very much |
| 21:18.09 | MothaWork | most of the time, I'm just writing drivers or doing hardware |
| 21:18.16 | MothaWork | not usually messing around in an environment where C won't work |
| 21:18.47 | MothaWork | going home |
| 21:20.32 | roxfan | funny, arm infocenter still claims that pdf is v5 |
| 21:20.35 | *** join/#edev flavioribeiro (n=flaviori@201.70.21.178) |
| 21:20.59 | *** join/#edev horizon (i=horizon@2.xcodes.de) |
| 21:21.21 | horizon | hi. |
| 21:28.13 | chouimat|work | going to the pub |
| 21:30.11 | ds2 | hmmm |
| 21:41.32 | *** part/#edev horizon (i=horizon@2.xcodes.de) |
| 23:19.26 | *** join/#edev landley (n=landley@m655e36d0.tmodns.net) |
| 23:19.26 | *** mode/#edev [+o landley] by ChanServ |
| 23:34.56 | *** join/#edev joe_bleau (n=joe_blea@adsl-69-151-118-19.dsl.fyvlar.swbell.net) |
| 23:49.29 | *** join/#edev ccjoe (n=joeg@75-169-8-56.slkc.qwest.net) |
