| 00:34.31 | joe_bleau | Microcontroller controlled LCD contrast...what are people using these days? |
| 00:34.56 | joe_bleau | I've used a dirt simple cap-diode inverter, followed by level shifted PWM before. |
| 00:35.05 | joe_bleau | High parts count, but cheap parts. |
| 00:50.17 | *** join/#edev linac (n=lin@168.123.61.59.broad.fz.fj.dynamic.163data.com.cn) |
| 01:01.05 | prpplague | joe_bleau: hmm, let me |
| 01:01.40 | prpplague | see |
| 01:02.32 | prpplague | joe_bleau: how about this |
| 01:02.34 | prpplague | Mouser Part #: 698-CAT32TDI-TE7 |
| 01:02.34 | prpplague | Manufacturer Part #: CAT32TDI-TE7 |
| 01:02.34 | prpplague | Manufacturer: Catalyst Semiconductor |
| 01:02.34 | prpplague | Description: LED Driver Integrated Circuit 2-7V Adj Curr 40mA |
| 01:06.16 | ds2 | or if you don't care about lead time, isn't there a maxim part? |
| 01:10.13 | prpplague | ~lart ds2 |
| 01:10.13 | purl | urinates on ds2 |
| 01:10.16 | prpplague | doh |
| 01:10.22 | prpplague | ds2: maxim is terrible |
| 01:10.30 | prpplague | ds2: you can;t count on them for anything |
| 01:12.50 | ds2 | free samples are great for one offs ;) |
| 01:16.32 | prpplague | ds2: indeed, but when you try to go to production its a terrible thing |
| 01:17.13 | ds2 | I have seen boards made with maxim watch dog parts that don't have too much of a sourcing problem so it isn't too horrible |
| 01:17.35 | ds2 | but their quantites was 50-100 an order |
| 01:19.07 | prpplague | ds2: yea we are talking 5 to 10 k per order |
| 01:20.31 | ds2 | always lineartech with a 10-20% more $$ version |
| 01:22.44 | prpplague | yea well, at 5 to 10k its not 20% more, its about the same in cost |
| 01:23.21 | ds2 | isn't all these problems a miracle purchasing has to preform? |
| 01:26.30 | prpplague | ds2: part of an engineers job is to choose parts that are economically viable |
| 01:27.15 | chouimat|cooking | prpplague: wow this is a nice lart ;) |
| 01:27.34 | prpplague | chouimat|cooking: indeed first time i've seen that one |
| 01:28.29 | chouimat|tv | prpplague: and I can tell you it work ... with a cat he stop peeing on everything after you do it once on him ;) |
| 01:28.42 | prpplague | hehe |
| 01:30.31 | ds2 | guess it is only the bigger places where purchasing forces the eng to use an approved list of parts |
| 01:31.32 | prpplague | ds2: we work hand and hand with purchasing to choose parts |
| 01:31.59 | ds2 | prpplague: do you guys have a dedicated CAD dept that maintains the component libraries, footprints, etc? |
| 01:32.03 | prpplague | ds2: i.e. "hello purchasing, i'm planning to use these parts.......how well can we purchase them?" |
| 01:32.30 | prpplague | ds2: not in the past, we just started about 6 months ago |
| 01:33.06 | ds2 | might be legacy infrastructure from his place |
| 01:33.40 | ds2 | hear him complaining about about being able to use exotic stuff (he's doing microwave analog stuff) |
| 01:34.59 | joe_bleau | I've used Maxim supervisor/battery switchovers without much trouble in 100-5k qty. |
| 01:35.00 | *** part/#edev pierrelux (n=pierre-l@144-125.sh.cgocable.ca) |
| 01:35.36 | joe_bleau | Now that they own Dallas, I guess you could say I've used their RTCs as well. |
| 01:36.10 | joe_bleau | Now this CAT32 appears to be an LED backlight controller more than an LCD contrast controller. |
| 01:36.30 | ds2 | speaking of RTCs, has anyone bought in on the TI marketing and used a MSP430 as a RTC? |
| 01:37.14 | joe_bleau | What's the advantage of the MSP430 in that application? |
| 01:37.28 | ds2 | customizable |
| 01:37.46 | ds2 | a "generic" part that uses as little power as a dedicated RTC chip |
| 01:38.13 | joe_bleau | If it's really low power, that is impressive. |
| 01:38.37 | ds2 | that's the claim by TI both at their talk and I think in their literature |
| 01:38.41 | joe_bleau | My current design really likes the built in battery switchover that some RTC chips have. |
| 01:39.55 | joe_bleau | Less than 1uA? |
| 01:40.17 | ds2 | that sounds about right, IIRC |
| 01:40.37 | joe_bleau | 500nA? Wow, that's not bad. |
| 01:40.52 | joe_bleau | I looked at some Seiko RTC chips the other night, but I don't remember what they pull. |
| 01:40.57 | ds2 | some very neat things powered off a CR2032 |
| 01:41.03 | joe_bleau | But that's not a fair comparison. |
| 01:41.22 | ds2 | if they are accurate, the MSP430 could make a simple LED watch |
| 01:42.02 | joe_bleau | That would really take the semi industry full circle, going back to LED watches! |
| 01:42.31 | joe_bleau | (Seiko claims 0.25uA for their super RTC, so TI claim is very impressive for a microcontroller.) |
| 01:43.28 | ds2 | I suspect the RTCs can run at a lower voltage then the TI; last I heard they can't do 2AA's but they can do a single Li cell |
| 01:44.08 | joe_bleau | Not surprising. I'm guessing RTCs are built on a more specialized process. |
| 01:44.38 | ds2 | not only would LED watches be a full circle, it would be an impressive leap; was reading about they doing it with all sorts of very advance technologies for their time now we are doing it w/hobbiest level tech. |
| 01:44.47 | joe_bleau | DS1307: 300nA typ, 500nA max (oscillator on). Runs down to 2.0V. |
| 01:45.28 | MonMotha | I really wish Atmel would get their picopower parts characterized |
| 01:45.39 | MonMotha | 1.8V, ~600nA supposedly, but the datasheet still says TBD |
| 01:46.33 | MonMotha | not quite to the level of the MSP430, but close |
| 01:46.39 | MonMotha | and slightly lower operating voltage, IIRC |
| 01:46.47 | ds2 | avr? |
| 01:47.05 | MonMotha | yeah |
| 01:47.26 | MonMotha | that's power down, RTC running |
| 01:47.38 | ds2 | that'd almost be right for recyclable computing --- powering systems off urine ;) |
| 01:47.39 | MonMotha | so you wake up on the RTC interrupt if you need to do anything (output a pulse, for example) |
| 01:48.37 | prpplague | ds2: hmm, isn't that what m$ does with their software? |
| 01:48.50 | MonMotha | they claim 500nA in power save with the RTC running for one part |
| 01:48.57 | MonMotha | at 1.8V, 600nA at 3V |
| 01:48.59 | ds2 | prplague: think they make it out of it, not power it ;) |
| 01:49.45 | MonMotha | active current is higher than MSP430 IIRC, about 500uA at 1MHz active |
| 01:50.08 | ds2 | any internal calibrated RC OSC? |
| 01:50.49 | MonMotha | yeah |
| 01:50.53 | MonMotha | but not for the 32kHz operation |
| 01:51.09 | MonMotha | which seems kinda dumb to me, but I'm guessing the RC osc is higher power than the 32kHz xtal one |
| 01:51.29 | ds2 | so one has to trade off pins available vs power usage?! |
| 01:51.51 | MonMotha | these aren't small devices, so pins available shouldn't be as big of a deal as the extra parts on the BOM |
| 01:52.04 | MonMotha | they ahve these in 40+ pin devices |
| 01:52.33 | MonMotha | but I agree, I'd like to see an internal RTC oscillator |
| 01:52.42 | ds2 | oh didn't know they made AVRs with that many pins |
| 01:52.49 | MonMotha | oh yeah |
| 01:52.52 | MonMotha | they have some HUGE AVRs |
| 01:52.56 | MonMotha | 60+ IO |
| 01:53.27 | ds2 | but the interesting ones, IMO, are those in 24 pin or smaller packages |
| 01:53.33 | MonMotha | yeah |
| 01:53.35 | ds2 | preferably in 8pin or smaller ;) |
| 01:53.37 | MonMotha | above that size, just use an ARM |
| 01:53.46 | ds2 | yep |
| 01:53.53 | MonMotha | the tinyX4 series is neat |
| 01:54.15 | MonMotha | er X5 |
| 01:54.25 | ds2 | is that the replacement for the tiny26? think that's the 8 pin one |
| 01:55.05 | MonMotha | 26 is a 20 pin package |
| 01:55.34 | MonMotha | I'm not sure why the tiny11 is no longer recommended for new designs |
| 01:55.38 | MonMotha | there's really no equivalent |
| 01:55.42 | MonMotha | that thing was CHEAP |
| 01:55.48 | ds2 | tiny13 |
| 01:55.53 | prpplague | i don't believe 11 is rohs |
| 01:56.07 | ds2 | %@#$#$@#!!@#!@@@ rohs... doing their best to make things less reliable |
| 01:56.25 | MonMotha | tiny13 isn't a comparable replacement, though it is about the closest they have and it could do anything an 11 could do |
| 01:56.30 | MonMotha | ah yes, rohs |
| 01:56.37 | MonMotha | the 11 is supposedly lead free, but may not be fully rohs |
| 01:56.50 | MonMotha | and yes, I fucking HATE lead free solder |
| 01:57.46 | MonMotha | there are several reasons lead was in solder |
| 01:57.50 | MonMotha | most of which are still unresolved |
| 01:58.00 | MonMotha | I expect a sudden rash of failures a few years from now due to tin whiskers |
| 01:58.04 | ds2 | lead is not that dangerous |
| 01:58.10 | MonMotha | no, it isn't |
| 01:58.17 | ds2 | but failures are 'good' it helps grow the economy ;) |
| 01:58.28 | MonMotha | the problem is, I can't get parts with leaded finishes anymore |
| 01:58.40 | MonMotha | which usually isn't a problem, but for BGAs, that pretty much means the whole board has to be a lead free process |
| 01:58.57 | ds2 | cross comtamination causes problems? |
| 01:59.26 | MonMotha | supposedly with such small interfaces, the two solders don't get along |
| 02:00.05 | MonMotha | certainly with larger stuff, QFPs and bigger, I've had plenty of luck at least manually soldering them down to lead free finished boards using lead free parts but leaded solder |
| 02:00.31 | ds2 | sounds like what needs to happen is someone design and build a cheap reballing machine |
| 02:01.57 | MonMotha | or we just need to stop using crappy solder |
| 02:02.17 | MonMotha | and what the fuck is causing weird keyboard repeat under heavy system load... |
| 02:02.19 | MonMotha | that's kernel level stuff |
| 02:02.30 | ds2 | but if you remove BGAs, reballing is needed so it'll solve multiple problems at once |
| 02:03.41 | MonMotha | I vote we just get rid of BGAs altogether :) |
| 02:03.51 | MonMotha | there's got to be a better way |
| 02:04.06 | joe_bleau | Occam process? |
| 02:05.05 | ds2 | yes, I'll go for that |
| 02:05.39 | ds2 | everyone should use bare dies instead |
| 02:08.23 | prpplague | ds2: actually we are looking into using flip chip die |
| 02:08.50 | prpplague | ds2: $250,000 machine investment |
| 02:08.54 | ds2 | prpplague: Ouch |
| 02:09.01 | ds2 | youmust be making a lot of them ;) |
| 02:09.26 | ds2 | do they actually do a bond wire on your board or??? |
| 02:09.30 | prpplague | well it will cut the cost of flash and sdram by a significant amount |
| 02:09.50 | prpplague | some parts with drop from $6.80 to $0.60 |
| 02:10.20 | ds2 | still, that seems like a long/very high volume payback |
| 02:11.43 | ds2 | Mallet - dip 8 sized board w/a MMU ARM + 16M Flash + 128M SDRAM ;) |
| 02:12.01 | joe_bleau | Is it batter to buy the machine and do it in-house, or have someone with the machine mount KGD for you? |
| 02:12.11 | prpplague | ds2: where? |
| 02:12.20 | prpplague | ds2: mallet? |
| 02:12.28 | joe_bleau | He's giving you a product idea |
| 02:12.30 | ds2 | prpplague: just kidding... figure that's roughly where you are headed ;) |
| 02:12.33 | prpplague | joe_bleau: still researching |
| 02:12.40 | ds2 | mallet == next gen of hammer :) |
| 02:12.42 | prpplague | ds2: ahh |
| 02:12.47 | joe_bleau | tack hammer? |
| 02:13.00 | prpplague | dip 8? that seems almost a waste |
| 02:13.07 | ds2 | dip8 sized, not 8 pin |
| 02:13.23 | ds2 | kind of like that RJ45 setup but lower profile |
| 02:13.45 | joe_bleau | I like your arm+RJ45 idea |
| 02:13.52 | ds2 | picolinux or something like that; runs linux in a RJ45 form factor |
| 02:14.07 | joe_bleau | x86 though, isn't it? |
| 02:14.10 | ds2 | joe_bleau: you can buy it |
| 02:14.19 | ds2 | no, ARM but runs MMUless linux |
| 02:14.27 | joe_bleau | oh, ok. |
| 02:14.35 | ds2 | it does have a ethernet though ;) |
| 02:14.59 | joe_bleau | I'll bet sledgehammer has 10/100 ethernet. |
| 02:15.06 | ds2 | http://www.linuxdevices.com/news/NS8386088053.html |
| 02:15.20 | ds2 | near the top off google even though I hate linuxdevices as a site ;) |
| 02:16.01 | joe_bleau | Ahh, it's based on those cute little Digi parts. |
| 02:16.46 | joe_bleau | I think I had the Lantronix Xport in mind. |
| 02:17.53 | ds2 | profile is too high ;) |
| 02:27.57 | *** join/#edev FireEgl (i=FireEgl@Sebastian.FireEgl.Com) |
| 02:29.39 | prpplague | joe_bleau: digi sucks |
| 02:29.55 | prpplague | joe_bleau: sledge hammer doesn't include any nick |
| 02:30.37 | joe_bleau | I like digi. They don't suck. |
| 02:30.47 | joe_bleau | But you may feel free to question my wisdom. |
| 02:30.56 | joe_bleau | After all, I run (and develop) for windows, too. |
| 02:31.45 | joe_bleau | How about a hockey puck. USB ports on the top, bottom, and 4 or so around the circumfrence. |
| 02:31.47 | prpplague | joe_bleau: yea try doing business with them |
| 02:32.09 | prpplague | ds2: dip8 but not 8 pin? i don't understand |
| 02:32.22 | joe_bleau | Don't need to, we just buy and use their stuff. |
| 02:32.42 | joe_bleau | Anyway, make one of the top or bottom ports a device, the others host/hub. |
| 02:32.58 | joe_bleau | They you wouldn't need any I/O at all. Just plug in USB devices to fit your needs. |
| 02:33.13 | joe_bleau | And you could stack the pucks up, too. (That's where the device port comes in.) |
| 02:33.33 | joe_bleau | Of course, it would boot from USB mass storage. |
| 02:33.38 | joe_bleau | Ethernet? Plug it in. |
| 02:33.48 | joe_bleau | Bluetooth? 802.11? Plug it. |
| 02:34.08 | joe_bleau | Maybe put a hole or two clear through for mouting. |
| 02:34.26 | joe_bleau | Arrange it so that you could couple several units and still screw 'em together in a stack. |
| 02:35.00 | prpplague | interesting idea |
| 02:35.18 | joe_bleau | Damn right. Think Digi would be interested in it? ;-) |
| 02:35.19 | prpplague | i can pretty much already do that with the hammer though |
| 02:35.35 | joe_bleau | Oh, encapsulate it too, so that it would be as tough as a hockey puck. |
| 02:35.45 | prpplague | joe_bleau: hehe, offer it to digi |
| 02:35.55 | joe_bleau | PuckARM? PuckPC? |
| 02:35.58 | joe_bleau | PuckIt? |
| 02:36.06 | prpplague | joe_bleau: puckit sounds interesting |
| 02:36.08 | joe_bleau | Jeez, this is too good. |
| 02:36.31 | joe_bleau | I know a guy at Digi, sort of. |
| 02:37.33 | prpplague | i had a working port of uclinux to their netsilicon arm7tdmi back in 2003 |
| 02:38.10 | prpplague | i offered it to them and them told me to go f myself that they weren;'t interesting in OSS |
| 02:38.46 | joe_bleau | Sorry. I don't think they really 'get' open source in general. |
| 02:41.07 | prpplague | joe_bleau: yea well the kicker was that they came by our offices about 6 months later trying to sell us netsilicon processors, using my uclinux distro |
| 03:09.16 | joe_bleau | So, they did get open source in the end, huh? |
| 03:09.51 | joe_bleau | I'll give you a week or so on PuckIt to decide before I mention it to the guy I sorta know. |
| 03:09.58 | joe_bleau | I'd love to see y'all do something with it. |
| 03:10.32 | joe_bleau | You could do it in clear potting compound, so that internal LEDs would be visible. That would be pretty. |
| 03:18.38 | ds2 | prpplague: same foot print as the plastic part of a DIP8 |
| 03:18.54 | ds2 | maybe a tiny QFP or something like that |
| 03:29.59 | prpplague | joe_bleau: too busy with other projects |
| 03:30.04 | *** join/#edev nighty (n=nighty@sushi.rural-networks.com) |
| 03:33.33 | joe_bleau | Yeah, I know that feeling. |
| 03:42.39 | *** join/#edev DrZimmerman (n=theo@77-56-240-9.dclient.hispeed.ch) |
| 04:38.53 | *** part/#edev joe_bleau (n=joe_blea@adsl-69-152-206-249.dsl.fyvlar.swbell.net) |
| 04:44.41 | *** join/#edev Off_Namuh72 (i=GPS@gateway/tor/x-8f09f4c7e6776dc9) |
| 06:26.46 | *** part/#edev nighty (n=nighty@sushi.rural-networks.com) |
| 07:35.39 | *** join/#edev howtoo (n=howtoo@195.159.62.66) |
| 08:25.19 | *** join/#edev Fire_Egl (i=FireEgl@adsl-220-213-203.bhm.bellsouth.net) |
| 08:26.33 | *** join/#edev dijenerate (n=dijenera@72.22.142.113) |
| 08:35.14 | *** join/#edev lyakh (n=lyakh@dynamic-unidsl-85-197-26-24.westend.de) |
| 08:43.49 | *** join/#edev pina (n=raf@unaffiliated/pina) |
| 09:26.53 | *** join/#edev FireEgl (i=FireEgl@Sebastian.FireEgl.Com) |
| 10:53.56 | *** join/#edev howtoo (n=howtoo@195.159.62.66) |
| 11:48.38 | *** join/#edev Off_Namuh72 (i=GPS@gateway/tor/x-3e614242a14600d6) |
| 12:17.50 | *** join/#edev Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) |
| 12:45.42 | *** join/#edev hw (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 13:19.07 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 13:23.02 | prpplague | ho ho hum |
| 13:29.59 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 13:29.59 | *** mode/#edev [+o GPSFan] by ChanServ |
| 13:38.12 | *** join/#edev dijenerate (n=dijenera@72.51.73.169) |
| 14:00.58 | chouimat | morning |
| 14:06.10 | samuelr | prpplague: ping |
| 14:06.19 | prpplague | samuelr: pong |
| 14:06.24 | prpplague | samuelr: whats cookin? |
| 14:06.55 | samuelr | Trying to dig myself out of a hole from the past couple of weeks :-) |
| 14:07.39 | samuelr | I know I've been out of touch, wanted to let you know I haven't forgotten about doing hammer stuff |
| 14:08.14 | prpplague | samuelr: np, been swampped myself with a new board bringup |
| 14:09.20 | samuelr | Heh. At least you've been occupied with fun stuff :-/ |
| 14:10.07 | samuelr | My 3 yr old has had mono. It's amazing how much time a sick little rufflebutt can take up. |
| 14:11.27 | prpplague | samuelr: ouch, sorry to hear that, i know i have two critters, not only time but money as well |
| 14:19.03 | wmat | prpplague: fwiw, we're likely going to move elinux.org to another host soon |
| 14:24.17 | prpplague | wmat: ok |
| 14:24.22 | prpplague | wmat: more problems? |
| 14:25.13 | wmat | prpplague: no, the same. Movial is a busy company, therefore it doesn't make sense to host with them as their trn around time for requests is too slow. |
| 14:25.30 | prpplague | wmat: ahh |
| 14:34.00 | *** join/#edev Crofton (n=balister@hc6521628.dhcp.vt.edu) |
| 14:50.08 | *** join/#edev CosmicPe1guin (n=nobody@163.181.251.103) |
| 14:50.53 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 14:51.05 | *** join/#edev CP|too (n=nobody@www.crouse-house.com) |
| 14:51.05 | *** mode/#edev [+o CP|too] by ChanServ |
| 14:55.11 | *** join/#edev chouimat (n=dieu@kde/developer/chouinard) |
| 15:34.15 | prpplague | CosmicPenguin: greetings |
| 15:34.20 | prpplague | CosmicPenguin: hows life at amd? |
| 15:34.28 | prpplague | CosmicPenguin: still going 100% on olpc? |
| 15:38.52 | CosmicPenguin | not really |
| 15:38.55 | CosmicPenguin | about 10% |
| 15:39.06 | prpplague | CosmicPenguin: oh? anything interesting you can talk about? |
| 15:39.19 | CosmicPenguin | Nothing interesting |
| 15:40.20 | prpplague | oh |
| 15:40.31 | prpplague | CosmicPenguin: attending any conferences soon? |
| 15:40.38 | CosmicPenguin | I doubt it |
| 15:41.11 | CosmicPenguin | LinuxBIOS meetings are in Denver in April |
| 15:41.21 | CosmicPenguin | and I'll probably try to do something for CELF |
| 15:41.31 | prpplague | CosmicPenguin: is it just me or does freenode and oftc seem dead lately? |
| 15:41.35 | prpplague | CosmicPenguin: cool |
| 15:41.42 | wmat | prpplague: dead how? |
| 15:41.45 | CosmicPenguin | I seem to stay pretty bush |
| 15:41.47 | CosmicPenguin | busy even |
| 15:41.57 | prpplague | CosmicPenguin: i'd really like to attend CELF next year |
| 15:42.05 | CosmicPenguin | you should do it |
| 15:42.08 | CosmicPenguin | I |
| 15:42.15 | prpplague | wmat: lack of active conversations about development |
| 15:42.21 | CosmicPenguin | I'm tossing around some ideas for making a toolkit for easily building LinuxBIOS payloads |
| 15:42.33 | CosmicPenguin | CELF would be a great place to do a tutorial for that |
| 15:42.35 | wmat | CosmicPenguin: what are you going to do for CELF? NOTE: I can get you paid for publishable papers for elinux.org |
| 15:42.40 | CosmicPenguin | but I'm not sure if it will end up getting together |
| 15:42.41 | prpplague | CosmicPenguin: pretty depressed about current dev work |
| 15:42.56 | CosmicPenguin | wmat: sounds good, but AMD won't let me do that |
| 15:43.12 | wmat | CosmicPenguin: can you accept prizes? |
| 15:43.22 | CosmicPenguin | I can't even accept a handshake |
| 15:43.47 | prpplague | hehe |
| 15:44.02 | wmat | CosmicPenguin: oh well, you'll have to do it for free then ;) |
| 15:45.15 | prpplague | wmat: the arm kernel people have all but disappeared from irc |
| 15:46.23 | prpplague | wmat: timriker , russdill, sjhill, andersee, pb , mallum |
| 15:46.35 | prpplague | wmat: hardly see any of those guys on irc anymore |
| 15:46.57 | prpplague | wmat: i think CosmicPenguin and i probably the last two from the original tuxscreen channel that hangs around |
| 15:47.09 | CosmicPenguin | thats because we haven't hit the big time yet like the other guys |
| 15:48.31 | wmat | heh |
| 15:48.56 | wmat | prpplague: Tim R. is in #elinux and andersee is stupidly busy, I here |
| 15:49.03 | wmat | not sure about the other guys |
| 15:50.28 | prpplague | wmat: yea, timr is still on irc but he never says much |
| 15:50.54 | CosmicPenguin | Tim lost having much to say after he left Lineo |
| 15:51.25 | prpplague | CosmicPenguin: actually i think it ended more when the TI stuff fell apart |
| 15:51.45 | CosmicPenguin | true |
| 15:52.13 | prpplague | CosmicPenguin: i reall think the TI think broke tim's spirit |
| 15:52.31 | CosmicPenguin | Understandably |
| 15:52.40 | CosmicPenguin | He put himself in a tough position to go there and lead that team |
| 15:54.25 | prpplague | CosmicPenguin: indeed |
| 15:55.10 | prpplague | CosmicPenguin: my problem is that it is getting increasingly frustrating not having any developers to talk to |
| 15:55.21 | prpplague | CosmicPenguin: no #arm-linux channel |
| 15:55.34 | prpplague | CosmicPenguin: i hate #kernelnewbies over on oftc |
| 15:55.52 | prpplague | CosmicPenguin: its basically worthless these days |
| 15:56.04 | prpplague | CosmicPenguin: handhelds.org channel is nothing but crap |
| 15:57.41 | prpplague | CosmicPenguin: all the serious devrs i know are doing olpc related stuff, so its all x86 |
| 15:57.55 | CosmicPenguin | The way it should be |
| 15:58.29 | prpplague | CosmicPenguin: and all the arm devrs i know are either working for MV or Timesys and can discuss hardly anything |
| 15:58.40 | prpplague | s/can/can't |
| 16:05.34 | *** join/#edev Off_Namuh72 (i=GPS@gateway/tor/x-28df6b8a4e84bce2) |
| 16:10.00 | Crofton | prpplague, you need to hang out with the #oe guys |
| 16:11.09 | prpplague | Crofton: the #oe guys always seemed more userland oriented |
| 16:11.25 | CosmicPenguin | tell that to RP |
| 16:11.35 | Crofton | true, but we do deal with kernel issues |
| 16:11.43 | Crofton | there are guys that know kernel stuff |
| 16:12.28 | prpplague | Crofton: yea, i'm not saying there probably aren't kernel people there, just saying that it seemed that discussing of kernel stuff on #oe was frowned upon |
| 16:12.48 | Crofton | well, go ask and move them over here :) |
| 16:13.01 | Crofton | just phrase it as an oe question :) |
| 16:17.15 | prpplague | Crofton: yea, i'm looking for more stuff on bootloader and initial kernel stuff, i.e. board bringups |
| 16:17.24 | prpplague | Crofton: kinda hard to fit that in #oe |
| 16:17.57 | Crofton | true |
| 16:18.11 | Crofton | comes up some, but not very often |
| 16:18.38 | Crofton | there is a guy russf who is working on a kb9202 problem at the moment |
| 16:18.45 | Crofton | but he is not a guru |
| 16:19.04 | Crofton | I'll send him over here though, you may be able to help him |
| 16:24.50 | prpplague | Crofton: heading to lunch bbiab |
| 16:28.52 | *** join/#edev ico2 (n=ico2@modem-2001.chimpanzee.dialup.pol.co.uk) |
| 16:32.47 | *** part/#edev samuelr (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 16:33.02 | *** join/#edev emaithy (n=eolhyte@41.251.66.138) |
| 16:33.09 | *** part/#edev emaithy (n=eolhyte@41.251.66.138) |
| 16:36.13 | Crofton | arg |
| 16:36.14 | Crofton | so close |
| 16:40.50 | *** join/#edev samuelr (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 16:45.15 | *** join/#edev ico2_ (n=ico2@modem-1820.cheetah.dialup.pol.co.uk) |
| 16:53.16 | *** join/#edev linac (n=lin@121.41.53.117) |
| 17:00.33 | prpplague | Crofton: back |
| 17:00.42 | Crofton | hi |
| 17:00.50 | Crofton | the guy with issues isn't around to day |
| 17:00.58 | Crofton | but I just got the SFFSDR to boot |
| 17:01.52 | Crofton | at least to the point where it tried initing the non-existent IDE controller :) |
| 17:01.52 | prpplague | fun fun |
| 17:01.52 | Crofton | yeah |
| 17:07.55 | prpplague | CosmicPenguin: hmm, running into a weird problem, trying to unload a module, hangs on unregister_netdev(); |
| 17:08.38 | CosmicPenguin | waiting for a spinlock? |
| 17:10.13 | prpplague | CosmicPenguin: doesn't appear so |
| 17:12.42 | prpplague | CosmicPenguin: ahh found it, the net_device var was getting set to null before the unregister |
| 17:36.22 | CosmicPenguin | thats bad |
| 17:38.18 | prpplague | CosmicPenguin: yea, still seems to be some issues though |
| 18:12.42 | prpplague | CosmicPenguin: hmm, still having problems with unregister_netdev() |
| 18:14.07 | *** join/#edev T0mW (n=Tom@24.238.86.158.res-cmts.sth.ptd.net) |
| 18:14.21 | T0mW | ~seen jhaluska |
| 18:14.26 | purl | T0mW: i haven't seen 'jhaluska' |
| 18:14.31 | prpplague | T0mW: hey |
| 18:14.37 | T0mW | prpplague: ho |
| 18:14.39 | prpplague | T0mW: whats the latest on your hammer lcd board? |
| 18:15.09 | T0mW | prpplague: I have to get some new PCs out to the stores this week, so I have been tied up. |
| 18:15.35 | T0mW | prpplague: parts + PCB + stencil are all here, just need about 8 hours to build it. |
| 18:15.41 | prpplague | T0mW: ahh ok |
| 18:16.33 | ds2 | wow someone is actually responding to requests hehe |
| 18:19.06 | chouimat | ds2: it's called: experience ;) |
| 18:19.32 | ds2 | I mean for the hammer board |
| 18:19.40 | ds2 | thought everyone ditched prpplague ;) |
| 18:20.09 | prpplague | ds2: no, we just have two slackers jbevren and Exiles |
| 18:20.15 | T0mW | heh |
| 18:20.17 | prpplague | ds2: and Exiles is on my shit list |
| 18:20.35 | chouimat | ds2: I'm not sure you really want to do this ... 1) he have your address and 2) he has guns :) |
| 18:20.47 | prpplague | hehe |
| 18:21.10 | ds2 | =) |
| 18:37.36 | *** join/#edev Off_Namuh72 (i=GPS@gateway/tor/x-16bc2e208d54b1e2) |
| 18:38.55 | *** join/#edev Crofton|work (n=balister@ossie.wireless.vt.edu) |
| 18:41.15 | *** join/#edev redblue (i=star@ppp214.108-253-207.mtl.mt.videotron.ca) |
| 18:41.16 | *** join/#edev ico2__ (n=ico2@modem-809.cougar.dialup.pol.co.uk) |
| 19:14.15 | prpplague | CosmicPenguin: weird, the refcnt seems to be causing a problem |
| 19:14.37 | CosmicPenguin | not going to zero? You double initalizing for some reason? |
| 19:14.51 | CosmicPenguin | rebooting between tests? |
| 19:17.04 | prpplague | CosmicPenguin: rebooting between tests |
| 19:17.22 | prpplague | CosmicPenguin: don't think i'm doing a double register_netdev |
| 19:17.29 | CosmicPenguin | somebody is |
| 19:17.37 | CosmicPenguin | put a watch on that variable |
| 19:21.01 | prpplague | CosmicPenguin: doing that now |
| 19:22.44 | prpplague | grumbles even |
| 19:24.04 | prpplague | CosmicPenguin: uh this is weird |
| 19:24.14 | prpplague | CosmicPenguin: added a printk and now it seems to be working |
| 19:25.07 | prpplague | must be a timming issue |
| 19:27.14 | CosmicPenguin | race! |
| 19:29.41 | prpplague | ok this is totally weird |
| 19:29.49 | prpplague | the problem seems to have vanished :(\ |
| 19:32.19 | *** join/#edev dfarnsworth (n=dfarnswo@router.zeniff.com) |
| 19:36.50 | prpplague | CosmicPenguin: i moved the unregister_netdev to the top of the remove function and it seems to be work, apparently something is getting hose in the cleanup |
| 19:38.37 | prpplague | GPSFan: got a new version of the enc28j60 driver using the spi subsystem |
| 19:38.44 | prpplague | GPSFan: haven't gotten it tested yet |
| 19:38.53 | prpplague | GPSFan: based on my original code |
| 19:53.33 | Crofton | looks like the default DV kernel tries to mount root over nfs |
| 19:53.54 | Crofton | I do not see any sign of "partitions" in NAND I could use as root |
| 19:54.41 | Crofton|work | physmap platform flash device: 01000000 at 02000000 |
| 19:54.41 | Crofton|work | physmap-flash physmap-flash.0: map_probe failed |
| 19:55.42 | Crofton | is this something I nee to setup in the kernel source? |
| 19:59.46 | *** join/#edev chouimat (n=dieu@kde/developer/chouinard) |
| 20:44.42 | GPSFan | prpplague: sounds good, can I get a copy? when you are ready. |
| 20:48.15 | prpplague | GPSFan: i'm sending you want i have, its for 2.6.19 and its not tested on the 24xx yet |
| 20:49.26 | GPSFan | prpplague: ok, I'll take a look at the code and see if I see anything gross. |
| 21:08.03 | *** join/#edev ico2__ (n=ico2@modem-1235.cheetah.dialup.pol.co.uk) |
| 21:11.27 | ds2 | Hmmmm 400MHz omap on the latest nokia |
| 21:15.06 | prpplague | omap seems to be picking up speed in the high end device market, but the small qty manufacturers are still headed towards samsung |
| 21:15.56 | ds2 | don't marvell hold the highest speed device ATM? |
| 21:17.01 | ds2 | those 600MHz PXAs |
| 21:17.54 | prpplague | ds2: highest clock speed yea, not nessarily highest performance |
| 21:18.13 | prpplague | ds2: s3c2440 at 400mhz does about the same performance as an xscale at 600mhz |
| 21:18.20 | ds2 | is the a difference? they aren't they all armv6 class devices? |
| 21:18.29 | ds2 | Oh blah that is irritating |
| 21:19.27 | prpplague | pxa is xscale , s3c24xx is arm9 |
| 21:19.50 | ds2 | xscale is effectively armv6, isn't it? |
| 21:20.08 | ds2 | or did dec screw with it a lot more before tossing to intel? |
| 21:22.14 | prpplague | ds2: xscale is armv5te but it has a different pipe structure than an arm9tdmi |
| 21:22.36 | ds2 | Oh |
| 21:23.23 | ds2 | am I correct in assuming a 600MHz pxa would pull more power then a 400MHz samsung? |
| 21:24.27 | prpplague | ds2: yea |
| 21:25.02 | prpplague | ds2: the xscale has a 7 stage pipeline whereas the arm9 is 5 stage |
| 21:25.51 | ds2 | this raises the stupid question - why are there so much xscale penetration |
| 21:27.03 | prpplague | ds2: because strongarm and xscale have always been easy to get and easy to intergrate |
| 21:27.28 | prpplague | ds2: problem is with intel selling xscale to marvell we have a new situation |
| 21:27.43 | prpplague | ds2: xscale is no longer easy to get and no longer easy to intergrate |
| 21:27.58 | prpplague | ds2: i.e. large mins, NDA's and other restrictions from marvell |
| 21:28.09 | ds2 | that's one thing I do not understand; how are people getting xscales back when intel had them? I couldn't get intel to even respond with the name of a distributor |
| 21:28.53 | prpplague | ds2: low end marketers were able to move them in small qty , i.e. 5k at a time |
| 21:29.13 | prpplague | ds2: we are still able to purchase sa-1110's in 5k qty |
| 21:29.51 | ds2 | must be a secret handshake thing cuz they chips didn't seem available despite everyone's claims |
| 21:29.57 | ds2 | docs were available |
| 21:31.44 | prpplague | ds2: we purchase from digikey |
| 21:31.53 | prpplague | ds2: for low volume xscale |
| 21:31.57 | prpplague | ds2: i.e. 4 or 5 |
| 21:32.03 | ds2 | whoa digikey has them now?! |
| 21:32.16 | prpplague | ds2: they don't stock them ,but they can get them |
| 21:32.36 | ds2 | how's that work in general? a lot of interesting stuff come up as nonstock at digikey |
| 21:34.08 | prpplague | ds2: they generall just have it shipped directly to you from a vendor wharehouse |
| 21:42.39 | *** join/#edev dfarnsworth (n=dfarnswo@65.200.49.156) |
| 21:44.45 | *** join/#edev Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) |
| 21:49.24 | ds2 | prpplague: so it'll be just a matter of place an order on the phone? |
| 21:50.13 | prpplague | ds2: or on your online account |
| 21:50.21 | prpplague | ds2: really why would you want to though? |
| 21:50.56 | prpplague | ds2: there are better choices, plus at the small volume, its easier and cheap to get a cpu module of some type that include all the hard to intergrate components like sdram and flash |
| 21:51.30 | ds2 | prpplague: no, that was a general digikey question... there are a lot of interesting parts listed as nonstock; wasn't sure if that means I need to buy 1239012231492349328 peices before I can get them |
| 21:52.08 | ds2 | prpplague: for CPUs, I'd rather start with something a little higher up like a pinless hammer (hint, hint!) or a gumstix |
| 21:52.37 | prpplague | pinless hammer? |
| 21:53.53 | ds2 | yeah, a hammer boards w/o the pins on it |
| 21:54.04 | ds2 | the DIP40 emulation pins |
| 21:54.26 | prpplague | ds2: why? |
| 21:54.53 | ds2 | so I can solder it onto a board w/minimal profile |
| 21:55.49 | prpplague | ds2: i don't see how you are going to get any better without the pins |
| 21:56.02 | prpplague | ds2: after all it is a two sided board |
| 22:05.52 | *** join/#edev chouimat (n=dieu@kde/developer/chouinard) |
| 22:31.52 | *** join/#edev CosmicPe1guin (n=nobody@163.181.251.103) |
| 22:32.49 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 22:37.22 | *** join/#edev redblue (i=star@ppp039.110-253-207.mtl.mt.videotron.ca) |
| 23:52.44 | *** join/#edev pAth0s (i=pathos@74-139-180-1.dhcp.insightbb.com) |