| 00:37.54 | *** join/#edev FireEgl (i=FireEgl@FireEgl.CJB.Net) |
| 00:54.10 | *** join/#edev linac (n=lin@121.41.52.23) |
| 02:17.30 | *** join/#edev joe_bleau (n=joe_blea@adsl-208-191-55-217.dsl.fyvlar.swbell.net) |
| 02:19.22 | *** part/#edev pierrelux (n=pierre-l@144-125.sh.cgocable.ca) |
| 02:28.24 | *** join/#edev redblue (n=star@ppp047.108-253-207.mtl.mt.videotron.ca) |
| 04:41.08 | *** join/#edev DrZimmerman (n=theo@2001:5c0:8fff:fffe:0:0:0:6af1) |
| 05:09.47 | *** join/#edev Crofton (n=balister@dslb-088-073-109-208.pools.arcor-ip.net) |
| 06:19.56 | joe_bleau | WTF? Luminary requires you to register to view a freakin' datasheet?! That's a bunch of crap. |
| 06:23.59 | MonMotha | joe_bleau: lots of places require that |
| 06:24.02 | MonMotha | then they spam you |
| 06:25.46 | joe_bleau | I don't *ever* remember registering to read a datasheet. |
| 06:26.14 | joe_bleau | And then (bbbb/bbbbbb), the doc won't open in acrobat! |
| 06:26.19 | joe_bleau | I'm really hot right now. |
| 06:26.32 | joe_bleau | Wonder if they have an indian call center staffed up right now. |
| 06:32.22 | MonMotha | most of when I've had to register it's for a click-wrap style NDA |
| 06:32.41 | MonMotha | most of which pretty much just say don't redistribute it verbatim, but rather direct people to the company |
| 06:32.47 | joe_bleau | Yeah, no NDA or terms of use. |
| 06:33.28 | MonMotha | I've been really busy (still am)...wow |
| 06:33.48 | MonMotha | we've been doing a board bringup at work, I sent another board off to be fabricated, and I'm designing a third |
| 06:34.03 | MonMotha | power supplies are fun |
| 06:34.14 | joe_bleau | I'm just starting up a small project. |
| 06:34.28 | joe_bleau | Need to have protos ready to show by january. |
| 06:34.53 | MonMotha | slightly longer timetable than what I have |
| 06:35.18 | joe_bleau | Do you have to do injection molding tooling for your project? |
| 06:35.25 | MonMotha | fortunately, for most digital stuff, I can churn out a 99% functional board (nothing is ever 100% the first time) within a few weeks |
| 06:35.33 | MonMotha | we probably will end up doing so, yes |
| 06:35.38 | MonMotha | that's a little bit down the line |
| 06:35.51 | MonMotha | and not 100% my responsibility as I'm not very mechanically inclined |
| 06:36.05 | MonMotha | though they are blocking on my board design, which is why I'm hurried on it |
| 06:36.24 | joe_bleau | We might get by with some stereo lithography rapid protos for the cases. But the tooling has to be close. |
| 06:36.39 | MonMotha | this device is kinda weird |
| 06:36.48 | MonMotha | it's basically going to end up being a giant brick of various electronics |
| 06:37.01 | MonMotha | it has a SMPS, bunches of analog IO, and a ARM7 micro |
| 06:37.15 | MonMotha | oh, and a couple relays |
| 06:37.45 | joe_bleau | Sounds fun. I'm looking at a 24 bit ADC, micro, and a display. |
| 06:38.39 | MonMotha | oh yeah, and a 20x2 character LC |
| 06:38.41 | MonMotha | LCD |
| 06:38.45 | MonMotha | and IEEE 802.15.4 |
| 06:38.57 | MonMotha | and ethernet, and USB, and RS-485, and RS-232, and CAN |
| 06:39.02 | joe_bleau | And the super secret datasheet has no spec for current draw in hibernate mode. Grr. |
| 06:39.12 | joe_bleau | CAN? Really? I've been wanting to play with CAN. |
| 06:39.13 | MonMotha | and it's 4x4 inches |
| 06:39.23 | MonMotha | well, from a hardware point of view, CAN is pretty straightforward |
| 06:39.30 | MonMotha | software wise, it's way overcomplicated for most things |
| 06:40.07 | joe_bleau | It seems like such a neat design, the bitwise arb. |
| 06:40.19 | joe_bleau | The little tiny packets seem odd. |
| 06:40.20 | MonMotha | IIRC, there's something like 295 parts on this 4x4 in. board |
| 06:40.38 | MonMotha | all passives are 0603, and I almost had to go to 0402s |
| 06:40.50 | MonMotha | completely double sided (assembly costs are kinda killer, unfortunately) |
| 06:40.56 | joe_bleau | I'll bet. |
| 06:41.11 | MonMotha | fortunately, this isn't an overly cost sensitive item |
| 06:41.22 | MonMotha | any time you deal with analog, you end up with tons of passives |
| 06:41.24 | MonMotha | just the way it is |
| 06:41.38 | MonMotha | that and I don't skimp on bypass caps |
| 06:41.47 | MonMotha | helps with EMI and is one of the reasons my boards tend to work correctly the first time |
| 06:41.55 | MonMotha | I figure I can always just non-pop them later |
| 06:42.29 | joe_bleau | There's a guy on usenet who argues people overuse bypass caps. |
| 06:42.39 | joe_bleau | And his boards usually work first time, or so he claims. |
| 06:42.56 | MonMotha | well, bypass caps are one of those things that it's hard to model the stuff accurately |
| 06:42.56 | joe_bleau | He likes multilayer as a bypass. |
| 06:43.04 | MonMotha | well, this is a 4 layer board |
| 06:43.18 | MonMotha | and the inner two are mostly dedicated to power/ground (though the power plane is a split plane) |
| 06:43.42 | MonMotha | on this project, though, the cost of all the bypass caps was so trivial, that I just chucked them all on there |
| 06:43.57 | MonMotha | in most cases, having extras doesn't hurt things (assuming inrush doesn't get out of hand) |
| 06:44.08 | MonMotha | and it's always easier to remove unnecessary ones later than add some in |
| 06:45.10 | joe_bleau | I can't decide if I want to go 5V or 3V for my main rail. |
| 06:46.17 | MonMotha | 3.3 unless you have a compelling reason not to |
| 06:46.29 | MonMotha | most ARM stuff is 3.3V IO |
| 06:46.40 | joe_bleau | My LCD is 5V. I'd like to keep it common with an existing product. |
| 06:46.52 | joe_bleau | 5V fits the ADC better, too. |
| 06:47.03 | MonMotha | well, then it sounds like you may have a compelling reason |
| 06:47.14 | joe_bleau | But it cuts down my micro choices. |
| 06:47.25 | MonMotha | on my board, the LCD and CAN were the only things that needed 5V, so the main rail is 3.3V |
| 06:47.33 | MonMotha | the IO on the micro are 5V tolerant |
| 06:47.39 | joe_bleau | There's ST with 5V, looks like luminary might be, and some of the H8 stuff is 5V. |
| 06:47.42 | MonMotha | and the LCD and CAN have TTL thresholds |
| 06:47.47 | joe_bleau | But yeah, more choices at 3.3. |
| 06:48.11 | MonMotha | Atmel's SAM7 series has 5V tolerant IO, though they only drive 3.3V high, and consume higher power when driven as inputs to 5V |
| 06:48.19 | MonMotha | that's a 1.2V core, 3.3V IO |
| 06:48.21 | joe_bleau | power ain't a problem. |
| 06:48.36 | MonMotha | there's an onboard LDO to generate the 1.2V, so you don't need a regulator |
| 06:48.50 | joe_bleau | Been there before with the LPC2105. |
| 06:48.58 | MonMotha | my board has 24V, 12V, 5V, 3.3V, and 1.2V rails |
| 06:48.59 | MonMotha | kinda nuts |
| 06:49.05 | joe_bleau | Too many freaking regulators these days. |
| 06:49.42 | MonMotha | 24V comes in from the stack base (generated by a flyback SMPS from 24-120V), switcher takes that down to 12V and 3.3V, the 5V is a LDO off 12V, and 1.2V is the SAM7 onboard LDO from 3.3V |
| 06:49.58 | MonMotha | switchers are great |
| 06:50.06 | joe_bleau | Why an LDO for 12 to 5? |
| 06:50.08 | MonMotha | and the buck topology is fairly easy to design for |
| 06:50.15 | MonMotha | because I only need a few mA at 5V |
| 06:50.20 | MonMotha | not worth using another SMPS |
| 06:50.42 | joe_bleau | Why not a standard regulator, with better performance over an LDO? |
| 06:50.45 | MonMotha | SMPSs are bigger and more expensive since you need an inductor and diode |
| 06:50.50 | joe_bleau | I only use LDO when I need 'em. |
| 06:51.02 | MonMotha | heh, because LDOs tend to have better performance than the non low dropout kind |
| 06:51.08 | joe_bleau | Really? |
| 06:51.10 | MonMotha | yeah |
| 06:51.33 | joe_bleau | I thought that PNP output was bad news. The datasheets always seem to be a little picky about output cap and ESR. |
| 06:51.36 | MonMotha | and they're often cheaper (which amazes me) |
| 06:51.53 | MonMotha | look up the NCP582 (I think) from ON |
| 06:51.58 | MonMotha | IIRC, it can run with no output cap |
| 06:52.39 | MonMotha | they recommend 0.1-0.47uF |
| 06:53.03 | MonMotha | well, they recommend a low ESR 1uF |
| 06:53.14 | MonMotha | but the qualification data says they're stable with less |
| 06:54.23 | MonMotha | ah, the 562, not 582 |
| 06:54.37 | joe_bleau | Does it oscillate when you put more capacitance on it? |
| 06:54.42 | MonMotha | <PROTECTED> |
| 06:55.21 | MonMotha | somewhere I had a plot of the safe operating area relating output capacitance and output series resistance |
| 06:55.33 | MonMotha | I've never had a 562 oscillate on me, and I've used them a fair bit |
| 06:56.22 | MonMotha | they pretty much seem to "just work" as long as you get at least a 0.1uF ceramic on the output |
| 06:56.54 | MonMotha | and even standard regulators require some output cap |
| 06:57.11 | MonMotha | switchers require larger output caps, of course (plus the additional hardware), but you get much higher efficiency |
| 06:57.42 | joe_bleau | I wonder what's up with the hints section...they do say to watch the impedance of the input lines. |
| 06:58.43 | MonMotha | I think that's mostly HF noise rejection related |
| 07:01.04 | MonMotha | I've found that it's also easier to get LDOs in tiny packages. Standard linear regulators tend to be targeted towards higher power applications |
| 07:01.09 | MonMotha | and as such come in larger packages |
| 07:01.12 | joe_bleau | Line and load reg are better with the old 7805. |
| 07:01.29 | MonMotha | 7805 isn't available in a SC82 :) |
| 07:02.02 | MonMotha | also, 7805 is usually spec'd with higher output cap...let me pull up the datasheet |
| 07:02.17 | MonMotha | which plot on the 562 are you looking at, the 0.1uF or the 1.0uF output cap? |
| 07:02.45 | MonMotha | actually, I guess those are just line/load |
| 07:02.55 | MonMotha | I don't know why they changed the output cap spec between those... |
| 07:05.31 | MonMotha | I should also point out that I'm not using a 562 for the 12V to 5V |
| 07:05.37 | MonMotha | (it's not rated for that) |
| 07:05.40 | joe_bleau | I just pulled up the MC78L05. |
| 07:05.50 | MonMotha | the 78L05 is a newer version |
| 07:05.57 | joe_bleau | Sorry, 78L05A. |
| 07:07.32 | MonMotha | a lot of people like the 1117s, nowadays in place of 7805s |
| 07:07.42 | joe_bleau | Yeah, I've used it. |
| 07:07.52 | joe_bleau | (The '2105 projects.) |
| 07:08.14 | ds2 | don't the 1117's require resistors for voltage setting? |
| 07:08.35 | MonMotha | they have them in fixed and adjustable varieties, just like the 7800 series |
| 07:08.50 | MonMotha | the 7800 series has been improved over time, too |
| 07:09.01 | MonMotha | if you pick one up from the early 90s, it'll suck way more than one purchased today |
| 07:09.06 | ds2 | n/m confused them with the mil grade lm317 |
| 07:09.11 | MonMotha | ah yes |
| 07:09.17 | MonMotha | the "virtually blowout proof" regulator |
| 07:09.43 | ds2 | love those to220 things... mess of alligator clips and you get a current regulated power supply |
| 07:10.21 | MonMotha | yeah, but when I'm chucking things on a 4x4 in. board with 294 other components, I want small |
| 07:10.22 | joe_bleau | I still consider LDOs to be a compromise compared to a standard linear regulator, when you don't need the low-dropout. |
| 07:10.49 | MonMotha | and small is something a TO-220 is decidedly not |
| 07:11.14 | ds2 | LDOs worry me... i remember them as not always unconditionally stable :/ |
| 07:11.29 | MonMotha | I've had more problems with 7805s oscillating than modern LDOs |
| 07:11.58 | MonMotha | though I'm usually using 7805s in less than ideal situations (i.e. solderless protoboards) while my LDOs are usually on custom PCBs that I've taken care to meet all stability criteria |
| 07:12.00 | ds2 | wow, I've abused the 7805's a lot (no caps, mega long wiring, etc) and never have them oscillate |
| 07:12.27 | ds2 | aren't LDOs sensitive to low esr components? |
| 07:12.35 | MonMotha | depends on the device |
| 07:12.57 | MonMotha | you'll notice that some of my preferred LDOs specifically state in the datasheet that they don't require any specific ESR |
| 07:13.15 | MonMotha | but yes, some of them are sensitive to that. Those usually have a safe operating region diagram (ESR vs. capacitance) in the datasheet |
| 07:13.24 | MonMotha | I tend to just avoid those to begin with |
| 07:13.40 | joe_bleau | National's AN-1148 is interesting, but dated. |
| 07:13.52 | ds2 | that's hard to do when you need a little power supply in 2 hour and the local place has 2 LDOs and no datasheets at the store :/ |
| 07:14.51 | MonMotha | true |
| 07:15.15 | MonMotha | If I need a power supply in 2 hours, I'll probably end up cannibalizing wall warts |
| 07:15.32 | MonMotha | since the local place doesn't sell much in the way of power electronics (not that won't take you more than 2 hours to find, anyway) |
| 07:15.36 | ds2 | don't find many 3.3V regulated warts :( |
| 07:15.51 | joe_bleau | It almost reads like they can optimize for low or moderate ESR, depending on what kind of cap you pair with it. |
| 07:16.01 | MonMotha | find a switch mode one, you can usually just adjust the feedback to change the output voltage |
| 07:16.20 | MonMotha | joe_bleau: I know a lot of the LDOs these days are specifically designed for ceramic output caps |
| 07:16.26 | MonMotha | since MLCCs have gotten to such high capacitances |
| 07:17.47 | ds2 | what are you building with that many discretes? |
| 07:18.19 | MonMotha | ds2: industrial analog signalling type stuff designed to handle a fairly wide variety of signal types |
| 07:18.34 | ds2 | ah analog |
| 07:18.41 | MonMotha | 0-10V, 4-20mA, PWM, as well as hooking various sensors (mostly RTDs and thermistors) directly |
| 07:18.46 | MonMotha | yeah |
| 07:19.04 | joe_bleau | Spiffy! Another industrial guy. |
| 07:19.05 | MonMotha | digital I can keep the BOM pretty small, but when you get into analog stuff you pretty much always end up with passives everywhere |
| 07:19.19 | MonMotha | joe_bleau: well, that's just what this project happened to entail |
| 07:19.28 | ds2 | they need a good analog equiv to a FPGA |
| 07:19.33 | MonMotha | yeah, it would be nice |
| 07:19.38 | joe_bleau | Are y'all doing any serious input protection? |
| 07:19.47 | MonMotha | and somewhat doable using CMOS switches |
| 07:19.54 | MonMotha | not really, no |
| 07:20.02 | joe_bleau | Transorbs, varistors, series resistance, those cute little self resetting fuses? |
| 07:20.02 | MonMotha | pretty much this is a "don't be a dumbass" type thing |
| 07:20.13 | joe_bleau | Oh. |
| 07:20.23 | MonMotha | if the installer toasts it, they get to buy another one |
| 07:20.24 | ds2 | CMOS switches are horrible, 50+ ohm on resistance and a bunch of other limitations |
| 07:20.34 | MonMotha | ds2: hence the somewhat |
| 07:20.48 | joe_bleau | I'm thinking more in terms of someone screws up the wiring after the fact, or something nearby blows up. |
| 07:20.49 | ds2 | =) |
| 07:21.03 | MonMotha | joe_bleau: yeah, it's likely that you'd toast some stuff if that happens |
| 07:21.09 | *** join/#edev Crofton (n=balister@dslb-088-073-109-208.pools.arcor-ip.net) |
| 07:21.13 | ds2 | oh have either of you seen the Actel FPGAs with an ARM core inside it? |
| 07:21.26 | MonMotha | there's no real room for any significant input protection, so there's not much |
| 07:21.34 | joe_bleau | Is it a hard core? |
| 07:21.39 | MonMotha | ds2: closest I have is a Virtex 2 Pro |
| 07:21.42 | MonMotha | dual PPC |
| 07:21.50 | MonMotha | ARM would probably be more useful |
| 07:22.12 | ds2 | MonMotha: those require mega cooling and an onboard diesel generator + tanker for mobile use ;) |
| 07:22.27 | joe_bleau | I'd love to play with the Xilinx embedded PPC stuff, but the tools are so pricey. |
| 07:22.38 | ds2 | joe_bleau: think so, since they claim to have ARM7TDMI versions and AFAIK, those only came in hard cores |
| 07:23.32 | joe_bleau | Here's a note from '05: "Capability-wise, the ARM core is a specialized soft IP core on the Actel device" |
| 07:23.52 | ds2 | oh... guess i misunderstood the guy at the booth |
| 07:24.08 | joe_bleau | This was '05...they may have something newer. |
| 07:24.35 | ds2 | they also have a cortex version but that is sold as FPGA IP so that makes it murky |
| 07:25.24 | joe_bleau | Cortex M1 soft core mentioned in '07. |
| 07:26.15 | joe_bleau | They're flash based, which is neat. |
| 07:26.38 | ds2 | they had a handheld batteyr power device w/it |
| 07:26.49 | MonMotha | ds2: I rarely see a V2P design which needs more than a heatsink on the chip |
| 07:27.02 | MonMotha | now, the diesel generator, well, you've got me there :) |
| 07:27.28 | ds2 | MonMotha: the vitrex ones are the big silver ones in a BGA right? |
| 07:27.37 | MonMotha | yes |
| 07:27.56 | MonMotha | unfortunately not available in QFP like the Spartans are |
| 07:28.26 | MonMotha | while QFP is worse from a thermal point of view and a connection density point of view, I can actually hand-solder the damn things |
| 07:29.23 | ds2 | ah... i misunderstood them... the actels are 'arm enabled' |
| 07:29.39 | ds2 | guess it was a 'too good to be true' thing |
| 07:30.00 | joe_bleau | Marketing! |
| 07:30.21 | joe_bleau | Still, it seems like someone should have an ARM hard core in an FPGA, right? |
| 07:30.30 | ds2 | still sorting through literature from the arm conference |
| 07:30.38 | joe_bleau | Was it fun? |
| 07:30.41 | ds2 | you'd think that |
| 07:31.05 | ds2 | yes, it was informative |
| 07:31.23 | ds2 | but I prefer ESC as this is a bit narrowly focused |
| 07:31.45 | joe_bleau | I'd love to go, but can't justify the $$$ and time. |
| 07:32.01 | ds2 | got free passes and it was local |
| 07:32.08 | joe_bleau | That would help! |
| 07:32.14 | ds2 | *nod* |
| 07:32.53 | ds2 | probally wouldn't have paid for it, it wasn't THAT useful |
| 07:33.12 | Crofton | an ARM plus FPGA would rock |
| 07:33.50 | joe_bleau | Assuming cheap tools to design with this ARM in an FPGA, right? |
| 07:34.06 | ds2 | sales guy claimed Linux tools downloadable from website |
| 07:34.21 | ds2 | not sure how much they want for the ARM IP though |
| 07:34.43 | Crofton | of course :) |
| 07:35.54 | ds2 | new to me but it is cool to see tools used to simulate power consumption so it can be optimized |
| 07:36.39 | ds2 | but notably absent in their presentation is an expected accuracy figure ;) |
| 07:45.44 | MonMotha | hum, I know atmel has AVR+FPGA |
| 07:45.49 | MonMotha | I dunno if they have RAM |
| 07:45.50 | MonMotha | ARM |
| 08:10.20 | MonMotha | gah, why do they put previews for ancient movies in the non-skippable track on DVDs? |
| 08:10.23 | MonMotha | just play the god damned movie |
| 08:10.36 | MonMotha | I need to hack my $12 DVD player to ignore that shit |
| 08:45.59 | *** join/#edev lyakh (n=lyakh@dynamic-unidsl-85-197-26-24.westend.de) |
| 09:17.10 | *** join/#edev Crofton (n=balister@dsl-62-220-14-162.berlikomm.net) |
| 10:22.53 | *** join/#edev orb1t (n=orb1t@213.154.199.214) |
| 10:56.32 | *** join/#edev pina (n=saturare@unaffiliated/pina) |
| 11:31.38 | *** join/#edev redblue (n=star@ppp227.108-253-207.mtl.mt.videotron.ca) |
| 12:27.55 | *** join/#edev steliosk (n=Stelios@ipa226.211.tellas.gr) |
| 12:44.08 | *** join/#edev ico2 (n=ico2@modem-127.chameleon.dialup.pol.co.uk) |
| 13:18.12 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 13:18.12 | *** mode/#edev [+o GPSFan] by ChanServ |
| 13:23.00 | *** join/#edev Off_Namuh72 (i=GPS@gateway/tor/x-66f136745a0d6703) |
| 13:32.37 | *** join/#edev ico2_ (n=ico2@modem-183.cheetah.dialup.pol.co.uk) |
| 15:32.48 | joe_bleau | Is gcc's support of thumb-2 any good? |
| 16:21.00 | *** join/#edev prpplague (n=dave@ppp-70-244-172-90.dsl.rcsntx.swbell.net) |
| 16:21.31 | prpplague | ho ho diddly ho |
| 16:21.39 | prpplague | Exiles: ping ping ping |
| 16:21.43 | prpplague | jbevren: ping ping ping |
| 16:30.01 | prpplague | davidc__: ping |
| 16:30.11 | davidc__ | prpplague: yup? |
| 16:30.40 | prpplague | davidc__: hey, on the new bluetooth module |
| 16:31.03 | prpplague | davidc__: why didn't you guys bring out a connection for the "connection gpio" ? |
| 16:31.19 | davidc__ | prpplague: dunno - wasn't my design |
| 16:31.24 | prpplague | davidc__: ahh |
| 16:31.30 | davidc__ | I'll ask my boss tomorrow |
| 16:31.36 | davidc__ | IIRC - it had to remain footprint compatible |
| 16:31.40 | davidc__ | which may have been a motivation |
| 16:31.55 | prpplague | davidc__: okie dokie, but the blueradio module has the "connection gpio" |
| 16:32.06 | prpplague | davidc__: which means it isn't 100% compatible |
| 16:32.14 | davidc__ | eh, I really dunno, I'll ask the guy who designed it |
| 16:32.27 | prpplague | davidc__: np, just thought you might know |
| 16:32.43 | prpplague | davidc__: whatcha going to be up to when you get back to CA? |
| 16:32.51 | davidc__ | prpplague: not 100% sure yet |
| 16:33.35 | prpplague | davidc__: ahh |
| 16:35.43 | prpplague | davidc__: well just because you are going back to ca doesn't mean you don't hang out in #edev anymore! hehe |
| 17:04.51 | joe_bleau | prpplague: how long does it take the hammer to go from power applied to a user app start? |
| 17:09.12 | prpplague | joe_bleau: right now its about 10 seconds |
| 17:09.21 | prpplague | joe_bleau: but that can be tweaked to around 3 |
| 17:09.58 | joe_bleau | 10? What is taking so long? |
| 17:14.17 | prpplague | joe_bleau: 10 second boot time is pretty darn quick |
| 17:16.19 | prpplague | joe_bleau: the main issues as far as boot time are the mounting of the jffs2 partition and the final mounting of the ramdisk |
| 17:16.43 | prpplague | joe_bleau: using a small ramdisk and mounting the jffs2 later down the init cycle |
| 17:17.04 | prpplague | joe_bleau: in addition there are a couple of checks for the difference between initramfs and ramdisk |
| 17:17.47 | prpplague | joe_bleau: with all the tweaks, the http://www.elinux.org/M5900 boots in about 2.4 seconds |
| 17:18.34 | joe_bleau | I don't remember the juicebox (the last embedded linux-like toy I spent any real time with) booting so slowly. |
| 17:18.52 | joe_bleau | Besides, if 10 was quick, why tweak for 3? ;-) |
| 17:19.11 | joe_bleau | Again, remember I'm coming at this with the microcontroller mindset. |
| 17:19.47 | prpplague | joe_bleau: well there are always differences of needs |
| 17:19.58 | prpplague | joe_bleau: 10 seconds is perfectly find most apps |
| 17:20.12 | prpplague | joe_bleau: but for items that need "instant on" you have to tweak |
| 17:22.25 | *** join/#edev pierrelux (n=pierre-l@144-125.sh.cgocable.ca) |
| 17:22.26 | prpplague | joe_bleau: the juicebox booted and ran primarily from flash |
| 17:22.32 | prpplague | joe_bleau: it used XIP |
| 17:22.54 | joe_bleau | I thought it copied all that stuff to sdram. Been too long, I guess. |
| 17:22.54 | prpplague | joe_bleau: which does work well for min ram/flash requirements, however it does impact performance |
| 17:23.12 | prpplague | joe_bleau: the kernel and roots were done as xip |
| 17:25.02 | *** join/#edev chouimat (n=dieu@kde/developer/chouinard) |
| 17:25.28 | joe_bleau | So with linux, XIP gives a quicker boot, but reduced performance once you're up? |
| 17:26.35 | chouimat | morning |
| 17:38.22 | prpplague | joe_bleau: correct |
| 17:38.35 | prpplague | chouimat: greetings |
| 17:39.30 | prpplague | joe_bleau: there is an enite section of the elinux wiki about boot time trade offs |
| 17:40.17 | joe_bleau | OK, sorry. |
| 17:41.26 | prpplague | joe_bleau: np, happy to answer questions any time |
| 17:41.27 | prpplague | :) |
| 17:42.00 | joe_bleau | Yeah, but you should have just said "RTFW" a few minutes ago, and I wouldn't have wasted your time... |
| 17:48.15 | prpplague | hehe, i like good conversation |
| 17:48.22 | prpplague | its the only way to learn some things |
| 18:13.40 | *** join/#edev chouimat (n=dieu@kde/developer/chouinard) |
| 18:43.09 | *** join/#edev ico2 (n=ico2@modem-1866.cougar.dialup.pol.co.uk) |
| 19:11.38 | *** join/#edev redblue (n=star@ppp181.108-253-207.mtl.mt.videotron.ca) |
| 19:44.22 | *** join/#edev ico2 (n=ico2@modem-1866.cougar.dialup.pol.co.uk) |
| 20:19.52 | ds2 | tick tock |
| 20:20.32 | joe_bleau | precisely. |
| 20:45.18 | *** join/#edev flyback (n=no@c-24-3-31-106.hsd1.pa.comcast.net) |
| 20:45.22 | flyback | m00\ |
| 20:45.45 | flyback | how well do you guys think the dht-walnut would work for a 10/100 lan sniffer between the router/firewall and the main switch |
| 20:45.50 | flyback | using a network tap |
| 20:47.34 | flyback | I also discovered that the 2 old antec/arris telco ups's one at my house, one at my folks both are no longer used by comcast for the phone system, output both +48v and +12 direct tap to the battery so I am thinking +12 high efficiency ups for low voltage devices like the dht |
| 20:47.37 | flyback | :P |
| 20:53.37 | agaffney | hmm, I've got one of those in my basement |
| 20:56.28 | flyback | yeah the new system uses a lithium cell in a all in one phone system and cable modem |
| 20:56.43 | flyback | and the one before blew out so I took the ups and battery when we canncelled the service yrs ago |
| 20:56.52 | flyback | so I got 2 units I just need batts |
| 20:57.06 | flyback | I figure why the hell take the inverter efficiency loss when I don't have to |
| 20:57.42 | flyback | my linksys and switch are +12 and the cable modem I have is +10 so I just need to make a simple voltage drop circuit |
| 20:58.38 | flyback | I got a bunch of these o2 cool battery fans from walmart now that they are getting cheap due to end of season |
| 20:58.47 | flyback | 11 inch dc brushed motor $7 |
| 20:58.58 | flyback | takes 8 d cells = 12 volts which is damn expensive |
| 20:59.10 | flyback | but they run on sla's just fine even worn out ones that are a little low in voltage |
| 20:59.22 | flyback | got me thinking about how I can improve efficiency of systems |
| 20:59.34 | flyback | and also systems that can handle power problems or running off grid |
| 21:01.01 | flyback | not so much from a global warming prespective because I am still not convinced, but I just like the challange of "how can we make this survive being put in some hell hole for 12 months" |
| 21:01.09 | flyback | or 3rd world |
| 21:01.36 | flyback | I would love to come up with simple designs for people to be able to generate a few watts of power using junk found around the city/village etc |
| 21:01.43 | flyback | and be able to do something useful work with it |
| 21:02.41 | flyback | agaffney you should build a small windmill |
| 21:02.51 | flyback | you can generate a few watts or more using some old steppers or servos |
| 21:03.05 | flyback | slap a low power mcu on it or one of those prpplague arm sticks |
| 21:03.18 | flyback | and be able to plant it out in a field somewhere and take measurements a few times a day |
| 21:03.20 | flyback | or photos :) |
| 21:05.21 | flyback | I wanna do verticle windmills more than horizontal/prop type but since I have 0 skills to make either, for now horizontal is being used because I can use old fan blades etc |
| 21:08.31 | flyback | verticle are awesome because they can be simplier, don't need to be pointed in the direction of the wind and they will run on extremely ;low mph winds |
| 21:09.00 | flyback | the disadvantage is they max out wind power extraction efficiecy faster (well the older ones did) but not a problem for small scale |
| 21:09.24 | agaffney | my next project involves my roomba, a netgear wgt634u running openwrt, and a usb->serial adapter |
| 21:09.33 | flyback | my room is being rebuilt with counter tops etc so I might just blow the dust off the old webpal and juiceboxes |
| 21:09.34 | flyback | :P |
| 21:09.36 | flyback | nice |
| 21:09.46 | flyback | did you see roomba released some new robot or something |
| 21:09.49 | agaffney | I just ordered a db9->7-pin mini-din connector and a 7-pin cable today |
| 21:09.51 | flyback | I think a geek edition but I can't remeber |
| 21:10.10 | agaffney | I know they released one that's just meant for playing with |
| 21:10.15 | agaffney | no vacuum or anything |
| 21:10.40 | agaffney | what I want to do is write some code that allows the roomba to "know" where it is in my house |
| 21:10.48 | agaffney | I also want to plug in a webcam |
| 21:10.55 | flyback | nice |
| 21:10.56 | agaffney | I'll need a usb hub, but that's no big deal |
| 21:11.01 | flyback | going to use some kinda gps system? |
| 21:11.02 | agaffney | the wgt634u only has 1 usb port |
| 21:11.12 | flyback | yeah hub is fine if you don't have a lot of bw need |
| 21:11.20 | agaffney | no, it would use obstacles it encounters to determine where it is |
| 21:11.22 | flyback | since you are just doing control you are probably fine |
| 21:11.26 | agaffney | which means I'd have to map out my rooms |
| 21:11.31 | agaffney | and the "fixed" obstacles in them |
| 21:11.47 | flyback | neat |
| 21:11.51 | agaffney | and when that fails, that's what the camera is for :) |
| 21:11.59 | flyback | www.futurlec.com |
| 21:12.06 | flyback | they have cheap everything including gas sensors |
| 21:12.13 | flyback | so it could swear at people farting in the room |
| 21:12.13 | flyback | :P |
| 21:12.48 | agaffney | heh |
| 21:12.49 | agaffney | I just want to chase my cats from work :P |
| 21:12.55 | flyback | prpplague I think already added their page to the main site |
| 21:13.03 | flyback | they have ttl to rs232 cards for < $5 |
| 21:13.08 | flyback | 2 port each +5 and +3.3 ones |
| 21:14.16 | agaffney | I still have lots of devices i need to find time to play with |
| 21:14.34 | flyback | yeah I know the feeling |
| 21:14.35 | agaffney | 2 walnuts, an efika, 2 netgear wgt634u |
| 21:14.42 | flyback | this flood though is kinda a fresh start |
| 21:14.50 | agaffney | flood? |
| 21:14.50 | flyback | room will no longer be a landfill |
| 21:14.55 | flyback | yeah washing machine repair I did |
| 21:15.04 | agaffney | heh |
| 21:15.10 | flyback | I didn't feel like putitng the cabinet back on it cause I wanted to touch up the rust |
| 21:15.16 | flyback | after I did so much work changing gearing etc |
| 21:15.23 | flyback | so I just clipped the fill tube to the tub |
| 21:15.35 | flyback | and one time I forgot to clip it and it kept trying to fill the tub |
| 21:15.37 | flyback | all over the floor |
| 21:15.53 | flyback | which wouldn't have been a problem if I had not also happen to block the sewer drain on the floor |
| 21:16.00 | flyback | came home to 3-6 inches of water |
| 21:16.11 | agaffney | heh, fun |
| 21:16.21 | flyback | I open the basement door |
| 21:16.26 | flyback | "what the f***?" |
| 21:16.44 | flyback | ruined a laser printer I had a bunch of old monitors |
| 21:17.01 | flyback | one from a pcjr (not much of a loss) one from a apple IIc and one from a decmate III |
| 21:17.06 | flyback | rest were just pc monitors |
| 21:19.35 | flyback | it's actually covered by insurance even if they were already dead |
| 21:19.52 | flyback | but it's still a pain in the ass, but the insurance did pay for a cleaning and repainting crew so my room will finally be a proper lab |
| 21:20.05 | flyback | not http://flyback.databurn.net like it was |
| 21:27.49 | flyback | even if I only get 1-2 hrs to play every night after work |
| 21:27.51 | flyback | I would be happy |
| 21:40.54 | *** join/#edev FireEgl (i=FireEgl@Sebastian.FireEgl.Com) |
| 21:56.11 | *** join/#edev ico2 (n=ico2@modem-2895.cheetah.dialup.pol.co.uk) |
| 22:31.17 | *** join/#edev diamat (n=cerne@2001:5c0:8fff:fffe:0:0:0:5b4f) |
| 22:38.32 | *** join/#edev chouimat (n=dieu@kde/developer/chouinard) |
| 22:46.25 | *** part/#edev pierrelux (n=pierre-l@144-125.sh.cgocable.ca) |
| 22:47.22 | *** join/#edev FireEgl (i=FireEgl@Sebastian.FireEgl.Com) |
| 22:48.04 | MonMotha | ok, DBUS officially sucks |
| 22:48.09 | MonMotha | sooooo much glue to accomplish so little |
| 22:48.22 | *** join/#edev Crofton (n=balister@dslb-088-073-079-088.pools.arcor-ip.net) |
| 23:14.01 | *** join/#edev Exiles (i=wired57@pimped.your.sister.out.while.you.were.incarcerated.info) |
| 23:20.51 | *** join/#edev Exiles (i=wired57@pimped.your.sister.out.while.you.were.incarcerated.info) |
| 23:27.26 | *** join/#edev Off_Namuh72 (i=GPS@gateway/tor/x-685101967d5002c4) |
| 23:31.01 | *** join/#edev FireEgl (i=FireEgl@Sebastian.FireEgl.Com) |