| 01:22.24 | *** join/#edev Ian_Daniher|xo (n=it@69.61.230.246) |
| 02:29.07 | *** join/#edev kergoth (n=kergoth@ip68-3-163-239.ph.ph.cox.net) |
| 03:35.44 | *** join/#edev JoeLlama (n=snork@unaffiliated/joellama) |
| 04:04.21 | *** join/#edev landley (n=landley@m495e36d0.tmodns.net) |
| 04:04.21 | *** mode/#edev [+o landley] by ChanServ |
| 04:29.45 | *** join/#edev DrZimmerman (n=theo@77-56-228-246.dclient.hispeed.ch) |
| 04:40.46 | *** join/#edev JoeLlama (n=snork@206.169.63.10) |
| 05:46.13 | *** join/#edev redb1ue (i=star@ppp026.108-253-207.mtl.mt.videotron.ca) |
| 05:47.43 | *** join/#edev harpal (n=Harpal@122.169.108.195) |
| 05:52.52 | *** join/#edev davidc__ (n=davidc__@S01060002b360aacd.vc.shawcable.net) |
| 05:57.13 | *** join/#edev landley (n=landley@cpe-70-116-30-223.austin.res.rr.com) |
| 05:57.13 | *** mode/#edev [+o landley] by ChanServ |
| 06:56.57 | *** join/#edev sleipnir (n=Sleipnir@p4FDD64C3.dip.t-dialin.net) |
| 07:36.18 | *** join/#edev lyakh (n=lyakh@p57BD2A35.dip0.t-ipconnect.de) |
| 07:57.36 | *** join/#edev Crofton (n=balister@12.234.232.66) |
| 08:14.15 | *** join/#edev [AD]Turbo (n=turbo@host162-220-dynamic.59-82-r.retail.telecomitalia.it) |
| 08:16.32 | [AD]Turbo | yo all |
| 08:22.06 | *** join/#edev boris_OmegA (n=boris@194.51.141.172) |
| 08:32.02 | methril | yo |
| 08:32.25 | methril | yo |
| 09:02.39 | *** join/#edev hw (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 09:23.29 | *** join/#edev KazaLite (n=kazaLite@116.58.8.130) |
| 09:32.24 | *** join/#edev harpal (n=Harpal@122.169.108.195) |
| 11:24.06 | *** join/#edev g1powermac (n=g1powerm@unaffiliated/g1powermac) |
| 11:49.33 | *** join/#edev roxfan2 (n=dunno@33.63-244-81.adsl-dyn.isp.belgacom.be) |
| 12:02.12 | *** join/#edev g1powermac_PB (n=g1powerm@unaffiliated/g1powermac) |
| 12:11.15 | *** join/#edev linac (n=lin@121.201.102.242) |
| 12:25.12 | *** join/#edev redblue (i=star@ppp122.108-253-207.mtl.mt.videotron.ca) |
| 12:30.52 | *** join/#edev Madsy (n=madsy@ti500710a342-0680.bb.online.no) |
| 14:09.01 | *** join/#edev kergoth (n=kergoth@ip68-3-163-239.ph.ph.cox.net) |
| 14:26.23 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 14:27.18 | prpplague | ho ho ho |
| 14:27.22 | prpplague | merry freakin monday |
| 14:32.18 | Madsy | I love Mondays |
| 14:32.49 | *** join/#edev dmead (n=dan@ip229.sensaphone.com) |
| 14:33.39 | prpplague | i hope Mondays is the name of a girl, otherwise you are plain crazy, hehe |
| 14:39.09 | *** join/#edev gandhijee (n=root@host-66-202-34-165.spr.choiceone.net) |
| 14:45.07 | *** join/#edev CosmicPenguin (n=nobody@www.crouse-house.com) |
| 14:49.54 | *** join/#edev samrobb (n=samrobb@eng.arriad.com) |
| 14:56.19 | prpplague | samrobb: greetings |
| 14:56.26 | prpplague | CosmicPenguin: greetings |
| 14:57.38 | CosmicPenguin | morning |
| 14:57.41 | *** join/#edev GPSFan (n=kenm@64.92.145.112) |
| 14:57.41 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:57.55 | samrobb | prpplague: Morning, folks. |
| 14:58.10 | prpplague | CosmicPenguin: got a Jay-Oh-Bee yet? |
| 14:58.21 | prpplague | CosmicPenguin: btw, how did your halloween stuff go? |
| 14:59.07 | *** join/#edev kergoth (n=kergoth@65.200.49.156) |
| 14:59.32 | prpplague | kergoth: ping |
| 15:03.55 | kergoth | prpplague: pong |
| 15:07.17 | prpplague | kergoth: hey, i know you are an audiophile, what kind of mp3 player do recommend/use? |
| 15:07.36 | prpplague | kergoth: (of course it needs to be supported under linux) |
| 15:07.56 | kergoth | used to be, not nearly as much as i used to be. at this point i generally just want something that works |
| 15:08.13 | kergoth | i have an ipod shuffle i listen to when i work out and occasionally when i'ma t work, thats about it |
| 15:08.17 | *** join/#edev chouimat|work (n=dieu@209.217.106.98) |
| 15:08.27 | prpplague | kergoth: ahh |
| 15:09.21 | prpplague | kergoth: i was thinking of finding something i could load rockbox on |
| 15:12.40 | kergoth | ah thats cool |
| 15:12.47 | kergoth | i had an old iriver that had rockbox on it |
| 15:13.00 | kergoth | wonders, idly, what he did with it |
| 15:20.53 | CosmicPenguin | prpplague: halloween went awesome |
| 15:21.01 | CosmicPenguin | I stil hven't blogged it all yet |
| 15:21.05 | CosmicPenguin | funny - I have the time |
| 15:21.23 | prpplague | hehe |
| 15:22.33 | prpplague | kergoth: i picked up a cheapy mp3 player from the dollar store the other day |
| 15:23.23 | kergoth | hehe. i can't believe how they cheap they are nowadays. keep seeing 2-4gb things for <$20 on woot and other places |
| 15:23.23 | kergoth | crazy |
| 15:23.23 | prpplague | kergoth: the interface on it is very cubersome, and they don't have the audio lines shielded so you can hear some of lcd clocking |
| 15:23.36 | kergoth | hahah, ouch |
| 15:24.14 | prpplague | kergoth: yea, i kinda wanted to jump in from the bottom of the line to see what i wanted in a good unit |
| 15:24.38 | prpplague | kergoth: writing a new firmware for some of these seems pretty interesting |
| 15:42.49 | *** join/#edev g1powermac_PB (n=g1powerm@unaffiliated/g1powermac) |
| 15:46.36 | *** join/#edev dbtid (i=h7scuigm@cpe-65-24-173-154.columbus.res.rr.com) |
| 15:47.19 | dbtid | anyone here familiar with the OMAP family? i'm wondering if anyone can give me a feel for integer and floating-point performance compared to something like a Freescale 5121. |
| 15:48.09 | prpplague | dbtid: Crofton might be able to give you some insight on that, or maybe ds2 |
| 15:48.16 | dbtid | prpplague: thanks. |
| 15:48.33 | dbtid | Crofton: ? |
| 15:48.35 | dbtid | ds2: ? |
| 15:49.07 | dbtid | some of the folks here are leaning heavily towards the 5121. i'm trying to keep the options open a little longer. |
| 15:50.54 | prpplague | dbtid: guess it all depends on what you want to do with it |
| 15:51.05 | prpplague | dbtid: i assume you will be running linux, correct? |
| 15:51.11 | dbtid | yes. |
| 15:51.35 | dbtid | i've been asked to look towards google's android architecture, too, if anyone has their $0.02... |
| 15:52.58 | prpplague | if android is part of your goals, then omap3 series is probably the way to go |
| 15:53.01 | prpplague | imho |
| 15:53.21 | CosmicPenguin | Really? |
| 15:53.25 | dbtid | i haven't dug deeply, so i don't know what android will work with (if there's even a limitation) |
| 15:55.34 | Crofton | dbtid, I don't know much about the freescale part |
| 15:55.46 | Crofton | no FPU on OMAP until OMAP3 |
| 15:56.00 | dbtid | i'm looking at OMAP35xx |
| 15:56.02 | Crofton | OMAP3 has single precision simd unit though |
| 16:01.06 | solar | prpplague: Re: mp3 players. Best to avoid using a mp3 player for audio in any commercial device. http://www.csamuel.org/2008/11/14/patent-trolls-attack-openmoko-project |
| 16:01.34 | CosmicPenguin | Yep - they hate to purge |
| 16:03.34 | prpplague | solar: naw, this is just my personal stuff |
| 16:04.30 | CosmicPenguin | s/hate/had/ |
| 16:05.18 | *** join/#edev kergoth (n=kergoth@65.200.49.156) |
| 16:26.13 | prpplague | solar: interesting, i wonder about the rockbox stuff |
| 16:26.43 | CosmicPenguin | I think you just wait until you get a C&D and then you react |
| 16:36.15 | prpplague | solar: hows the 8050 dev going? |
| 17:05.05 | solar | prpplague: I think we left off last thursday with you planning on testing a tarball on friday. But this was a busy weekend so I really have not got much of anything done sense then |
| 17:05.36 | solar | this is going to be a busy week also. |
| 17:05.39 | MothaWork | solar: well, unfortunately the public pretty much demands "MP3", but if you're using it for uses not exposed to the user, then Vorbis seems like a very viable alternative to MP3 |
| 17:08.06 | solar | MonMotha: sadly yes. But as a vendor the risks have to be taking carefully into account before pushing a commercial product using it. |
| 17:11.01 | prpplague | solar: np, busy for me as well |
| 17:11.21 | prpplague | solar: we've finalized all the hardware fixes and sending off the boards for assembly |
| 17:11.32 | prpplague | solar: so the next rev should be pretty stable |
| 17:11.52 | prpplague | solar: we are getting 10 built up, so i'll be able to send some additional ones out |
| 17:13.26 | *** join/#edev likewise (n=likewise@atwork-193.r-212.178.107.atwork.nl) |
| 17:15.24 | solar | ROHS etc..? I think we are near the point where it's time to tidy up some of the initscripts. Make sure all hw-devices work and call it done for gpe. So any remaining time can be spent on something that looks a little nicer |
| 17:16.08 | solar | sadly linux+embedded+gui's imo are still way behind WM ;/ |
| 17:16.27 | prpplague | solar: indeed |
| 17:16.37 | prpplague | solar: yea, all the RoHS stuff is done |
| 17:16.59 | prpplague | solar: my main concerned is the wifi config utility |
| 17:17.10 | prpplague | solar: that needs to be solid |
| 17:18.08 | solar | oh touchscreen calibration. Thats where we left off. You were going to take a peek at it. |
| 17:19.21 | solar | bbiaf heading into work |
| 17:19.30 | prpplague | yea, i'm not having any problems with calibration on any of my other builds, and i've not tested your tarball yet |
| 17:29.47 | *** join/#edev sleipnir (n=Sleipnir@p4FDD64C3.dip.t-dialin.net) |
| 17:31.06 | prpplague | sleipnir: greetings |
| 17:32.45 | sleipnir | heho |
| 17:33.30 | prpplague | sleipnir: what brings you to #edev? |
| 17:35.04 | sleipnir | I tested some stuff of tslib on the brand-new openmoko freerunner gentoo overlay |
| 17:35.21 | sleipnir | was last week or so with kergoth |
| 17:36.04 | prpplague | ahh right i remember now |
| 17:36.51 | sleipnir | something wrong with me beeing here? :-S |
| 17:37.25 | CosmicPenguin | yeah, kergoth is alergic to bees.. :) |
| 17:37.45 | sleipnir | argh sorry s/bee/be/ |
| 17:37.46 | prpplague | sleipnir: nope, i just like to try to know who's here and what kind of help/assistance they are looking for |
| 17:38.24 | sleipnir | i'm looking for testing tslib then ;-) |
| 17:40.30 | *** join/#edev Ei5GTB_ (n=Paul@78.16.200.239) |
| 17:48.26 | kergoth | sleipnir: wouldnt mind a test of current svn (update last thursday), specifically ts_calibrate. committed a patch i found at montavista that clears the ts event buffer between calibration points in ts_calibrate. should fix that annoying thing where an accidental double tap on one before the next point results in having to restart the calibration |
| 17:48.30 | *** join/#edev lyakh (n=lyakh@p57BD2A35.dip0.t-ipconnect.de) |
| 17:54.49 | *** join/#edev davidc__ (n=davidc__@S01060002b360aacd.vc.shawcable.net) |
| 18:14.15 | sleipnir | kergoth: i think i can't manage to test it today, would be nice if you could remind me tomorrow... |
| 18:14.39 | kergoth | not a problem, tahnks |
| 18:51.28 | MothaWork | prpplague: so, what are your thoughs on RoHS? :) |
| 18:59.03 | prpplague | MonMotha: imho, it is silly and worthless, but it isn't like my opinion counts for anything |
| 19:00.36 | prpplague | weird i haven't gotten a single email in over 2 hours |
| 19:03.04 | MothaWork | prpplague: good to know I'm not alone :) |
| 19:03.41 | MothaWork | prpplague: I applaud their efforts, but I think many of the measures are a bit premature, and the documentation requirements can be nothing short of onerous |
| 19:04.29 | *** join/#edev RobotGuy (n=Dale@pool-96-225-210-142.ptldor.fios.verizon.net) |
| 19:21.16 | *** join/#edev DrZimmerman (n=theo@77-56-225-121.dclient.hispeed.ch) |
| 19:23.52 | solar | kergoth: Note. After the latest version of tslib.. I notice and I'm not sure why but things were unable to find ts_open in -lts (xorg-server for example was unable to link to it) pkg-config files looked right however. |
| 19:24.16 | kergoth | that's odd... ts_open is there |
| 19:24.16 | kergoth | hmm |
| 19:24.19 | ds2 | morning |
| 19:24.40 | dbtid | has anyone here done anything with google's Android? Got any comments? openmoko competition? |
| 19:24.54 | ds2 | sure |
| 19:25.03 | ds2 | wouldn't want to use it though |
| 19:25.06 | ds2 | =) |
| 19:25.50 | keesj | dbtid: I ran it on the n810 once :p |
| 19:26.51 | *** join/#edev beewoolie (n=beewooli@66.147.194.238) |
| 19:27.31 | ds2 | heh my hack for handle is still on youtube...wish they'd expire them |
| 19:29.25 | RobotGuy | solar: Have you seen http://www.societyofrobots.com/axon/ ?? |
| 19:30.17 | prpplague | beewoolie: greetings earthling |
| 19:32.17 | ds2 | prpplague: did the libs make it to some site last week? |
| 19:35.41 | beewoolie | prpplague: Hey man. |
| 19:36.02 | beewoolie | prpplague: wierd. I sent a message, but it didn't show up. |
| 19:36.29 | beewoolie | prpplague: That's some wicked lag. |
| 19:37.54 | prpplague | ds2: no, i got the stuff all prepared, but didn't have time to upload it |
| 19:38.01 | ds2 | 'k |
| 19:38.10 | prpplague | ds2: trying to finish up with some 8050's to send off to get wince ported to it |
| 19:38.41 | prpplague | ds2: should have some time tomorrow to do it |
| 19:38.53 | prpplague | lordy lordy |
| 19:39.06 | prpplague | really needs to find another job |
| 19:39.20 | ds2 | its okay, I am just following up since I blew all of friday trying to find an alternative to a laptop that would fail NEBS noise level tests :( |
| 19:50.06 | prpplague | ds2: ahh |
| 19:50.45 | ds2 | off hand, anyone know what's the max pixel clock on the S3C2410 and at what granularity can it be changed at? |
| 19:51.37 | prpplague | ds2: max pix clock iirc is the pclk |
| 19:54.00 | ds2 | and what's that in MHz? |
| 19:55.37 | prpplague | ds2: sorry its hclk, and it all depends on what freq you are running the device with |
| 19:55.50 | prpplague | ds2: in most cases thats around 101.25MHz |
| 19:55.52 | ds2 | prpplague: on the hammer specically |
| 19:56.02 | ds2 | Oh NICE |
| 19:56.19 | ds2 | very very very nice |
| 19:57.15 | prpplague | ds2: ahh just double checked, max clock val is HCLK/2 |
| 19:57.16 | ds2 | prpplague: any idea on the granularity of that clock? |
| 19:57.26 | prpplague | ds2: so that would make it 50MHz |
| 19:58.08 | prpplague | ds2: clk = HCLK/(divisor x 2) |
| 19:59.31 | ds2 | so close |
| 19:59.41 | solar | RobotGuy: nope. Never seen that board. |
| 19:59.54 | ds2 | prpplague: was wondering if I can use the hammer to emit a tracking signal on 3M or 2M |
| 20:00.14 | ds2 | unfortunately, the odd harmonics aren't in a useful range :( |
| 20:00.26 | ds2 | back to proper analog modules |
| 20:00.35 | prpplague | ds2: what kind of tracking signal? |
| 20:00.58 | *** join/#edev beewoolie (n=beewooli@66.147.194.238) |
| 20:01.38 | ds2 | prpplague: anykind... remember I was talking about having the hammer capture slow frame ratevideo on a model rocket or plane?Ｉãããã I just want to be ab if I can't find it |
| 20:02.03 | ds2 | bah arggg foreign lang input sprinkled on the keyboard :( |
| 20:02.33 | ds2 | I just want to be able to DF it if I can't find it |
| 20:02.37 | prpplague | ds2: no reasons you couldn't just use the pwm to generate that |
| 20:02.47 | ds2 | so a tone on 3M (FM BCB) or 2M ham would work |
| 20:02.51 | prpplague | ds2: assuming i understand what you are doing |
| 20:03.07 | ds2 | prpplague: I need a suitable carrier and looks like I'll need an external PLL or xtal |
| 20:03.48 | prpplague | ds2: how did you plan to use the lcd pixclock for a carrier? |
| 20:04.59 | ds2 | prpplague: had it been 101.x MHz, I'd gate it with a 1KHz square wave, then band pass it for 100MHz out; buffer that with a MPF102 JFET and have a wire as antenna |
| 20:05.18 | ds2 | an analog tuning Broadcast FM radio should pick it up |
| 20:05.37 | prpplague | looks |
| 20:06.24 | ds2 | but since the clock is max HCLK/(divisor * 2), I can't find suitable 3rd harmonic; anything else would have too little energy |
| 20:06.44 | ds2 | and build a doubler would be as much work as getting a proper analog module |
| 20:08.34 | prpplague | ds2: what is the target freq you need as an input? |
| 20:11.29 | ds2 | prpplague: easiest is if I can generate a 88MHz up to 108MHz; otherwise, 1/3 of that value is the output I want |
| 20:11.57 | ds2 | even better is if I can do 144-148MHz (1/3 of that should work) |
| 20:12.31 | prpplague | ds2: so 1/3 of 88MHz is acceptable? |
| 20:12.36 | ds2 | yes |
| 20:12.55 | prpplague | ds2: then the pwm on the s3c2410 should be able to do that |
| 20:13.04 | ds2 | 1/5 might be but I donno if the rise times on there can keep up |
| 20:13.17 | ds2 | prpplague: eh? elaborate.... |
| 20:13.34 | ds2 | though the timers were all in the KHz range? |
| 20:14.03 | *** join/#edev mbuf (n=user@c-7974e455.17-501-64736c14.cust.bredbandsbolaget.se) |
| 20:14.50 | prpplague | ds2: you can generate up to 33MHz with the pwm |
| 20:15.13 | ds2 | prpplague: what granularity ? |
| 20:15.26 | *** join/#edev prpplague_home (n=prpplagu@ppp-70-244-169-145.dsl.rcsntx.swbell.net) |
| 20:16.50 | prpplague | ds2: one sec, let me do a quick calc |
| 20:30.06 | prpplague | ds2: 16bit granularity |
| 20:31.28 | ds2 | hmmmm think we have a misunderstanding here |
| 20:31.58 | joe_bleau | If PWM maxes out at 33MHz, wouldn't the next step down be 16.something MHz? |
| 20:32.00 | ds2 | thinks how to ask this as apparently RF terminology and software terminology are different |
| 20:32.36 | ds2 | Is the PWM its own PLL for the base clock or does it use a divisor |
| 20:33.22 | ds2 | if it is a PLL on a loop withsay 32KHz xtal, I should be able to step it as such |
| 20:33.32 | prpplague | ds2: if the clock is 33.25MHz based, you can divide it by 65535 to get the interval of each step |
| 20:34.02 | prpplague | ds2: each step would be 507Hz |
| 20:34.16 | joe_bleau | I see where you're going ds2. Tweak the master clock rate to put the max PLL rate where you want it. |
| 20:35.34 | joe_bleau | Adjusting the PWM high and low time won't help, it'll be *way* too course. |
| 20:35.34 | prpplague | ds2: the pwm takes the PCLK and then uses the registers to furture divide it down |
| 20:35.59 | ds2 | prpplague: so you are saying I can generate 33.25MHz-507Hz*N ? |
| 20:36.19 | ds2 | dividing means I can only do 33.25/2, 33.25/3, etc |
| 20:37.27 | prpplague | ds2: no, the first divisor puts you in the range, then you can tweak it based on the resolution |
| 20:37.42 | prpplague | ds2: the formula would be |
| 20:39.22 | prpplague | ds2: (PCLK/1/2) / N , where N is a value from 0 to 65535 |
| 20:40.32 | prpplague | ds2: (66.25/1/2) / 500 = 66.25 KHz |
| 20:40.35 | ds2 | confused |
| 20:41.06 | ds2 | I need an output in the range of 33MHz |
| 20:41.56 | prpplague | ds2: np, you can use the PCLK/1/2 , then you can adjust the freq by 507Hz for each step |
| 20:41.59 | *** join/#edev sunrunner20 (n=jamesdoe@pool-71-123-184-40.dllstx.fios.verizon.net) |
| 20:42.09 | joe_bleau | You should be able to get a 33 MHz square wave from the PWM. |
| 20:42.19 | ds2 | then how do I get 33.125Hz - 507Hz then? |
| 20:42.21 | joe_bleau | You can't easily tune it in 507 Hz increments through the PWM module. |
| 20:42.25 | sunrunner20 | lol |
| 20:42.27 | sunrunner20 | When you say to remove the coating around the copper wire do you mean just at the tip or the whole wire that is used to coil around the magnet? |
| 20:42.30 | sunrunner20 | I have no experiences with copper wire whatsoever. |
| 20:42.34 | sunrunner20 | whoops, whould have been one post |
| 20:42.50 | sunrunner20 | but that is _SO_ worth the two :) |
| 20:43.02 | ds2 | unless i read it wrong,Ｉcan only do 33.125MHz or 3.125MHz/2 but nothing in between |
| 20:43.19 | ds2 | or 33.125MHz |
| 20:43.34 | joe_bleau | But ds2, I think you can tweak the PCLK to adjust the PWM output frequency. |
| 20:43.42 | prpplague | ds2: you are confusing the prescaller with the divisor |
| 20:44.10 | ds2 | prpplague: does it matter? they are all dividing it down |
| 20:44.42 | ds2 | let me ask this instead... what is the next highest output I can get besides 33.125MHz? |
| 20:45.25 | prpplague | ds2: 33.25 is the max running a PCLK of 66.5MHz |
| 20:45.46 | joe_bleau | If the PWM counts at 33.125MHz, does that mean you can get a 33MHz square wave out, or 16MHz square wave out? |
| 20:46.00 | joe_bleau | I assume you have to program the PWM to toggle when count hits azero. |
| 20:46.36 | ds2 | prpplague: what is the 2nd highest I can get |
| 20:46.45 | prpplague | joe_bleau: 33.25MHz provides a square wave |
| 20:46.56 | prpplague | ds2: 33.25MHz - 507Hz |
| 20:47.11 | ds2 | Ok |
| 20:47.21 | ds2 | then I have indeed misunderstood... I shall have to read the docs |
| 20:47.33 | joe_bleau | I still don't see how you're going to get 507 Hz. |
| 20:47.41 | prpplague | ds2: or 33.25MHz / 65535 = 507Hz |
| 20:47.49 | joe_bleau | If you program the down counter with 2 instead of 1, your frequency will drop in half. |
| 20:48.04 | joe_bleau | You only get 507Hz resolution when you change the downcounter from max to max-1. |
| 20:48.52 | joe_bleau | Think it thought: program the downcounter with 1. |
| 20:49.02 | joe_bleau | 1 tick later the PWM output pin flips polarity. |
| 20:49.16 | joe_bleau | Then program it with two. Now it takes two of the 66MHz ticks to count down. |
| 20:50.48 | ds2 | unless there is a loop somewhere turning this into a big pll |
| 20:51.29 | joe_bleau | That's why I was suggesting to look back to the main clock circuit. |
| 20:51.40 | joe_bleau | If you can find a PLL in front of the PWM module, you have a chance. |
| 20:51.52 | ds2 | *nod* hence the need to study the docs |
| 20:51.57 | joe_bleau | Run the PWM at max frequency, tweak the upstream PLL to move the PWM frequency around a bit. |
| 20:52.16 | joe_bleau | I'm looking at the datasheet, and it's not clear how much control you have over pclk. |
| 20:52.22 | ds2 | all else fails, a DDS module ;) |
| 20:52.38 | ds2 | or figure out the rules and use a ISM band chip |
| 20:53.07 | joe_bleau | LC oscillator? |
| 20:53.24 | ds2 | SAW based IIRC for the ISM |
| 20:53.31 | prpplague | joe_bleau: problem is the PCLK for the s3c2410 is fix |
| 20:53.35 | ds2 | the DDS requires an xtal reference |
| 20:53.35 | prpplague | joe_bleau: fixed |
| 20:54.11 | joe_bleau | It's not derived from MPLL? |
| 20:55.17 | prpplague | joe_bleau: system xtal is 12MHz, which is feed into the MPLL, and generates FCLK, HCLK, and PCLK, which then feeds the different devices |
| 20:55.24 | prpplague | joe_bleau: the pwm uses PCLK as the input |
| 20:56.29 | joe_bleau | Right, so where does PCLK come from? Isn't if just FCLK divided down? |
| 20:56.40 | joe_bleau | And doesn't FLCK come from MPLL in normal operation? |
| 20:57.28 | prpplague | joe_bleau: MPLL is configured to generate FLCK , HCLK can either be FCLK or FCLK/2 , PCLK can either be HCLK or HCLK/2 |
| 20:57.47 | joe_bleau | So if you adjust MPLL, doesn't PCLK change too? |
| 20:58.08 | ds2 | probally everything else would too |
| 20:58.22 | prpplague | joe_bleau: thats the point, you can't adjust MPLL without having to adjust _everything_ |
| 20:58.25 | joe_bleau | And it looks like there's an 8 bit divider (M) in the loop generating MPLL. |
| 20:58.30 | ds2 | UARTs changing baudrates would suck |
| 20:58.39 | joe_bleau | Exactly, you'd tweak everything. |
| 20:58.55 | prpplague | that would be a massive undertaking |
| 20:59.41 | joe_bleau | If he's not using the UARTs while the transmitter is active, they won't care. |
| 21:00.07 | joe_bleau | Maybe the ticker runs 10% fast or slow--does that cause a big problem? |
| 21:00.07 | ds2 | the entire system's notion of time would break |
| 21:00.24 | ds2 | and i will be using USB so that nails down a 48MHz clock somewhere |
| 21:01.03 | ds2 | external module it is |
| 21:01.21 | ds2 | more external things == long before I have something ready to fly :/ |
| 21:02.16 | joe_bleau | Weird. What if you used the ADC to sample audio. |
| 21:02.24 | joe_bleau | Then adjust the PLL to deviate the master clock. |
| 21:02.36 | joe_bleau | You could build a software controlled FM transmitter. |
| 21:03.11 | ds2 | that is what I am trying to get to but it doesn't look practical |
| 21:03.22 | ds2 | and all I need is a tone for RDF use |
| 21:03.37 | joe_bleau | Or have a lookup table for sine waves, then do software DDS to tweak the PLL. |
| 21:03.59 | joe_bleau | FM transmitter for test tones or modulated digital data. |
| 21:04.34 | joe_bleau | Wonder how sluggish the PLL loop filter is |
| 21:04.48 | ds2 | easier to usean external DDS |
| 21:05.26 | joe_bleau | DDS seems like overkill if you don't need the agility. |
| 21:06.19 | joe_bleau | LC oscillator with a varicap, driven from the PWM output? |
| 21:06.21 | prpplague | ds2: the best solution would be use either a i2s mclk or one of the clkout lines on the s3c2410, but neither are brought out on the hammer |
| 21:06.37 | joe_bleau | Tune LC to the FM band, then use PWM to generate your FM tone. |
| 21:06.49 | ds2 | prpplague: I don't think the I2S stuff goes fast enough anyways |
| 21:07.09 | ds2 | joe_bleau: I am not going down the road of fighting temp co. on an LC OSC :) |
| 21:07.20 | joe_bleau | How long do you intend to run this? |
| 21:07.28 | joe_bleau | Can follow the freq with the receiver? |
| 21:07.33 | ds2 | 1hr max |
| 21:07.43 | ds2 | no, this is used when the unit gets lost and i have to go find it |
| 21:10.11 | prpplague | joe_bleau: ahh indeed you are right, i was thinking that it was the divisor and not the count |
| 21:11.14 | joe_bleau | Reminds me of the new style PWM I once "invented" |
| 21:13.30 | prpplague | is not looking forward to working with wince |
| 21:13.47 | joe_bleau | Is it that bad? |
| 21:13.50 | ds2 | prpplague: when does it begin? |
| 21:14.18 | prpplague | last week |
| 21:14.53 | prpplague | joe_bleau: from what i hear, yea |
| 21:14.57 | prpplague | joe_bleau: you do wince dev? |
| 21:15.25 | joe_bleau | No, but I wouldn't panic based on rumors. |
| 21:15.46 | kergoth | there's a difference between panicing and not looking forward to it |
| 21:16.42 | prpplague | yea, not really worried if can do it or not, its just that my mindset has always been to work towards promoting linux, so thats the kind of dev i've been training/learning about |
| 21:17.03 | prpplague | to learn a whole new style is going to be problematic |
| 21:18.33 | prpplague | joe_bleau: plus i am totally disappointed with managment taking this direction |
| 21:19.35 | joe_bleau | Sure, that's understandable. |
| 21:20.08 | prpplague | in the last 8 months they have pretty much pissed off every one of our linux/oss customers |
| 21:21.24 | prpplague | we've already had 2 incidents in the month where the people in sales have told customers that we don't provide source code for any of products |
| 21:21.38 | prpplague | s/the month/the last month |
| 21:22.36 | joe_bleau | Do you have contact with the unhappy customers directly? |
| 21:23.06 | prpplague | i usually talk to the programmers/engineers |
| 21:23.10 | prpplague | so i hear about it all |
| 21:23.35 | joe_bleau | So the sales farce put you in the middle with their lies? That's dirty, but not unexpected. |
| 21:24.23 | joe_bleau | Now I'm mad. Gonna go watch the sales conference scene from Glengarry Glen Ross. |
| 21:24.50 | prpplague | joe_bleau: well it goes beyond that |
| 21:25.23 | prpplague | joe_bleau: the problem is the sales department is all new people, within the last 8 months, and none of them understand how to market to linux/oss customers |
| 21:25.50 | prpplague | joe_bleau: so the sales people are complaining that we need to switch to wince since it is easier for them to sell |
| 21:26.00 | joe_bleau | Sounds like a training problem. |
| 21:26.54 | joe_bleau | I'd do it as an experiment, then swing the axe if the sales didn't skyrocket, quickly. |
| 21:26.54 | prpplague | indeed |
| 21:27.04 | *** join/#edev RobotGUI (n=Dale@pool-96-225-210-142.ptldor.fios.verizon.net) |
| 21:27.15 | prpplague | is not in management and as such as no say in it |
| 21:33.44 | joe_bleau | No say at all? If you're valuable to the company, I suspect you have more say than you realize. |
| 21:34.11 | joe_bleau | Seniority as well. Surely you are respected by the powers that be. |
| 21:35.31 | prpplague | joe_bleau: nope, no say at all, the company management has been changing alot in last 18 months |
| 21:44.56 | *** join/#edev g1powermac_PB (n=g1powerm@unaffiliated/g1powermac) |
| 21:45.23 | prpplague | ho ho hum |
| 21:48.23 | g1powermac_PB | hey prpplague |
| 21:48.38 | prpplague | g1powermac_PB: hey bud, how ya feeling? |
| 21:48.54 | g1powermac_PB | getting better, finally :-) |
| 21:49.23 | prpplague | good deal |
| 21:49.44 | prpplague | g1powermac_PB: done any more with that quickbooks sdk? |
| 21:49.58 | g1powermac_PB | though exhausted, haven't been able to sleep well these past few days |
| 21:50.10 | g1powermac_PB | been waking up at like 5am |
| 21:50.31 | g1powermac_PB | not yet, but should be able to this week |
| 21:50.33 | prpplague | bummer |
| 21:51.08 | g1powermac_PB | just need to finish up a commercial I've been working on for the store first |
| 21:52.20 | prpplague | ahh ok |
| 21:53.25 | prpplague | is totally bummed out today |
| 21:53.32 | g1powermac_PB | oh? |
| 21:55.46 | prpplague | lots of stuff going on |
| 21:56.34 | chouimat|work | prpplague I just hope you don't suffer from insomnia too :) |
| 21:57.11 | g1powermac_PB | prpplague, meh, fun |
| 21:57.21 | g1powermac_PB | ugh, insomnia is a real pain |
| 21:57.31 | prpplague | chouimat|work: naw, have a good beer and i'm fine to go to bed |
| 21:57.33 | g1powermac_PB | I've suffered with it on and off for the past year and a half |
| 21:58.01 | g1powermac_PB | either can't fall asleep or just wake up at insanely early times thinking its later |
| 21:58.02 | chouimat|work | prpplague nice even if I take sleeping pills I wake up at 3am :( |
| 21:58.18 | g1powermac_PB | chouimat|work: yea, thats the worst |
| 21:58.46 | chouimat|work | g1powermac_PB: yup ... especially when you went to bed drunk at 1am |
| 21:58.57 | prpplague | i've got so much going on right now, its a wonder i've not gone nuts |
| 21:59.08 | g1powermac_PB | chouimat|work: my sleep schedule has been shifting really strangely this past 6 months |
| 21:59.30 | chouimat|work | prpplague you're nuts ... it's just it doesn't show up yet ... it's like some disease ... |
| 21:59.34 | g1powermac_PB | chouimat|work: I use to never wake up before 8 am, go to bed usually after 11pm |
| 21:59.40 | prpplague | chouimat|work: indeed |
| 21:59.53 | g1powermac_PB | chouimat|work: now I consistently wake up before 7am and fall asleep around 10pm |
| 22:00.38 | prpplague | i went to see my father over the weekend who is terminally ill, left me pretty upset |
| 22:00.51 | chouimat|work | prpplague ouch |
| 22:00.55 | g1powermac_PB | prpplague, oh |
| 22:01.03 | landley | prpplague: that sort of thing is why I left timesys. |
| 22:01.17 | prpplague | landley: the management stuff? |
| 22:01.31 | landley | Not specifically switching to wince, but management deciding "we have a good product now, it's selling well, management's done its job and can go home, it's all sales from here on in". |
| 22:01.44 | landley | And the person in charge of sales doing a power grab. Tried to get _me_ to report to her more than once. |
| 22:02.07 | joe_bleau | How'd that work out? |
| 22:02.08 | prpplague | landley: ahh |
| 22:02.12 | landley | When I left, ther were two remaining engineers (one who couldn't leave due to visa concerns, and jeremiah). |
| 22:02.34 | prpplague | landley: yea, sjhill was very unhappy at timesys |
| 22:03.14 | landley | Plus the CEO quit suddenly once the extent of the damage became apparent, and the venture capitalists stepped in to run the company. |
| 22:03.42 | landley | When your engineering team of over a dozen people is down to 6 and two of the remaining ones give their two weeks notice on the same day... |
| 22:03.47 | landley | You're kind of screwed. |
| 22:04.11 | landley | I'm told it's all better now. |
| 22:04.39 | landley | They hired a bunch of fresh graduates from carnegie mellon, a school where I never _could_ find the Linux department. |
| 22:04.51 | prpplague | hehe |
| 22:04.52 | landley | (But where they named a large building after bill gates because he gave them money to do so.) |
| 22:04.55 | chouimat|work | it's a virtual one |
| 22:06.51 | prpplague | landley: yea, my main problem is the $$, when i took this job i knew it was about $20k less than it should be, but it was an opportunity to work with alot of people in other areas, but now they've changed all that |
| 22:07.23 | prpplague | landley: so settling for $20k less and no extras, i.e. training/classes/conferences |
| 22:07.28 | prpplague | landley: isn't cutting it |
| 22:09.36 | landley | prpplague: shop around your resume then. |
| 22:10.11 | prpplague | landley: yea, dfw just isn't a hub of OSS dev, gonna take awhile to find something that fits |
| 22:10.37 | landley | I went up to work at timesys for the people; the money was ok but not great (noticeably less than I could have earned elsewhere, and the city was more expensive), and I really didn't like the city. Moved back to Austin after the people were gone. |
| 22:10.52 | CosmicPenguin | Tivos can now order pizzas from Dominos |
| 22:10.53 | landley | Wave a resume at Cannonical, they're coming out with an arm version of Ubuntu. |
| 22:10.56 | CosmicPenguin | hah! |
| 22:11.12 | CosmicPenguin | yep - netbooks for everybody |
| 22:11.12 | landley | http://arm.com/news/23761.html |
| 22:11.49 | kergoth | isn't sure how he feels about his job at the moment |
| 22:12.19 | prpplague | landley: yea someone else mention cannonical too |
| 22:13.26 | landley | Worth a shot. David Mandala (head of Ubuntu Mobile and EMbedded) telecommutes from Dallas. Or one of the suburbs of Ft Worth, anyway. |
| 22:13.36 | kergoth | thought greg k-h's presentation ripping canonical about their conributions was pretty funny |
| 22:13.54 | landley | He's davidm over on #ubuntu-mobile |
| 22:14.06 | landley | kergoth: when was this, the plumber's conference? |
| 22:14.09 | kergoth | aye |
| 22:14.17 | kergoth | wasn't there, sadly, just downloaded the videos |
| 22:14.43 | prpplague | landley: yea, i've spoken to david a few times |
| 22:15.01 | landley | kergoth: where do I download the videos? |
| 22:15.06 | prpplague | landley: not sure how much kernel work they'd need, more than likely they need more userland stuff |
| 22:15.16 | landley | shrugs. |
| 22:15.42 | CosmicPenguin | kergoth: yeah it was pretty good |
| 22:15.53 | kergoth | tries to remember where he saw it |
| 22:15.58 | kergoth | oh, i think its on google video, actually |
| 22:16.01 | CosmicPenguin | can't argue the truth though - I took it back to AMD as an example of why depending on the distributions was flawed |
| 22:19.12 | prpplague | landley: i think my problem is i'm the type of person that really does't like changes in a work environment |
| 22:19.28 | landley | prpplague: join the club. :) |
| 22:20.02 | *** join/#edev kergoth (n=kergoth@65.200.49.156) |
| 22:21.20 | *** join/#edev Ian_Daniher|xo (n=it@69.61.230.246) |
| 22:26.40 | agaffney | prpplague: I'd say that's quite common :P |
| 22:26.47 | agaffney | I switched jobs just because of that |
| 22:26.53 | agaffney | of course, they were evil, evil changes |
| 22:33.06 | prpplague | plus i have a serious inferiority complex about my skills |
| 22:34.38 | landley | prpplague: you know more than me about numerous things. :) |
| 22:35.54 | chouimat|work | ok time to go home ... or maybe a stop at the pub first |
| 22:36.07 | prpplague | landley: what i worry about is that it would take someone else a week to do something and it might take me 2 weeks |
| 22:42.41 | prpplague | landley: of course i always comfort myself knowing that CosmicPenguin will take 4 weeks |
| 22:42.45 | prpplague | jokes with CosmicPenguin |
| 22:43.09 | kergoth | prpplague: perhaps, but the best way to learn is to dive in like that, whether it takes a bit longer or no. there's plenty i'd take way too long to do |
| 22:45.25 | landley | prpplague: there's always somebody better than you at any given task, sure. |
| 22:45.32 | kergoth | indeed |
| 22:45.44 | landley | But I've generally found that A) they're busy, B) there are other things they suck at. |
| 22:45.45 | kergoth | can't let that interfere with your life or your goals |
| 22:45.51 | landley | Mix of skills, sort of thing. |
| 22:46.22 | prpplague | yea, that's what i've always tried to do |
| 22:47.01 | kergoth | also, there's a lot to be said for the ability to admit when you don't know something. most companies would rather have someone who's willing to ask for help when needed, but doesn't necessarily know as much off the top of their head, then someone who thinks they're all that, and just battles a difficult problem for weeks when they could've done it in half the time by talking to their coworkers |
| 22:47.07 | kergoth | imo anyway |
| 22:47.27 | prpplague | kergoth: yea, i've seen that alot before |
| 22:47.27 | beewoolie | kergoth: amen |
| 23:17.36 | *** join/#edev redblue (i=star@ppp059.108-253-207.mtl.mt.videotron.ca) |