| 00:35.25 | *** join/#edev bsdfox__ (n=h36sa@c-71-197-73-68.hsd1.ca.comcast.net) |
| 00:45.48 | *** join/#edev roxfan2 (n=dunno@124.58-242-81.adsl-dyn.isp.belgacom.be) |
| 02:20.21 | *** join/#edev CoR__ (n=quassel@111.69.241.202) |
| 02:31.42 | *** join/#edev SpeedEvil (n=user@87.115.66.220) |
| 02:57.05 | *** join/#edev lyakh_ (n=lyakh@p57BD1AFD.dip0.t-ipconnect.de) |
| 03:57.45 | *** join/#edev JoeLlama (n=snork@206.169.63.10) |
| 04:06.01 | monkey_ | Mirell: ? |
| 04:06.24 | Mirell | monkey_: Yes? |
| 04:07.07 | *** part/#edev flyback (n=yukon@c-24-3-254-46.hsd1.pa.comcast.net) |
| 04:07.16 | *** join/#edev f1yback (n=yukon@c-24-3-254-46.hsd1.pa.comcast.net) |
| 04:07.20 | f1yback | I need to bring this up again |
| 04:07.21 | monkey_ | i need a hand with hw |
| 04:07.26 | monkey_ | brb, |
| 04:07.31 | f1yback | http://cgi.ebay.com/BROKEN-Polaroid-7-Portable-DVD-Player-Red-PDX-0073_W0QQitemZ260499009070QQcmdZViewItemQQptZDVD_Players_Recorders?hash=item3ca6f36e2e |
| 04:07.36 | f1yback | why am I pasting this in this channel |
| 04:07.39 | f1yback | because of this |
| 04:07.45 | Mirell | f1yback: I was about to ask... |
| 04:07.54 | f1yback | http://www.getlofi.com/?p=1233 |
| 04:08.10 | f1yback | = cheap embedded display for old devices that had composite video out |
| 04:08.25 | f1yback | I really need to get prpp to clear that old ban on me |
| 04:08.43 | f1yback | anyways there you go |
| 04:08.52 | f1yback | has 6 although not all of them have video in |
| 04:09.02 | monkey_ | I'm just really effin confused |
| 04:09.02 | f1yback | there is a constant supply on ebay |
| 04:09.31 | f1yback | just make sure before you bid, it's a model that has video in (if it says it supports xbox, etc it does for sure) and that the breakage is listed and not screen problems |
| 04:09.40 | f1yback | usually #1 problem is "won't read disc" |
| 04:10.15 | f1yback | i'm getting a mini-itx this week with tv out |
| 04:10.39 | f1yback | and as soon as I get a pci video card with tv out that supports large endian, one is going on my dht-walnut board |
| 04:10.48 | f1yback | yeah I am the guy that told this channel about that |
| 04:10.54 | f1yback | well the one selling for $20 :P |
| 04:11.28 | f1yback | Mirell, there you go, sorry if you though I was just spamming ebay or something |
| 04:12.13 | f1yback | damn I wish my webpal had not died |
| 04:12.18 | f1yback | this would have been a perfect combo :( |
| 04:14.30 | f1yback | btw that page |
| 04:14.42 | f1yback | you don't have to hack anything to use a portable dvd player as a cheap display |
| 04:14.50 | f1yback | he just did that to remove all the dead pieces to shrink the footprint |
| 04:15.44 | f1yback | crawls back into his hole |
| 04:26.46 | monkey_ | hey Mirell do I find on the schematic the port that a accessory is plugged into and use that address from the chip manual? |
| 04:32.42 | Mirell | monkey_: I am not at all the guy to ask on hardware stuff |
| 04:33.11 | monkey_ | well, it's not really a hardware question. i'm supposed to scan the buttons and if button 1 is pushed i'm supposed to turn on an led |
| 04:33.15 | monkey_ | other button, the lcd backlight |
| 05:07.53 | *** join/#edev mbuf (n=mbuf@118.102.130.6) |
| 05:17.27 | *** join/#edev JoeLlama (n=snork@206.169.63.10) |
| 05:46.56 | *** join/#edev chiques (n=working_@pool-71-189-111-152.lsanca.fios.verizon.net) |
| 05:49.11 | *** join/#edev linac (n=lin@59.61.124.12) |
| 06:00.18 | *** join/#edev Madsy (n=Madsy@fu/coder/madsy) |
| 06:44.07 | *** join/#edev Madsy^ (n=Madsy@ti0207a340-0474.bb.online.no) |
| 06:55.31 | *** join/#edev tsjsieb (n=tsjsieb@dejongbeheer.nl) |
| 06:55.47 | *** join/#edev lyakh (n=lyakh@p57BD1AFD.dip0.t-ipconnect.de) |
| 07:17.39 | *** join/#edev hw__ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 07:17.57 | *** join/#edev JoeLlama (n=snork@206.169.63.10) |
| 09:32.13 | *** join/#edev florian_kc (n=fuchs@Maemo/community/contributor/florian) |
| 09:56.26 | *** join/#edev flo_lap (n=fuchs@Maemo/community/contributor/florian) |
| 10:35.40 | *** join/#edev Redhatter (n=stuartl@gentoo/developer/redhatter) |
| 10:38.14 | *** join/#edev Redhatter (n=vk4msl@2001:388:f000:0:0:0:0:71) |
| 11:06.53 | *** join/#edev kazaa_lite (n=msaleem@cpc6-slam5-2-0-cust171.2-4.cable.virginmedia.com) |
| 11:21.53 | *** join/#edev pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB) |
| 12:21.39 | *** join/#edev linac (n=lin@114.114.136.164) |
| 12:31.06 | *** join/#edev Redhatter (n=vk4msl@2001:388:f000:0:0:0:0:71) |
| 12:48.08 | *** join/#edev kazaa_lite (n=msaleem@94-193-98-124.zone7.bethere.co.uk) |
| 13:05.22 | *** part/#edev kazaa_lite (n=msaleem@94-193-98-124.zone7.bethere.co.uk) |
| 13:32.12 | *** join/#edev SpeedEvil (i=1000@tor/regular/SpeedEvil) |
| 13:32.36 | *** join/#edev SpeedEvil (i=1000@tor/regular/SpeedEvil) |
| 13:43.40 | *** join/#edev SpeedEvil1 (i=1000@tor/regular/SpeedEvil) |
| 13:51.12 | *** join/#edev mne (n=mne@unaffiliated/mne) |
| 13:51.54 | mne | Hi. I have the uart of a sam9260 connected to a DSP board. At 480600 baud an lower rates data transfer works just fine. But at 921600 it doesn't work anymore. All the rates I tried are supported by both the DSP and the sam9260 controller. The baud error is under 1.8% (according to the sam9260 datasheet the error should not be below 5%). I also looked at the signal coming from the sam9260 with a scope: It looks fine and the timing seems to be correct. The cables |
| 13:51.54 | mne | <PROTECTED> |
| 13:52.40 | SpeedEvil | I would compare the sending speeds at 921600 |
| 13:52.47 | SpeedEvil | and see if they are mutually within 5% or so |
| 13:55.05 | mne | well, the calculated error for the 921600 baud rate is below 1,7%. From the measurement results it is hard to tell. I guess I could measure the time for one transfered packet and then divide it by the number of bits (including start and stop bits). This would give me the period time which can be used to calculate the frequency |
| 13:56.14 | SpeedEvil | I was meaning look at it on the scope |
| 13:56.28 | mne | But the time of the whole packet it exactly half the time of the packet sent at 460800, so it should be fine |
| 13:56.44 | SpeedEvil | hav eyou done raw tests? |
| 13:56.54 | SpeedEvil | or can you not control one side |
| 13:56.59 | SpeedEvil | I mean stuff like send AAAAAAAAAAAAAAaaa |
| 13:57.00 | SpeedEvil | ... |
| 13:57.10 | SpeedEvil | Without any possible packet overhead |
| 13:57.26 | mne | ah, ok. Unfortunately the DSP chip is a proprietary chip, so I cannot control anything there |
| 13:57.34 | SpeedEvil | you are following whatever protocol closely - may there be any delay time? |
| 13:58.15 | mne | a delay time between what exactly ? |
| 13:58.21 | SpeedEvil | or if you add a delay after each packet to delay the packet to the equivalent of the original sending rate, does it work |
| 13:58.33 | mne | no |
| 13:58.59 | mne | The thing is that I sent one packet to the DSP and then wait for the response. But the DSP never sends a response |
| 13:59.15 | SpeedEvil | ah |
| 14:00.12 | SpeedEvil | Sorry - nothing comes to mind. |
| 14:00.19 | mne | At the moment my guess is that there is something wrong on the DSP side. But before writing mails to the manufacturer I would like to ensure that it really is no problem on my side. |
| 14:01.01 | mne | Hmm, thanks anyway |
| 14:03.23 | SpeedEvil | Other than perhaps hook it up to a PC serial port that can do that speed and verify that you're actually getting all the bits in the right order |
| 14:04.18 | *** join/#edev GPSFan (n=kenm@64.92.145.112) |
| 14:04.18 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:17.08 | prpplague | GPSFan: greetings |
| 14:17.30 | GPSFan | prpplague: morning.. |
| 14:18.22 | mne | SpeedEvil, I tried that already but so far it seems that I have no serial ports here that support a baudrate of 921600. The only thing I have is a USB<-->RS232 dongle that "supports" that baud rate. But the signal coming out of it looks terrible on the scope. |
| 14:22.16 | SpeedEvil | ah |
| 14:22.46 | SpeedEvil | mne: you can observe on a scope the repeating bit-pattern for AAAAAAAAAAAAAAA/... and pick out any errors moderately easily |
| 14:24.48 | mne | hmm, let me just test that .. |
| 14:27.25 | mne | SpeedEvil, it looks just fine. The signal is consistent |
| 14:28.03 | SpeedEvil | :/ |
| 14:28.20 | SpeedEvil | It's a 5v/3.3V signal? |
| 14:30.52 | mne | 3.3V |
| 14:32.12 | SpeedEvil | And you're sure it's not a marginal thresholds thing? |
| 14:33.42 | mne | Well, both the low and the high signal are in the specified thresholds. Besides the signal levels are the same for different baud rates and as it is working with baud rates up to 460800 I doubt that it is a threshold thing |
| 14:34.11 | mne | It's really strange .. |
| 14:34.37 | mne | Maybe I should just write a mail to the manufacturer |
| 14:35.17 | SpeedEvil | mne: if the thresholds are marginal, then you ight get it barely making at low speed, and just missing at low. |
| 14:35.23 | SpeedEvil | at high |
| 14:39.29 | mne | Hmm. The high level is exactly 3.3V. The low level is about 600mV. The DSP is based on a TMS320F2810 core. The datasheet says that a low level is everything below 800mV. |
| 14:40.23 | mne | a high level is 2-3.3V |
| 14:43.50 | mne | Do you think the low level of 600mV might be a problem ? If so, how can I pull down the signal to GND ? The DSP input pins have internal pull-ups |
| 14:44.04 | SpeedEvil | 600 is a fair bit under 800 |
| 14:44.12 | mne | yes |
| 14:44.14 | SpeedEvil | however - it only needs 200mV of ground-bounce |
| 14:44.20 | SpeedEvil | I would try 1K to ground |
| 14:44.55 | mne | Hmm ok. So the internal pull-ups would be "overpowered" |
| 14:45.24 | mne | But I need to verify first how much current my at91 controller can drive |
| 14:47.09 | mne | at 3.3V a 1k resistor would give I = U/R ==> I = 3.3/1000 = 3,3mA, right ? |
| 14:47.18 | mne | that shouldn't be a problem |
| 14:49.46 | mne | the controller can drive up to 16mA on the corresponding pin. So let me get a 1k resistor any try :) |
| 14:54.35 | SpeedEvil | it will cause the voltage to drop a little |
| 14:54.36 | SpeedEvil | yes |
| 14:54.56 | SpeedEvil | Also more decoupling on the DSP may help |
| 14:55.05 | SpeedEvil | marmite sandwitch |
| 14:55.06 | SpeedEvil | oops |
| 14:59.28 | mne | With the 1k resistor added the result and the voltage levels is still the same |
| 14:59.43 | mne | The low level is still at ~600mV |
| 14:59.48 | SpeedEvil | hmm. |
| 15:00.03 | SpeedEvil | I'd next try putting a 500 ohm in series with the tx too |
| 15:00.29 | mne | for what reason ? |
| 15:00.36 | mne | I'm just curious |
| 15:02.41 | SpeedEvil | To act as a voltage divider, to drop the voltage a bit without overpowering the driver |
| 15:07.17 | mne | now my scope doesn't measure anything anymore (if it is attached after the 500 ohm resistor) ;) At 460800 it still works but not at 921600. |
| 15:08.13 | SpeedEvil | odd |
| 15:09.50 | mne | How come that I can not pull the signal down to GND ? GND of the controller PCB and the DSP PCB are directly connected. |
| 15:10.45 | SpeedEvil | something must be pulling it up |
| 15:11.12 | SpeedEvil | If there is 200mV of ground-bounce in the circuit, that means you will get occasional coirruptions of the rs232 signal |
| 15:12.42 | mne | Can I measure ground bounce somehow ? As GND of both PCBs is connected I guess I won't be able to measure it by just attaching the scope to the GND layers of the PCBs .. |
| 15:14.01 | mne | The strange thing is that the 1k resistor did not pull down the signal a bit. Not even for 10mV :( |
| 15:17.02 | SpeedEvil | off |
| 15:17.03 | SpeedEvil | odd |
| 15:17.05 | mne | On the other side, wouldn't be 200mV ground bounce much ? I do not have any high current consumers on the boards and all voltage inputs on the ICs have decoupling caps |
| 15:17.27 | SpeedEvil | the DSP will use a widely varying amount of current. |
| 15:18.04 | mne | but is has 100nf decoupling caps on all voltage inputs |
| 15:18.05 | SpeedEvil | If this locally disrupts the ground - it can change by amps/us easily - by only 200mV - this will cause bit errors. |
| 15:18.22 | SpeedEvil | I would try first working out how to drop that 600mV. |
| 15:18.30 | SpeedEvil | Of course htis could be a complete red herring. |
| 15:19.04 | mne | sure, but it seems to be reasonable at least |
| 15:23.09 | mne | At the moment the DSP board uses the power coming from the voltage regulator on the controller board. I will connected it to an external power supply, maybe that makes things different ? |
| 15:25.33 | mne | nope, still the same levels |
| 15:27.23 | SpeedEvil | afk, sorry, good luck |
| 15:28.01 | mne | thanks. see you |
| 15:35.42 | *** join/#edev SpeedEvil (i=1000@tor/regular/SpeedEvil) |
| 15:55.39 | prpplague | doh out of printer paper |
| 15:56.11 | chouimat | morning |
| 16:00.22 | *** join/#edev Geep (n=JoeBelow@bas21-toronto12-1242364376.dsl.bell.ca) |
| 16:44.39 | *** join/#edev linac (n=lin@114.114.132.20) |
| 17:04.49 | *** join/#edev mne (n=mne@unaffiliated/mne) |
| 17:15.25 | *** join/#edev mbuf (n=mbuf@122.169.187.35) |
| 17:32.49 | *** join/#edev SpeedEvil (n=user@tor/regular/SpeedEvil) |
| 17:56.34 | *** join/#edev CosmicPenguin (n=nobody@66-162-99-237.static.twtelecom.net) |
| 18:15.12 | *** join/#edev chiques (n=working_@pool-71-189-111-152.lsanca.fios.verizon.net) |
| 18:18.05 | Mirell | Huzzah, I get 4 weeks of vacation |
| 18:19.29 | prpplague | wow |
| 18:19.52 | prpplague | Mirell: the most i've ever taken was 6 days |
| 18:20.02 | Mirell | That's just sad. |
| 18:20.05 | prpplague | indeed |
| 18:23.25 | Mirell | Did you at least cash out the unused vacation? |
| 18:24.01 | prpplague | yea they had to pay me for that |
| 18:24.36 | Mirell | Okay, so you're not completely stark raving mad. |
| 18:34.16 | chouimat | Mirell: I'm still not sure about this ;) |
| 18:34.49 | CosmicPenguin | Yeah, the jury is still deliberating |
| 18:49.32 | chiques | Are there any transient simulators for Ubuntu (Linux) that are free or for sale? |
| 19:23.31 | *** join/#edev _new_Dev_ (n=blabla@41.227.203.176) |
| 19:24.21 | _new_Dev_ | Hello! |
| 19:33.05 | *** join/#edev florian (n=fuchs@Maemo/community/contributor/florian) |
| 20:00.58 | *** join/#edev ulfsen (n=ulf@f053084190.adsl.alicedsl.de) |
| 20:11.10 | *** join/#edev Ian_Daniher (n=it@nat-pool-128-88.olin.edu) |
| 20:57.25 | *** join/#edev likewise (n=chatzill@82-171-51-231.ip.telfort.nl) |
| 21:53.38 | *** join/#edev daowee_ (n=daowee@ua-83-227-154-166.cust.bredbandsbolaget.se) |
| 22:20.03 | *** join/#edev kergoth (n=kergoth@ip24-251-170-95.ph.ph.cox.net) |
| 22:26.02 | *** join/#edev landley (n=landley@rrcs-24-227-244-49.sw.biz.rr.com) |
| 22:26.02 | *** mode/#edev [+o landley] by ChanServ |
| 22:26.38 | landley | waves at prpplague |
| 22:26.45 | prpplague | landley: greetings |
| 22:26.59 | landley | How goes stuff? |
| 22:30.36 | prpplague | landley: busy busy, working on a big project |
| 22:34.21 | prpplague | landley: how are things on your end? |
| 22:34.32 | prpplague | landley: i read that you got an interview with google |
| 22:34.37 | prpplague | landley: when is that suppose to happen? |
| 22:38.14 | *** join/#edev flavioribeiro (n=flavio@187.64.44.60) |
| 22:42.45 | landley | prpplague: thursday. |
| 22:47.34 | SpeedEvil | Good luck. |
| 22:54.00 | *** join/#edev adyer (n=adyer@173-15-79-94-Illinois.hfc.comcastbusiness.net) |
| 23:00.19 | landley | SpeedEvil: thanks. |
| 23:36.22 | Mirell | Hmm... |
| 23:44.51 | landley | Mirell: hmmm? |
| 23:50.22 | *** join/#edev dieter__ (n=dieter@dslb-088-067-189-143.pools.arcor-ip.net) |