| 00:22.30 | Dr{Who} | wow that was a long meeting.. |
| 00:22.34 | Dr{Who} | my head hurts.. |
| 00:25.22 | Mirell | Heh |
| 00:35.08 | Dr{Who} | on a friday too.. dam. i try to avoid that if at all possible.. |
| 00:41.18 | *** join/#edev davidc__ (n=davidc__@S01060002b360aacd.vc.shawcable.net) |
| 00:48.17 | *** join/#edev davidc__ (n=davidc__@S01060002b360aacd.vc.shawcable.net) |
| 01:02.11 | Mirell | Indeed. |
| 01:02.31 | Mirell | I just spent about three hours figuring out how to get rid of all my credit card debt in a year. |
| 01:02.45 | Mirell | It involves around eight balance transfers at various times, and living on teh cheap, but yay. |
| 01:03.28 | Dr{Who} | ya i hear that.. im still wondering how things will shape up in the next few years. |
| 01:20.05 | *** join/#edev chouimat (n=dieu@r2351064.cidc.net) |
| 01:59.00 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 02:15.29 | Dr{Who} | hay robot.. |
| 02:15.47 | RobotGuy | Dr{Who}: Greets |
| 02:16.35 | Dr{Who} | well i have a c compiler kinda working.. |
| 02:16.45 | RobotGuy | Cool |
| 02:16.50 | Dr{Who} | i had to blow up my init a LOT!! i was shocked how fast it grew just adding a few peices |
| 02:17.07 | RobotGuy | I had forgotten to upload my jffs2 image the last time I built it. |
| 02:17.21 | Dr{Who} | ahh that was the console issue? |
| 02:17.27 | RobotGuy | Dr{Who}: That's why I went to JFFS2 on Flash. |
| 02:17.30 | RobotGuy | Dr{Who}: Yes. |
| 02:18.13 | Dr{Who} | Ya i may take that route.. i modified my kernel to use the 0x440000 format option |
| 02:18.50 | Dr{Who} | cat /proc/meminfo |
| 02:18.56 | Dr{Who} | what does your first 2 lines look like. |
| 02:19.08 | Dr{Who} | MemTotal: 30820 kB |
| 02:19.08 | Dr{Who} | MemFree: 20588 kB |
| 02:21.56 | RobotGuy | MemTotal: 30396 kB |
| 02:21.57 | RobotGuy | MemFree: 25076 kB |
| 02:22.37 | Dr{Who} | cat /proc/mtd |
| 02:22.46 | Dr{Who} | dev: size erasesize name |
| 02:22.46 | Dr{Who} | mtd0: 00440000 00020000 "System" |
| 02:22.46 | Dr{Who} | mtd1: 00bc0000 00020000 "JFFS2 |
| 02:26.59 | RobotGuy | dev: size erasesize name |
| 02:26.59 | RobotGuy | mtd0: 00140000 00020000 "System" |
| 02:26.59 | RobotGuy | mtd1: 00ec0000 00020000 "JFFS2" |
| 02:28.24 | RobotGuy | I need to create at least two packages for buildroot. Never done that before. They will both require Python. |
| 02:30.36 | Dr{Who} | its not too bad.. you just find a package that is close to what you want copy it and make changes to the .mk and Config.in file |
| 02:31.10 | Dr{Who} | then you link it into your build root by modifying the Config.in /packages/ |
| 02:31.57 | Dr{Who} | "most" of the make files are no problem. the problems are in building multiple bin's say its a package that needs to install several bin's |
| 02:31.58 | Dr{Who} | or libs. |
| 02:32.04 | Dr{Who} | that gets tricky.. |
| 02:32.21 | Dr{Who} | python??? |
| 02:32.30 | Dr{Who} | ohhh.. explain.. |
| 02:32.33 | RobotGuy | Yes, PySerial and PyRo for Python |
| 02:32.57 | Dr{Who} | oh well hmm thats interesting because its not exactly compiled to use uClibc |
| 02:33.11 | RobotGuy | Dr{Who}: I know - it's a bit different. |
| 02:33.27 | Dr{Who} | you could just make that on your host and then have it install into a /someplace/ folder |
| 02:33.37 | Dr{Who} | and then just add the files to your jffs2 |
| 02:33.42 | RobotGuy | Dr{Who}: I will need a module for SQLite access also for Python. |
| 02:34.08 | RobotGuy | Dr{Who}: I'd prefer to do it within buildroot and supply a patch to the team. |
| 02:34.15 | Dr{Who} | hmm that may be harder. is the SQLite python module native python or does it require a c module |
| 02:34.28 | RobotGuy | I think it is mostly copying stuff to site-packages |
| 02:34.49 | Dr{Who} | not sure about that. i have not seen any packages that dont actually build using the cross compiler.. |
| 02:34.59 | RobotGuy | And in the case of PyRo, not including those two modules. |
| 02:35.42 | RobotGuy | Dr{Who}: I am not sure about the SQLite module for Python. |
| 02:35.49 | Dr{Who} | goes looking at packages.. |
| 02:37.55 | RobotGuy | I'll be adding Phenny, as well as DJango or Pylons also. |
| 02:38.37 | Dr{Who} | ok sinse you are not doing a real make use the package customeize as your template.. it is just a series of commands.. |
| 02:38.52 | RobotGuy | Dr{Who}: Oh, OK |
| 02:38.54 | Dr{Who} | hmm |
| 02:39.04 | Dr{Who} | but it does not download the package.. |
| 02:40.00 | Dr{Who} | if you look at almost all packages you can see quickly how they setup variables to hold the values such as the download URL etc |
| 02:40.19 | Dr{Who} | the idea is that its then easy to modify the package to change the version you will download |
| 02:40.37 | RobotGuy | Dr{Who}: I follow that much |
| 02:42.18 | RobotGuy | I've had to change a few version numbers and download sites. |
| 02:42.18 | Dr{Who} | its standard makefile type scripting.. it follows dependencies.. |
| 02:42.37 | RobotGuy | I don't grok standard Makefile scripting. |
| 02:42.39 | Dr{Who} | the parent make will call 'packagename-source' and then 'packagename' |
| 02:42.50 | Dr{Who} | or well make 'packagename-source' then make 'package' |
| 02:43.17 | Dr{Who} | wipe-source: $(DL_DIR)/$(WIPE_SOURCE) $(DL_DIR)/$(WIPE_PATCH) |
| 02:43.21 | Dr{Who} | is in the wipe package. |
| 02:44.15 | Dr{Who} | this will force the make to look for "$(DL_DIR)/$(WIPE_SOURCE)" |
| 02:44.31 | Dr{Who} | that will then have make find |
| 02:44.32 | Dr{Who} | $(DL_DIR)/$(WIPE_SOURCE): |
| 02:44.33 | Dr{Who} | <PROTECTED> |
| 02:44.41 | Dr{Who} | and wala you now have a file downloaded. |
| 02:44.51 | Dr{Who} | then the second part of that line. |
| 02:45.00 | Dr{Who} | <PROTECTED> |
| 02:45.27 | Dr{Who} | will then find the section that matches that name.. |
| 02:45.49 | Dr{Who} | dont be confused by $ those are varibles.. but the make resolves them to find the section to run. |
| 02:46.25 | Dr{Who} | then the parent make for buildroot will call |
| 02:46.27 | Dr{Who} | make wipe |
| 02:46.34 | Dr{Who} | and it will find this |
| 02:46.35 | Dr{Who} | wipe: uclibc $(TARGET_DIR)/$(WIPE_TARGET_BINARY) |
| 02:46.57 | Dr{Who} | uclibc is the first thing.. ok does it exist? well it was compiled already so it will be valid and move on. |
| 02:47.25 | Dr{Who} | next it will do this section |
| 02:47.26 | Dr{Who} | $(WIPE_DIR)/$(WIPE_BINARY): $(WIPE_DIR)/.configured |
| 02:47.26 | Dr{Who} | <PROTECTED> |
| 02:47.26 | Dr{Who} | <PROTECTED> |
| 02:47.58 | Dr{Who} | and so on and so on |
| 02:48.02 | Dr{Who} | following the chain. |
| 02:48.33 | Dr{Who} | thats how a makefile works.. well as best as i understand. I am not a big fan of makefiles. : c ) |
| 02:49.08 | Dr{Who} | the only thing to remember on buildroot's makes is the few hard coded entry points? or do more exist? |
| 02:49.13 | Dr{Who} | packagename-source |
| 02:49.17 | Dr{Who} | packagename |
| 02:49.27 | RobotGuy | :) Thanks. I learn what I need to about make when I absolutely can't avoid it. |
| 02:49.31 | Dr{Who} | probably a clean? |
| 02:49.44 | RobotGuy | Yeah, there is a clean. |
| 02:49.53 | Dr{Who} | ya so clean and dirclean |
| 02:51.02 | Dr{Who} | so at buildroot level if you call 'make wipe' |
| 02:51.07 | Dr{Who} | it will just make that package. |
| 02:51.14 | Dr{Who} | a make wipe-clean will clean it |
| 02:51.29 | Dr{Who} | or a make clean will run make xxxxx-clean on all packages. |
| 02:52.25 | Dr{Who} | this is cute.. |
| 02:52.27 | Dr{Who} | $(WIPE_DIR)/.configured: $(WIPE_DIR)/.unpacked |
| 02:52.27 | Dr{Who} | <PROTECTED> |
| 02:52.43 | Dr{Who} | so once it runs this it will touch a file called .configured |
| 02:53.06 | Dr{Who} | and next time it wont run this because .configured exists so the make will not process that section. |
| 02:54.03 | Dr{Who} | anyway you can build anything you want including your python you just have to break it down into steps and embed your bash code if needed into your .mk |
| 02:55.16 | Dr{Who} | ok i have to head to the store get foood for the family and hope they will do the cooking :( i dont want to shop and cook today.. |
| 02:55.40 | RobotGuy | OK |
| 02:56.18 | Dr{Who} | i will be at the comic book convention on sunday my daughter will have a booth.. but i hope to be on some this weekend. |
| 05:36.30 | *** join/#edev DrZimmerman (n=theo@tbaumgartner.broker.freenet6.net) |
| 05:37.24 | *** join/#edev dirk2 (n=dirk@F3242.f.strato-dslnet.de) |
| 05:57.13 | *** join/#edev bsdfox_ (n=h36sa@c-71-197-73-68.hsd1.ca.comcast.net) |
| 06:01.39 | *** join/#edev drath_ (i=vmaster@p5B07CE3C.dip.t-dialin.net) |
| 06:10.08 | *** join/#edev ccjoe (n=ccjoe@75-169-37-101.slkc.qwest.net) |
| 06:17.24 | *** join/#edev ccjoe (n=ccjoe@75-169-37-101.slkc.qwest.net) |
| 06:40.10 | *** join/#edev Appleman1234 (n=Appleman@CPE-124-186-229-83.qld.bigpond.net.au) |
| 06:58.51 | *** join/#edev linac (n=lin@121.41.47.12) |
| 08:18.19 | *** join/#edev ccjoe (n=ccjoe@75-169-37-101.slkc.qwest.net) [NETSPLIT VICTIM] |
| 08:22.24 | *** join/#edev lyakh (n=lyakh@p57BD358C.dip0.t-ipconnect.de) |
| 08:23.33 | *** join/#edev Madsy (n=madsy@ti500710a342-1985.bb.online.no) |
| 08:43.32 | *** join/#edev ccjoe (n=ccjoe@75-169-37-101.slkc.qwest.net) [NETSPLIT VICTIM] |
| 08:56.35 | *** part/#edev dirk2 (n=dirk@F3242.f.strato-dslnet.de) |
| 09:00.11 | *** join/#edev dir1 (n=dirk3@F3242.f.strato-dslnet.de) |
| 09:38.37 | *** join/#edev ccjoe (n=ccjoe@75-169-37-101.slkc.qwest.net) [NETSPLIT VICTIM] |
| 09:49.20 | *** join/#edev [AD]Turbo (n=turbo@79.22.241.24) |
| 09:50.24 | [AD]Turbo | hi all |
| 11:20.56 | *** join/#edev Appleman1234 (n=Appleman@CPE-124-186-229-83.qld.bigpond.net.au) |
| 12:12.51 | *** join/#edev rodserling (n=rodserli@unaffiliated/rodserling) |
| 12:14.37 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 13:05.30 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 13:05.30 | *** mode/#edev [+o GPSFan] by ChanServ |
| 13:36.10 | *** join/#edev Appleman1234 (n=Appleman@CPE-124-186-229-83.qld.bigpond.net.au) |
| 14:06.01 | *** join/#edev chouimat (n=dieu@r2351064.cidc.net) |
| 14:10.30 | *** join/#edev bbradley (n=bbradley@82-32-35-42.cable.ubr05.azte.blueyonder.co.uk) |
| 14:52.58 | *** join/#edev ico2 (n=ico2@modem-818.chimpanzee.dialup.pol.co.uk) |
| 15:11.27 | *** join/#edev dirk2 (n=dirk@F321a.f.strato-dslnet.de) |
| 16:04.56 | *** join/#edev ico2_ (n=ico2@modem-3559.chameleon.dialup.pol.co.uk) |
| 17:20.54 | *** join/#edev adyer (n=adyer@adsl-69-223-13-78.dsl.chcgil.ameritech.net) |
| 17:37.13 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 17:56.31 | RobotGuy | Hmmmmm, I have no usb-storage module and it isn't in USB Support. |
| 17:57.41 | ohsix | usb-storage is the old driver thing, i wouln't be surprised if it had been removed on your platform |
| 18:06.02 | RobotGuy | ohsix: Well, that does not help me because I have a USB Thumb Drive |
| 18:06.12 | *** join/#edev ico2_ (n=ico2@modem-2431.cougar.dialup.pol.co.uk) |
| 18:06.27 | RobotGuy | It's in kernel 2.6.22 |
| 18:06.57 | ohsix | it's the dumb slow driver, theres a fast nice driver |
| 18:07.21 | RobotGuy | ohsix: What is it called? |
| 18:10.13 | RobotGuy | usb-storage is still in the kernel - I just didn't have scsi disk support enabled. |
| 18:10.15 | ohsix | looks like a lot of usb stuff got folded under "usb support" latelt |
| 18:11.35 | RobotGuy | Maybe. I know it's been this way since 2.6.22 at least |
| 18:13.08 | ohsix | odd, with the plain config it didn't even let me expand the usb section |
| 18:14.27 | ohsix | i was thinking of BLK_DEV_UB anyways, and its still there, nevermind me |
| 18:15.09 | RobotGuy | I had to enabled scsi support and scsi-disk before I got to do usb-storage |
| 18:15.25 | ohsix | right, as it says on the menu :] |
| 18:15.45 | RobotGuy | It was the other way around - selecting usb-storage would enable scsi and scsi-disk |
| 18:17.40 | ohsix | i dont recall having any problems or changes in that stuff since .20 |
| 18:23.05 | RobotGuy | Kernel configs always confuse me. |
| 18:29.57 | *** part/#edev dirk2 (n=dirk@F321a.f.strato-dslnet.de) |
| 18:50.52 | *** join/#edev steliosk (n=Stelios@athedsl-105743.home.otenet.gr) |
| 19:28.10 | *** join/#edev Sienna (n=Sienna@pool-71-125-183-65.bltmmd.east.verizon.net) |
| 19:28.55 | Sienna | eep! Landley and prpplauge aren't here... |
| 19:29.02 | Sienna | so RobotGuy, that means you get to help! |
| 19:29.58 | Sienna | oh, maybe you won't need too |
| 19:30.03 | Sienna | i see a prompt! |
| 19:30.53 | RobotGuy | What do I get to help with? |
| 19:31.03 | Sienna | oh, hmm... I am guessing you cannot have both the serial port and flyswatter plugged into the hammer at the same time :| |
| 19:31.06 | RobotGuy | I found my usb-storage module. |
| 19:31.17 | Sienna | i am fumbling through my first day of hammer play |
| 19:31.24 | RobotGuy | Not to the Flyswatter. One or the other. |
| 19:32.06 | Sienna | oh good, they do colors on the terminal |
| 19:32.58 | Sienna | I was thinking about the RDP, and such, and I said "hey, I am going to plug in the hammer for the hell of it |
| 19:33.33 | Sienna | and then I realized I was in linux, and had no idea how to use minicom, so I rebooted to windows just to use terraterm quickly |
| 19:34.26 | Sienna | and I had the flyswatter and serial port all plugged in, hoping to give me the most connections |
| 19:34.26 | Sienna | which wasn't working of course |
| 19:34.50 | Sienna | as soon as I took out the flyswatter, i was able to get a root terminal, which was all I really wanted to see, to make sure my hammer worked |
| 19:35.12 | Sienna | what is the LED on the hammer for near pin 1, and why isn't it lit? |
| 19:35.47 | RobotGuy | Sienna: The LEDs are programmable devices. You have to tell them what to do. |
| 19:36.08 | Sienna | oh, i was thinking it would power on to say "hey, i am good!" |
| 19:37.05 | RobotGuy | Sienna: You must write the required Hello World program. :) |
| 19:37.16 | Sienna | how do i write hello world? |
| 19:37.16 | RobotGuy | I haven't even done that yet. |
| 19:37.21 | Sienna | oh darn |
| 19:39.57 | RobotGuy | I'm tweaking my JFFS2 image for W.A.L.T.E.R. - got Python and SQLite there now. IN FLASH. :) |
| 19:42.08 | Sienna | i have no idea what JFFS2 is :P |
| 19:42.22 | Sienna | right now I am furiously treading water... so much to learn |
| 19:42.25 | RobotGuy | Sienna: You will learn one day, young one. :) |
| 19:42.33 | Sienna | i was happy to see the root prompt *nods |
| 19:42.47 | RobotGuy | Sienna: Been there, done that, haven't got a t-shirt yet though. |
| 19:43.10 | *** join/#edev roxfan (n=dunno@75.195-240-81.adsl-dyn.isp.belgacom.be) |
| 19:43.12 | *** join/#edev lyakh (n=lyakh@p57BD358C.dip0.t-ipconnect.de) |
| 19:43.19 | RobotGuy | JFFS2 is the file format for the Flash. |
| 19:44.15 | Sienna | ok |
| 19:44.18 | RobotGuy | I created a few bash scripts to automate my builds. |
| 19:45.03 | Sienna | That FT2232 chip that prplauge was recomending in the forums, I only glanced at the datasheet before I rebooted, but it looks like it would be good for the RDP |
| 19:45.20 | Sienna | now I need to know how to make a USB hub.... |
| 19:45.40 | RobotGuy | Sienna: I have not looked at it yet. Anything that provides more I/O and serial ports is good. |
| 19:45.56 | RobotGuy | Sienna: I need to know that too. It can't be that hard. |
| 19:46.55 | RobotGuy | I need to get some of my 3.3V I2C stuff connected to Hammer so I can experiment. |
| 19:47.15 | Sienna | The best part about the FT2232 chip? It said it has a VCC line and translaters built in for each of the two ports |
| 19:47.30 | RobotGuy | Sienna: Cool! I need to go look at that. |
| 19:47.31 | Sienna | which means a really easy way to do either 3.3 or 5v serial! |
| 19:47.56 | RobotGuy | I need 3.3V -> 5V TTL serial to connect to my SSC-32. |
| 19:48.35 | RobotGuy | Did I mention I have all the servos for the arm on W.A.L.T.E.R. and the Little Grip? |
| 19:48.40 | Sienna | yep |
| 19:49.12 | RobotGuy | Three HS-645MG's should at least lift and move the arm (now top deck mounted). |
| 19:49.51 | RobotGuy | I'm going to have to borrow some 3mm hardware from MsBrat. |
| 19:50.59 | RobotGuy | I wonder if I can just tap a battery into the current power lines on the carrier board without hurting anything. |
| 19:51.15 | RobotGuy | And switch between the batter and wall wart, |
| 19:51.38 | RobotGuy | s/batter/battery/ |
| 19:58.04 | Sienna | i can turn the led on and off! |
| 19:59.05 | RobotGuy | w000 h00 you! |
| 20:02.28 | RobotGuy | uploads a 12 MB jffs2 image to hammer |
| 20:06.07 | Sienna | ok, going to be a long time studying this platform! |
| 20:06.22 | Sienna | but I can make leds light and buzzers sound, thanks to the wiki |
| 20:06.53 | RobotGuy | Sienna: I have two tutorials on the Wiki so far. :) |
| 20:07.04 | Sienna | *clap* |
| 20:07.23 | RobotGuy | :D |
| 20:07.57 | RobotGuy | I'll have a WiFi tutorial on there as soon as I get a working wifi router. Got two that don't work. |
| 20:08.55 | RobotGuy | I already have WiFi configured on the Hammer side. Just have to change the AP address when I get a new router. |
| 20:09.19 | RobotGuy | thinks about looking inside the bricked LinkSys WRT54GS. |
| 20:10.52 | RobotGuy | I'm also going to have a Bluetooth tutorial on there soon, hopefully. |
| 20:12.01 | Sienna | cool |
| 20:15.09 | Sienna | gah! horrible site opens when you guess maxim's website wrong |
| 20:18.25 | MonMotha | if you're referring to Maxim Magazine, then that's hardly one of the more horrible things on the Internet |
| 20:19.07 | Sienna | yeah well, i live sheltered ;D |
| 20:19.21 | MonMotha | apparently |
| 20:19.30 | MonMotha | you can buy Maxim Magazine on the shelf at most bookstores |
| 20:24.05 | Crofton | that's not so horrible ... |
| 20:24.17 | Crofton | there are far more disturbing sites around |
| 20:24.23 | MonMotha | indeed |
| 20:25.46 | Sienna | ok, i didn't look at it too closely |
| 20:26.04 | Sienna | it looked like a pr0n site, so i jumped quickly |
| 20:26.16 | *** join/#edev emil|k (i=emil@e179234196.adsl.alicedsl.de) |
| 20:27.15 | RobotGuy | MonMotha: Even at a lot of grocery stores like SafeWay |
| 20:27.51 | adyer | Sienna: btw it's www.maximic.com that you want :-) |
| 20:28.06 | adyer | or maxim-ic.com |
| 20:28.39 | adyer | Sienna: maxim's motto: one should be enough for anyone! |
| 20:28.42 | MonMotha | Sienna: it's not |
| 20:29.06 | MonMotha | Maxim Magazine is one of those faux "men's health" magazines that basically features fluffy articles and scantily clad "hot" women |
| 20:29.10 | RobotGuy | One woman for each man. :) |
| 20:29.18 | Sienna | sorry if i assumed! as i said, didn't read to close |
| 20:29.21 | MonMotha | scantily clad, but everything covered |
| 20:29.53 | adyer | RobotGuy: no, I was referring to maxim's performance on delivering samples vs. orders ;-) |
| 20:30.19 | RobotGuy | adyer: I have never had problems getting samples from Maxim. |
| 20:30.39 | adyer | RobotGuy: exactly - one is enough for anyone |
| 20:30.57 | adyer | RobotGuy: now if you want 1000 in 4 weeks you're SOL |
| 20:31.02 | RobotGuy | I just haven't found my one yet. |
| 20:31.40 | RobotGuy | There aren't many GeekGirls to choose from. |
| 20:32.12 | Sienna | hmm... I don't know if I like all this USB power management stuff that hubs have to do |
| 20:32.40 | RobotGuy | Sienna: Isn't some of that stuff optional? |
| 20:32.59 | Sienna | i have no idea |
| 20:33.22 | Sienna | how much power does a wifi or bluetooth usb module consume? |
| 20:33.56 | adyer | Sienna: usually a few 100 mW in transmit, less in idle |
| 20:34.25 | Sienna | i would like to put a hub on to use a FT2232 without blocking the port, but i don't know how to find power for 500mA usb ports like that |
| 20:34.47 | Sienna | actually, 100mW, thats 20mA @ 5V right |
| 20:34.56 | RobotGuy | Is 10K OK for a pull-up on a reset line? |
| 20:35.03 | Sienna | which means they will work on the 100mA buss powered types? |
| 20:35.38 | Sienna | why wouldn't it be RobotGuy? |
| 20:35.49 | RobotGuy | I was just wondering. |
| 20:36.03 | RobotGuy | I have to add a pull-up to the reset line on my serial Bluetooth module. |
| 20:36.28 | RobotGuy | I'm adding Bluetooth to the console port. |
| 20:36.40 | adyer | RobotGuy: that's an 'it depends' answer. You might want to go lower if you have a lot of load on the line |
| 20:36.41 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 20:36.55 | RobotGuy | adyer: Just one module so far. |
| 20:37.30 | RobotGuy | There will be at least two or three more chips that will need a reset. |
| 20:37.55 | adyer | RobotGuy: then I'd guess it's okay. datasheet should have input specs to figure out for sure |
| 20:38.29 | RobotGuy | Ahh, datasheet. Good idea! I am braindead apparently. |
| 20:39.07 | RobotGuy | I have my two tutorials up on my wiki now. |
| 20:39.46 | adyer | feh. one of the sites for buildroot source package is down :-( |
| 20:40.17 | Sienna | ok, for the RDP, anyone object if I tried to limit usb to 100mA? |
| 20:40.21 | Sienna | (per port) |
| 20:42.01 | Sienna | RobotGuy, that was really directed at you ;D |
| 20:43.03 | RobotGuy | Sienna: No objections to the limit as long as it handled all the possible modules we can connect. :) |
| 20:43.18 | adyer | Sienna: you will limit the number of devices that will work - IIRC all usb devices start in low power mode and then negotiate to the higher setting in s/w |
| 20:43.48 | RobotGuy | I have two different USB WiFi adapters, USB Bluetooth, 2 USB Thumb Drives. |
| 20:43.54 | adyer | Sienna: host is supposed to keep track of pwer availabel to each 'leaf' |
| 20:44.25 | RobotGuy | adyer: How would it do that if it's connected to a powered Hub? |
| 20:44.33 | Sienna | i know i might limit the power available... but if we go higher, it might require a dedicated 1A regulator just for two usb ports |
| 20:44.56 | Sienna | s/power/devices |
| 20:45.02 | adyer | RobotGuy: when the hub connects to the host, it tells it what config it has and how much power it can supply where |
| 20:45.17 | RobotGuy | Sienna: Hmmmmm.. |
| 20:45.44 | RobotGuy | adyer: Ah, OK. We are trying to get at least one USB Hub put on the Hammer-RDP. |
| 20:46.25 | RobotGuy | Now I am wondering if that is a good idea. |
| 20:46.38 | Sienna | What i am trying to do is turn the Hammer's usb port into two FT2232 devices, and then have to more usb ports for the user |
| 20:46.51 | Sienna | if I need to, i might ax user ports |
| 20:46.52 | RobotGuy | Maybe it would be better to just gut a purchased USB Hub and use that. |
| 20:47.04 | adyer | RobotGuy: you can mock it up with most 4-port hubs - they will run in 'unpowered' (ie using power from upstream) if you dont plug in the AC adapter |
| 20:47.14 | RobotGuy | Sienna: Two more full host ports? |
| 20:47.30 | Sienna | i am looking at the TUSB2046 |
| 20:47.34 | Sienna | 4 port usb hub |
| 20:47.34 | RobotGuy | adyer: I would power the hub(s) from the onboard battery. |
| 20:47.53 | RobotGuy | Sienna: Ah, yes, nice chip. |
| 20:48.24 | Sienna | can i have a 'host' provide a 100mA usb port to a downstream self powered hub that in turn can power all the hungry devices? |
| 20:48.35 | RobotGuy | There was an article in Nuts & Volts about power distribution for projects liker robots. |
| 20:49.22 | adyer | Sienna: I believe root host ports usually can source the full 500mA, so you can run a hub + 4 devices @100mA from that |
| 20:49.41 | Sienna | umm.. no adyer, hmm... |
| 20:49.50 | Sienna | the hammer has one root host port |
| 20:50.04 | adyer | Sienna: ahh, was thinking about PC |
| 20:50.09 | RobotGuy | If we have to have a dedicated regulator, would that be a real problem? |
| 20:50.12 | Sienna | i was thinking about embedding a 4 port hub on the board |
| 20:50.25 | RobotGuy | Sienna: Hammer has two host ports - both can be used host or slave. |
| 20:50.50 | Sienna | if 'my' hub can only source 100mA per port, can the downstream hub self power and get back to 500mA per port? |
| 20:51.10 | adyer | Sienna: I believe so |
| 20:51.27 | Sienna | i had hoped so to... so that might be what I choose |
| 20:51.39 | RobotGuy | SO we are talking up to 2A max for a single 4 port hub. |
| 20:51.42 | Sienna | RobotGuy, where did you see that hammer can configure both ports? |
| 20:52.08 | RobotGuy | Sienna: prplague told me. You can verify it with the s3c2410 datasheet. |
| 20:52.25 | Sienna | oh, have to go read the hammer schematic then :| |
| 20:52.29 | RobotGuy | Both USB ports are either Host or Slave. |
| 20:52.33 | adyer | RobotGuy: the 2A is part of why the wall wart power supplies for USB hubs tend to be kinda big |
| 20:53.24 | Sienna | actually.... might be able to provide a screw block for external usb power? |
| 20:53.35 | Sienna | that way if they want to use it, they source it! |
| 20:53.35 | RobotGuy | We may need a separate battery supply connector just for USB if we use the onboard hub, like what is done on the Lynxmotion Bot Board II for Electronics and Servo power. |
| 20:53.50 | RobotGuy | Sienna: I think that would be the way to go. |
| 20:54.12 | RobotGuy | Power connections for Electronics, Servo, USB. |
| 20:54.37 | Sienna | geez.. all this board real estate being claimed! |
| 20:54.58 | RobotGuy | Sienna: We need to find out how much each requested feature requires. |
| 20:55.08 | adyer | Sienna: don't forget about usb power switching, overcurrent detection and fusing :-) |
| 20:55.34 | Sienna | adyer, the TI datasheet covers that and the other TI parts needed for that |
| 20:57.07 | RobotGuy | I'm OK with just having basic functions on the RDP - Power connects for Electronics and USB (no real need to connect servos direct to Hammer), Four port USB Hub, 3 Pin headers for all Hammer pins, Stacking connections for add-on boards. |
| 20:57.28 | Sienna | RobotGuy, you are NOT going to get three row pins for much |
| 20:57.44 | RobotGuy | Oh, and headers for I2C and SPI busses. |
| 20:57.47 | Sienna | it would be WAY to big |
| 20:58.23 | RobotGuy | The three row right angle pins do not take up that much room though. |
| 20:58.34 | Sienna | yes they do |
| 20:58.38 | Sienna | so much so |
| 20:59.04 | RobotGuy | How much space does the 4 port USB hub need? |
| 20:59.10 | Sienna | especially if you want to bring them in from the edge of the board so that the mechanics of the robot don't short them |
| 20:59.25 | Sienna | it looks like 5 or six chips |
| 20:59.25 | RobotGuy | Sienna: Let the pins hang off the board. |
| 20:59.51 | Sienna | you are tempting a short, thats unwise |
| 21:00.52 | RobotGuy | I don't agree. It's all in how the RDP is mounted on the robot or other project. If it can mount on a Lynxmotion EC-02 carrier, then no problem. |
| 21:01.35 | Sienna | There will be some three pin connections. |
| 21:01.48 | adyer | Sienna: this explains the pwr. stuff for USB pretty well http://www.micrel.com/_PDF/App-Notes/an-17.pdf |
| 21:01.49 | Sienna | but things like I2C and SPI will not be on them |
| 21:01.51 | RobotGuy | There has to be 3 pin connection for sensors |
| 21:02.03 | Sienna | The RDP prolly won't have sensor connections... |
| 21:02.09 | Sienna | where would you hook them to the hammer>? |
| 21:02.20 | Sienna | thanks adyer |
| 21:02.43 | RobotGuy | Sienna: Without 3 pin connections, it won't be RDP. |
| 21:03.00 | RobotGuy | Edges of the board on two or three sides. |
| 21:03.18 | RobotGuy | Use a header for the reset switch. |
| 21:03.33 | Sienna | no, where on the Hammer are going to hook up sensors? |
| 21:03.43 | Sienna | about the only thing available is the LCD lines |
| 21:03.48 | Sienna | and those won't do analog |
| 21:03.49 | RobotGuy | To 3 pin headers, GPIOs. |
| 21:04.20 | RobotGuy | Functions of the pins are not dedicated - depends on how Hammer is initalized for a given project. |
| 21:04.24 | Sienna | what GPIOs RobotGuy? thats what I am asking. Everthing on the hammer is serial in some format |
| 21:04.38 | RobotGuy | This should be a general type of board. |
| 21:05.10 | RobotGuy | GPIOs, Analog I/Os, etc, has to be accessible. |
| 21:05.23 | Sienna | you have two analog ins.... |
| 21:05.27 | RobotGuy | Otherwise we do not have a standalone RDP. |
| 21:06.01 | Sienna | and if I do it the way you seem to want, its going to be a 5" x 5" board, to cover everything |
| 21:06.23 | RobotGuy | What are you including in that? It should not have to be that large. |
| 21:06.50 | Sienna | i don't know yet exactly! |
| 21:07.00 | Sienna | i was banking on serial going up the stack |
| 21:07.12 | Sienna | The hammer to me isn't a GPIO processors |
| 21:07.14 | RobotGuy | OK. That's why we need to define how much space each feature requires on the board. :D |
| 21:07.15 | Sienna | -s |
| 21:07.35 | RobotGuy | We are thinking of hammer in two different ways. |
| 21:07.47 | Sienna | i was going to develop a atmel board that stacked onto the hammer and use it for GPIO |
| 21:08.14 | RobotGuy | That can certainly still be done. What are you considering for Hammer? We need to get on the same page. :D:D |
| 21:08.44 | Sienna | I was thinking the RDP would be a serial oriented device... thats the easist way for me to descibe it |
| 21:09.06 | Sienna | if you need GPIO, which we do, i was thinking that a coprossesor would be an addon |
| 21:09.13 | RobotGuy | I want a general purpose board that brings out all hammer functions to pins or headers or such, as required, not unlike the carrier board but smaller. |
| 21:09.50 | Sienna | yes, the my rdp would still bring out all the hammer pins to two single row headers (the stack connections) |
| 21:10.02 | RobotGuy | I would much prefer not having to have an add-on board where having direct access to features of Hammer is possible. |
| 21:10.32 | RobotGuy | Everything can stay 3.3V and anything for 5V can be an add-on board. |
| 21:10.36 | Sienna | what I wasn't planning on was bringing all of the hammer pins to 3 pin connections |
| 21:10.55 | RobotGuy | That was one of the first things I have on my feature request list. :D |
| 21:11.07 | Sienna | you are more then welcome to design your own version of the rdp :D |
| 21:11.22 | RobotGuy | I'm not able to do that. |
| 21:11.35 | Sienna | neither can i, but i am learning |
| 21:12.07 | RobotGuy | I don't have the Eagle parts I would need and have no patience with the Eagle part editor. |
| 21:12.21 | Sienna | I am thinking, as a base set of functions, if they can fit: Two FT2232 chips (will provide four 3.3 or 5v serial / spi ports) |
| 21:12.40 | Sienna | prolly a 8 to 1 mux, connected to ain0 |
| 21:13.07 | Sienna | prolly lcd data lines be 8 bit GPIO |
| 21:13.24 | Sienna | everything else is connected to headers for serial connections |
| 21:13.29 | RobotGuy | LCD/GPIO lines to 3 pin headers. :) |
| 21:13.36 | Sienna | (but thats all my thoughts) |
| 21:13.43 | RobotGuy | I2C and SPI bus headers direct from Hammer. |
| 21:13.46 | Sienna | yes, lcd to 3pin |
| 21:14.05 | Sienna | mux to three pin as well |
| 21:14.11 | RobotGuy | OK, if we scrap the on board USB Hub, where does that get us? |
| 21:14.20 | Sienna | you lose the FT2232 chips |
| 21:14.28 | *** part/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 21:14.34 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 21:14.39 | Sienna | you lose the FT2232 chips |
| 21:14.46 | RobotGuy | What about more serial ports then? |
| 21:14.47 | Sienna | so you are limited to the hammers 3.3V serial |
| 21:14.53 | RobotGuy | Am I missing something? |
| 21:15.25 | Sienna | ? |
| 21:15.51 | RobotGuy | <Sienna> I am thinking, as a base set of functions, if they can fit: Two FT2232 chips (will provide four 3.3 or 5v serial / spi ports) |
| 21:16.13 | Sienna | you asked what happens if we scrap the usb hub |
| 21:16.23 | RobotGuy | Yes |
| 21:16.25 | Sienna | ft2232 are usb to serial chips |
| 21:16.58 | RobotGuy | Ahhh, OK. One FT2232 per USB? |
| 21:17.03 | Sienna | yep |
| 21:17.24 | RobotGuy | SO we don't have a USB Hub with that arrangement, right? |
| 21:17.25 | Sienna | each ft2232 provides two generic (ttl / spi or 8bit I/O) ports |
| 21:18.11 | RobotGuy | I am confused |
| 21:18.48 | RobotGuy | No USB Hub to connect to stuff like thumb drives, etc? |
| 21:19.20 | Sienna | if we want the ft2232 to do the 3.3V to 5V serial translation, i am going to thikn we want a hub |
| 21:19.34 | Sienna | otherwise it will take up one of the hammers usb ports |
| 21:19.51 | RobotGuy | I am thinking of "hub" as in connecting stuff like wifi adapters, thumb drives, etc. |
| 21:20.19 | Sienna | yes. I was looking at a TI 4 port hub - two connected internally to FT2232 chips, two ports user available |
| 21:20.31 | Sienna | plus the other usb port a hammer has |
| 21:21.12 | RobotGuy | OH! The two remaining ports on the TI chip could still be USB hubs then or connected to external USB Hubs? |
| 21:21.22 | Sienna | yes |
| 21:21.33 | RobotGuy | Ah, OK, I understand now. :D |
| 21:22.35 | RobotGuy | OK, how about this, put any USB Hubs on an add-on board. TI chip on RDP handles the FTD2232 and connections to external hubs. Same could be done externally for the other Hammer USB port. |
| 21:22.57 | RobotGuy | Perhaps add a jumper to select host or slave operation of the second Hammer USB port. |
| 21:23.35 | Sienna | i can't add an FT2232 to the RDP without dedicating a hammer usb port to ut |
| 21:23.36 | Sienna | it* |
| 21:24.11 | RobotGuy | Sienna: I know that. I am saying do that with one Hammer USB port. Leave the second Hammer USB port for add-ons. |
| 21:24.45 | RobotGuy | No 3 pin headers for dedicated function pins. |
| 21:25.18 | RobotGuy | Only bring out the GPIO and Analog Input type functions to 3 pin headers. |
| 21:27.05 | Sienna | ok, so only use one FT2232, and dedicate one USB to it |
| 21:28.11 | RobotGuy | One FTD2232 provides two ports, right? |
| 21:28.58 | RobotGuy | It still gets us two addtional serial ports, if I undrstand correctly. |
| 21:29.42 | Sienna | So, i have a list right now at 1 octal analog mux, octal i/o (lcd pins), 2 3.3V TTL Serial, 2 3.3V SPI Serial, 1 I2C 3.3V bus, 2 3.3V/5V TTL/SPI configurable serial |
| 21:31.10 | RobotGuy | Yes, I think that looks good. Second USB port just left for stack boards - I don't want to dedicate a connector to this one because it could be either slave or host on an add-on board. |
| 21:31.46 | RobotGuy | I mean the second Hammer USB port. |
| 21:32.12 | Sienna | how about if I remove one each of the 3.3V TTL and SPI, and just leave those to be exposed to the stack? |
| 21:33.22 | RobotGuy | I'd rather have both ttls from the FTD2232's on the RDP if there is room. So we would then still have two USB slave ports (or is it three) coming from the TI hub chip? |
| 21:34.10 | Sienna | nono, the ones i want to remove aren't the FTDI's, but the hamemrs |
| 21:35.23 | RobotGuy | OK, leave the second Hammer USB port to the stack. Everything on the TI chip will be brought to connectors on the RDP in one form or another, right (to FTD2232 or slave USB connector)? |
| 21:36.18 | Sienna | now i am confused... what TI chip? the FT2232 would take one the Hammer's usb ports, dedicated |
| 21:36.27 | Sienna | the other Hammer usb port goes on the stack |
| 21:36.33 | RobotGuy | The TI Hub chip you mentioned. |
| 21:36.59 | RobotGuy | I am really confused now. :( |
| 21:36.59 | Sienna | Yes, that could use the other hammer usb chip to make a hub on a stack board for other things |
| 21:37.34 | RobotGuy | I thought a USB port from the TI chip was going to the FTD2232 chip(s). |
| 21:38.04 | Sienna | no, we removed the ti chip from the rdp |
| 21:38.25 | RobotGuy | -sigh- Didn't realize we did that. |
| 21:38.46 | RobotGuy | Can't we leave the TI chip on and use one port from it for an FTD2232? |
| 21:39.03 | Sienna | we could, but that takes up a lot of room, thats what we were discussin |
| 21:39.05 | Sienna | g |
| 21:39.17 | RobotGuy | How much room? |
| 21:39.32 | RobotGuy | Can you draw a picture? |
| 21:39.39 | Sienna | we don't know yet, five or six chips worth |
| 21:39.51 | Sienna | and no, i can't picture it cause I don't have all the schematics and such |
| 21:40.24 | RobotGuy | That many chips to just support a four port hub (TI chip)? |
| 21:40.31 | Sienna | yep |
| 21:40.39 | Sienna | most of them deal with transients and power |
| 21:41.18 | RobotGuy | OK then. just route one Hammer USB port to an FTD2232 and the second port for stack boards only. I'm not real happy with this, but oh well. |
| 21:41.44 | RobotGuy | I was really hoping to have at least a couple slave USB ports right on the RDP. |
| 21:41.48 | Sienna | I will try to put a hub on there, but no guarentees |
| 21:42.05 | RobotGuy | No problem - scrap the hub if it will not fit. |
| 21:42.25 | RobotGuy | We can always do a stack board with the other Hammer USB port. |
| 21:42.45 | Sienna | for the three pin, i think this is how it will come out: 8 analog in (3.3V), 8 digital I/O (3.3V), one TX (3.3V/5V) one RX (3.3V/5V) |
| 21:43.13 | RobotGuy | Bring both Hammer serial ports to 3 pin connectors. |
| 21:43.41 | Sienna | i think the rest of the serial connections are going to be double row right angle on the sides of the hammer |
| 21:44.09 | RobotGuy | Two Hammer serial and two FTD2232 serial ports, right? |
| 21:44.32 | Sienna | one of the FTDI's serial would be the 3pin version (so you can jump it 3.3V or 5V) |
| 21:45.06 | RobotGuy | Other FTDI serial would be?? |
| 21:45.15 | Sienna | on the sides of the board |
| 21:46.11 | Sienna | the reason i want to put the ftdi on the 3 pin is simply to hook it to the 5V ssc32 |
| 21:46.16 | RobotGuy | FTDI ports should not be on 3 pin connectors but should be jumperable 3.3V/5V. |
| 21:46.44 | Sienna | ok, you would rather see the 3pins be all 3.3V? |
| 21:47.38 | RobotGuy | You can use one double row right angle for both FTD2232 serial ports. I'm OK with everything remaining 3.3V though and leave level translations for external circuits (stack boards or right off the RDP connectors. |
| 21:48.15 | Sienna | ok |
| 21:48.45 | RobotGuy | Sure I would still like 3 pin headers to be 3.3V/5V jumperable, but not sure this is reasonable. |
| 21:49.02 | RobotGuy | Let's not even consider servo control direct from the RDP. |
| 21:49.09 | Sienna | i wouldn't... cause then someone is going to put a 5V sensor on it, and fry things |
| 21:49.26 | RobotGuy | At least for now if we can't have 3.3/5V jumpers for power. |
| 21:49.46 | RobotGuy | Let's just have it a total 3.3V board then. |
| 21:50.22 | RobotGuy | I would like the SPI and I2C pins to be dedicated for those uses. |
| 21:50.30 | Sienna | ok, that might mean we can do away with the FTDI |
| 21:50.47 | RobotGuy | I'd like the extra serial ports. |
| 21:51.00 | RobotGuy | If they will fit with everything else. |
| 21:51.05 | Sienna | ok |
| 21:51.18 | RobotGuy | We can always do FTDI on a stack board. |
| 21:51.31 | Sienna | yeah, thats what i was thinking |
| 21:51.38 | RobotGuy | I'm OK with that. |
| 21:51.40 | Sienna | the stacking thing is going to be huge |
| 21:51.54 | RobotGuy | Yes, that is extremely important for RDP. |
| 21:53.23 | RobotGuy | IF there is a bit of room left, perhaps we could add a small hardware RTC but it would be useless without battery backup so I think this is for a combo stack board. I am getting a good vision of what it should have on it now. |
| 21:53.35 | Sienna | i am thinking about making the RDP +5VDC input only, would that kill you? |
| 21:54.11 | RobotGuy | Make the 3 pin I/O headers jumperable then - 3.3V or 5V. |
| 21:54.35 | Sienna | are you sure about that... remember, somone is going to jump a 5V source to the 3V hammer |
| 21:54.35 | RobotGuy | Like what is done for Servo/5V on the Lyxmotion Bot Board II |
| 21:55.00 | RobotGuy | Sienna: We can't be responsible for stuff like that - let the user beware. |
| 21:55.19 | Sienna | i treat the user with care... because doing otherwise creates computer service costs |
| 21:55.30 | RobotGuy | I know this is gonna bite me at some point. :) |
| 21:56.14 | RobotGuy | Can each pin be protected some way depending on how it is jumpered? |
| 21:56.25 | Sienna | not that i know of |
| 21:56.32 | Sienna | or that i would add |
| 21:56.34 | RobotGuy | Making sure not more than 3.3V ever gets to the Hammer? |
| 21:56.55 | RobotGuy | Simple series resistor perhaps? |
| 21:57.25 | RobotGuy | Dang, probably would not work. |
| 21:57.26 | Sienna | i would prefer keeping the rdp 3.3v |
| 21:57.38 | Sienna | but it is something i will be looking at |
| 21:57.43 | RobotGuy | Well, we have to decide and stick to it then. It' |
| 21:57.53 | RobotGuy | It's either all 3.3V or all 5V |
| 21:58.10 | Sienna | then its all 3.3V |
| 21:58.21 | RobotGuy | OK, all 3.3V it is then. |
| 21:58.45 | Sienna | now, can you live with only allowing 5V regulated in? |
| 21:59.26 | RobotGuy | Sienna: Has to be able to work with standard battery voltages - at least 6V and 7.2V. |
| 21:59.56 | Sienna | batterys like that will range from 4.8V to almost 9V |
| 22:00.01 | Sienna | depending on charge state |
| 22:00.07 | RobotGuy | We have to deal with that then. |
| 22:00.08 | Sienna | that would roast a 3.3V regulator |
| 22:00.38 | RobotGuy | Run it all to a 5V regulator and the output from the 5V to a 3.3V regulator. |
| 22:00.59 | Sienna | why can't I force the user to have an external 5V regulator? |
| 22:01.24 | RobotGuy | Sienna: Because that does not allow a standalone Hammer RDP. |
| 22:01.57 | Sienna | define standalone? Cause none of my current bots allow battery voltage to touch VL... I always have a DE 6V regulator there |
| 22:02.37 | RobotGuy | Standalone means the RDP has to be able accept power from a standard battery or other easily portable power source. |
| 22:02.50 | *** join/#edev landley (n=landley@66-90-245-215.dyn.grandenetworks.net) |
| 22:02.50 | *** mode/#edev [+o landley] by ChanServ |
| 22:02.59 | Sienna | hey landley |
| 22:03.08 | RobotGuy | waves at landley |
| 22:03.34 | Sienna | but batteries are horrible.. and when you go with a 12V main battery, there is no way you are pluggin that into a RDP or other board |
| 22:03.46 | Sienna | you would overheat any linear regulator on the planet |
| 22:04.12 | adyer | Sienna: depends on the load and heatsinking... |
| 22:04.21 | Sienna | and i am not even going to try to suggest they design a switchmode regulator to allow the user to hook 9V or 12V to the RDP |
| 22:04.52 | Sienna | adyer, i am thinking a load of about 1A at 3.3V |
| 22:05.03 | Sienna | and heatsinking, well, its a tiny board to begin with |
| 22:05.51 | adyer | Sienna: so what's so bad about a switcher - if you don't care about efficiency, it's not THAT bad |
| 22:05.51 | landley | Hi Sienna, RobotGuy. |
| 22:06.30 | RobotGuy | The Lynxmotion Bot Board II and SSC-32 have their own on board power regulation. |
| 22:06.38 | Sienna | adyer, don't get me wrong, I LOVE switchers. I use Dimension engineering's products all the time. I just wouldn't want to ask TCT to design one... cause i understand its not trivial |
| 22:06.46 | landley | Sienna: some years back the Sharper Image catalog had a fuel cell you could fill with... butane, I think, and use it to power your laptop. |
| 22:07.04 | adyer | Sienna: LOL, it's not trivial, but it's not hard. |
| 22:07.14 | landley | Not the world's cheapest little toy, but fuel cells with some capacitors might be a reasonable way to power a laptop. |
| 22:07.23 | Sienna | RobotGuy, yes they do, and if you notice, they derate it by 50% to allow for the heat |
| 22:07.36 | landley | They have the advantage that half their energy comes from the oxygen in the atmosphere, so the weight/energy ratio in them is much higher than batteries. |
| 22:07.54 | RobotGuy | Sienna: We could not very well derate it, so would have to heatsink the regulator(s). |
| 22:08.24 | Sienna | i really don't think we have the space to heatsink a 3.3V LDO from a 12V source |
| 22:08.37 | RobotGuy | From a 6V to 7,2V source. |
| 22:08.50 | Sienna | from a 9V source... thats how people power logic now |
| 22:08.52 | RobotGuy | Or 4.8V to 9V source |
| 22:08.56 | landley | Sienna: google for "fuel cell hobbyist". |
| 22:08.58 | adyer | Sienna: something like LM3100 is pretty easy - intergrated fets, 4.5-36V in, 3.3V @ 1.5A out |
| 22:09.13 | RobotGuy | I power logic from a 6V to 7.2V battery. |
| 22:09.56 | RobotGuy | So the range would b 4.8V to 9V or so. maximum. |
| 22:12.11 | Sienna | adyer, thats actually a pretty good suggestion... how big are those (registration stopped me temporarily) |
| 22:12.25 | Sienna | and is there any worry about those and the hammer interfering with each other? |
| 22:12.40 | adyer | Sienna: 20 pin TSSOP - the inductor is usually bigger than the part |
| 22:13.29 | MonMotha | there's tons of good SMPS solutions out there, and most makers have great app notes on how to use them |
| 22:13.29 | RobotGuy | Wouldn't an inductor on the RDP create a whole set of new problems with magnetic fields? |
| 22:13.58 | adyer | Sienna: not too much, you have to keep the layout tight, use the right kind of caps and pay attention to where the switching currents go, but because the fets are on-board, it's not bad. |
| 22:14.04 | MonMotha | National, LT, ON, TI, and our good friend Maxim |
| 22:14.33 | MonMotha | ON is often great about giving entire design spreadsheets |
| 22:14.39 | MonMotha | basically, put your numbers in, and it spits out a design |
| 22:14.56 | RobotGuy | Do they provide samples? |
| 22:15.02 | MonMotha | ON? yes, but you have to pay for them now :( |
| 22:15.04 | Sienna | hmm... 82% efficiency, thats enough right :P |
| 22:15.22 | MonMotha | well, the parts themselves are free, but there's S&H |
| 22:15.33 | MonMotha | it's cheaper to jus tbuy from Mouser |
| 22:15.45 | MonMotha | TI and National I know do free samples |
| 22:15.47 | RobotGuy | Hmmmm, I wonder if Microchip has anything similar. |
| 22:15.56 | MonMotha | but TI's smps app notes have never impressed me |
| 22:16.03 | MonMotha | National's are decent, though |
| 22:16.22 | MonMotha | LT's are fairly detailed about design, but tend to leave me still somewhat nebulous in terms of actual operation |
| 22:16.25 | adyer | Sienna: as monmotha says, there are lots of good regulator app notes. There was even a company that mad a switcher in the same package as a 7808, so you could just drop it in. |
| 22:16.38 | RobotGuy | Something got messed up with my National login. |
| 22:16.57 | MonMotha | Fairchild also has some nifty stuff, though it tends to be a bit more domain specific |
| 22:17.21 | Sienna | hmm.. ok fine, maybe it is reasonable then to add a switcher to an RDP... but it just makes more work people could do themselves :P |
| 22:17.26 | MonMotha | like they have a flyback regulator where everything you need except the transformer and optocoupler is contained in a 4/5 lead TO-220 package |
| 22:17.57 | RobotGuy | Sienna: RDP must have it's own power regulation. I'm not going to give in easily on this one. :) |
| 22:18.40 | MonMotha | I'm a fan of the NCP3163 from ON. Great design tools, high-power, leaded package (it has an EP which you don't really have to solder down for lower power applications), and the control loop seems very difficult to "confuse" |
| 22:18.53 | MonMotha | it does tend to "sing" a little bit, though |
| 22:18.57 | RobotGuy | RDP has to be able to handle the same input voltages as the Lynxmotion Bot Board II and SSC-32 boards. |
| 22:19.22 | Sienna | then it will have to have a switcher, because I don't see it working with a small linear |
| 22:19.34 | RobotGuy | Sienna: Nor do I, |
| 22:19.35 | Sienna | (being the expert that I am of course! :look) |
| 22:19.51 | adyer | Sienna: check these out: http://www.recom-power.com/pdf/innoline/R-78Bxx-1.0_L.pdf |
| 22:20.01 | MonMotha | the only time I use linear regulators these days are low-dropout applications or uber low-power applications (like 50mA or less) |
| 22:20.18 | RobotGuy | Well, I have not eaten all day so I guess I better go get some food. |
| 22:20.20 | MonMotha | or when the behavior of a switcher just won't cut it in terms of regulation or noise |
| 22:20.38 | Sienna | adyer, awesome! |
| 22:21.52 | RobotGuy | I will be back in a bit. |
| 22:22.46 | RobotGuy-Out | is away: Gone away for now. |
| 22:23.04 | adyer | MonMotha: I would add the case where something needs to be dirt cheap |
| 22:23.32 | MonMotha | adyer: I guess, but you'd be amazed how cheap you can get switchers, and heatsinks aren't cheap, if a linear solution would need one |
| 22:24.02 | Sienna | eep! Not so awesome! $15 ea? darnitall |
| 22:24.13 | MonMotha | if it's low enough power to not need a heatsink and efficiency isn't a concern but cost is, then I guess a linear would be acceptable |
| 22:24.44 | MonMotha | seems like most of the situations I've run into where cost is a concern, efficiency is also a concern (portable, semi-disposable devices) |
| 22:24.59 | adyer | MonMotha: some heatsinks are cheap :-) espcially if you don't isolate them :-) |
| 22:25.05 | MonMotha | I guess |
| 22:25.19 | MonMotha | reminds me of the heatsink on my arcade monitor on the smps FET |
| 22:25.27 | MonMotha | yeah, that heatsink is "hot" in more than one way :) |
| 22:25.38 | Sienna | here, you to fight it out. We have the RDP, it will need to provide 5V at 250mA, and 3.3V prolly at least 500mA... and over a 6v to 12V input range |
| 22:25.39 | MonMotha | that thing is right on the drain of that FET |
| 22:25.45 | adyer | MonMotha: no user servicable parts inside :-) |
| 22:25.52 | MonMotha | on an arcade monitor? far from true |
| 22:25.56 | MonMotha | they are intended to be serviced |
| 22:26.24 | adyer | MonMotha: I used to design arcade games, I'm very familiar with their problems :-) |
| 22:26.28 | MonMotha | adyer: ah |
| 22:26.45 | MonMotha | I try to avoid working on monitors, but I'll do it if I have to |
| 22:26.56 | MonMotha | I've resurrected several game boards, though |
| 22:27.02 | adyer | MonMotha: guy in the next office over did Mortal Kombat :-) |
| 22:27.05 | Sienna | blessed be! The dishwasher finished... *silence is golden* |
| 22:27.14 | MonMotha | really...perhaps he could help me with my sound board on MK2? :) |
| 22:27.21 | MonMotha | or NBA Jam, similar to original MK |
| 22:27.48 | MonMotha | oh Midway, how I miss thee |
| 22:27.59 | adyer | MonMotha: doesn't MAME do all those nowadays :-) |
| 22:28.13 | MonMotha | adyer: sure, but it's not the same |
| 22:28.38 | adyer | MonMotha: Midway was fun while it lasted... |
| 22:28.55 | MonMotha | speaking of...I could go play NFL Blitz |
| 22:29.13 | adyer | MonMotha: cool. That's one of mine :-) |
| 22:29.18 | MonMotha | really... |
| 22:29.43 | MonMotha | know much about the sound stuff on it? I've been meaning to hack it :) |
| 22:29.57 | MonMotha | I've got all the textures |
| 22:30.04 | MonMotha | but the sound files have alluded me so far |
| 22:31.00 | adyer | MonMotha: that sound system was grafted on to the arcade stuff from an old pinball design. There is an asic with security stuff in between the main logic and sound land that I don't know about |
| 22:31.31 | MonMotha | well, the sound appears to basically be the old "DCS" stuff from back in the MK2 heyday |
| 22:31.51 | adyer | MonMotha: the sound DSP, main proc, and a microcontroller all talked to the asic |
| 22:32.04 | MonMotha | I'm just mostly looking to hack up the "BNK" files to change the sounds to be (even) sillier |
| 22:33.11 | adyer | MonMotha: IIRC that version used dram and the sounds were loaded from the host side, so they should be on the disk somewhere |
| 22:34.30 | MonMotha | they are |
| 22:34.37 | MonMotha | the "BNK" files on disk appear to be sound |
| 22:34.47 | MonMotha | I'm just curious about how I might go take them apart |
| 22:34.57 | MonMotha | they seem to basically get chucked straight at the sound system |
| 22:35.36 | adyer | MonMotha: I don't know the internal format of the sounds, they had a whole group of people just doing sounds & tools |
| 22:35.43 | MonMotha | and hum, my hard drive looks like it may be messed up |
| 22:35.54 | MonMotha | definitely loaded the wrong textures there |
| 22:38.36 | adyer | MonMotha: brb |
| 22:39.57 | MonMotha | there we go, just needed reset :) |
| 22:46.10 | *** join/#edev landley_ (n=landley@66-90-245-215.dyn.grandenetworks.net) |
| 22:46.24 | adyer | lol - shock mount http://www.silentpcreview.com/files/images/no-vibe-3.jpg |
| 22:47.05 | MonMotha | hum, I do know that my image for Blitz 2k Gold is bad |
| 22:47.13 | MonMotha | it won't take any settings or remember player profiles :( |
| 22:48.52 | MonMotha | adyer: and I must thank Midway for not taking the Capcom approach to "copy protection" |
| 22:48.59 | MonMotha | it's why their game boards are still usable at this point |
| 22:49.13 | adyer | MonMotha: not familiar with what capcom did |
| 22:49.24 | MonMotha | the term commonly used is "suicide battery" |
| 22:49.42 | adyer | MonMotha: ouch |
| 22:49.53 | MonMotha | they had an ASIC to decrypt the ROMs on the fly, and it used an external battery to back up the keys stored in some RAM |
| 22:50.07 | MonMotha | the batteries last about 5-8 years, depending on how much time the board spends unpowered |
| 22:50.56 | adyer | MonMotha: lol, we did that with a virtex-4 FPGA for another companies' video game project. total PITA |
| 22:51.07 | MonMotha | yeah, I can imagine |
| 22:51.18 | MonMotha | I really wish the FPGA makers would just put some actual flash in the package |
| 22:51.20 | MonMotha | even OTP |
| 22:51.34 | MonMotha | doesn't need to even be on the same die. Those things are so expensive that multi-die packaging wouldn't be a big deal |
| 22:52.07 | adyer | MonMotha: worst was the super cheapo batty socket they wanted to use, needed a pad on the PCB, and if it wasn't really clean, gunk would grow after a while and open the connection of short it out. |
| 22:52.55 | MonMotha | that's great. Virtex 4 FPGA (several $100 at a minimum, I'm guessing) and a 20 cent battery socket |
| 22:53.05 | adyer | MonMotha: they worry about someone decapping it and probing the wires. I think some of the new xilinx parts do have it tho as an option |
| 22:53.29 | MonMotha | adyer: well, there's the S3A series (I think that's what it is) |
| 22:53.33 | MonMotha | fully NV FPGA |
| 22:53.47 | MonMotha | well, it's a standard SRAM arch, but it has flash internal to populate the SRAM |
| 22:53.57 | MonMotha | so no external programming source (that can be probed) needed |
| 22:54.04 | adyer | MonMotha: Lattice did it first, IIRC |
| 22:55.31 | MonMotha | I guess my view is that if an attacker has the resources to decap your chip, you have other avenues of attack against them (mostly legal) |
| 22:55.54 | MonMotha | I realize that there are companies that will provide such services for you, but still |
| 22:56.43 | adyer | MonMotha: lots of them especially for off-shore cloners. |
| 22:56.55 | MonMotha | I guess |
| 22:57.18 | MonMotha | I've just never had a product where the hardware was so sensitive that it would be impossible to go after cloners through legal means |
| 22:57.34 | adyer | MonMotha: strangely enough, one of the big reasons for all the security stuff was to keep the distibutors honest when a game upgrade or new version came out |
| 22:57.46 | MonMotha | I can imagine that |
| 22:57.57 | MonMotha | most of them would have ROM programmers back in their service departments anyway |
| 22:58.01 | MonMotha | and EPROMs aren't that expensive |
| 22:58.09 | adyer | MonMotha: or disk duplicators |
| 22:58.12 | MonMotha | same idea, yeah |
| 22:58.35 | MonMotha | I had to resurrect my Blitz 2k Gold from the MAME HDD image, which probably has something to do with why it doesn't work quite right |
| 22:58.45 | MonMotha | hard drive was bad when I got it |
| 22:58.59 | MonMotha | my Blitz 99 fortunately worked (and yes, I have an image of that one :) |
| 22:59.08 | MonMotha | as did my Blitz 97 |
| 22:59.50 | adyer | MonMotha: IIRC the old quantums were good, newer ones not so good :-( |
| 22:59.58 | MonMotha | that would explain that scenario.. |
| 23:00.12 | *** mode/#edev [+o landley] by ChanServ |
| 23:00.15 | MonMotha | oh, and I just heard that my hardware passed FCC Part 15b, so that's excellent |
| 23:00.42 | MonMotha | apparently failed an edge-band test on the top channel of the transmitted, probably because the ref osc is running high (which can be tuned in software) |
| 23:33.30 | *** join/#edev linac (n=lin@121.41.47.12) |
| 23:35.16 | *** part/#edev adyer (n=adyer@adsl-69-223-13-78.dsl.chcgil.ameritech.net) |
| 23:58.38 | RobotGuy-Out | returns. |
