| 00:01.33 | *** join/#edev redblue (i=star@ppp153.109-253-207.mtl.mt.videotron.ca) |
| 00:29.03 | *** join/#edev dbtid (i=k2y0fv18@cpe-65-24-173-154.columbus.res.rr.com) |
| 01:08.30 | *** join/#edev Appleman1234 (n=Appleman@CPE-121-222-178-251.qld.bigpond.net.au) |
| 01:22.53 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 02:04.29 | RobotGuy | It's not the amount of flash you have - it's how you use it.. |
| 03:57.49 | *** join/#edev DrZimmerman (n=theo@tbaumgartner.broker.freenet6.net) |
| 03:59.50 | *** join/#edev Appleman1234 (n=Appleman@CPE-124-186-42-216.qld.bigpond.net.au) |
| 04:03.08 | solar | and putting python on an embedded device is the most questionable thing ever ? |
| 04:03.20 | *** join/#edev Redhatter (n=vk4fsjl@2001:388:f000:0:0:0:0:279) |
| 04:03.37 | solar | next to cough... ppl that think they need perl/php |
| 04:05.47 | *** join/#edev mbuf (n=shakthim@61.16.248.242) |
| 04:17.24 | *** join/#edev Sascha (n=sisawess@cssun.rrze.uni-erlangen.de) [NETSPLIT VICTIM] |
| 04:17.24 | *** join/#edev dfarnsworth (n=dfarnswo@65.200.49.156) [NETSPLIT VICTIM] |
| 04:23.29 | *** join/#edev dfarnsworth (n=dfarnswo@65.200.49.156) [NETSPLIT VICTIM] |
| 04:25.06 | *** join/#edev Sascha (n=sisawess@cssun.rrze.uni-erlangen.de) |
| 05:04.14 | ds2 | what a day |
| 05:06.09 | RobotGuy | I just need Python. :) |
| 05:06.30 | RobotGuy | Why is that questionable? |
| 05:08.01 | ds2 | it is python. enuff said ;) |
| 05:11.23 | RobotGuy | There is nothing wrong with using Python for an embedded system. |
| 05:11.37 | ds2 | there is something wrong with using Python. ;) |
| 05:11.46 | solar | RobotGuy: In your env it's ok.. But if a vendor attempted to pull that off as a base requirement they would be shot |
| 05:12.17 | ds2 | shot? you are so nice to your vendors |
| 05:12.17 | RobotGuy | Vendors use Python. |
| 05:12.25 | ds2 | I'd make sure they are drawn and quartered ;) |
| 05:12.56 | RobotGuy | What am I supposed to use - straight C? |
| 05:13.07 | ds2 | feel free to use assembly too ;) |
| 05:13.11 | solar | ds2 wins at being a better dictator :) |
| 05:13.22 | ds2 | nah.. perl, awk, or sh is fine =) |
| 05:13.24 | RobotGuy | I don't pay attention to ds2 |
| 05:13.34 | solar | RobotGuy: naa. Use what works best for the job at hand. |
| 05:14.11 | solar | RobotGuy: I think a few of us (myself included) always look for the lowest level soultion to accomplish a task |
| 05:14.58 | solar | so vs using python.. I'd start thinking. What are my goals with pytyon. What am I trying to accomplish.. |
| 05:15.23 | solar | But then.. It all boils down to resources... If you have them then by all means use them. |
| 05:15.27 | RobotGuy | solar: Everything I need is in Python. Web framework, IRC bot, etc. |
| 05:16.10 | RobotGuy | No matter what I use, somebody is going to take exception to it. |
| 05:16.11 | solar | RobotGuy: yeah.. I guess I'm just a little displeased with how many support files has to go along with that inter lang |
| 05:16.35 | kergoth | if i wanted a scripting language, say, for configuration, on an embedded device, i'd use lua, not python |
| 05:16.37 | solar | RobotGuy: As noted " Use what works best for the job at hand. |
| 05:17.08 | RobotGuy | kergoth: I don't know lua, and I think I already tried to nail it. |
| 05:17.25 | RobotGuy | I don't want to learn another new language right now. |
| 05:17.52 | kergoth | i know people who use python on embedded systems fine, just not what i'd use personally :) |
| 05:17.54 | solar | kergoth: see that logitech link/device posted the other day which makes use of lua for it's primary applet lang? |
| 05:18.08 | kergoth | nope, interesting |
| 05:18.17 | solar | sec let me see if I can dig it up. |
| 05:18.33 | solar | pretty sexy device actually |
| 05:18.44 | RobotGuy | I think I finally found a web server and the other components I need that work together. |
| 05:19.26 | solar | kergoth: http://wiki.slimdevices.com/index.php/SqueezeboxController |
| 05:20.00 | solar | RobotGuy: and still all fitting within the existing flash? |
| 05:20.14 | RobotGuy | solar: Yes. :) |
| 05:20.30 | solar | wonderful |
| 05:20.36 | kergoth | huh, interesting |
| 05:20.53 | solar | kergoth: you still doing much OE these days? |
| 05:21.09 | *** join/#edev Jx_ (i=Jx_@81.180.252.139) |
| 05:21.33 | solar | I had a chance to talk to one of your guys for a while at the last CELF (I forget his name [Mike?] but he was a really nice guy) |
| 05:21.44 | kergoth | naw, havent really been active in a couple years. if i was to get back into it, i'd want to revamp / rewrite everything and start from scratch... looking at it from a new perspective, i just don't like the way it went. i still think its one of the better solutions at the moment, but suboptimal |
| 05:21.45 | solar | and it was cool to see another FOSS vendor at the show |
| 05:21.48 | kergoth | ah, nice |
| 05:22.06 | kergoth | couldve been michael lauer aka mickey, he's one of the current core devs |
| 05:22.49 | solar | he had a bit of a PR head+style to him.. Good rep for the community |
| 05:34.56 | RobotGuy | kergoth: Why not do something new, the right way? |
| 05:35.39 | kergoth | if i had the motivation to do it, i would. still pretty burned out on programming in general. ruined the hobby |
| 05:35.48 | RobotGuy | Bummer |
| 05:36.24 | RobotGuy | I'm just getting excited about programming again, but what I am attempting to do is pretty major (at least for me). |
| 05:41.04 | kergoth | nice :) |
| 05:41.24 | RobotGuy | At one time, I was working on an admin aid for FreeBSD to make it easier to find stuff in Ports. I had everything all parsed out and ready to shove into a database. All in Python. |
| 05:42.56 | RobotGuy | kergoth: I will get it done, eventually, but would like to have at least the minimal behaviors coded and tested before Sept. I don't think I am going to make it by then then though. |
| 06:01.33 | *** join/#edev drath_ (i=vmaster@p5B07C12B.dip.t-dialin.net) |
| 06:34.09 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 06:40.01 | solar | kergoth: you went to work for MV right? they were also at the show (older gentlemen mid 50's.. Not real linux savy but nice and strong hardware knowledge).. |
| 06:57.49 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 07:15.41 | mehtril | yo |
| 07:15.45 | mehtril | yo yo :) |
| 07:26.32 | solar | evening and nite.. |
| 07:28.16 | mehtril | morning here :) |
| 07:29.52 | solar | would suggest moving but that would sound kinda odd |
| 07:30.45 | solar | jokes aside. have a nice day/nite |
| 07:31.19 | mehtril | the same fr you :) |
| 07:32.16 | *** join/#edev [AD]Turbo (n=turbo@host250-0-dynamic.11-87-r.retail.telecomitalia.it) |
| 07:34.35 | [AD]Turbo | yo all |
| 08:59.10 | mehtril | some u-boot JTAG load guru here? |
| 10:45.49 | *** join/#edev g1powermac_PB (n=g1powerm@unaffiliated/g1powermac) |
| 11:29.32 | *** join/#edev g1powermac (n=g1powerm@unaffiliated/g1powermac) |
| 11:34.20 | *** join/#edev sjhill (n=sjhill@real.realitydiluted.com) |
| 11:34.20 | *** mode/#edev [+o sjhill] by ChanServ |
| 12:23.51 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 12:23.51 | *** mode/#edev [+o GPSFan] by ChanServ |
| 13:04.25 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 13:23.48 | RobotGuy | g1powermac: Mornin' |
| 13:28.26 | *** join/#edev T0mW (n=Tom@24.115.218.135.res-cmts.sth.ptd.net) |
| 13:33.42 | RobotGuy | T0mW: Hi |
| 13:36.54 | *** join/#edev MothaWork (n=monmotha@mail.forwardengineering.com) |
| 13:40.08 | T0mW | RobotGuy: good morning. I'm trying wake up drinking decaf |
| 13:40.22 | *** join/#edev chouimat|work (n=dieu@209.217.106.98) |
| 13:40.36 | RobotGuy | I'm just considering whether to go out in the rain or not. |
| 13:40.53 | T0mW | looks outside at the sunshine |
| 13:41.04 | RobotGuy | I'm in Oregon. :) |
| 13:41.07 | T0mW | lol |
| 13:41.22 | RobotGuy | That should explain everything. :) |
| 13:41.24 | T0mW | that does happen there a lot, I am in Pennsylvania |
| 13:41.31 | chouimat|work | same here |
| 13:41.36 | chouimat|work | in ottawa |
| 13:42.06 | T0mW | chouimat|work: rain, or that sh!t happens a lot in Ottawa? |
| 13:42.29 | chouimat|work | rain ... and sh!t too ... the federal parlement is here |
| 13:42.36 | T0mW | LOL |
| 13:42.50 | RobotGuy | T0mW: Have you done anything with Gadget Ethernet? |
| 13:43.14 | T0mW | RobotGuy: well, I just tinkered with it. It does work |
| 13:43.26 | RobotGuy | I can't get the host side working. |
| 13:43.42 | T0mW | hmmm, linux host? |
| 13:43.46 | RobotGuy | Yes |
| 13:44.11 | T0mW | well, you have to set an ip address (ifconfig) the client (gadget). |
| 13:44.22 | RobotGuy | I have it configured on Hammer. |
| 13:44.23 | T0mW | s/the/on the |
| 13:44.42 | g1powermac | heya RobotGuy |
| 13:44.46 | T0mW | run ifconfig on the hammer and see if you get a USB device listed there. |
| 13:45.04 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 13:45.07 | RobotGuy | T0mW: I do have it on Hammer, all configured. It's the HOST side I can't get working. |
| 13:45.45 | T0mW | did you try using arp to set the hardware (MAC) number to an IP address of the client? |
| 13:46.19 | RobotGuy | T0mW: ?? |
| 13:46.31 | T0mW | example: arp -s 192.168.2.2 00:01:02:03:04:05 |
| 13:46.50 | RobotGuy | I don't have usb0 device on the host. |
| 13:47.13 | T0mW | at first, I couldn't get the gadget to work, so I first thought it was some issue with arp on the host. |
| 13:47.45 | T0mW | I saw the MAC on the client and added it to the arp table of the host as the ip address I wanted the client to be at. |
| 13:47.48 | RobotGuy | I don't understand about arp |
| 13:48.41 | T0mW | then I finally figured out that I needed to the ifconfig on the client (gadget) to set the ip of the usb port. |
| 13:48.41 | T0mW | oHHHH |
| 13:48.41 | T0mW | you also have to set the ip address on the host usb port device as well. |
| 13:48.53 | T0mW | don't bother with arp, just use ifconfig. |
| 13:49.03 | RobotGuy | I know. I just don't get the usb0 device on the host. |
| 13:49.07 | T0mW | do this: |
| 13:49.14 | T0mW | tail -f /var/log/messages |
| 13:49.28 | T0mW | then plug and unplug the hammer from the usb of the host |
| 13:49.40 | T0mW | you should see the host recognizing a new usb device |
| 13:50.05 | T0mW | messages will then tell you the device name of the new usb device. |
| 13:50.38 | T0mW | there is probably a cleaner way to setup this, but I just wanted to verify that the ethernet gadget worked. |
| 13:53.24 | RobotGuy | It doesn't work no matter what I do. :( |
| 13:53.37 | T0mW | RobotGuy: if you use the base board for the hammer, the one with the prototyping grid, power supply etc., that board should have a properly wired slave port |
| 13:53.54 | RobotGuy | I'm using a Nail right now. |
| 13:54.03 | T0mW | if you don't see the client (tail messages), then you don't have something configured within the kernel on the client. |
| 13:54.22 | RobotGuy | Client = Hammer, right? |
| 13:54.30 | T0mW | RobotGuy: do you know that the slave port is wired differently than a usb host port? |
| 13:54.40 | RobotGuy | I don't know. |
| 13:54.41 | T0mW | correct, client == hammer. |
| 13:54.53 | RobotGuy | I have usb0 on Hammer and it is configured. |
| 13:55.31 | T0mW | a host port has a 15K resistor on each data line wired to GND |
| 13:56.20 | T0mW | a slave port has a 470K to GND on D-, then a 1.5K to VCC on D+ |
| 13:56.49 | RobotGuy | I just need to get the host side configured. |
| 13:56.57 | T0mW | I'm assuming that the nail board does not have the actaul connectors on the it? |
| 13:57.09 | RobotGuy | Connectors for what? |
| 13:57.13 | T0mW | RobotGuy: USB |
| 13:57.25 | RobotGuy | It just has two USB connectors, one host and one slave |
| 13:57.29 | T0mW | ok ok |
| 13:58.15 | RobotGuy | Slave connects to PC for JTAG, etc. |
| 13:58.37 | T0mW | then, if you have the nail configured to respond as a gadget, looking at the messages log on the host should show that the host OHCI / UHCI USB sees the new client when you plug it into the system. |
| 13:59.33 | T0mW | It may not yet configure it automagically, but it should still see some USB device, assign as USB address to it, then identify it (read manufactuerer ID, name strings, ...). |
| 14:00.34 | T0mW | the nail slave port is where the gadget driver would be, the type-A host connector cannot behave as a slave there. |
| 14:01.25 | T0mW | RobotGuy: jtag plugs into the usb slave port? |
| 14:01.39 | RobotGuy | I know. I don't always get the third USB device. Yes, JTAG on the slave port. |
| 14:01.46 | T0mW | odd |
| 14:01.59 | T0mW | maybe it plugs into that for power |
| 14:02.33 | RobotGuy | Nail does get power from USB, and same port gives JTAG and serial console. |
| 14:02.37 | T0mW | seems odd though as the jtag header should be able to provide power to a simple bitbanger jtag. |
| 14:02.47 | RobotGuy | It should also give Gadget Ethernet. |
| 14:03.06 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 14:03.06 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 14:04.58 | T0mW | I have a couple of Hammer boards here (and some SODIMMs), maybe you should ask one of the other guys who have a nail board? |
| 14:05.15 | T0mW | gadget does work. |
| 14:07.39 | RobotGuy | I know. It just doesn't work for me. :( |
| 14:08.36 | RobotGuy | Well, I am going to get some food. BIAB. |
| 14:12.07 | MothaWork | speaking of USB, has anybody ever gotten OTG to work on anything? |
| 14:12.37 | CosmicPenguin | I saw it work once, but that might have been a dream |
| 14:12.50 | MothaWork | well, it's a hack and a half, that's for sure |
| 14:26.47 | MothaWork | why oh why couldn't we all have just decided upon IEEE1394 |
| 14:28.25 | ohsix | most cameras do otg for printer connections |
| 14:28.42 | ohsix | my old a75 does |
| 14:28.54 | MothaWork | and it actually works? |
| 14:29.03 | ohsix | yea |
| 14:29.25 | chouimat|work | otg? as in on the go? |
| 14:30.19 | MothaWork | yes, on the go |
| 14:32.52 | ohsix | firewire everywhere would be fub with the host dma stuff |
| 14:33.39 | MothaWork | well, USB is fine for what it was originally intended: keyboards, mice, UPSs, etc. |
| 14:33.46 | MothaWork | really low data rate, dumb peripherals |
| 14:34.13 | MothaWork | even using it for things like audio devices is pushing it, though isosychronous mode does seem to work |
| 14:34.16 | chouimat|work | usb hdd sucks |
| 14:34.37 | ohsix | i like mine :] |
| 14:35.01 | MothaWork | anything involving higher data rates (mass storage) or "smart" devices (pdas, modern printers, digital cameras, etc.) should really use a smarter bus like IEEE1394 |
| 14:35.22 | MothaWork | high-speed USB should never have come into existence |
| 14:35.32 | ohsix | how do you rank which is smarter |
| 14:35.38 | MothaWork | my CPU shouldn't have to babysit my hard drive |
| 14:35.57 | MothaWork | I'd say anything with an embedded processor capable of "substantial processing power" qualifies as smart |
| 14:36.20 | MothaWork | that describes just about all modern printers, digital cameras, cell phones, and of course PDAs |
| 14:36.41 | MothaWork | basically, anything with a micro that is capable of DMA |
| 14:36.44 | ohsix | i meant the bus |
| 14:36.49 | MothaWork | and there are 8-bit micros out there with DMA controllers |
| 14:37.17 | MothaWork | the 1394 bus architecture is cleaner, especially wrt peer-to-peer traffic |
| 14:37.26 | MothaWork | USB is extremely master/slave oriented |
| 14:37.32 | MothaWork | which is why OTG is so hackish |
| 14:37.53 | MothaWork | OTOH, that cleaner arch ends up requiring more brains on every device |
| 14:38.01 | MothaWork | while USB gets to offload most of the brains to the host |
| 14:38.08 | MothaWork | makes your mouse cheap, which is fine |
| 14:39.12 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 14:39.12 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:40.55 | MothaWork | what I would have liked to see is something along the lines of a "reduced function" 1394 controller interface. Basically, yank all the DMA stuff and cut it down to the bare minimum required to make things work. Force it to be an edge node on the bus, for example. |
| 14:41.08 | MothaWork | I realize that's hard, but it would have been nice and would have eliminated the need for USB OTG |
| 14:41.29 | MothaWork | since if I had a 1394 keyboard (reduced function), and a 1394 capable cell phone, I could have just plugged it in |
| 14:41.54 | MothaWork | as it is, I had to wait several years for OTG to get standardized and (it seems, universally poorly) implemented |
| 14:42.16 | MothaWork | all so that I can plug a keyboard into a PDA, which is a quite reasonable thing (from a user POV) to do |
| 14:43.13 | MothaWork | I'll admit that I don't have a ton of experience with 1394, but everything I've seen makes it look like a dream compared to USB |
| 14:43.24 | MothaWork | and I do have a fair bit of experience with USB |
| 14:44.17 | ohsix | theres the problem with device power, and firewire takes a lot of juice in general |
| 14:44.51 | MothaWork | you mean for the bus signaling itself or provided power for the other end |
| 14:44.52 | MothaWork | ? |
| 14:44.59 | ohsix | with usb, what the host has to do has a lot to do with the controller |
| 14:45.08 | ohsix | ehhh |
| 14:45.42 | ohsix | you only get host power from the 6 pin connectors with firewire |
| 14:45.59 | MothaWork | I'm not really trying to argue that USB has no place, just that several modern uses of it outright suck, OTG is a hack, and IEEE1394 seems to get it "more right" in most situations |
| 14:46.35 | MothaWork | and yes, power is only provided on the 6-pin "big" 1394 connector. Are you referring to that power or for power required for actual bus signaling? |
| 14:46.55 | ohsix | me either, but its not very black & white, otg is pretty low key |
| 14:47.15 | ohsix | i was referring to both, and the host controller |
| 14:47.29 | MothaWork | well, the issue is that the functionality provided by OTG is becoming more and more important, and yet it's so hackish and never seems to work right from a "black box" end user POV |
| 14:47.37 | ohsix | net power use for portable widgets |
| 14:48.11 | ohsix | works pretty awesome for cameras :D |
| 14:48.33 | MothaWork | well, I'm guessing that if you allowed a reduced function host controller for 1394 (possibly amending the specification to allow for it), you could probably get power consumption lower. Now, the bus signaling would be unaffected of course |
| 14:48.38 | ohsix | only thing i've really seen it in |
| 14:48.39 | MothaWork | I've never had OTG work for anything |
| 14:48.58 | MothaWork | and I tend to try to use it for simple things like hooking up a keyboard to a cell phone |
| 14:49.04 | ohsix | the devices need to have common classes they support |
| 14:49.10 | MothaWork | perhaps they just don't intend that, though it seems a very, very obvious use |
| 14:51.09 | ohsix | firewire would need a lot of bolstering as far as device class/peer functions actually work if it was going to take up the role |
| 14:51.48 | ohsix | usb is pretty kitchen-sinky, it has all that crap already |
| 14:52.29 | MothaWork | yes, that is something the 1394 crowd would need to address |
| 14:52.41 | MothaWork | I think that the reason it has never really been addressed is that USB dominated that field from the get-go |
| 14:52.49 | MothaWork | the USB HID spec is pretty decent |
| 14:52.54 | MothaWork | I'm not sure sure I like some of the others |
| 14:53.08 | MothaWork | and I think at this point 1394 has no hope of displacing USB |
| 14:53.25 | MothaWork | I kinda wish it had done so back when there was an actual "contest" |
| 14:54.06 | ohsix | there never really was, the firewire proponents used/marketed it at what it was designed for |
| 14:55.11 | MothaWork | you mean what Apple intended it to be used for |
| 14:55.36 | MothaWork | IEEE1394 is basically the IEEE equivalent to USB similar to how IEEE802.15.4 is the IEEE equivalent to Bluetooth |
| 14:55.53 | MothaWork | they don't solve a 100% intersecting set of problems, and they don't approach it the same way |
| 14:56.00 | ohsix | sony was a big proponent too |
| 14:56.02 | MothaWork | but they are targeted at roughly the same problem se |
| 14:56.03 | MothaWork | t |
| 14:56.06 | MothaWork | true |
| 14:56.13 | ohsix | huge, even |
| 14:56.20 | MothaWork | both of them really only saw it being used for interactive media applications |
| 14:56.37 | MothaWork | while that was a big intended use by IEEE, it wasn't the only intended use |
| 14:57.43 | ohsix | it took over scsi for peripherals and stuff, at least thats what apple did with it |
| 14:58.05 | ohsix | in that sense the host controllers are pretty cheap and power efficient :] |
| 14:58.38 | MothaWork | true, though in the PC world people tended to stick with ATA/IDE |
| 14:58.51 | MothaWork | likely because they were cheap enough to begin with |
| 14:58.59 | *** join/#edev sunrunner20 (n=jamesdoe@pool-71-170-40-118.dllstx.fios.verizon.net) |
| 14:59.08 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 14:59.27 | ohsix | anything that touched mpeg-ts had a firewire port to get it onto a pc |
| 14:59.50 | MothaWork | that's going away these days |
| 15:00.20 | ohsix | then theres the STB firewire mandate |
| 15:00.26 | MothaWork | but yes, real-time media was one of the few places that 1394 took off - probably because that's the primary consumer application both Apple and Sony promoted it for |
| 15:00.39 | MothaWork | yeah - which all the providers promptly disable |
| 15:00.56 | MothaWork | or ship STBs without it anyway and give you the runaround when you ask |
| 15:01.02 | ohsix | well they've differentiated a bit, but pretty much any dv cam still has one |
| 15:01.14 | MothaWork | there are DV cams being shipped now with only USB |
| 15:01.23 | MothaWork | especially common on HDD based cameras |
| 15:01.31 | MothaWork | but even some using DV/MiniDV |
| 15:01.44 | ohsix | well that fits the disk better |
| 15:01.58 | MothaWork | not really; they still stream it off as MPEG-TS or DV |
| 15:02.14 | MothaWork | most of the time they do not show up as mass storage |
| 15:02.24 | ohsix | the dv cams with usb can be used as multiple device classes usually, webcam and whatnot |
| 15:02.47 | MothaWork | you could usually do that with the 1394 connection also |
| 15:03.00 | MothaWork | just have it dump the DV stream live onto the line and snap what frames you want |
| 15:03.11 | ohsix | but you'd need a software stack for it on the host |
| 15:03.21 | MothaWork | which all capture programs supported |
| 15:03.38 | MothaWork | heck, I think my friend's DV cam shipped with a Video4Windows driver |
| 15:04.02 | ohsix | that would be weird |
| 15:04.43 | ohsix | vfw was axed long before usb |
| 15:06.03 | MothaWork | well, whatever the current Windows video subsystem is called, then |
| 15:06.13 | MothaWork | basically, it shows up in the Windows TV support stuff |
| 15:06.19 | ohsix | now all you need is a dshow miniport if it doesn't already have a device class, and ones with vfw drivers got wrapped by dshow proxies, and vice versa for using new sources in old software |
| 15:06.22 | MothaWork | as terrible as that stuff is |
| 15:06.45 | MothaWork | I know on Linux, 1394 based cameras show up as V4L devices with no additional drivers |
| 15:06.50 | MothaWork | so yes, you can use them as a webcam |
| 15:07.38 | MothaWork | I really don't touch windows enough anymore to know how they do that stuff |
| 15:07.53 | ohsix | likely from mpeg-ts, usb can have multirole devices, they can be different things at the same time |
| 15:08.04 | MothaWork | oh yes, I'm aware |
| 15:08.13 | MothaWork | I've done the hardware and software side of several USB devices |
| 15:08.57 | MothaWork | the alternate configurations and multiple interfaces stuff can get quite confusing if you actually try to use it as they seem to intend (and nobody does) |
| 15:09.02 | ohsix | it only means you can pick & choose, and wedge custom stuff in there easily |
| 15:09.24 | MothaWork | but I've had devices that show up as HID + storage. That's fun |
| 15:09.52 | ohsix | hid & audio, for the volume control keys, popular combo |
| 15:09.56 | MothaWork | indeed |
| 15:10.06 | MothaWork | this was a storage device with a "backup" type button on it |
| 15:10.32 | MothaWork | that sort of scenario works fairly well |
| 15:10.40 | ohsix | hid is the best, 90% of stuff has buttons |
| 15:11.01 | MothaWork | what doesn't seem to work well is when you try to use alternate configurations to choose between mutually exclusive featuresets |
| 15:11.09 | MothaWork | i.e. I can present this as a gamepad or a keyboard |
| 15:11.44 | MothaWork | I think the spec uses like an ISDN modem that can also act as a raw serial port or some similar nonsense |
| 15:12.03 | MothaWork | it works OK in Linux, though you need a driver to pick the configurations. Windows seems to get very confused |
| 15:12.09 | ohsix | well thats basically just a flag to tell the host what it might do, nothing stopping you from having letters on a gamepad or analog controls on a keyboard |
| 15:12.16 | MothaWork | though that may just be that I almost never write drivers for Windows |
| 15:12.31 | MothaWork | oh, I'm aware of how the HID spec works. |
| 15:12.58 | MothaWork | This really is "I can present these buttons either in their true form where they are special function keys on a gamepad, or I can present it as a keyboard that types letters and you map them using my special software" |
| 15:13.08 | MothaWork | and of course, you probably don't want both |
| 15:13.15 | MothaWork | at the same time, at least |
| 15:13.26 | MothaWork | HID is actually one of the few parts of USB I think is really rather good |
| 15:13.27 | ohsix | not on windows, in general |
| 15:13.37 | MothaWork | it's amazingly complete and still fairly simple |
| 15:13.58 | MothaWork | complicated usage descriptors can run into somewhat undefined territory, but otherwise it works fine |
| 15:14.57 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 15:19.18 | ohsix | the way firewire works makes it pretty hard to virtualize too |
| 15:28.56 | *** join/#edev SpeedEvil (n=fjfjfjf@tor/regular/SpeedEvil) |
| 15:30.09 | SpeedEvil | On a sort-of-ontopic announce, that others have probably made - http://us.direct.openmoko.com/products/neo-freerunner - on sale now! (the store seems to imply there are 430 left if you enter 1000 into the quantity box) |
| 15:31.03 | SpeedEvil | (Linux phone with actual open-source everything. (no, it's not a user-ready phone really yet, but it is arm + LCD+GPS+GSM+SD+BT)) |
| 15:38.28 | Crofton | I want a quad band :( |
| 15:39.22 | RobotGuy | I like the Freerunner. I just wish I could afford it. I think I'd rather get a Pandora once it is clear about the software situation. |
| 15:39.53 | SpeedEvil | Crofton: yeah - it's triband only. |
| 15:40.09 | SpeedEvil | Crofton: though you can pick the bands. (but they are sold out of one) |
| 15:42.41 | RobotGuy | I'd need the 850 version |
| 15:43.31 | SpeedEvil | There are unfortuante hardware issues ATM. |
| 15:43.55 | SpeedEvil | Primarily the 2D/3D graphics accel/mpeg decoder has turned out to be almost a wash. |
| 15:44.27 | RobotGuy | The question is: Is it still better than what is usually available? |
| 15:44.52 | SpeedEvil | It is connected over a 7mbytes/second bus, and DMA doesn't work, so if you're using 3.5M/second transfer, then you lose 50% CPU |
| 15:44.55 | SpeedEvil | Well... |
| 15:45.04 | CosmicPenguin | ahhh - pixil |
| 15:45.08 | CosmicPenguin | we hardly knew yee |
| 15:45.25 | SpeedEvil | If you are talking about opensource phones encouraged by the manufacturer to hack on it - clearly yes, as there are almost no others. |
| 15:46.06 | SpeedEvil | If you are talking about simple hardware - that's a lot less clear at this price point than when it was first announced. |
| 15:47.18 | SpeedEvil | Then add hackable phones that are not encouraged by the maker, and have no docs, and some hardware may not work with linux, and it gets a lot more muddy. |
| 15:48.08 | RobotGuy | I wonder how carriers in the USA will react to the OpenMoko phones. |
| 15:48.58 | MothaWork | RobotGuy: by ensuring that service in major areas is available only on bands not supported by the OpenMoko RF side :) |
| 15:49.48 | SpeedEvil | RobotGuy: on GSM carriers, you pull out your existing SIM, and slap it in. |
| 15:49.56 | RobotGuy | MothaWork: Freerunner will be available in two different three band versions - 850 and 900 are the bands that are different between versions. |
| 15:50.06 | RobotGuy | I am on AT&T (GSM). |
| 15:50.12 | SpeedEvil | 900MHz is not used in the US. |
| 15:50.18 | RobotGuy | Yes it is. |
| 15:50.21 | SpeedEvil | So if you're solely in the US, you don't need it. |
| 15:50.25 | SpeedEvil | It is? |
| 15:50.34 | RobotGuy | 900 is used by T-Mobile and a few others. |
| 15:50.34 | solar | is all about tct warming up to the idea of making a better phone but leaving a slot open for telit chips that you select and put in yourself. |
| 15:50.54 | SpeedEvil | RobotGuy: sure you're not thinking about 850? |
| 15:50.59 | RobotGuy | A Gentoo Embedded Open Source phone! |
| 15:51.10 | RobotGuy | I know the US carriers. |
| 15:51.26 | RobotGuy | Unless T-Mobile has switched recently, they use the 900 MHz band. |
| 15:51.35 | SpeedEvil | RobotGuy: I mean GSM |
| 15:51.44 | RobotGuy | I know what you mean. |
| 15:52.01 | mnemoc | doh, the french distributor is much "cheaper" |
| 15:52.10 | RobotGuy | My old T-Mobile phone (Nokia 6600) is 900 |
| 15:52.33 | RobotGuy | My new Blackjack II (AT&T) is 850 |
| 15:52.46 | RobotGuy | Actually, it's quad band, I think. |
| 15:52.47 | SpeedEvil | 900MHz only? |
| 15:52.57 | SpeedEvil | Are you sure you're not thinking of 1900? |
| 15:53.02 | RobotGuy | SpeedEvil: No, 900 MHz is the third band. |
| 15:53.13 | RobotGuy | SpeedEvil: No, 900 MHz is the third band. |
| 15:55.46 | SpeedEvil | RobotGuy: the 6600 supported 1900MHz too - which I see references to t-mobile using. |
| 15:55.56 | RobotGuy | SpeedEvil: Go to the T-Mobile site and check for yourself if youn don't believe me. |
| 15:56.02 | RobotGuy | The 6600 is triband. |
| 15:57.14 | SpeedEvil | Yes, I know - it supports 900MHz. |
| 15:57.32 | SpeedEvil | I haven't seen any reference other than yours to the US using 900MHz on GSM |
| 15:59.03 | SpeedEvil | And http://en.wikipedia.org/wiki/T-Mobile_(USA) agrees (1900) I can't see anything on the t-mobile site, but it's being slow. |
| 15:59.13 | RobotGuy | Whatever. Go check for yourself. I know my phones. |
| 15:59.40 | ohsix | e-fite |
| 16:00.06 | SpeedEvil | RobotGuy: I'm not saying you don't - I'm just wondering how you came to the conclusion it was operating on 900MHz. |
| 16:00.22 | SpeedEvil | Observation of course trumps wikipedia. |
| 16:00.37 | RobotGuy | The phone's specs state the operating bands. |
| 16:00.38 | MothaWork | quad band phones show up in the USA all the time, but last I heard the USA did not have one of the bands allocated to PCS or mobile telephone at all |
| 16:00.43 | RobotGuy | Real simple. |
| 16:01.04 | MothaWork | just because the phone is capable of operating on a band doesn't mean that it does so in the USA if there is no service on those bands |
| 16:01.11 | RobotGuy | T-Mobile appears to be switching to quad band service. |
| 16:01.30 | MothaWork | they got the FCC to allocate that 4th band to PCS or mobile telephony? |
| 16:01.37 | RobotGuy | Of course there has to be service for a phone to operate on a particular band. |
| 16:01.51 | MothaWork | perhaps you mean that T-Mobile is selling quad band phones |
| 16:02.08 | MothaWork | i.e. they are selling handsets capable of operating internationally, which can be quite handy |
| 16:02.08 | RobotGuy | MothaWork: Yes, they are selling quad band phones. |
| 16:02.21 | SpeedEvil | RobotGuy: but if the phone supports 900/1800/1900, and there is only a 1900 network, it won't use 1800 or 900 |
| 16:02.39 | RobotGuy | SpeedEvil: I'm not stupid! I know that, sheesh. |
| 16:03.00 | SpeedEvil | https://support.t-mobile.com/knowbase/root/public/tm22037.htm |
| 16:03.07 | *** part/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 16:03.13 | SpeedEvil | T-Mobiles domestic roaming partners all operate on the GSM 1900 band. |
| 16:03.24 | ohsix | he's a bit fragile |
| 16:04.04 | SpeedEvil | And that page lists only one frequency supported by all phoens. |
| 16:04.05 | SpeedEvil | Oh. |
| 16:04.08 | ohsix | last time i was roaming it was on 850, south eastern oregon |
| 16:04.08 | SpeedEvil | Oh well. |
| 16:04.24 | SpeedEvil | Yeah - they have a 'we are adding 850 in some areas page' |
| 16:04.30 | ohsix | i dont have my phone here but iirc its on 900 at home |
| 16:04.58 | MothaWork | looks like there is a VERY small portion of the international 900MHz GSM band allocated to that service in the USA |
| 16:04.58 | kergoth | morning |
| 16:05.13 | MothaWork | only 5MHz of spectrum |
| 16:05.40 | MothaWork | I'm not sure that it's widely used |
| 16:05.55 | MothaWork | that may be the oddball Nextel allocation |
| 16:59.58 | SpeedEvil | For those times you need refreshmen when hacking USB devices. usbwine.com |
| 17:30.27 | *** join/#edev axarob (n=ebash@cpc2-barn4-0-0-cust162.brnt.cable.ntl.com) |
| 17:49.08 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 17:58.37 | MothaWork | I think hacking on USB devices (at least the USB portion thereof) may sometimes require something stronger than mere wine |
| 18:03.40 | *** join/#edev Redhatter (n=vk4fsjl@2001:388:f000:0:0:0:0:279) |
| 18:06.27 | CosmicPenguin | crowbars come to mind |
| 18:37.34 | g1powermac | meh, sledge hammers work better :-) |
| 18:46.55 | *** join/#edev g1powermac_PB (n=g1powerm@unaffiliated/g1powermac) |
| 18:50.49 | *** join/#edev axarob (n=ebash@cpc2-barn4-0-0-cust162.brnt.cable.ntl.com) |
| 18:55.27 | SpeedEvil | On that topic |
| 18:55.48 | SpeedEvil | can the hammer do a transparent usb snooper? |
| 18:57.20 | CosmicPenguin | I don't think it would be transparent |
| 18:57.36 | SpeedEvil | I mean pretending to be a hub |
| 18:57.48 | CosmicPenguin | yeah - it could be a hub |
| 18:57.50 | CosmicPenguin | probably |
| 18:58.01 | CosmicPenguin | if nothing else, it could be a repeater |
| 19:01.00 | *** join/#edev dijenerate (n=dijenera@72.51.69.246) |
| 19:05.01 | *** join/#edev RobotGuy-Out (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 19:42.28 | kergoth | so what are people using the hammer/nail/etc for, just learning tools, or prototyping? |
| 19:45.33 | solar | I've not found a legit use for it yet. But I'm having fun connecting misc devices to it and getting those to work. |
| 19:45.43 | kergoth | hehe |
| 19:47.01 | solar | currently focused on getting http://blog.tmcnet.com/blog/tom-keating/digital-cameras/logitech-quickcam-orbit-af-review.asp and |
| 19:47.26 | solar | http://www.provantage.com/siig-ce-s00022-s1~7SII2026.htm |
| 19:48.37 | solar | sound I think might be working.. But silly mpg123 links to libltdl and thats been giving me a few problems cuz I flush all *.la files. |
| 19:48.48 | solar | so need to find a diff userland app to test with. |
| 21:49.54 | *** join/#edev CosmicPe1guin (n=nobody@163.181.251.103) |
| 21:53.35 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 22:06.56 | *** join/#edev sunrunner20|prom (n=jamesdoe@pool-96-226-38-134.dllstx.fios.verizon.net) |
| 22:41.14 | *** join/#edev roxfan2 (n=dunno@253.0-200-80.adsl-dyn.isp.belgacom.be) |
| 22:58.33 | *** join/#edev ce_geek (n=ce_geek@c-24-61-92-197.hsd1.nh.comcast.net) |
| 23:18.12 | *** join/#edev cajaka (i=dirc@chil-nrp1-cs-1345.vdsl.bright.net) |