IRC log for #utah on 20150810

05:54.07*** join/#utah infobot (ibot@69-58-76-73.ut.vivintwireless.net)
05:54.07*** topic/#utah is Utah related discussion including user groups | logs @ http://ibot.rikers.org/%23utah/ | See also ##lds
05:54.07*** mode/#utah [+o infobot] by ChanServ
05:58.18*** join/#utah OldCoder (~OldCoder@unaffiliated/oldcoder)
06:09.19*** join/#utah davido__ (~davido@c-50-168-241-168.hsd1.ut.comcast.net)
06:12.31*** join/#utah davido__ (~davido@c-50-168-241-168.hsd1.ut.comcast.net)
06:15.57*** join/#utah spangborn (~spangborn@216.51.232.249)
06:47.06*** join/#utah Migs (~migs@205.122.50.108)
07:12.12*** join/#utah weasy (~weasy@2601:681:4501:57c2:70e9:b443:305d:7ecf)
12:41.20*** join/#utah hydroplane (~hydroplan@2601:681:4502:56e0:29b8:43ca:dc3b:a737)
13:10.49*** join/#utah emcnabb (emcnabb@nat/redhat/x-nwxlhewxrhlmgxqb)
13:10.49*** mode/#utah [+v emcnabb] by ChanServ
13:14.06*** join/#utah pdayton (~Adium@50.255.199.81)
13:56.50*** join/#utah jalbretsen (~jalbretse@sonic.itransact.com)
14:17.46*** join/#utah nage (~nathan.ge@unaffiliated/nage)
14:24.36*** join/#utah josephscott (~josephsco@192.0.67.251)
14:42.41*** join/#utah davido_work (~doswald@216.50.142.194)
14:44.20*** join/#utah resplin (~richard@68.69.160.206)
14:46.20*** join/#utah kendsnyder (~Adium@68.69.174.66)
15:47.21leviTodPunk: PoE works because Ethernet over UTP uses differential signaling, which means that while the sender is pulling one line high it's pulling the other line low by the same amount. The receiver subtracts the signal on one line from the other; therefore any common-mode (i.e. same on both wires of the pair) noise or DC bias is eliminated. One pair is given the high-level DC bias while another is given a
15:47.22levilow-level DC bias; these don't affect the signals, but they effectively provide a DC voltage between the two pairs.
15:49.50leviTodPunk: Condenser microphones that plug in via XLR cables work in a similar way, except that XLR has a third conductor for ground rather than separately biasing two pairs.
16:09.13*** join/#utah emcnabb (emcnabb@nat/redhat/x-odgghhumagpkabnn)
16:09.13*** mode/#utah [+v emcnabb] by ChanServ
16:17.52Tod-Worklevi, that's essentially how ethernet-over-power works too, if I recall correctly
16:20.24*** join/#utah fugufish (~fugufish@67.207.38.212)
16:24.20Tod-WorkI have totally lied, that's not at all how it works
16:30.36leviIIRC, ethernet-over-power just overlays a high-frequency signal on top of the 60Hz AC carrier. Normal AC-powered devices filter that stuff out or aren't affected by it.
16:32.18leviSo a high-pass filter gets you your data and a low-pass filter gets you a (relatively) clean 60Hz sine wave for power.
16:33.56leviActually dealing with the high-frequency signal works more like WiFi than Ethernet, again IIRC.
16:36.19leviHaving looked at the schematics and done some experiments with some HomePlug AV modules, I can tell you that they use the same sort of line drivers that ADSL modems do. And internally, this one was running basically the same Linux-on-MIPS that powers WRT-compatible routers.
16:36.53Tod-WorkWent to Lowe's yesterday and found I can fit a Raspberry Pi (again, as a worst-case scenario) completely inside a 2-gang box for $1
16:39.36leviNo doubt the cheapest part of the project. :)
16:40.02Tod-WorkMost likely, yeah
16:40.17Tod-Workgotta get a couple spools of cat6 sometime this week
16:46.24leviSpeaking of MIPS wireless router chips, you can get modules with MediaTeck MT7620n on them for cheap, and it's got I2S and SPI to talk to a codec and smart display module.
16:46.54levihttp://www.cnx-software.com/2015/01/23/vonets-vm300-wi-fi-ethernet-iot-board-with-mediatek-mt7620-runs-openwrt/
16:48.00leviThis would give you an AP in every room you've got an intercom panel. ;)
16:51.04leviCool, both the datasheet and programmer's reference manual are available online.
16:56.27leviThe embedded ethernet switch in it is pretty powerful. It's got a powerful packet recognition engine that you can program, which would let you put your audio packets in high-priority queues automatically.
17:30.07*** join/#utah hydroplane (~hydroplan@c-24-10-151-20.hsd1.ut.comcast.net)
17:51.11*** join/#utah kendsnyder (~Adium@68.69.174.66)
18:53.17*** join/#utah peno (~irc4me@c-73-131-204-209.hsd1.ut.comcast.net)
18:55.38*** join/#utah hydroplane (~hydroplan@c-24-10-151-20.hsd1.ut.comcast.net)
19:04.45*** join/#utah cstead (~cstead@50-198-176-49-static.hfc.comcastbusiness.net)
19:12.11*** join/#utah josephscott (~josephsco@192.0.67.251)
19:16.53*** join/#utah hydroplane (~hydroplan@c-24-10-151-20.hsd1.ut.comcast.net)
19:17.11*** join/#utah yukaia (~yukaia@74-81-251-245.static.logl001.digis.net)
19:19.41Tod-Worklevi, that would be pretty cool.  The touch screen would have to be bitbanged via GPIO, but the display could be pretty simple overall
19:19.51Tod-Workjust a lot less flexible on display stuff and more flexible on network stuff
19:20.54*** join/#utah hydroplane (~hydroplan@c-24-10-151-20.hsd1.ut.comcast.net)
19:28.37leviWell, if you used the displays I was talking about before, they're controlled via SPI and are plenty fancy.
19:29.36Tod-Worktouchscreen
19:29.38Tod-Worktouch
19:29.39Tod-Workscreen
19:29.44leviUm, yes.
19:29.48leviThey're touchscreens.
19:30.02Tod-Workthe display is obviously spi, but the touchscreen portion is probably not?
19:30.05leviApparently you didn't watch the cool video demos!
19:30.42Tod-WorkThis is the EVE display, yes?
19:30.51Tod-WorkBecause I watched the demo video for that on adafruit
19:30.53leviYes.
19:31.35Tod-WorkI know it HAS a touchscreen, mind you.  I'd have to interface with it and write the driver though, that's my point
19:31.49leviNo, I think you misunderstand how it works.
19:31.58Tod-WorkI've only ever used SPI for output
19:32.22leviIt captures touchscreen events internally and associates them with widgets; you poll SPI for widget events.
19:32.32Tod-WorkO.O
19:32.39leviOr something like that, anyway.
19:32.48Tod-WorkThat's.... that's weird.  Kinda cool, but definitely different
19:32.52Tod-Workhardware level widgets
19:32.59Tod-Work(well, as far as the abstraction is concerned)
19:33.01leviIt's essentially a 2-d GPU.
19:33.58leviThis is the programmer's manual: http://www.ftdichip.com/Support/Documents/ProgramGuides/FT800%20Programmers%20Guide.pdf
19:34.02Tod-WorkI won't have time or cash to play with one of these for like, a year-ish, so this is all speculative at the moment, but it's in my bookmark list
19:34.48Tod-Work83 registers...
19:35.06leviBTW, I'm pretty sure I saw a video of some guy controlling one of these with a microcontroller running Forth code.
19:36.43Tod-WorkThat might excite John, but this project is less about the coding for me and more about the outcome.  There's some other electronics projects I'm more interested in doing with Forth.  An intercom system with additional audio/video capabilities I'm more interested in just getting up and running in a way I can integrate with the rest of my home automation
19:37.37Tod-WorkIf I get Phone Janitor to the point we're all doing only it full time, I'll probably start diving into these.
19:37.56leviDepending on how you were going to program the GUI before, this might be a much faster way to do it.
19:38.14Tod-WorkIt is not.
19:38.52Tod-WorkI was just going to do a framebuffer or use a desktop toolkit.  Probably do it in Python.
19:39.02Tod-Workthe Pi has plenty of libs for both
19:39.16leviWell, I guess if you're sticking a Pi in it. :P
19:39.55Tod-WorkIt's $35 plus a DC-DC module thing.  That's pretty good as a price point given the display and speaker/mic will be more
19:40.00Tod-Workwell, the speaker MIGHT be
19:40.10leviThe Pi is a great ultra-cheap PC, but it's a terrible microcontroller.
19:40.12Tod-Work3D print an encasing
19:40.19Tod-Workoh, yeah
19:40.28bregaladlevi: totally
19:40.37Tod-Workthis intercom isn't going to be a salable product
19:40.58Tod-WorkI just want a mini computer I can do intercom stuff with.  Fast and easy.
19:41.00bregaladit doesn't come with any DC switching at all
19:41.11bregaladnot to mention AC switching
19:41.26Tod-WorkThen I can hook in the cameras around the house, play audio to the room from the server, etc
19:41.34Tod-Workflexible as hell
19:41.48leviThe Pi has got graphics down, but you've got to add a bunch of peripherals to get anything else done. And most peripherals go over USB, and it has a terrible USB implementation.
19:42.09Tod-WorkThe display, the mic, and the speaker are not USB
19:42.16Tod-Workwell, display might be depending on the display
19:42.21Tod-Workbut the one I'm looking at is just HDMI
19:43.05Tod-WorkThis FT800 is pretty intense
19:44.41leviI just see a lot of projects using Pi that start accumulating extra junk; USB hub, USB soundcard, etc. and start getting crappy performance out of them due to the Pi's USB limitations.
19:44.50bregaladlevi: what don't you like about the pi's usb implementation?
19:45.17Tod-WorkYeah, I'd be keeping it very simple.  This is pretty much exactly the kind of thing the Pi is suited to
19:45.32Tod-Workthe most complicated addition is literally the PSU
19:45.50bregaladall of the projects I'm thinking about are for diy automation, like running DC switches mostly
19:46.05bregaladwhich the pi doesn't even have
19:46.06Tod-WorkI'd use an actual MCU for that
19:46.11Tod-Worknot onboard, no
19:46.24Tod-Workyou tend to use relays or somesuch through GPIO or whatnot
19:46.34Tod-Workbut an MCU is a better target for that
19:49.11leviThe USB controller on the Pi is like *half* of what a normal USB controller supplies. Software is expected to take care of a bunch of stuff, and it must do so with real-time response.
19:51.59leviAnyway, it's important to note that the onboard Ethernet is actually connected via USB, so heavy Ethernet streaming could get glitchy if there's a bunch of other load on the system.
19:53.10leviAnyway, if the Pi works for you, it's probably the cheapest way to go.
19:53.33Tod-WorkI have a friend who uses Pis around his house for every TV as just a home theater frontend off an NFS share of video files
19:54.01Tod-Workworks great if you have it doing JUST that.  When people start trying to make it a media center hub that controls every other thing under the sun, less so
19:54.20leviThey're okay at that because most of the die space is used by the video subsytems.
19:54.23Tod-WorkA BBB might be better for my purposes at the end of the day, but we'll see when we get there.
19:54.54Tod-WorkThe landscape will probably be changed a bit, too
19:55.22Tod-WorkCHIP is going to be available soon, there will be better tools for a lot of MCUs and kits, and who knows, maybe Arduino will have killed itself finally
19:55.30Tod-Work(I can dream)
19:55.36leviThere are actually a bunch of little ARM-based boards that might work.
19:56.19leviArduino Zero is apparently an Atmel Sam D20, which is a Cortex M0.
19:59.04leviVery few of these are as cheap as Pi, though.
20:00.00Tod-Workyeah
20:00.04Tod-Worksad, that
20:00.23bregaladbut you end up paying more money for pi addons to get the functionality you need anyway
20:00.46bregalads/more/more money than just for the pi/
20:01.02spangbornHaha.
20:01.05spangbornYo dawg.
20:01.48leviThese things have traditionally been software development platforms to validate that your application will work with the MCU while you get a custom board built up.
20:02.21*** join/#utah youtah (~youtah@166.70.79.226)
20:02.23leviThe pricing changes dramatically when you're just putting the components you need on a custom board for mass production.
20:02.30youtahMooooooo
20:02.35leviBut when you're building 5 of something....
20:04.38bregalad~rails youtah
20:04.38infobotyoutah refactors to 0.561392778550896 picolines of code with the latest rails magic
20:05.01klyssomeone told me that you can get ten or so dies printed for under ten bucks each at pcbexpress, haven't checked it out yet.
20:05.10bregalads/rails/$web_framework/g
20:05.14leviEveryone watch out, I've turned on my scope! :P
20:05.20bregaladinfobot: I did, actually, thanks
20:05.20infobotYou did, actually, thanks?
20:07.45Tod-Workinfobot, I did a bad bad thing.
20:07.46infobotYou did a bad bad thing.?
20:07.50Tod-WorkYes
20:14.51harleypigTod-Work: Well?! Don't leave us hangin'!
20:17.39spangbornAnybody want a drink? https://twitter.com/nickmartin/status/630833596140507136
20:18.01youtahSo. Sleepy.
20:18.34youtahShouldn't have drank that tang
20:18.38youtahfrom the river
20:19.09bregaladthe tang river?
20:19.17penofull of minerals and vitamins
20:19.26youtah^ yes
20:19.27bregaladaren't the moon's rivers made of tang?
20:20.07youtah"The U.S. Geological Survey reported the size of the spill to be more than 3 million gallons, compared with the initial EPA estimate of 1 million gallons."
20:20.09youtahvia CNN
20:20.52RyanEfull of *metals*, not *minerals*.
20:20.55RyanEdifference there...
20:21.17youtahLead is a vitamin
20:21.24youtahand so is mercury
20:21.25RyanEThe EPA has a bit of incentive to say it was only 1M gallons, since they *caused* the spill...
20:21.58RyanEactually, neither of them are vitamins.
20:22.05RyanEvitamin = vital amino acid.
20:22.30youtahFred Flintsone isn't a vital amino acid either...
20:22.43youtahCheck... Mate...
20:23.37youtahThe believe it to be primarily iron, zinc and copper actually
20:23.45youtahno lead or mercury
20:23.49youtahwell, that they know of
20:30.44RyanEagain, in EPAs best interest to downplay it... :)
20:33.47youtahOh of course
20:33.56youtahI heard that the EPA is going to fine them.
20:33.58youtah:|
20:34.08spangbornEPA fine-ception
20:35.52*** join/#utah sjansen (~sjansen@50-204-233-22-static.hfc.comcastbusiness.net)
20:36.08youtahand then they'll all get bonuses because they were able to find a spill and successfully fine the offender?
20:36.24sjansenHappy S3-pocalypse everyone!
20:36.39youtahS3-wut?
20:37.23sjansenAWS S3 is... having a bad day.
20:37.29youtahreading about it now
20:37.39youtahlooks like a DNS issue
20:38.01sjansenIt's more than that.
20:38.11youtahis just digging in
20:38.25youtahYou have a good link?
20:38.33youtahsave me searching time?
20:39.08sjansenNot really. But we're not a Java shop, so I'm 100% certain it isn't being caused by JVM DNS TTL issues.
20:39.25sjansenhttps://twitter.com/search?f=tweets&vertical=default&q=aws%20s3
20:39.45*** join/#utah tristanbob_l (~tristanbo@137.190.208.10)
20:40.17sjansenSadly, that's the best source. AWS support doesn't know much yet.
20:41.29sjansenIt's entirely possible that Amazon's own services are getting bitten by JVM DNS TTL, but they'll never admit to such.
20:43.40programmerqyuck.
20:43.45programmerq(re: orange river)
21:01.25*** join/#utah emcnabb (~emcnabb@71.65.240.17)
21:01.26*** mode/#utah [+v emcnabb] by ChanServ
21:07.55leviA link I saw on that page: http://talkingpointsmemo.com/livewire/todd-courser-audio-blackmail
21:08.12leviThat's just all kinds of screwed up, of an entirely different kind than the orange river.
21:09.03Tod-Workhttps://youtu.be/WQSRdQls10w
21:11.58*** join/#utah metacortex (~metacorte@193.105.134.246)
21:13.09Tod-Worklevi, makes sense
21:13.16Tod-Worknot RATIONAL sense, but sense just the same
21:13.48Tod-Work"I'm desparate, here's this crazy scheme my desperate mind has come up with because I'm both crazy and not all that bright"
21:14.05RyanEI think Thanksgiving Point's museum has that color exhibit up now.
21:16.34bregalad~napoleon sjansen
21:16.34infobotACTION makes sjansen buy a lotto ticket for Napoleon
21:22.54RyanEHad to check whether today was April 1st: http://consumerist.com/2015/08/10/google-reorganizing-itself-under-new-company-called-alphabet/
21:24.53Tod-Workright?
21:26.18*** join/#utah yukaia (~yukaia@74-81-251-245.static.logl001.digis.net)
21:27.16leviWeird.
21:28.52leviAlthough I guess the weird thing is not that they're spinning out companies unrelated to their core business, but that they're putting their core business along with the spinoffs under a new umbrella.
21:29.24Tod-Workon a .xyz domain
21:29.35Tod-Workguess it's better than a .io
21:37.33*** join/#utah metacort_ (~metacorte@193.105.134.246)
21:54.49*** join/#utah dkelson (~dkelson@40.141.117.2)
22:13.53*** join/#utah hydroplane (~hydroplan@c-24-10-151-20.hsd1.ut.comcast.net)
22:28.02*** join/#utah hydroplane (~hydroplan@2601:681:4502:56e0:2405:c64e:d41d:a029)
22:31.54leviYay, DMA is working again.  Now to see if I can get scatter-gather mode working...
23:36.16*** join/#utah yukaia (~yukaia@74-81-251-245.static.logl001.digis.net)
23:42.33*** join/#utah Migs (~migs@205.122.50.108)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.