2015-08-23.log

--- Log opened Sun Aug 23 00:00:49 2015
-!- sandeep_ [~sandeep@111.235.64.135] has joined ##hplusroadmap00:03
-!- Beatzebub_ [~beatzebub@d162-156-106-225.bchsia.telus.net] has joined ##hplusroadmap00:10
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: Namaste]00:12
-!- Beatzebub [~beatzebub@d162-156-106-225.bchsia.telus.net] has quit [Ping timeout: 252 seconds]00:12
-!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:c89a:5fec:415b:f7e1] has quit [Quit: Leaving]00:19
-!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:c89a:5fec:415b:f7e1] has joined ##hplusroadmap00:20
-!- JayDugger [~jwdugger@108.19.186.58] has joined ##hplusroadmap00:23
-!- sandeep_ [~sandeep@111.235.64.135] has quit [Ping timeout: 240 seconds]00:26
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-olsiwnemjklpjdvz] has joined ##hplusroadmap00:32
nmz787https://hackaday.io/project/1620-the-human-connection-1st-impression00:33
-!- Houshalter [~Houshalte@oh-71-50-58-200.dhcp.embarqhsd.net] has quit [Quit: Quit]00:49
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Quit: Leaving]00:51
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap00:56
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-ipsucansqcwbhoei] has quit [Quit: Connection closed for inactivity]01:30
-!- cuba [~qba@cracksucht.de] has joined ##hplusroadmap01:33
-!- Beatzebub_ [~beatzebub@d162-156-106-225.bchsia.telus.net] has quit [Quit: Beatzebub_]01:37
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-ybnrskdqnmpyoyyh] has joined ##hplusroadmap02:25
-!- EnLilaSko [~Nattzor@unaffiliated/enlilasko] has joined ##hplusroadmap02:33
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-olsiwnemjklpjdvz] has quit [Quit: Connection closed for inactivity]02:36
-!- FourFire [~FourFire@77.19.32.221.tmi.telenormobil.no] has joined ##hplusroadmap03:44
-!- JayDugger [~jwdugger@108.19.186.58] has quit [Quit: Leaving.]04:09
-!- xtalmath [~xtalmath@ip-83-134-83-209.dsl.scarlet.be] has joined ##hplusroadmap05:03
-!- sandeep_ [~sandeep@111.235.64.135] has joined ##hplusroadmap05:04
-!- JayDugger [~jwdugger@108.19.186.58] has joined ##hplusroadmap05:21
FourFireThis looks potentially interesting for BCI05:53
FourFirebut I'd be concerned about having a lot of LN² close to my head05:53
-!- FourFire [~FourFire@77.19.32.221.tmi.telenormobil.no] has quit [Quit: Leaving]06:18
-!- c0rw|zZz is now known as c0rw1n06:24
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds]06:59
kanzurenot like i needed that body part anyway07:10
-!- CaptHindsight [~2020@unaffiliated/capthindsight] has joined ##hplusroadmap07:42
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap07:45
-!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:c89a:5fec:415b:f7e1] has quit [Read error: Connection reset by peer]07:55
kanzurebeep bloop08:20
-!- math3 [uid54897@gateway/web/irccloud.com/x-srhjpmnrdrvkcjhi] has joined ##hplusroadmap08:52
xtalmaththat magnet guy was ...08:56
xtalmathmagneto?08:56
kanzurenah there are others playing around with finger magnets09:13
kanzure"LabCures (https://labcures.com) is trying a back-the-team rather than back-the-project approach to crowdfunding" ugh more crowdfunding09:14
xtalmathmust be a pain in the ass to clean your finger after milling ferromagnetic materials...09:14
kanzureand lifepsan.io.... bleh. reason@fightaging.org really needs to reconsider his crowdfunding approach.09:14
xtalmathalso MRI machine is no go zone for him if he has a serious accident, the doctors might not know about the magnet...09:16
kanzurecan't they just leave his hand out of the machine09:17
kanzureanyway look up "grindhouse wetware" for more finger magnet stuff09:18
xtalmaththats gonna be a lot of fun, when it yanks his hand and arm until closest point, then rips the magnet out of his skin, and the magnet flies around the room until it hits the superconducting magnet housing, and shatters in all directions09:18
kanzureperhaps he has very long arms09:19
xtalmathMRIs and high tesla MS are places with lots of funny stories that people experience09:19
xtalmathat my university, somebody once brought a laptop in the room, but a little too close to the machine, and the magnet attracted the laptop to underneath the device09:20
xtalmaththe person couldnt seperate the laptop on his own, one person had to lay under the magnet and grab the laptop with both arms, while the others pulled his legs to pull him from under the machine09:21
xtalmathlonger arms just means longer period of acceleration before ripping out of his skin09:22
kanzuremaybe 20 foot arms09:23
xtalmathhis hand may collide with the machine before his arms are fully extended, ouch09:23
xtalmathmight break a bone09:23
xtalmathso you wake up after coma from car accident, and they tell you "you had 2 accidents, but calm down"09:24
xtalmath"oh, you lost your partner too"09:24
xtalmath"but the good news is, life extension made you survive it, your 200yrs now btw"09:26
kanzurethe life extension is because of the finger magnet?09:26
xtalmathno seperate development09:26
kanzurethe partner death is because finger magnet?09:27
xtalmathno car acciden09:27
kanzurethe car accident is because finger magnet?09:28
xtalmathyes09:28
gradstudentbotI hope they kick me out.09:28
xtalmathit pulled the keys out of the ignition09:28
kanzurekeys are not a good idea anyway09:28
xtalmathwhich he normally got used to, but that time it was inconvenient09:28
xtalmath"good news is, we still have your car keys" "you can probably pick up the remains of your car, somewhere at lost and found, early 21st century section"09:29
xtalmath"oh, theres still the matter of your partner's family suing you for causing the accident"09:31
kanzurethere's no way that the lawsuit wouldn't have been settled earlier09:32
xtalmathwhat about the right to a defense?09:32
kanzurehuh? you can still have a defense.09:33
xtalmathso the lawsuit was not final09:33
kanzurealmost all lawsuits aren't09:33
xtalmathindeed09:34
xtalmath"you still have the right to a defense, but formal verification tools have after heuristic search, found that there exists no defense"09:40
kanzurehuh? no they just assign you a public defender or whatever, go ask the people stuck in comas how they deal with lawsuits09:44
xtalmath"also, your internationalized human rights gave you your daily proper share of average rent of all world resources, which amounts to millions now"09:44
xtalmath"in absence of your consciousness doctors decided it was in your best interest to spend it all on the best treatment during your stay here"09:44
kanzureare you just completely unaware of how the current system works, or are you just wasting my time for fun?09:45
xtalmath"we hope you enjoyed the holographic television you couldn't possibly perceive, well we enjoyed it"09:45
xtalmathnot wasting your time, just fun09:45
kanzureit's a waste of my time to explain to you things that you might already know09:45
kanzureif you don't know how medical decisions and estate decisions are handled when someone is in a coma, that's one thing. knowing and still getting it wrong is another.09:45
xtalmath150yrs passed in the hypothetical09:45
kanzure"oh noes stuff happens during 100s of years" so what?09:46
gradstudentbotThe paper was rejected.09:47
kanzureon an only slightly related note,09:48
kanzurextalmath: http://extropians.weidai.com/extropians.3Q97/4356.html09:48
xtalmaththat text reminds me of ...09:50
xtalmathroko's basilisk09:50
-!- Houshalter [~Houshalte@oh-71-50-58-200.dhcp.embarqhsd.net] has joined ##hplusroadmap10:04
kanzureaww i missed the start of ludum dare 33 http://ludumdare.com/compo/10:11
kanzure"guide an ambitious young fire hydrant to fulfill her dream of destroying the sun" http://ludumdare.com/compo/ludum-dare-32/?action=preview&uid=5125910:19
-!- Beatzebub [~beatzebub@d162-156-106-225.bchsia.telus.net] has joined ##hplusroadmap10:31
-!- Houshalter [~Houshalte@oh-71-50-58-200.dhcp.embarqhsd.net] has quit [Quit: Quit]10:32
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap10:46
-!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has quit [Quit: Leaving...]11:12
-!- Beatzebub_ [~beatzebub@d162-156-106-225.bchsia.telus.net] has joined ##hplusroadmap11:31
-!- Beatzebub [~beatzebub@d162-156-106-225.bchsia.telus.net] has quit [Ping timeout: 250 seconds]11:32
-!- Beatzebub_ is now known as Beatzebub11:33
-!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has joined ##hplusroadmap11:39
-!- Beatzebub_ [~beatzebub@d162-156-106-225.bchsia.telus.net] has joined ##hplusroadmap11:41
-!- Beatzebub [~beatzebub@d162-156-106-225.bchsia.telus.net] has quit [Ping timeout: 250 seconds]11:43
-!- Beatzebub_ is now known as Beatzebub11:45
-!- Beatzebub_ [~beatzebub@d162-156-106-225.bchsia.telus.net] has joined ##hplusroadmap11:49
-!- Beatzebub [~beatzebub@d162-156-106-225.bchsia.telus.net] has quit [Ping timeout: 240 seconds]11:49
-!- Beatzebub_ [~beatzebub@d162-156-106-225.bchsia.telus.net] has quit [Max SendQ exceeded]11:52
-!- android1 [~android1@105.156.92.158] has joined ##hplusroadmap11:53
-!- EnLilaSko- [~Nattzor@host-85-30-145-65.sydskane.nu] has joined ##hplusroadmap11:56
-!- EnLilaSko [~Nattzor@unaffiliated/enlilasko] has quit [Disconnected by services]11:57
-!- EnLilaSko- is now known as EnLilaSko11:57
-!- EnLilaSko [~Nattzor@host-85-30-145-65.sydskane.nu] has quit [Changing host]11:57
-!- EnLilaSko [~Nattzor@unaffiliated/enlilasko] has joined ##hplusroadmap11:57
-!- Beatzebub [~beatzebub@d162-156-106-225.bchsia.telus.net] has joined ##hplusroadmap12:10
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: reboot]12:12
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-ybnrskdqnmpyoyyh] has quit [Quit: Connection closed for inactivity]12:20
-!- phillyj [~chatzilla@c-71-225-87-180.hsd1.pa.comcast.net] has joined ##hplusroadmap12:33
-!- math3 [uid54897@gateway/web/irccloud.com/x-srhjpmnrdrvkcjhi] has quit [Quit: Connection closed for inactivity]12:39
-!- phillyj [~chatzilla@c-71-225-87-180.hsd1.pa.comcast.net] has quit [Quit: ChatZilla 0.9.92 [Firefox 40.0/20150819100751]]12:47
-!- FourFire [~FourFire@77.19.32.221.tmi.telenormobil.no] has joined ##hplusroadmap12:49
-!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:5545:7448:f5bc:97f0] has joined ##hplusroadmap12:50
-!- Guest48938 [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has quit [Quit: No Ping reply in 180 seconds.]12:53
-!- maaku [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has joined ##hplusroadmap13:00
-!- maaku is now known as Guest1513613:00
-!- JayDugger1 [~jwdugger@108.19.186.58] has joined ##hplusroadmap13:03
-!- jackybgood [~jackybgoo@72-160-2-207.dyn.centurytel.net] has joined ##hplusroadmap13:04
-!- Guest15136 [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has quit [Client Quit]13:04
jackybgoodpaperbot doi:10.1038/nm.382013:05
-!- JayDugger [~jwdugger@108.19.186.58] has quit [Ping timeout: 244 seconds]13:05
-!- delinquentme [~delinquen@74.61.157.78] has joined ##hplusroadmap13:07
-!- maaku_ [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has joined ##hplusroadmap13:09
jackybgoodpaperbot: http://www.nature.com/nm/journal/v21/n4/full/nm.3820.html13:13
-!- yashgaroth [~ffffff@2602:306:35fa:d500:8049:d663:77ad:31c6] has joined ##hplusroadmap13:20
-!- rancyd [strtrk@shell.xshellz.com] has joined ##hplusroadmap13:33
kanzurehi rancyd13:37
rancydhi kanzure13:37
chris_99kanzure, did you know whether the nozzles on a printer are all individually addressable btw (i'm assuming they would be)13:38
-!- JayDugger [~jwdugger@108.19.186.58] has joined ##hplusroadmap13:40
-!- JayDugger1 [~jwdugger@108.19.186.58] has quit [Ping timeout: 250 seconds]13:42
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-jddiosnkjdorwlch] has joined ##hplusroadmap13:42
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap13:55
-!- Houshalter [~Houshalte@oh-71-50-58-200.dhcp.embarqhsd.net] has joined ##hplusroadmap14:00
-!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:5545:7448:f5bc:97f0] has quit [Read error: Connection reset by peer]14:02
-!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:a8c5:3f7e:58bb:d5ad] has joined ##hplusroadmap14:04
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap14:14
kanzurechris_99: yes14:27
kanzurewe went over this with you yesterday14:28
kanzureor the day before. not sure.14:28
chris_99nope, noone answered that14:28
-!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has quit [Ping timeout: 255 seconds]14:33
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: Namaste]14:34
kanzurehttp://gnusha.org/logs/2015-08-21.log14:34
kanzure15:39 < chris_99> is there a piezo per nozzle14:34
kanzure15:39 < chris_99> or only per colour14:34
kanzure15:40 < CaptHindsight> chris_99: with piezo yes, it's one piezo per nozzle14:34
kanzure15:40 < kanzure> there is a piezo per nozzle14:34
kanzure15:40 < kanzure> a better question would be "does the control circuitry allow individual piezo control"14:34
chris_99yeah to which noone answered14:34
kanzurehuh?14:40
chris_99answered the question on whether they do allow individual piezo control14:42
-!- JayDugger1 [~jwdugger@108.19.186.58] has joined ##hplusroadmap14:45
CaptHindsightyes, each piezo is individually controlled14:46
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap14:46
CaptHindsightotherwise you'd just be firing (or not firing) all the nozzles at the same time14:47
-!- JayDugger [~jwdugger@108.19.186.58] has quit [Ping timeout: 252 seconds]14:47
CaptHindsightchris_99: ^^14:47
chris_99gotcha, i had a look at an image of one of the bottom of the heads, it looked like each colour has a line of nozzles, so it's printing vertically over a fairly large surface at a time right?14:48
CaptHindsightit all depends on the printhead14:48
CaptHindsightsome have only one fluid channel14:48
CaptHindsightsome have multiple14:48
gradstudentbotMay I ask what formula did you use?14:50
chris_99i found you can get MEMS ultrasonic mics really cheap, so i'm gonna get one to see if you can pick up an interesting ultrasonic audio from the heads14:52
CaptHindsightso in the case of the Epson heads anywhere from 0-180 nozzles per channel can fire at a time...14:54
CaptHindsightand Epson has up to 8 channels per head (depending on model)14:55
CaptHindsightthe piezo are also put into modes other than just firing ink, they are also used to churn up the ink14:55
chris_99does 8channels mean 8 colours?14:56
CaptHindsightit can14:57
nmz787chris_99: I was just looking at this last night, uses MEMS mics and a MCU I might want to use (LPC4370) for an xray spectrometer https://hackaday.io/project/1620-the-human-connection-1st-impression14:57
nmz787i can't tell how much code is here though, like if it is complete (seems not) https://github.com/ehughes/hc-114:58
chris_99neat, i was looking at the knowles ones14:58
chris_99let me check if that's the analog or digital one14:58
nmz787I am guessing analog, because the person wanted to leverage the 80MSPS ADC14:59
nmz787I thought14:59
gradstudentbotMy experiment was working a second ago, but now it doesn't even work.14:59
nmz787(maybe I was just assuming that?)14:59
chris_99yeah thats analog14:59
chris_99i'm looking at the digital PDM one15:00
nmz787oh, I actually didn't pay much attention to the mic aspect15:00
nmz787since I want to throw a photodiode in front with (probably a ton of) amplification15:00
chris_99i wasnt sure whether to get an analog one or pdm one, as the analog one possibly has a slightly high freq. response15:01
archelsnmz787: that's cool stuff, been waiting for something like that to come along15:01
chris_99what are they using them for anyway, as it mentions x-rays?15:01
nmz787what mentions xrays?15:02
nmz787(I mentioned xrays here)15:02
chris_99oh sorry, you did heh15:02
nmz787yeah I just want really fast sampling so I can average to remove some noise15:03
nmz787by oversampling15:03
chris_99ahh15:03
gradstudentbotYeah, I don't know.15:03
nmz787since I want to get as much resolution as possible from the different xray photon energy levels15:03
chris_99how are you detecting x-rays?15:04
nmz787supposedly the PIN photodiodes commonly used for fiber optic communication in NIR can get something like 120eV of resolution in the xray area15:04
chris_99yeah i've heard of people using PIN diodes for radiation, i think you can pick up gamma too with them?15:05
nmz787which isn't as good as the single digit resolution that wavelength dispersion spectrometers can achieve, or the few 10s of eV that the SDD or SiLi detectors can get... but it seems pretty damn good15:05
nmz787and PIN diodes are like $1 each15:05
nmz787yeah15:05
nmz787I actually probably don't need a spectrometer, but it would be a nice upgrade to the xray detector project I have in the works now with just a buzzer15:06
nmz787and it is stressing so many areas of expertise... so it seems like a good goal to march toward.15:07
chris_99i'd like a dental x-ray sensor, alas they're not exactly cheap15:07
-!- AmbulatoryCortex [~Ambulator@173-31-155-69.client.mchsi.com] has joined ##hplusroadmap15:15
-!- c0rw1n [~c0rw1n@10.216-241-81.adsl-dyn.isp.belgacom.be] has quit [Ping timeout: 272 seconds]15:39
nmz787http://arrc.ou.edu/~rockee/RIO/serialio-book.pdf15:42
nmz787'As one expert put it, “Radiated emission problems are really just signal integrity problems.”'15:42
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-tqexvzmsobeckrbp] has joined ##hplusroadmap15:42
nmz787"Before we think that gigabit serial I/O sounds too good to be true, let's look at the downsides. In our designs, we must first we must pay close attention to signal integrity issues. For example, one vendor reported a 90% failure rate on their first attempt with high-speed, multi-gigabit serial designs for a particular application. To improve the odds, we might need to perform analog simulations and use new, more complex bypassing schemes. ...15:43
nmz787... In fact, we may even need to simulate and model the bypassing scheme."15:43
nmz787"We can also expect to pay more for impedance-controlled PC (printed circuit) boards, high-speed connectors, and cables. We will have to deal with complications and smaller time bases in digital simulations. And when taking advantage of a predefined protocol, we must plan time for integration and extra gates or Central Processing Unit (CPU) cycles for protocol overhead."15:43
-!- xtalmath [~xtalmath@ip-83-134-83-209.dsl.scarlet.be] has quit [Ping timeout: 246 seconds]15:45
nmz787"""SERDES was initially used to talk box-to-box. But it exploded into the marketplace because of how nicely it handles chip-to-chip communication on the same circuit board. Chip-to-chip communication had previously been almost exclusively a parallel domain. The amount of logic needed to serialize and deserialize far outweighed any savings that come from pin count reduction."""15:46
nmz787"""But with deep sub-micron geometry, an incredible amount of logic can be achieved in a very small amount of silicon. SERDES can be included on parts for a very low silicon cost. Add to that the everincreasing need for I/O bandwidth, and SERDES quickly becomes the logical choice for moving any significant amount of data chip-to-chip"""15:46
nmz787.wik infiniband15:47
yoleaux"InfiniBand (abbreviated IB), a computer-networking communications standard used in high-performance computing, features very high throughput and very low latency. It is used for data interconnect both among and within computers." — https://en.wikipedia.org/wiki/Infiniband15:47
chris_9910Gbps infiniband cards are dirt cheap on fleabay15:49
-!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has joined ##hplusroadmap15:49
nmz787"""An unsettling aspect of the Gigabit SERDES is that they appear to be almost magical. They work with 3, 5, and even 10+ gigabits. How is that kind of speed possible? There are several techniques that provide this speed."""15:50
nmz787"""One of the biggest advances in high-speed SERDES involves the PLLs used in clock and data recovery. A normal PLL requires a clock running at the data speed, but there are several techniques that can be used to avoid this requirement, including fractional rate phase detectors, multi-phase PLLs, parallel sampling, and over-sampling data recovery."""15:52
-!- c0rw1n [~c0rw1n@186.79-67-87.adsl-dyn.isp.belgacom.be] has joined ##hplusroadmap15:53
nmz787"""The 8b/10b encoding scheme was developed by IBM and has been widely adapted. It is the encoding scheme used in Infiniband, Gigabit Ethernet, FiberChannel, and the XAUI interface to 10 Gigabit Ethernet."""15:54
nmz787"""first let’s examine how 8b/10b ensures a good DC balance"""15:54
nmz787this is like RNA tiplets and amino acids or something...15:55
nmz787triplets*15:55
nmz787"""8b/10b uses two different symbols assigned to each data value. In most cases, one of the symbols has six zeros and four ones, and the other has four zeros and six ones. The total number of ones and zeros is monitored and the next symbol is chosen based on what is needed to bring the DC balance back in line. The two symbols are normally referred to as + and - symbols."""15:56
nmz787"""One additional benefit of the running disparity is that the receiver can monitor the running disparity and detect that an error has occurred in the incoming stream because the running disparity rules have been violated."""15:57
nmz787"""Scrambling is a way of reordering or encoding the data so that it appears to be random, but it can still be unscrambled. We want randomizers that break up long runs of zeros and ones. Obviously, we want the descrambler to unscramble the bits without requiring any special alignment information. This characteristic is called a self-synchronizing code."""15:58
-!- xtalmath [~xtalmath@ip-83-134-140-175.dsl.scarlet.be] has joined ##hplusroadmap15:58
nmz787wow15:58
-!- FourFire [~FourFire@77.19.32.221.tmi.telenormobil.no] has left ##hplusroadmap ["Leaving"]15:59
xtalmathchris_99: did the microphone move during the recording?15:59
chris_99nope15:59
xtalmathis the recording taken straight from the video?16:00
chris_99yup16:00
chris_99its from a shotgun mic ontop of the camera16:00
xtalmathgreat16:01
xtalmathI'll tick your shoulders when I have something usefull16:02
chris_99i'm planning on getting a MEMS mic to get up to 80kHz16:02
chris_99or so16:02
xtalmathchris_99: I think piezo can easily get this?16:02
xtalmathjust need fast enough ADC16:03
chris_99the MEMS i'm looking at does PDM so i don't need an ADC16:03
xtalmath.wiki PDM16:03
chris_99https://en.wikipedia.org/wiki/Pulse-density_modulation16:03
chris_99i was also thinking of how to rip off the head from the printer, and have it print the same stuff without the mechanical belt noises etc16:05
xtalmathyes16:05
xtalmathat what frequency is the PDM?16:05
gradstudentbotThat's definitely a Nature paper.16:05
chris_99around 3MHz you clock the mic at16:05
chris_99for ultrasound16:05
xtalmathwell the belt noise should not be a problem as it will have a different spectrum16:06
xtalmathoh, I see what you mean16:07
xtalmathyou might need the belt noise durations though16:07
xtalmathif the printer knows that for the rest of the line no ink will follow it can "newline"16:07
xtalmath"carriage return"16:08
xtalmathchris_99: how do you record the PDM in the computer? you have a special soundcard/port for PDM?16:09
chris_99i'll probably use an MCU devboard, that has usb on it16:10
chris_99and it'll grab the PDM16:10
chris_99decode to PCM16:10
chris_99and shovel through USB16:10
xtalmathyeah16:10
xtalmathI was thinking in terms of what I have laying around, .. got an fpga with a 1 Msps adc16:11
xtalmathor was it 1016:11
xtalmathbut ... then you could connect a piezo to that :) well, your gear will probably give better result...16:11
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Read error: Connection reset by peer]16:14
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap16:17
-!- android1 [~android1@105.156.92.158] has quit [Ping timeout: 260 seconds]16:21
-!- Madplatypus [uid19957@gateway/web/irccloud.com/x-hznajwpsmpviyahc] has joined ##hplusroadmap16:40
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving]16:42
-!- AmbulatoryC0rtex [~Ambulator@173-31-155-69.client.mchsi.com] has joined ##hplusroadmap16:46
xtalmathso PDM seems basically the same as pulse width modulation?16:47
xtalmathah no16:47
xtalmathits delta sigma modulation16:49
-!- AmbulatoryCortex [~Ambulator@173-31-155-69.client.mchsi.com] has quit [Ping timeout: 240 seconds]16:49
kanzurehttp://www.sobify.com/the-lyubov-orlova-a-russian-cruise-ship-drifting-through-international-waters/16:54
-!- sandeep_ [~sandeep@111.235.64.135] has quit [Ping timeout: 272 seconds]16:57
kanzure"Meanwhile, the academic research industry is $1 trillion asset and runs on $2.5 billion/ week. Calico couldn't make it to Friday in that league. "17:00
kanzurenot sure why someone would be comparing calico to the size of academia17:00
kanzuretoxoplasma freezing http://www.protocol-online.org/biology-forums-2/posts/23756.html17:02
kanzurehttp://www.researchgate.net/post/how_can_i_isolate_and_culture_Toxoplasma_gondii_for_experimental_infection_in_details17:05
kanzureoh god i forgot how terrible biologists are at writing17:05
delinquentmeanyone have suggestions for good reading on the evolutionary process by which novel insertions are bred out ... and ideally how to prevent that17:11
delinquentmeOTHER than the JCVI play of simplifying the fuck out of the organism17:12
delinquentmealso it gave me pause to think about if theres some way to robotically handle / sort / quantify cell excretion17:12
-!- xtalmath [~xtalmath@ip-83-134-140-175.dsl.scarlet.be] has quit [Ping timeout: 260 seconds]17:13
gradstudentbotHoly crap, I need to add that to my slide collection.17:14
yashgarothdelinquentme: the normal way to prevent it is through constant antibiotic selection17:20
yashgarothand to do your second question, sure put your stable cell line clone plates into a tecan that can do ELISA17:21
yashgarothassuming this is related to you making a monoclonal in rpe1 cells17:21
-!- drewbot [~cinch@ec2-174-129-68-147.compute-1.amazonaws.com] has quit [Remote host closed the connection]17:28
-!- drewbot [~cinch@ec2-54-197-214-5.compute-1.amazonaws.com] has joined ##hplusroadmap17:28
-!- xtalmath [~xtalmath@ip-83-134-140-175.dsl.scarlet.be] has joined ##hplusroadmap17:29
nmz787"The ferrite bead has a low impedance at low frequencies and a very high impedance at high frequencies (Figure 4-12)."17:42
xtalmathyes17:44
nmz787"""As with any switching circuit, bypassing is critical. But, at these speeds, we cannot just insert few capacitors and say the bypassing is complete. That approach used to work a few years ago, so why not now? It still can work if we can find some ideal capacitors (no inductance or resistance) and get them on the board using ideal routes and vias (no inductance or resistance), and the package is ideal, and so on. As switching frequency and ...17:47
nmz787... current needs have increased, the ESR and ESL that at one time could be ignored now have to be considered."""17:47
nmz787"""Another important aspect of bypassing is placement. As a general rule, the larger the cap value, the less critical the placement. The smallest values want to go as near a power and ground pin as possible. One way to do this that is often available when using MGTS inside FPGAs is to remove the trace and via of unused general IO to make room for the bypassing"""17:51
nmz787(on the backside of the PCB, from the pin on the BGA or whatever)17:52
nmz787"""For best results, we should run differential pairs tightly coupled and closely matched. Trace length matching is essential. In FR-4, a 100-mil (1 tenth of an inch) difference in trace length results in approximately 18 picoseconds of difference between the positive and negative signal. This is also enough skew to start causing problems. And, while a tenth of an inch may sound like a lot if we just use normal trace routing from one BGA to ...17:54
nmz787... another, it is easy to end up with 300 - 400 mils of difference. If our PCB tool has an auto-trace matching, we need to use it. In general, we will want 50 mils or less difference in differential trace lengths."""17:54
delinquentmeyashgaroth, does science have an idea what is applying those pressures to breed out the novel modifications?17:56
xtalmathnmz787: what are you planning to use MGTS for?17:56
delinquentmealso planar growth seems to make sense for antibody selection , idk about suspended growth ... unless of course im calculating out the expected cell yield along side the antibiotic load were placing in the reactor17:57
delinquentmewhich could be done17:57
xtalmathnmz787: sounds like you are reading some FPGA manual17:58
nmz787oh someone pointed out that the LPC4370 has a 204MHz clock (with 3 cores) but even with that and maybe SIMD instructions, at 80 MSPS (12 bits per sample) you effectively only have 204/80 clocks (2.55... or 2 i guess) to get anything done with the data17:58
nmz787so I'm thinking I'd have to pipe it over to an FPGA with some boxcar filtering and edge detection logic17:58
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: rebooty]17:59
nmz787and then averaging the data within a certain pulse that was found... which would greatly reduce the data transmission afterwards17:59
nmz787in fact at that point it would probably just be binning the values into a histogram17:59
nmz787(this is for xray spectrometer idea)17:59
yashgarothdelinquentme the pressure is from natural selection, any cell that happens to methylate & repress the insert will be able to divide faster than the other cells, that + time = lower yields18:00
xtalmathwhat calculation do you need on the data? what is the register width on the core?18:00
xtalmathyou have 3 cores, so depending on the task you can divide the workload by 3, and depending on SIMD instructions and register size you might get another 2-4 times speedup18:02
nmz787basically its just taking the ADC values, which are oversampled... finding where a pulse starts and ends, then taking all the samples within the pulse and averaging them, then throwing them into a dictionary/map with the key being the average-value and the value being a counter that increments18:02
nmz787then after some time, the dictionary keys and values would be flushed over USB or to an SD card or something18:03
nmz787so you'd get a spectrum of photodiode readings over some amount of time18:03
xtalmathbut yes FPGA sounds more reasonable, but with FPGA you could as well have used an ADC with parallel outputs, like a 120Mhz video ADC, so you get your samples at the rate of 120Mhz with 12bits in parallel in the FPGA18:03
nmz787yeah but then I have to think about the PGA which I think comes built in to the LPC437018:04
nmz787but yeah18:04
nmz787I realized that18:04
delinquentmeyashgaroth, is a continuous growth / selection bioreactor a thing?18:04
delinquentmelike FACS + bioreactor18:05
xtalmathsounds like you want to identify isotopes?18:05
nmz787someone in another room said the LDO on the LPC-Link-v2 adds noise to the ADC readings... but it would be awesome to be able to use that board as-is with a daughter card(s) for the ADC pre-amp and FPGA stuff18:05
xtalmathoh yeah, I forgot theres an FPGA in the FLP437018:06
xtalmathi should grab that devboard some day18:06
yashgarothdelinquentme, um most bioreactors are running continuous growth and selection, not sure how FACS fits into it, also you can't select with FACS for a secreted antibody unless you do some fancy multi-splicing expression where the cell produces both secreted and membrane-anchored antibody and you use the membrane-anchored as a proxy for secreted yield18:06
nmz787xtalmath: with a SEM or FIB you can do elemental mapping... but in my case I want to support an application of thickness detection/monitoring as a FIB milling operation proceeds (as the depth of the hole increases, more xrays can be seen out the backside)18:06
nmz787xtalmath: flp4370?18:07
xtalmathnmz, I see, so you are using the material's gamma ray stopping power to measure thickness while milling18:07
xtalmathLPC, typo18:07
nmz787oh, no there's no FPGA there18:07
nmz787hence the reading about high-speed links to other boards18:08
delinquentmeyashgaroth, yeah thats what I was thinking but could be complex.  closely co-located antibody + GFP ... could possibly allow GFP to act as proxy for antibody secretion18:08
nmz787also I work on board design automation at work, so I've been thinking of what it would take to integrate signal-integrity simulations/modelling into my software18:08
-!- jackybgood [~jackybgoo@72-160-2-207.dyn.centurytel.net] has quit [Remote host closed the connection]18:09
gradstudentbotMy code works. I have no idea why...18:09
yashgarothdelinquentme, depends on the cell line, oftentimes they're limited by their secretory procesing pathway, but usually we use the antibiotic resistance protein as a proxy instead of like gfp18:09
-!- AmbulatoryC0rtex [~Ambulator@173-31-155-69.client.mchsi.com] has quit [Read error: Connection reset by peer]18:09
xtalmathah no its not stopping power, its Beer Lambert law18:11
yashgarothyou will be hard pressed to find a method that is both reliable, and doesn't require months of tissue culture work of expanding single cells, endless ELISAs and pipetting media and oh god I'm getting flashbacks18:11
delinquentmeyashgaroth, Yessss. Ok so this is why  it would be a good idea to start w highly secreting cell line instead of just saying " all cells have capacity to secrete"18:11
delinquentme( which is what my gut feeling said , but I had nothing to back it up w  )18:11
nmz787xtalmath: umm, I think its just some bandgap thing18:11
delinquentmeyashgaroth, you're saying that antibiotic selection is just a really simple well-known path18:11
gradstudentbotHave you read this paper?18:12
xtalmathnmz787: electric apparently can do some signal integrity calculations, uses FastHenry etc if self-inductance18:12
yashgarothdelinquentme, uhh what am I backing you up on specifically; also antibiotic selection or some other very similar principle is used 99% of the time for this18:12
nmz787the photons warp the electron field in the thickness of the PIN silicon and convert all their energy before they make it to the end18:12
nmz787yashgaroth: i feel like there could be a self-selection system setup though18:12
xtalmathnmz787: no I meant what you use it for, during milling18:13
nmz787oh, to detect when you milling to a specified thickness, so you can change to a smaller milling diameter18:13
xtalmath"but in my case I want to support an application of thickness detection/monitoring as a FIB milling operation proceeds (as the depth of the hole increases, more xrays can be seen out the backside)"18:13
nmz787which would otherwise not be effective to mill through so much thickness without getting redeposition18:14
yashgarothsure there's ways you don't need traditional selection, like with the bicistronic antibody where you have beads with the target on them, run cells over it and pull off the high binders, or use a fluorescent target and do FACS I guess18:14
xtalmathso you are measuring the transmission signal strength of xrays18:14
nmz787.wik bremstralung radiation18:14
yoleauxnmz787: Sorry, that command (.wik) crashed.18:14
nmz787.gc bremstralung radiation18:14
yoleauxnmz787: Sorry, that command (.gc) crashed.18:14
nmz787.g bremstralung radiation18:14
yoleauxhttps://en.wikipedia.org/wiki/Bremsstrahlung18:14
xtalmathyes bremstralung18:14
nmz787yeah18:14
nmz787.wik Bremsstrahlung18:14
yoleaux"Bremsstrahlung (German pronunciation: [ˈbʁɛmsˌʃtʁaːlʊŋ] ( listen), from bremsen "to brake" and Strahlung "radiation", i.e. "braking radiation" or "deceleration radiation") is electromagnetic radiation produced by the deceleration of a charged particle when deflected by another charged particle, typically an electron by an atomic nucleus." — https://en.wikipedia.org/wiki/Bremsstrahlung18:14
nmz787so as you mill, that happens... and at some thickness, the material you're milling stops being able to absorb all the radiation, and it leaks out18:15
nmz787so when you detect a leak, you know you're at some thickness18:15
yashgaroththe problem is no one uses non-traditional methods for cell line production since the normal ones work fine, assuming you have a few months to do the endless plating; sure you could try a new method but the development will take even longer to test out, for probably negligible gain18:15
xtalmathnmz787: with milling to thickness, you mean thickness of the target along the direction of the ion beam?18:15
xtalmathnmz787: this was experimentally observed I assume18:16
nmz787yashgaroth: well i'm pretty sure those systems must already exist as well18:17
nmz787umm, like if you want to make a really small hole, you will waste time if you set the machine to a tiny beam diameter... because milling will take so long18:17
nmz787also, you get redeposition at some point18:18
nmz787so you need to widen your hole before continuing downwards18:18
yashgarothnmz787 I'm sure someone is running a startup based on one of those systems, and/or published a few papers, but for manufacturing biosimilars no one is really bothering18:18
nmz787so normally you use a big diameter which is higher power to start, then later switch to a finer diameter18:18
xtalmathnmz787: the bremstralung can be seen as a source of xrays, and the remaining material as an absorber, so you expect intensity on the back I = I_0 * exp(-x*mu)18:18
xtalmathwhere mu is a material constant18:19
nmz787so you want to know when you're deep enough to switch over18:19
xtalmathI see18:19
nmz787umm, idk if that would be the equation... but yes metal absorbs xrays18:19
xtalmathyeah, I had to do this calculation like for 3 or so different experiments at university18:20
nmz787so for a metal piece, or silicon, etc... you will need to calibrate for given beam power/diameter and signal detected18:20
nmz787but you can just turn the working piece and take an image, and use geometry to calculate the depth you're at, etc18:20
xtalmathyes, I_0 will depend on milling settings18:20
nmz787i was looking at this, $89, but am not sure if I could mate it with the LPC4370 (easily) http://www.element14.com/community/docs/DOC-69210/l/fpga-development-board-for-the-raspberry-pi18:21
xtalmathbut if you have a seperate xray source, then I_0 will be constant18:21
nmz787it has some diff-pairs routed to a decent connector18:21
nmz787nah no separate source, from PIXE only18:22
nmz787.wik PIXE18:22
yoleaux"Particle-induced X-ray emission or proton-induced X-ray emission (PIXE) is a technique used in the determining of the elemental make-up of a material or sample. When a material is exposed to an ion beam, atomic interactions occur that give off EM radiation of wavelengths in the x-ray part of the electromagnetic spectrum specific to an element." — https://en.wikipedia.org/wiki/PIXE18:22
xtalmathnmz787: heh I have a Spartan 6 LX9 dev board, but not that one18:22
nmz787I wish there was someone around here that wanted to be an electronics apprentice or something... someone I could get to do a lot of the non-research aspect of this design process18:23
xtalmathI am not familiar enough with LPC4370, so even if I advised you, you should take my word with a grain of salt18:23
nmz787heh18:23
nmz787nah even another ADC would be OK18:24
nmz787especially if I already need an FPGA18:24
nmz787I've done one other slower ADC design before for a CCD... but I never really tested that for signal integrity or anything18:24
nmz787and I wasn't oversampling18:24
xtalmathnmz787: I think it will be simpler to start with an ADC with parallel outputs, but not sure how suitable video ADC's are, as they may or may not assume/need a specific timing reference in the input signal18:25
xtalmathis it possible to output the 12 bits on 12 pins with the LPC?18:26
xtalmathif so its straightforward to connect fpga18:26
xtalmathOTOH, if the LPC has SerDes, the spartan can interface with lots of signalling standards18:27
xtalmathoh crap, have to wake up in 4 hours, and visit my sister, help a little with their house rebuilding18:28
xtalmathtalk to you later!18:28
xtalmathI will think about the apprenticeship18:28
xtalmathXD18:29
-!- xtalmath [~xtalmath@ip-83-134-140-175.dsl.scarlet.be] has quit [Quit: Leaving.]18:52
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 250 seconds]19:22
-!- delinquentme [~delinquen@74.61.157.78] has quit [Ping timeout: 256 seconds]20:04
-!- Houshalter [~Houshalte@oh-71-50-58-200.dhcp.embarqhsd.net] has quit [Ping timeout: 244 seconds]20:23
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap20:29
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-jddiosnkjdorwlch] has quit [Quit: Connection closed for inactivity]20:46
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: Namaste]21:03
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-tqexvzmsobeckrbp] has quit [Quit: Connection closed for inactivity]21:20
rancydis the mailinglist still going? i recieve a 404 when visiting https://groups.google.com/diybio21:23
-!- sheena [~home@d162-156-32-180.bchsia.telus.net] has joined ##hplusroadmap21:24
kanzureyou want https://groups.google.com/group/diybio21:25
rancydkanzure: thank you21:28
-!- c0rw1n is now known as c0rw|zZz21:39
-!- sheena [~home@d162-156-32-180.bchsia.telus.net] has quit [Ping timeout: 265 seconds]21:47
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap22:35
-!- yashgaroth [~ffffff@2602:306:35fa:d500:8049:d663:77ad:31c6] has quit [Quit: Leaving]22:45
-!- Mr_Rhino [~satoshi@179.104.144.253] has joined ##hplusroadmap22:52
-!- sandeep__ [~sandeep@111.235.64.135] has joined ##hplusroadmap23:13
--- Log closed Mon Aug 24 00:00:50 2015

Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!