Back
[00:00:05] <orlock> oh heh, took me a moment
[00:00:24] <orlock> to figure out what PR was
[00:00:31] <orlock> i'm still here for another hour or so
[00:00:44] <DagoRed> lol
[00:00:59] <DagoRed> rue_house: one of these days you need to put tobber on git
[00:01:58] <rue_house> tobbor is acidblood with custom modules
[00:02:13] <DagoRed> yay!
[00:02:20] <DagoRed> rue_house: oh cool
[00:02:36] <DagoRed> still... it would be fun to dive deep with tobber.
[00:02:59] <DagoRed> Specifically including a live scripting interface for it so it can adapt at run time.
[00:03:08] <orlock> DagoRed: i only used that for the first time last night for testing, i actually changed the code so that it uses millis() instead of micros()
[00:03:21] <DagoRed> ahh
[00:03:27] <DagoRed> not bad orlock
[00:03:31] <orlock> DagoRed: to avoid the overflow i'm seeing
[00:04:09] <DagoRed> yup
[00:04:24] <DagoRed> no biggy, just small and if your name is on it... fuck it.
[00:04:28] <orlock> (so i can keep displaying the timer information even after the micros() timers hit the bug
[00:04:35] <orlock> ST2147200016
[00:04:36] <orlock> PT2147173044
[00:04:44] <orlock> then
[00:04:45] <orlock> ST-2146765824
[00:04:45] <orlock> PT2147451580
[00:04:58] <orlock> with 6 steps in between
[00:05:21] <orlock> anyway - think thats fixed it
[00:05:25] <orlock> i can only see clouds
[00:05:37] <DagoRed> understood
[00:06:11] <DagoRed> Also that issue sounds like a rollover issue.
[00:06:44] <orlock> yeah
[00:06:55] <orlock> look at lines 278 + 279
[00:07:18] <orlock> though that comment should say <0 or negative
[00:07:22] <orlock> rather than wrap around to 0
[00:07:38] <orlock> but the code is right
[00:07:43] <rue_house> why is time a signed value
[00:07:53] <DagoRed> my thoughts exactly
[00:07:54] <rue_house> shouldn't time be unsigned
[00:08:14] <orlock> what happens when an unsigned overflows?
[00:08:24] <DagoRed> goes to zero
[00:08:28] <rue_house> it goes to zero
[00:09:06] <rue_house> but it hits twice the value before it does it
[00:09:20] <orlock> yeah
[00:09:26] <orlock> 70 minutes instead of 35
[00:10:05] <orlock> but i can see if its wrapped around with a signed int
[00:10:08] <orlock> as it will be -ve
[00:10:21] <orlock> if its unsigned, it will be positive
[00:10:46] <orlock> how do i tell if its wrapped around or not?
[00:12:42] <orlock> i guess previousStepDEC_Micros should always be smaller than currentDEC_Micros, except for what they have just been made equal
[00:13:09] <orlock> so if its not, then it hasnt wrapped
[00:13:52] <orlock> i guess the real reason - because its easier to see if theres still a problem after 35 minutes than 70 minutes
[00:14:33] <DagoRed> orlock: you need to add a counter for rollover.
[00:16:16] <orlock> do i?
[00:16:28] <DagoRed> It would be beneficial and take little code.
[00:53:14] <anniepoo> 8cD anybody out there need a programmer? We're needing a shot in the arm to make the biz go, so I'm just going to work a job for a bit.
[00:55:26] <DagoRed> anniepoo: what kind of business?
[00:55:32] <DagoRed> Also... where?
[00:55:51] <anniepoo> I'm in eastern Oregon, but could temporarily move.
[00:56:20] <DagoRed> Oh you're looking for work?
[00:56:25] <DagoRed> I'd hit up Datalogic.
[00:56:30] <DagoRed> .g datalogic eugene
[00:56:31] <makepi> DagoRed:
http://www.datalogic.com/
[00:56:39] <anniepoo> I'm a software engineer. I've been an engineer for 38 years, I could do most anything
[00:56:54] <anniepoo> thanks!
[00:57:03] <DagoRed> Yeah... contact them. ADS is over there and they are pretty awesome.
[00:57:09] <DagoRed> Telford, PA... not so much.
[00:58:03] <anniepoo> I just need a job job for a year or so. Raising money at the stage our startup is at isn't really making sense - we don't need all that much, I can just go earn it
[00:58:25] <DagoRed> so boot strapping. Yeah hit them up.
[00:59:14] <DagoRed> Or... David Eadline has a book on hadoop. Read that book and say you're a hadoop expert and you'll make big money while working at home on distributed systems.
[00:59:15] <anniepoo> ok, sounds good
[00:59:24] <DagoRed> .g david eadline hadoop book
[00:59:25] <makepi> DagoRed:
https://www.linkedin.com/in/deadline
[00:59:36] <anniepoo> thanks
[01:00:48] <DagoRed> http://www.clustermonkey.net/Hadoop2-Quick-Start-Guide/
[13:45:48] <veverak> hi folks
[13:46:00] <veverak> I suppose that raspberries 3.3V regulator can power esp8266 ?
[13:46:15] <adam789654123> i just "discovered" difraction last night
[13:46:36] <adam789654123> i was trying to measure a shadow, and was mystified that i could move the thing
[13:46:45] <adam789654123> pretty cool
[13:46:48] <adam789654123> :)
[13:48:25] <adam789654123> sorry veverak
[13:48:26] <adam789654123> hello
[13:48:30] <Snert_> veverak: it might depend on which generation and flavor of pi. I always check the docs. I think so, but an 8266 can pull quite abit.
[13:48:30] <adam789654123> <veverak> I suppose that raspberries 3.3V regulator can power esp8266 ?
[13:48:40] <veverak> Snert_: rpi3
[13:49:22] <Snert_> even the 8266 docs warn you against using the onboard 3.3v - saying it might not work depending on what other loads you may/may not have.
[13:49:44] <veverak> wat?
[13:49:45] <veverak> :)
[13:49:57] <veverak> Snert_: eps8266 afaik doesn't have onboard 3.3v regulator
[13:50:03] <veverak> it's supposed to be powered by 3.3v
[13:50:10] <Snert_> I'd use it but keep an eye out and 1st suspect if there's a problem would be lack of juice.
[13:50:47] <Snert_> the 8266 docs warn you about usung the pi onboard regulator for the above reasons.
[13:50:55] <veverak> oh, that way
[13:50:56] <veverak> ok
[13:50:59] <Snert_> yea.
[13:51:36] <Snert_> I mean, I'd use it. but with cognizance and if there's a problem you know what to suspect first.
[13:51:42] <deshipu> pi zero has a stronger regulator
[13:51:57] <deshipu> pi3 too, iirc
[13:53:20] <Snert_> I just see so many problems over on #esp8266 related to juice starvation is all.
[13:53:42] <deshipu> yeah, it's number one problem
[13:53:57] <deshipu> mostly when people try to power them from usb2ttl
[13:54:08] <veverak> :))))
[13:54:17] <veverak> hmm
[13:54:19] <veverak> come to think of it
[13:54:31] <veverak> for my use case it makes much more sense to separate regulator
[13:54:36] <veverak> esp should be backup serial connection
[13:54:44] <veverak> backup powered from the device it back's up sounds bullshity
[13:57:43] <Snert_> and going with external juice would be the first thing to do if there's a problem anyways. For proof if it's a power problem or not.
[13:58:14] <Snert_> you might get away with onboard 3.3v.
[14:06:21] <veverak> http://imgur.com/fhmVUZD
[14:06:23] <veverak> beta version
[14:06:23] <veverak> ;)
[14:06:53] <veverak> but still things to do, find a way how to make lines for serial wider after it leaves header
[14:07:09] <veverak> aaand reroute 3.3v line so it's not under the esps antena
[14:52:49] <DagoRed> Nice veverak
[15:31:19] <wpo> anybody it# going to ieee southeastcon?
[15:32:00] <DagoRed> I wish.
[17:48:58] <rue_house> "If you weren't disabled when you parked here, you will be when you leave."
[18:00:34] <z64555> yeah, thats a job for police and the parking lot attendent, not concerned citizens
[20:22:40] <doomlord> google sale of boston dynamics - impressive demos but too ambitious / too far from comercially viable products?
[20:27:13] <z64555> I can see a pack robot potentially viable
[20:27:28] <z64555> basically an expense donkey/mule
[20:31:06] <doomlord> maybe the whole thing of walking robots today is just too ambitious
[20:31:17] <doomlord> stick to wheels
[20:40:58] <z64555> how are they going to progress without some big funding
[20:45:35] <doomlord> i suspect it's waiting on advances in motors ?
[20:45:54] <doomlord> need more fundemental research into electric muscles
[20:46:10] <doomlord> wheras that company was solving the software (admitedly impressive) for putting everything together
[20:46:17] <doomlord> software/design
[20:46:28] <doomlord> I am dissapointed , i would like to know more
[20:51:39] <DagoRed> Boston dynamics is fine. The issue is we don't have enough people dying right now to justify the cost.