<prpplague>
implr: yea, i thought about that, just worry about the details of getting on the train
<prpplague>
implr: WAW is $1700
<cr1901_modern>
pie_: I don't think your grounded socket is working then :P; instead of being at constant potential like it's supposed to, your chassis gnd is being referenced to neutral plus noise (I think).
<prpplague>
implr: 18 hours
<cr1901_modern>
I'd be curious to see multimeter measurements tho :P
<implr>
that's not that bad
<implr>
I'm in SF right now and I'll be returning in a week, SFO-FRA-WAW, 11h + 4h layover + 1.5h
<prpplague>
implr / kc8apf it was more of a mental exercise in checking the routes
<kc8apf>
ah, that was the flight. SFO->FRA->GDN
<kc8apf>
15h or so
<implr>
direct flights from west coast to anywhere in Poland are quite rare and/or expensive
<kc8apf>
need to buy those tickets
<pie_>
cr1901_modern, yeah im suspicious now
<pie_>
sigh
<q3k>
yeah, flying into GDN isn't worth it unless you're already in mainland europe
<q3k>
WAW is much better connected
<q3k>
but really, land in KAT or WAW or KRK or GDN, doesn't matter, take the train
soylentyellow_ has joined ##openfpga
soylentyellow has quit [Ping timeout: 256 seconds]
Miyu has quit [Ping timeout: 240 seconds]
<rqou>
extremely weird and random question: anybody here (whitequark?) know what the decomposition temperature of PEDOT:PSS is?
digshadow has quit [Ping timeout: 260 seconds]
Miyu has joined ##openfpga
<cr1901_modern>
Miyu: (2:05:04 PM) ***Miyu has painful memories of compiling code on RPi systems
<cr1901_modern>
I once compiled gcc on RPi 1 Model A w/ 256MB of memory. It took 3 days (thanks to it going to swap).
<Miyu>
heh :D
<Miyu>
yeah, I was just compiling a small service the last time (took ~5-10 seconds on my 2013-era i7 laptop), and it was taking ~1+ minute
<cr1901_modern>
I couldn't use distcc for this because the GNU C++ ABI changed between the stock RPi compiler and the cross-compiler I had
<Miyu>
fun :)
<cr1901_modern>
So, yes I very much sympathize w/ your pain :)
<Miyu>
:D
<Miyu>
yeah, I'd definitely try the cross-compiling route next time
<Miyu>
I was also confident/stupid enough to think that I could do the debugging on the device
<Miyu>
made for very long and painful debug/develop/compile/test cycles
<Miyu>
ended up writing a simple integration test instead
<Miyu>
mocking up the I2C device and GPIO
<qu1j0t3>
:D
<cr1901_modern>
Sounds like a blast (and by blast I mean not)
<Miyu>
yup
<Miyu>
though it did make me love integration tests even more :)
<Miyu>
pretty simple GPIO mock, just using timers to simulate the switches being triggered
* cr1901_modern
spent about 30 seconds looking for wiringPi
<Miyu>
:)
<cr1901_modern>
At least on Net, there's ioctls for accessing peripherals. In practice, all RPi GPIO I've seen accesses /dev/mem directly
* Miyu
heads off to board her plane :)
* Miyu
nods
<cr1901_modern>
Cool have fun
<Miyu>
thanks :)
<Miyu>
flying SEA to FRA
<Miyu>
10.5 hours :D
<Miyu>
o/~
<cr1901_modern>
ouch
<rqou>
everybody is having bad experiences with planes these few weeks wtf
<rqou>
is this just travel season?
Miyu has quit [Ping timeout: 272 seconds]
<azonenberg_work>
wait, miyu was in SEA?
<azonenberg_work>
first i heard, lol
<rqou>
do you know them irl or something?
<rqou>
hey azonenberg_work you work with weird materials right? do you know anything about PEDOT:PSS?
soylentyellow__ has joined ##openfpga
<azonenberg_work>
We've known each other on IRC for... quite a few years
<azonenberg_work>
Never managed to be in the same place in meatspace
<rqou>
yeah, meatspace is hard
soylentyellow_ has quit [Ping timeout: 256 seconds]
soylentyellow__ has quit [Client Quit]
soylentyellow has joined ##openfpga
GenTooMan has quit [Quit: Leaving]
<rqou>
azonenberg_work: what is part 90?
<azonenberg_work>
rqou: private licensed land mobile
<azonenberg_work>
Although most part 90 radios are able to work on the 2m amateur band too
<azonenberg_work>
part 90 VHF radios*
<rqou>
does that include police/public safety?
<azonenberg_work>
Yes
<azonenberg_work>
Which is my intended use (I want one radio that can run on both the licensed SAR channel and the local ham repeaters)
<rqou>
ah ok
<rqou>
hmm, is that actually allowed?
<azonenberg_work>
You can run basically any radio on amateur bands
<azonenberg_work>
there's no part 97 equipment certification needed
<azonenberg_work>
So if the radio is certified for the licensed band you run it on, and you have a ham license, you're golden
<rqou>
i do know a chinese company tried to pull a fast one on the FCC and make a ham+GMRS radio
<azonenberg_work>
Yes, which really annoys me re not being allowed
<rqou>
which the FCC subsequently caught and revoked
<azonenberg_work>
If i have the necessary licenses i should be good to go
<rqou>
so why is ham+part 90 ok?
<azonenberg_work>
Well, GMRS has a rule that you cannot have a VFO
<azonenberg_work>
Most part 90 stuff does too, I think? But if you use a factory programming cable to load one specific amateur frequency into a part 90 radio
<rqou>
which this chinese radio doesn't, if you boot into the gmrs mode
<azonenberg_work>
it still doesnt have a VFO
<azonenberg_work>
all i know is that running part 90 radios on ham frequencies is common practice and legal
<rqou>
of course
<azonenberg_work>
And running a ham-only radio on part 90 frequencies, even if you are a licensed user of the band, is not legal
<rqou>
yeah, dumb
<azonenberg_work>
Anyway it needs to be totally weatherproof (duh), and have a headset able to be worn under a bike helmet or rescue helmet
<rqou>
(inb4 the germans jump in and tell us nothing is allowed there :P )
<azonenberg_work>
I'm trying to move away from shoulder mics because they tend to get water clogged up in them
<azonenberg_work>
even if waterproof (meaning, no damage when wet) the signal can be unintelligible
<azonenberg_work>
So i wanted something i could put under a hood to keep dry
<azonenberg_work>
and then a chest or ear mounted PTT or something
<rqou>
makes sense
<rqou>
FCC type approval is such a huge pain in the ass
<azonenberg_work>
Yes, it is
<rqou>
especially in the current era of "everything is (or should be) an sdr"
<azonenberg_work>
Yeah
<azonenberg_work>
i have a GMRS license, a ham license, and am a member of an organization with a part 90 license
<azonenberg_work>
Why can i not get one radio that (legally) can be used on all those bands?
<balrog>
azonenberg_work: because most people don't have all those licenses
<rqou>
um, afaik this is actually quite common?
<azonenberg_work>
balrog: yes, but?
<azonenberg_work>
What is that supposed to mean
<rqou>
there's a decent amount of azonenberg-types who do SAR and radios
<balrog>
rqou: they FCC doesn't care about that "edge case" (to them)
<azonenberg_work>
balrog: The point is, the FCC seems to have a very whitelist-oriented policy when it comes to what hardware is legal
<balrog>
azonenberg_work: yes
<azonenberg_work>
As opposed to simply saying, whatever hits the antenna has to comply with the rules of the license/band you are transmitting under
<azonenberg_work>
And how that signal was generated is irrelevant
<rqou>
i mean, the ham rules work like that
<azonenberg_work>
Exactly
<rqou>
(other than weird archaic classifications of phone/RTTY/etc.)
<azonenberg_work>
I'd even be OK with a rule that says that you cannot market such a multi-band radio as anything other than a ham radio
<azonenberg_work>
But that, if you have a ham license and buy said radio, you can legally transmit on any band you have permission to transmit on
<azonenberg_work>
within power/modulation restrictions obviously
<rqou>
the problem with that nowadays is called "China" :P
<azonenberg_work>
well i mean baofeng exists
<azonenberg_work>
and flies in the face of all these rules
<rqou>
although stupid shit from China already happens, so idk
<azonenberg_work>
I'm not asking for such devices to exist
<rqou>
um.. baofeng is part 15 certified?
<azonenberg_work>
i'm asking for them to be legal to use :p
<rqou>
which is all they technically need to do?
<azonenberg_work>
i believe some baofeng models could transmit on FRS frequencies
<azonenberg_work>
Which is one of the things that got them in trouble
<rqou>
oh yeah they can
<azonenberg_work>
since it had a VFO and removable antenna
<rqou>
but they don't market as a FRS radio?
<balrog>
azonenberg_work: isn't it legal for a radio to be able to have a VFO or channels, but require software and programming cable to switch between them?
<azonenberg_work>
balrog: Dont know
<azonenberg_work>
rqou: doesnt matter if they market it as such
<azonenberg_work>
aiui its supposed to be *incapable* of transmitting on frs
<rqou>
i don't think so?
<rqou>
afaik lots of ham equipment can transmit outside the proper bands
<rqou>
it's up to you not to do that
<azonenberg_work>
I dont remember the details, other than that the FCC slapped them with some fines over something
<rqou>
i know a chinese company did try to pull a fast one on the fcc by sending the same radio through the approval process twice, once for gmrs and once as just a receiver that had a VFO
<rqou>
they then put both fcc ids on the radio
<rqou>
the fcc noticed and was not amused
Bike has quit [Quit: Lost terminal]
s1dev has joined ##openfpga
<rqou>
azonenberg_work: is there an easy way to draw a graphic shape in kicad that is a rectangle with multiple circles punched out?
<azonenberg_work>
rqou: hmmm
<azonenberg_work>
Not that i know of
<gruetzkopf>
i have stuff in the works to make the motorola jedi family illegal ;)
<gruetzkopf>
(at least the bigger models with a full keyboard)
azonenberg_work has quit [Ping timeout: 240 seconds]
azonenberg_work has joined ##openfpga
<lain>
lol I have a baofeng uv-5r
<cr1901_modern>
lain: If I ever didn't drag my feet w/ using my license, that would be also what I get
<lain>
we got it to monitor the ferry around here, since they use simple NFM walkies in a band the uv-5r supports, but I intend to get my ham license and use it for that as well
<lain>
from our testing, the transmit is clean. it seems some of the uv-5r's have faulty tx filters and just spew harmonics, but ours does not, thankfully
dmaarouf has joined ##openfpga
<dmaarouf>
Hi, anyone know where I can find a copy of RapidWright?
dmaarouf has quit [Quit: Page closed]
digshadow has joined ##openfpga
_whitelogger has joined ##openfpga
<Lord_Nightmare>
lain: i have a baofeng uv-5r as well. how would i be able to tell if it is one of the faulty ones?
<Lord_Nightmare>
is it a specific firmware range?
<lain>
Lord_Nightmare: I'm honestly not sure other than just testing it. the sense I got from watching youtube videos of people testing it was that some of the units just have faulty filters somehow (manufacturing defect?)
<Lord_Nightmare>
ah. and my brother with his nice radio rig is halfway across the country, so i can't easily check. drat.
rohitksingh has joined ##openfpga
rohitksingh has quit [Quit: Leaving.]
rohitksingh has joined ##openfpga
azonenberg_work has quit [Ping timeout: 240 seconds]
azonenberg_work has joined ##openfpga
s1dev has quit [Ping timeout: 256 seconds]
rohitksingh has quit [Quit: Leaving.]
gnufan has quit [Ping timeout: 244 seconds]
gnufan has joined ##openfpga
ayjay_t has quit [Read error: Connection reset by peer]
<pie_>
Guys, good or dumb idea: application users should get to have history trees instead of just undo/redo
<pie_>
(and merging too)
X-Scale has quit [Ping timeout: 244 seconds]
Miyu has joined ##openfpga
<whitequark>
awygle: wtf
<whitequark>
it's very good that we still haven't fabbed revb and revc
<whitequark>
because i think i'm observing a signal integrity issue clearly unrelated to fxma
<whitequark>
awygle: any chance i could interest you in writing an usb benchmark/stresstest applet? it's been on my list for a while but i'm kind of swamped
<whitequark>
thinking of generating patterns for maximum crosstalk here mostly
<whitequark>
tl;dr i filled the low 2 bits of rgb grabber pixel samples with an up counter and all hell broke loose
<whitequark>
yep and if I fill the almost-high 2 bits then all hell breaks loose -even worse-
<whitequark>
which makes zero sense whatsoever from pure digital logic perspective
<whitequark>
hm, thaaat is really interesting
<whitequark>
i have a synchronous reset here and one register without synchronous reset. resetless..
<whitequark>
actually, two.
<whitequark>
anyway, when i added back reset to it, this nastiness stopped. i wonder if everything's going to work better if i remove the second too.
azonenberg_work has quit [Ping timeout: 268 seconds]
<whitequark>
I wonder if there's something about resets I don't know about
<whitequark>
these appear to be normal synchronous resets, why the hell don't they work properl
azonenberg_work has joined ##openfpga
digshadow has quit [Ping timeout: 265 seconds]
<azonenberg_work>
whitequark: if a glasgow can make its way to my lab once i'm set up
<azonenberg_work>
My scope might be fast enough to debug some of that?
digshadow has joined ##openfpga
X-Scale has joined ##openfpga
digshadow has quit [Ping timeout: 260 seconds]
<kc8apf>
azonenberg_work: cr123a was a pretty difficult requirement for radios. Vertex Standard (now Motorola *sigh) and Icom both have part 90 radios with IP57 or IP67 with behind-the-head earpieces, etc. They've all moved to Li-ion batteries or AAs as a fallback
<azonenberg_work>
Yes
<azonenberg_work>
kc8apf: I'm wondering how difficult 3d printing / machining a custom cr123a battery pack would be
<azonenberg_work>
With the same voltage as the li-ion pack
<kc8apf>
most of these have 7.5V +/- 20% input jacks for charging
<whitequark>
azonenberg_work: yeah sure, i imagine at least awygle's one is already not that far
<whitequark>
but this is really puzzling
<whitequark>
this entire design is a SINGLE clock domain and it passes timing by a margin of 1/3
<azonenberg_work>
kc8apf: Hmmm, so you think just building a primary cell pack with a cord to feed into it
<azonenberg_work>
that would actually work nicely, keep the cr123s in my backpack
<azonenberg_work>
and use it to top off the li-ion pack overnight if we're out long enough to drain the primary battery (which will easily last a single day of operations)
<kc8apf>
azonenberg_work: yup. Need to check that they haven't moved away from separate charging plugs. That was a trend for a while.
<azonenberg_work>
That would let me keep the li-ion pack as the primary power source since its rechargeable, and since >90% of our missions are <24 hours that would let me cut battery costs
<kc8apf>
even buying the AA adapter and running a cable out would work
<azonenberg_work>
While still having a single type of spare battery for everything but the GPS (which still runs on AAs, but only two of them so 4 batteries is 2 changes which will easily last through a 48h mission)
<azonenberg_work>
kc8apf: no, because that would require that i have cables on me while on the move
<awygle>
whitequark: hm, odd. fully synchronous resets on a free-running clock?
<kc8apf>
ah, true
<azonenberg_work>
I'm ok with a primary cell charger while i'm sleeping
<azonenberg_work>
and a li-ion pack on the radio itself
<azonenberg_work>
But i dont want to have two batteries when i'm on the trail
<azonenberg_work>
or worse yet, off trial
<azonenberg_work>
cords get snagged on things
<azonenberg_work>
kc8apf: and i'm actually eyeing a really nice looking Kenwood
<whitequark>
awygle: the clock is sourced from cy7c
<azonenberg_work>
the NX-5200, it has a nice big LCD on it so i can finally move away from 6 character segmented channel names
<azonenberg_work>
As well as P25 and NXDN digital capability, not that i need that now but nice to futureproof
<whitequark>
hmmm I do have a suspect
<azonenberg_work>
if the unit goes digital eventually
<whitequark>
when I wrote the FX2 interfacing code I didn't add any synchronizers toit
<whitequark>
because I never managed to get it to pass timing *and* function at all with arachne
<azonenberg_work>
whitequark: welp
<awygle>
oh, yeah, that'd do it probably
<whitequark>
so I imagine nextpnr just places all that logic however it likes and it gets placed in fun ways from bitstream to bitstream
<azonenberg_work>
kc8apf: and the radio is actually IP68 rated (1 meter submersion for 2 hours)
<whitequark>
(or arachne for that matter)
<kc8apf>
azonenberg_work: yeah, that's a nice radio
<azonenberg_work>
whitequark: cross clock support is one thing i really want to see in the open tools
<kc8apf>
I tend to go with Yaesu personally
<azonenberg_work>
kc8apf: yeah i own a vx-6r
<azonenberg_work>
But its not part 90
<whitequark>
azonenberg_work: i want to constrain phase of a parallel bus
<whitequark>
seems related
<whitequark>
also, I now extremely see why parallel buses died
<kc8apf>
I have an vx-8r. They sold the 8r as a vertex standard model as well.
<azonenberg_work>
Lol
<azonenberg_work>
:p
<azonenberg_work>
whitequark: interesting, can you cheat by doing placement constraints and hoping for the best?
<kc8apf>
so it can be modified to be part 90
<whitequark>
azonenberg_work: well right now I cheat by not adding synchronizers
<azonenberg_work>
Lol
<whitequark>
it almost completely works almost all of the time. simpler designs just never break
<azonenberg_work>
i meant by having synchronizers and LOCing them to adjacent slices
<whitequark>
whereas if I add registers the whole design just falls apart
<awygle>
wait how bad can your phase difference actually be? isn't it a 48 MHz bus or something?
<azonenberg_work>
then hoping the routes are direct
<whitequark>
or at least did before
<azonenberg_work>
Lol
<azonenberg_work>
awygle: you underestimate how bad paths can be if the toolchain isnt optimizing
<azonenberg_work>
kc8apf: yeah i want something actually legal
<whitequark>
everything also broke
<awygle>
is the data edge == the clock edge?
<whitequark>
uhhhhhhh
<whitequark>
take a look at the diagrams in the cypress manual.
<awygle>
yeah ok
<whitequark>
they have some profoundly weird s/h timings
<whitequark>
including in one case a setup timing slightly longer than clock period
<whitequark>
i'm not actually sure how that even works
<azonenberg_work>
kc8apf: the vx-459 looks like the only vertex standard submersible radio left post motorola merger
<kc8apf>
sadly
<awygle>
"slave fifo synchronous write"?
<whitequark>
in fx2.py i summarized them athough i could have mangled some detail i couldn't understand
<azonenberg_work>
and i like the display on that kenwood better
<whitequark>
yes
<awygle>
hm ok, so "no", clock lags data by 9.2ns. love that 0ns hold time though wow
<whitequark>
and clock lags write strobe by 18ns
<whitequark>
and read strobe by 10ns
<whitequark>
like what??
<whitequark>
what the hell kinda silicon do you have
<whitequark>
awygle: i also have a problem that i can't just add registers
<whitequark>
i have to change logic pretty severely
<whitequark>
because i have to monitor FIFO fullness
<whitequark>
and then there's FIFOADR 25ns setup time
<awygle>
yes, that is very weird. but synchronization shouldn't be a problem, if everything's clocked on IFCLK and the first thing the data hits are the IOB registers, i don't see how you'd even need synchronizers unless the ice40 setup time is >9.2ns
<whitequark>
so for one you can see an expression in fx2.py i had to rewrite
<whitequark>
to make a combinatorial feedback path shorter
<whitequark>
or it wouldn't clear timings
<awygle>
hm ok. i can't dive into this at the level it needs right now (at work). i will try and look into it this weekend but i'm still unpacking so no promises unfortunately
<azonenberg_work>
although that looks like a knockoff so idk
<azonenberg_work>
It looks like if you don't mind a slower charge (overnight) you can also just stick a wallwart (or wallwart-lookalike cr123 pack) right into the barrel jack on the radio
digshadow has joined ##openfpga
<azonenberg_work>
actually no there is no barrel jack
<azonenberg_work>
the official "wallwart" charger looks just like that but with kenwood branding
s1dev has joined ##openfpga
cr1901_modern has quit [Read error: Connection reset by peer]
cr1901_modern has joined ##openfpga
oeuf has quit [Read error: Connection reset by peer]
oeuf has joined ##openfpga
<rqou>
ugh friday means tons of h. sapiens larvae in the cafeteria for lunch
<jn__>
bring-your-kid-to-work day?
cr1901_modern has quit [Read error: Connection reset by peer]
<rqou>
nah, just Friday in general
cr1901_modern has joined ##openfpga
<rqou>
despite all the "zomg millennials aren't producing enough crotchspawn" articles there are a ton of people with kids here
cr1901_modern has quit [Read error: Connection reset by peer]
cr1901_modern has joined ##openfpga
s1dev has quit [Remote host closed the connection]
s1dev has joined ##openfpga
<qu1j0t3>
probably varies by socioeconomic bracket
<rqou>
soooo, i started following some people on birbsite that I met at the dcfurs party, aaaaaand now i understand why "likes are now florps/stochastic retweets" annoys people
uovo has joined ##openfpga
oeuf has quit [Ping timeout: 268 seconds]
<azonenberg_work>
lol
s1dev has quit [Ping timeout: 256 seconds]
<pie_>
crotchspawn
<pie_>
lmao god that sounds horrible
wpwrak has quit [Read error: Connection reset by peer]
wpwrak has joined ##openfpga
<azonenberg_work>
rqou: getting spammed with random furry pr0n?
<gruetzkopf>
do my old motorola GP-300 and GP1200 (MT(S) 2000/2100?) count as part 90 radios?
<gruetzkopf>
especially the 5-tone models are great (enter out-of-band frequency in programming software, hold shift while pressing enter, done)
s1dev has joined ##openfpga
s1dev has quit [Ping timeout: 268 seconds]
<pointfree>
This PIC24 CTMU "Charge-TIme Measurement Unit" looks like what I need to measure picoseconds for TDR. It would be nice if a PSoC had something like it.
<pointfree>
The CTMU is often used for capacitive touch, so maybe the PSoC capsense can do some of what the CTMU does. Hm.
<azonenberg_work>
pointfree: how are you doing TDR with charge time measurement?
<azonenberg_work>
it might work for a limited case looking for one peak or something
<pointfree>
azonenberg_work: I'm just exploring options right now. Microchip has some appnotes on using the PIC's CTMU for TDR.
<azonenberg_work>
interesting
<azonenberg_work>
I mean it really depends on what you want to measure
<azonenberg_work>
how preciselyt
<azonenberg_work>
how fast
<azonenberg_work>
single ended or differential
<azonenberg_work>
etc
<azonenberg_work>
For example if its cat5 or 100 ohm differential cable in general, a lot of ethernet PHYs have a basic tdr built in
<azonenberg_work>
ksz9031 is somewhat limited in performance but acording to datasheet does work
<azonenberg_work>
the dp83867 seems to have a much more capable one but i havent tried using it yet
<azonenberg_work>
(up to 4 peaks with fairly detailed height and position)
<pointfree>
azonenberg_work: I'm also thinking about using an ECL gate or a really fast 30GHz GaAs MESFET to effectively measure race conditions.
<azonenberg_work>
fast comparator
<azonenberg_work>
see my blog
<pointfree>
azonenberg_work: you did something like this for io characterization with starshipraider?
<kc8apf>
azonenberg_work: there's a few variants of the KSC-32 charger. The one I linked is bulk and designed for automotive use. I'd be worried about the radio falling out of the desktop charger and having a dead battery pack in the morning.