<awygle>
anyone know what happened to that thing where you could put ads in your github Readme?
<rqou>
wtf?
<awygle>
I feel like I read something about github shutting it down, but I can't find the official statement.
Bike has quit [Quit: Lost terminal]
FabM has joined ##openfpga
<rqou>
hey azonenberg_work, how's the house btw?
pie_ has quit [Ping timeout: 264 seconds]
<azonenberg_work>
rqou: we have one room 100% finished for in-wall stuff (ready for rough electrical inspection, still needs insulation and sheetrock)
<azonenberg_work>
Goal is to finish room #2 to that extent by saturday - we have to pull out a bunch of old insulation, frame out a new stud wall over some bare concrete, then put power in the new wall plus the wall it shares with the garage
<azonenberg_work>
then room #3 ideally on sunday, it needs a stud wall over some concrete, power in that wall, and a jumper to connect two strings of outlets that i already have installed
<azonenberg_work>
oh, and data wiring still
<azonenberg_work>
During the week we'll probably start work on the future office room, that was entirely tabled while waiting for a sub to jackhammer out the concrete slab and install a french drain
<azonenberg_work>
They finished that earlier today so once the concrete cures enough that i can build a wall on top, i can start framing out that room
<azonenberg_work>
in the meantime i can rip out the old electrical outlets on the one wall i kept (interior wall, all of the exterior walls had the studs removed)
<azonenberg_work>
and start putting power on that
<rqou>
wait, why does your office need a drain?
<azonenberg_work>
After that, i have to rip out sheetrock on one side of the stairs and install a light switch for the already-installed overhead light box
<azonenberg_work>
and the first floor is ready for electrical inspection
<azonenberg_work>
Because we're at the bottom of a hill in a very wet part of the country with underground springs all over the place
<azonenberg_work>
About 5 gallons of water per hour are trying to get into the basement
<azonenberg_work>
The sump pump couldnt suck it out fast enough
<rqou>
wait, is the office in the basement?
<azonenberg_work>
Yes
<azonenberg_work>
well, first floor / basement
<azonenberg_work>
its below grade on one side and at grade on the other
<rqou>
ah ok
<rqou>
no earthquake concerns?
<azonenberg_work>
There's earthquake concerns in the entire pacific northwest :p
<azonenberg_work>
not more so in this area afaik
<rqou>
right, but usually basements aren't too popular in earthquake-prone regions?
<azonenberg_work>
every house i've seen here has one
<azonenberg_work>
because here's so many hills
<azonenberg_work>
If you build a house flat and the ground is sloped, you have two options
<azonenberg_work>
have the uphill side underground, or the downhill side floating above ground
<azonenberg_work>
the former is more stable
<rqou>
ok
<rqou>
i guess the bay is just a lot more flat
<azonenberg_work>
yeah
<azonenberg_work>
The office room has grade level around chest height
<rqou>
also, at least over in the PNW you actually have housing :P
<azonenberg_work>
meanwhile, on the other end of the house
<azonenberg_work>
the lab has grade level at floor level
<azonenberg_work>
its a garage
<azonenberg_work>
s/is/was/
<azonenberg_work>
Lol, barely
<azonenberg_work>
seattle real estate market is NUTS
<rqou>
still less than the bay
<azonenberg_work>
bainbridge was nuts as well, median home price there is up to 800K
<rqou>
still less than the bay
<rqou>
:P
<azonenberg_work>
not by much
<azonenberg_work>
seattle is worse
<rqou>
still less than SF :P
<azonenberg_work>
they had a condemned hulk ready for the bulldozer selling for $575
<azonenberg_work>
The land was probably worth 600+ alone, they discounted it because you had to level the building first
<rqou>
btw an article is currently floating around the bird website about a guy who wants to build an apartment building in SF
<rqou>
it's taken 4 years and $1m and construction isn't allowed to start yet
<azonenberg_work>
if it makes you feel any better
<azonenberg_work>
bainbridge just imposed a moratorium on further development as well
<azonenberg_work>
for some months until some new zoning laws are approved
<rqou>
O_o
<rqou>
they got bit by the nimby disease too?
<azonenberg_work>
the only building permits they're granting are existing applications in process
<azonenberg_work>
and renovations to existing single family residences
<azonenberg_work>
Meanwhile out in poulsbo where we're moving
<azonenberg_work>
the city has a population of 10,290
<azonenberg_work>
according to the sign at the border
<azonenberg_work>
We're a bit closer to downtown than i really wanted, and our new street is a relatively quiet residential road that people have taken to using as a shortcut between two other parts of town
<azonenberg_work>
So we get lots of thru traffic
<azonenberg_work>
But it's affordable, we own the place, and i can build an evil lair finally :p
tnt has quit [*.net *.split]
flaviusb has quit [*.net *.split]
<rqou>
better than me cramming evil lairs in my parents' house :P
RaivisR_ has quit [*.net *.split]
azonenberg has quit [*.net *.split]
moho1 has quit [*.net *.split]
kc8apf has quit [*.net *.split]
<azonenberg_work>
Lol
<azonenberg_work>
well its gonna be another year or so until i actually have it to the point i'm happy, i think
<azonenberg_work>
considering that a lot of the gear i still want to get is expensive
<azonenberg_work>
the proposed new UPS is $8K
<azonenberg_work>
23U / 1100 pounds of batteries
<azonenberg_work>
6 kVA peak output, will run 5 kW for 75 minutes or 700W for 8 hours
stefanct has quit [Ping timeout: 260 seconds]
<rqou>
lol much bigger than my 900VA consumer-grade device :P
<rqou>
<troll> you should get solar panels </troll>
<azonenberg_work>
i have a 1500 now
<azonenberg_work>
but its not enough
<rqou>
PNW = no sun ever :P
<azonenberg_work>
the new one will be hard wired to a 50A subpanel
<rqou>
solar feeding a HVDC grid would be "interesting"
<rqou>
(assuming you have sun of course, which you don't)
stefanct has joined ##openfpga
stefanct has joined ##openfpga
stefanct has quit [Changing host]
<azonenberg_work>
soooo this is interesting
<azonenberg_work>
i need to investigate
<azonenberg_work>
apparently my NAS is running in 100baseT mode instead of gig
<azonenberg_work>
I wonder how long this has been going on
<azonenberg_work>
that would explain things being slow lately...
<rqou>
lol
<rqou>
no 10g nas for you?
<azonenberg_work>
not yet
<azonenberg_work>
i wonder if this is a bad crimp or something
<azonenberg_work>
yeah TDR shows pair D open at 2m length
* azonenberg_work
goes out to garage
<azonenberg_work>
wonder if loose cable or bad crimp or what
RaivisR_ has joined ##openfpga
moho1 has joined ##openfpga
kc8apf has joined ##openfpga
azonenberg_work has quit [Ping timeout: 260 seconds]
openfpga-bb has quit [Ping timeout: 276 seconds]
flaviusb has joined ##openfpga
tnt has joined ##openfpga
openfpga-bb has joined ##openfpga
anuejn has quit [Ping timeout: 240 seconds]
sielicki has quit [Ping timeout: 256 seconds]
Wallbraker has quit [Ping timeout: 240 seconds]
nrossi[m] has quit [Ping timeout: 256 seconds]
pointfree1 has quit [Ping timeout: 256 seconds]
azonenberg has joined ##openfpga
AlexDaniel` has quit [Ping timeout: 276 seconds]
azonenberg_work has joined ##openfpga
<azonenberg_work>
Weko
<azonenberg_work>
Welp*
<azonenberg_work>
In the process of swapping the (confirmed bad) cable between the nas and the switch
<azonenberg_work>
i unplugged the cable to my border router without noticing
<azonenberg_work>
Lol
<azonenberg_work>
now that THAT link is up...
<azonenberg_work>
i'm finally NFS'ing at 1gbps
<rqou>
lol
<rqou>
wait NFS?
<azonenberg_work>
yes
<rqou>
you should consider switching to 9p
<rqou>
no stupid kernel mode server implementing an awful protocol
<azonenberg_work>
lol
<azonenberg_work>
nfs works just fine if i actually run it at full speed
<azonenberg_work>
I need to stick a 10g card in the nas after i move
<azonenberg_work>
they arent too pricey and i'll have fiber everywhere anyway
<rqou>
oh yeah, 9p is much more resilient when run across slow links
<azonenberg_work>
my problem was that i was trying to pull a vivado install image onto another box
<azonenberg_work>
i needed to saturate it :p
<rqou>
ah
<rqou>
btw idk if all of the glue has been written but you can use 9p as rootfs over a serial link
<rqou>
this is useful if you're ever bringing up a deeply embedded box
tnt has quit [*.net *.split]
flaviusb has quit [*.net *.split]
RaivisR_ has quit [*.net *.split]
moho1 has quit [*.net *.split]
kc8apf has quit [*.net *.split]
stefanct has quit [Ping timeout: 255 seconds]
stefanct has joined ##openfpga
stefanct has joined ##openfpga
stefanct has quit [Changing host]
<awygle>
nfs4 is finally pretty good
<rqou>
hmm maybe
<rqou>
iirc nfs4 is basically a rewrite?
<rqou>
the mechanisms for configuring the kernel-mode server is still annoying
<awygle>
Yes, and it was terrible for a while but now it's finally decent
<rqou>
9p's server is just a simple userspace binary
<awygle>
"we made it stateful for speed!" "it's 40% slower"
<rqou>
you can do things like run it in a user namespace too
kc8apf has joined ##openfpga
RaivisR_ has joined ##openfpga
moho1 has joined ##openfpga
<azonenberg_work>
yeah i do everything with nfs4
<rqou>
ugh, currently studying power electronics, and magnetic materials are super finicky and annoying
<rqou>
why do these things actually have nontrivial physical size and weight? :P
<awygle>
mu
<rqou>
also TIL skin effect really does matter :P :P
<rqou>
(i'm mostly a digital guy)
<awygle>
digital just means lying to nature
<rqou>
i'm going to get so screwed when i get around to actually building my "learn how to build high-speed digital" board
<awygle>
... I'm gonna go to bed.
<awygle>
PCBs are soothing because there's an unambiguous right way to do everything
<rqou>
i should force you to spend the rest of your life carefully winding inductors :P
<awygle>
I built an inductor winding robot
mumptai_ has quit [Quit: Verlassend]
<awygle>
For magnetorquers
<rqou>
wait, you've actually built designs with nontrivial magnetics?
<rqou>
did you also take EE113?
<rqou>
wait wait wait
<awygle>
Nnnno I don't think so
<rqou>
"The magnetorquer creates a magnetic field that interfaces with an ambient magnetic field, usually Earth's"
tnt has joined ##openfpga
<awygle>
Yup
<rqou>
wtf who thought of this and why does this work?
flaviusb has joined ##openfpga
<awygle>
I SPIN DA SPACE SHIP
<awygle>
It's absolutely required so you can dump your momentum bias from your reaction wheels
<awygle>
It's just a backwards compass nbd
<sorear>
isn't it a regular compass?
<awygle>
You're using it for the opposite effect but the mechanism is the same
<awygle>
I wanted to build a hemholtz (sp?) cage but never got around to it
<awygle>
Anyway. Sleeeep.
<rqou>
oh wtf
<rqou>
magnetic core loss numbers can be measured in W/lb
<rqou>
WHAT ARE THESE UNITS?!
<rqou>
what is wrong with american engineers?
<rqou>
azonenberg_work: ^
<rqou>
how do you like W/lb as a unit?
<azonenberg_work>
lol watts per pound?
<azonenberg_work>
wut
<rqou>
yup
<rqou>
sane people would measure the loss in W/cm^3 instead
<azonenberg_work>
or even W/in^3
<azonenberg_work>
But W/lb is useless without knowing the density of the core
<azonenberg_work>
and even if i did, i cant imagine why power per unit mass, vs volume, would be useful to me
<rqou>
the textbook says "The specific gravities of magnetic steel alloys are all approximately 8"
<rqou>
:P
<rqou>
magnetics comes from "that" era of engineering
<rqou>
where really skilled MEs have all of these figures memorized
<azonenberg_work>
lol
<rqou>
also it comes from the era where power transformers actually weighed a non-negligible number of pounds :P
<rqou>
well, they still do
<rqou>
but people find ways to use different transformers instead
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
<jn>
"The specific gravities of magnetic steel alloys are all approximately 8". unitless. :P
<rqou>
specific gravity is unitless
<rqou>
it's a ratio to the density of water
<jn>
oh, ok, fair point
<sorear>
although as a small counterpoint I'm annoyed that people measure conductivity on a volumetric basis, when unless you're winding a magnet you probably don't actually care about the volume of the wires
<rqou>
this book uses ohms per length
<rqou>
semiconductor people normally use ohms per square
<rqou>
oh yeah, ohms per square is also a fun unit :P
Wallbraker has joined ##openfpga
user10032 has joined ##openfpga
<sorear>
teachers love the fact that Cu has the second highest volumetric conductivity but this is almost completely irrelevant
<whitequark>
rqou: huh? I have that figure memorized, it's very handy
<rqou>
which figure?
<whitequark>
pretty much all non-exotic steels have a density of around 8
<whitequark>
whether stainless or carbon
<rqou>
ah
<rqou>
i guess my mechanical engineering skills are just really weak
<eduardo__>
rqou: I think only those people out-ninja you which focus on their shit.
<eduardo__>
rqou: like azonenberg says NO to most things he could possibly work on.
sielicki has joined ##openfpga
pointfree1 has joined ##openfpga
nrossi[m] has joined ##openfpga
anuejn has joined ##openfpga
AlexDaniel` has joined ##openfpga
m_t has joined ##openfpga
m_w has joined ##openfpga
m_w has quit [Ping timeout: 240 seconds]
FabM has quit [Quit: ChatZilla 0.9.93 [Firefox 52.5.0/20171114221957]]
user10032 has quit [Remote host closed the connection]
rohitksingh_work has quit [Ping timeout: 276 seconds]
futarisIRCcloud has joined ##openfpga
m_t has quit [Quit: Leaving]
pie_ has joined ##openfpga
<pie_>
i like how ice40 sticks on ebay are more expensive than from lattice
<pie_>
well, unless shipping, i dunno
genii has joined ##openfpga
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
pie_ has quit [Ping timeout: 264 seconds]
azonenberg_work has quit [Ping timeout: 255 seconds]
<awygle>
I assumed, foolishly, that that unit involved Webers rather than watts
ondrej2 has quit [Quit: Leaving]
awygle has quit [Read error: Connection reset by peer]
keith-man has quit [Ping timeout: 252 seconds]
awygle has joined ##openfpga
azonenberg_work has joined ##openfpga
mumptai has joined ##openfpga
mumptai has quit [Remote host closed the connection]
gnufan has quit [Ping timeout: 256 seconds]
gnufan has joined ##openfpga
jhol has quit [Quit: Coyote finally caught me]
jhol has joined ##openfpga
m_w has joined ##openfpga
user10032 has joined ##openfpga
m_w_ has joined ##openfpga
rqou has quit [Ping timeout: 276 seconds]
m_w has quit [Ping timeout: 256 seconds]
xdeller has joined ##openfpga
rohitksingh has joined ##openfpga
<qu1j0t3>
might interest somebody @Broadcom: Tomahawk® 3 – the world’s first 12.8 Tb/s Ethernet switch chip – is now sampling. │
<mithro>
awygle: obessing about the routing infrastructure inside Artix-7 (it's infecting my dreams now...)
<genii>
The Nanopi M3 place had a datasheet of that chip
<awygle>
ah, you have the routing infrastructure version of Tetris Syndrome
<mithro>
awygle: Yeah
<genii>
I think Samsung S5P6818, based on Exynos 8818
<mithro>
awygle: I'm ***so*** close to getting vpr to route something -- but I keep getting bitten by the amount of technical debt I've already accumulated
<awygle>
i've had variations on that theme. most notably 802.11a
<awygle>
mithro: ah that's always rough :/ i usually hit a point where ignoring the tech debt eats up so much of my brain that i have to stop and clean up to make progress
<awygle>
mithro: that's what i've been doing at $DAYJOB today and yesterday. no more magic numbers or unnecessary copies!
eduardo_ has joined ##openfpga
<awygle>
actual abstractions instead of huge flat arrays with hardcoded indexes! it's beautiful! they should have sent a poet!
eduardo__ has quit [Ping timeout: 276 seconds]
rqou has joined ##openfpga
<qu1j0t3>
:)
Bike has joined ##openfpga
<rqou>
friggin linode downtime
<rqou>
also friggin problem set that i need to stop procrastinating on