<sorear> exactly naked is physically possible but frowned upon by the law even if it's 40C
m_t has quit [Quit: Leaving]
<zkms> for people who care about electromagnetic emanations, it's a feature, but most people who install metallised windows only care about energy efficiency so it's a bug
<awygle> sorear: strongly agree
<zkms> since it kills their cell signals lol
azonenberg_work has quit [Ping timeout: 240 seconds]
<shapr> sorear: men wearing only shorts and flip flops isn't that rare around here
<sorear> i kind of assumed cell signals mostly went through wood, oops
<q3k> have y'all seen the beijing bikini?
<q3k> it's fantastic to see IRL.
<zkms> sorear: they do but if you put low-e windows in a brick structure you're fucked
<q3k> also apparently it helps you cool down quite a lot but I wasn't brave enough to try it.
<zkms> it kills the SNR and that drops the MCS which makes the radio operate longer for a given amount of data which kills race-to-idle and that kills your cell phone battery ^_^
<sorear> gender-invariant indecent exposure laws in the US when
<sorear> although SF banned public nudity not very long ago, so they're going in the wrong direction
<whitequark> zkms: good thing the US doesn't build houses out of real materials
<sorear> i've recently moved to the part of the US where brick buildings will stay up for longer than 20 years and it's amazing how many of them there are
<zkms> whitequark: i was in a hotel once with low-e windows and it was some kind of concrete/brick/masonry structure and if i opened the door or window i got like 20megabits/second better download speed on LTE
<whitequark> lol
<q3k> the only place i've ever went to that had frustratingly slow LTE was a hotel in warsaw. in the building that hosts poland's largest IXP. that has a ton of LTE BTSen on the roof.
<q3k> and the hotel wifi was unbearingly slow, too.
<zkms> q3k: le cordonnier le plus mal chaussé
<sorear> so it sounds like I should try to make 35c3 plans
<awygle> the choices here appear to be a) matchsticks b) brutalist concrete monoliths c) literal glass hosues
<q3k> zkms: never heard that expression :)
<sorear> i *like* brutalist concrete monoliths
<q3k> so do i
<awygle> i get the first two on a fault line but why the hell the glass
<awygle> i think brutalism can be done well but mostly isn't
<sorear> because the walls are non-structural and architecture is made of fads
<q3k> sorear: please make 35c3 plans
<awygle> but i famously Have No Taste so *shrug*
<awygle> sorear: gonna be non-structural flying shards of glass when the big one hits
<sorear> do architects ever do like, market research
<sorear> or do they do and normies actually like this glass monstrosities :|
<q3k> fashion? try designing something that stands out too much and you'll be automatically weird and worst.
<q3k> *worse
<rqou> honestly I'd rather have a house that was just a giant warehouse with no internal walls except for a bedroom
<whitequark> rqou: that's called "studio"
<rqou> except those aren't giant
<rqou> they're usually really tiny
<q3k> rqou: that's what i'm looking for in munich
<q3k> rqou: but it's only really a thing for commercial/industrial settings
<pie_> inb4 youre not allowed to live in a warehouse because reasons
<q3k> rqou: and if you want to live in one that's also cleared for housing you gotta pay the hipster loft tax
<q3k> actually not sure how it's like in germany
<q3k> i guess if you get an anmeldung you're fine?
<q3k> i know most industrial/commercial leases i've seene elsewhere explicitely prohibit people from living there
<q3k> not sure about when you actually own the place
<rqou> in the us it tends to be prohibited by zoning
<rqou> although it's not too clear what the penalty is if you own it
<q3k> i think zoning is not nearly as anal in european countries as it is in the us
<q3k> at least observing how often you end up with old industrial buildings right next to apartment blocks
<rqou> there's also the Ghost Ship; that certainly didn't help
<q3k> yeah :/
<q3k> well, there's good and bad ways to go about this
<q3k> if you live by your own volition in a building that's not cleared for fire codes then it's your own stupidity, but you should maybe be allowed to do it?
<q3k> but if you bring others there then yeah :/
<whitequark> ghost ship?
<zkms> zoning and land-use restrictions in US are completely pathologically broken it's pretty fucked
<whitequark> oh
<sorear> the ghost ship seems to be a fairly clear example of "if you make something illegal, it will be provided by the market in the shoddiest way possible"
<whitequark> >In March 2017, emails from 2014 were reported to have described serious electrical problems in the building.
* whitequark twitches
<sorear> in CA even if something is legal you still can't build it because of CEQA and discretionary review fuckery, which I don't understand very well
* pie_ detaches whitequark from the exposed wire with a broom
<awygle> in king county apparently if something is legal you can build it but the permit might cost 8k for a 3k shed.
<awygle> I still might try to build someday though because there are no houses available
<sorear> supposedly king county is one of the better places for "is it actually possible to get approval to build something", although low bar
cr1901 has quit [Ping timeout: 268 seconds]
<sorear> and if you try to do anything about this you get called a "developer shill"
cr1901 has joined ##openfpga
s1dev has quit [Ping timeout: 264 seconds]
s1dev has joined ##openfpga
<rqou> wheee, <redacted> (not project chibi board) is now on Rev B-eco1
<rqou> i am good engineer
<prpplague> hehe
* whitequark opens steam
wpwrak has quit [Read error: Connection reset by peer]
<pie_> rqou, did kanna protect from bugs btw
wpwrak has joined ##openfpga
Miyu has quit [Ping timeout: 265 seconds]
<shapr> whitequark: I like the nickervision studios casual games
<shapr> I don't understand why people would play a house flipping game
<shapr> I should check my mail, see if my exapunks limited edition copy arrived.
<sorear> ~ nation of temporarily embarrassed millionaires ~
<shapr> Anyone here plays zachtronics games?
<shapr> now that I think about, PCB design in a game might be a bit silly if you're reverse engineering FPGAs
<shapr> I do wish exapunks didn't use javascript for in game networking code
<s1dev> The other day I was thinking about writing SAT formulas to proof that my train system in Factorio couldn't deadlock :P
<shapr> haha, did you?
<shapr> TLA+ ?
<s1dev> I ended up scrapping the world instead
<s1dev> I would have just written a CNF formula and dumped it into MapleSAT or something
<marcan> q3k: not doing CCC this year btw, it's kind of losing its appeal for me, I'm not giving a talk, and I want to have a no-travel new year's for once
<marcan> might go to camp next year though, never tried that
<marcan> I miss the CCC berlin days tbh
<whitequark> shapr: no, it's a house *renovation* game
<shapr> that's not my kind of game
<awygle> "fixed horrible electrical work - ten points"
<shapr> but neither is ice fishing, or driving a semi through europe
<shapr> awygle: look like that is in fact true
<awygle> "tried to hang sheet rock by yourself - minus four points"
<shapr> I wonder if you lose points if you don't connect the ground wire?
<whitequark> awygle: so, the reviews say that the customers do not in fact give a fuck if you did anything up to code
<whitequark> which is depressingly accurate to reality
<shapr> beat me to it
<awygle> lol this is why I don't play zachtronics games either. Too much reality
<shapr> ha
<whitequark> same
<shapr> I'd love to have a "tune your bitstream by hand" step for yosys
<whitequark> that sounds horrifying
<whitequark> like work, but with a game designer's idea of an UI
<shapr> is there not an FPGA demo scene?
<whitequark> which has an 1% chance to be a breakthrough in FPGA UIs and a 99% chance to be an abomination the world has never seen
<whitequark> not that I know of
<whitequark> the idea has certainly surfaced on this channel more than once
<zkms> FPGA toolchains are terrifying enough without manually tuning bitstreams :|
<awygle> frankly given the state of fpga UI odds might be better than that
<shapr> they're a lot like compilers, but with way more fun constraints
<awygle> but yeah if I'm going to push traces I'm either gonna get paid or make a cool thing or both
* whitequark . o ( homeowner association simulator )
<awygle> "convince people you have any kind of authority - five points. convince local government to _give_ you authority - fill in your own point value"
<whitequark> take the mechanics from something nice like Black Closet but replace the narrative with the absolute most mind numbing low stakes politcs you can come up with
<shapr> for some reason, I'm good at convincing people I have authority and know what I'm doing.
<marcan> re: zachtronics, I made a CTF level based on one of their games :P
<marcan> nc marcan.st 10847
<whitequark> marcan: that just drops the connection
<marcan> hm.
<marcan> what did I break
<pie_> (besides everything? xD)
<marcan> oh, fixed
<marcan> python versions
<rqou> pie_: the project chibi board powers up
<rqou> full bringup isn't done yet
* pie_ looks at this black closet thing
<rqou> kanna didn't protect from all bugs since there's a DFM issue on it
s1dev has quit [Ping timeout: 268 seconds]
<pie_> need a bigger kanna (tm)
* shapr opens a kanna worms
<pie_> lol shapr pls
<shapr> :-D
<pie_> you kanna change the laws of physics capn
* pie_ goes to reverse some polarities
<marcan> who is kanna besides a dragon and one of my server hostnames?
<pie_> in ##openfpga? just that
<shapr> no idea, sorry
<sorear> kanna refers to the dragon character, yes
<marcan> good.
<marcan> kanna got upgraded btw
<marcan> ⚠ root@kanna:~ -# lscpu | grep 'Model name|CPU\(s\):'
<marcan> CPU(s): 32
<marcan> Model name: Intel(R) Xeon(R) Silver 4108 CPU @ 1.80GHz
<marcan> shiny new server :>
<shapr> nice!
<marcan> (tohru also, same hardware)
<gruetzkopf> oh, nice. can you get at all the 48 PCIe lanes?
<marcan> it's a 1U Lenovo SR530, not sure?
<gruetzkopf> scrolling over the specsheet, for CPU0 you might get close, for CPU1 16 lanes
<gruetzkopf> pretty good for a 2s 1u box
<marcan> yeah, sounds like it
<marcan> good price too, we paid €11500 for both servers, with 3x1TB SATA + 2x240GB SSD + 2x10Gbe + 4x1Gbe, dual PSU and the BMC remote KVM license thing
<marcan> and 64G RAM
<gruetzkopf> not sure if one can get below that on 1u
<gruetzkopf> did you get 2 CPUs?
<marcan> yeah
<marcan> 2x 4108
<marcan> (hence 16 cores total, 32 threads)
digshadow has quit [Ping timeout: 240 seconds]
<gruetzkopf> i'm really impressed with the power consumption on modern boxes, one of my dell R530 (dual E5-26xx, 64G ram, 8 spinning disks) claims below 120W in average operation
<gruetzkopf> replaced 3 far-too-old X3650m1
balrog has joined ##openfpga
<marcan> gruetzkopf: tohru claims 122W input / 105W output right now
<marcan> well it was 117W before I started a minor CPU-consuming test :P
digshadow has joined ##openfpga
<gruetzkopf> yeah, the old stuff would have been close to 1kW
<gruetzkopf> few servers at that location, and even the ~1GWh/year for HVAC is considered a minor expense there
balrog has quit [Ping timeout: 248 seconds]
<marcan> gruetzkopf: actually elma, which inherited tohru's old hardware, is claiming 90 watts consumption right now
<marcan> but that's single CPU
<marcan> (quad core)
<marcan> HP DL380 G6 (1xE5520)
<pie_> you people and your fancy specs...
<marcan> pie_: I didn't pay for these :P
<gruetzkopf> my DL120Gen7 died, sadly
<pie_> :p
<marcan> the new boxes replaced G5s
<gruetzkopf> should be easy to fix
<marcan> which were seriously old
<gruetzkopf> (easy to fix == replacing some fan driver FETs
<marcan> now I can replace my package server with the same VM I use for the event servers
<marcan> until now it needed to be different because the G5s had different CPU features, too old :P
<marcan> speaking of event servers, this is where they live now: https://twitter.com/marcan42/status/1023612400887648257
<marcan> I didn't get a good photo of the rear at the end, but this is it with most of the cable management done: https://photos.app.goo.gl/GbcMhPSNuSWGaGqy5
<gruetzkopf> neat
<gruetzkopf> far better than the stack of boxes that piles up under the WHF assembly table every year
s1dev has joined ##openfpga
<marcan> :D
<gruetzkopf> (last year it was: IBM x3650 gen1 (equivalent to HP DL380gen5), Sun T1000, Sun T2000, a Powermac G5 dual 2.5, a sun FC SAN thing and 48-port switch, iirc
<marcan> cute
<gruetzkopf> (well, my part of the pile)
<marcan> lol
<marcan> looks legit
<gruetzkopf> top-to-bottom toshiba libretto, mac se 30 next to a sun mips cube, a 8088 portable and i believe some 32bit sparc thing on the bottom
<gruetzkopf> prince of persia does *not* run well on that amber-screened portable, but it does run
Bike has quit [Quit: Lost terminal]
<gruetzkopf> (proven after totally not soldering together a DB25M to DB9F nullmodem on site for INTERLNK)
<marcan> :D
digshadow has quit [Ping timeout: 268 seconds]
<shapr> My laptop is a quad core Xeon(R) CPU E3-1505M with 64GB of ram. Battery life is okay until something misbehaves. As soon as a thread gets busy, battery life just goes away.
<shapr> big rust compiles eat 1% battery per minute
<whitequark> wow, xeon
<whitequark> there's like 1 vendor on market which does xeon laptops
<jn__> Dell?
<whitequark> yeah I think
<shapr> thinkpads also have xeons
soylentyellow__ has quit [Ping timeout: 244 seconds]
digshadow has joined ##openfpga
balrog has joined ##openfpga
<kc8apf> Ugh. Why does Cargo _require_ a git repo for the package index?!?!?! It's the only protocol that insists on using POST to _read_ the repo.
<awygle> that is a bummer. it should support like, mercurial at least
<awygle> (complains awygle with no plans to do anything about this almost-certainly-fixable problem)
<kc8apf> or, not use a source control system for serving the index
<awygle> but source control is Good
<kc8apf> like every other package manager out there
<pie_> nixos doesnt serve its "index" over git but it might as well since half the time you end up wanting to edit it and make PRs upstream anyway xD
<pie_> well ok its not really the index
<pie_> it derives contnet addressable hashes for everything and if the result of your expression is in the binary cache then you get it
<pie_> *content
<kc8apf> when you upload a crate to crates.io, it updates a json file in https://github.com/rust-lang/crates.io-index
<kc8apf> Cargo fetches that repo to figure out what packages exist
<kc8apf> the crates themselves are served as tar.gz bundles
<awygle> I wonder if cork is esd safe
<kc8apf> yes but flammable
<kc8apf> see early Google rack designs
<rqou> bendy bendy
<awygle> huh really? cool
<awygle> I was thinking of surfacing a bench and/or maybe a case for something, not sticking a Xeon in it
<kc8apf> for a bench, I'd go with ESD rubber. Don't want solder to start a fire.
<awygle> that's true I suppose
<awygle> esd rubber is expensive and also ugly :-P
rohitksingh has joined ##openfpga
<pie_> wtf cork is esd sage?
<pie_> safe?
<rqou> probably because it's pretty low on the triboelectric series?
<daveshah> afaik it will always have a bit of moisture in it too?
<daveshah> that should also dissipate charges
Miyu has joined ##openfpga
Miyu has quit [Ping timeout: 248 seconds]
rohitksingh has quit [Ping timeout: 240 seconds]
digshadow has quit [Ping timeout: 240 seconds]
rohitksingh has joined ##openfpga
rohitksingh has quit [Quit: Leaving.]
rohitksingh has joined ##openfpga
pie_ has quit [Ping timeout: 244 seconds]
rohitksingh has quit [Quit: Leaving.]
s1dev has quit [Ping timeout: 240 seconds]
rohitksingh has joined ##openfpga
s1dev has joined ##openfpga
indy has quit [Ping timeout: 240 seconds]
GenTooMan has joined ##openfpga
kuldeep has quit [Ping timeout: 255 seconds]
kuldeep has joined ##openfpga
indy has joined ##openfpga
GenTooMan has quit [Quit: Leaving]
GenTooMan has joined ##openfpga
knielsen_ is now known as knielsen
<q3k> marcan: aw, that sucks
<q3k> marcan: but I get it
<q3k> marcan: come to camp though!
<marcan> yeah, I'm going to try to make that happen :)
<marcan> how do tickets for that work?
<q3k> i think they appearify half a year early
<q3k> not sure if they are going to be voucher based, as there hasn't been a CCCamp since the ticket shortage at the congress
<q3k> SHA2017 was voucher-free
<marcan> right
<marcan> is capacity a general issue like the congress?
<q3k> ticket availability wasn't an issue, but physical space abailability was (ie. get early to get good tent spots)
<marcan> ah
<q3k> i was at CCCamp once before, and that was 3 years ago, so who knows how it'll be like next year
<q3k> from what I've heard from people that attended the one 7 years ago it was very simliar
pie_ has joined ##openfpga
<q3k> oh, and - ESPR & dragonsector are probably going to be setting up a larger hacking/party tent at the next camp
<q3k> like two big wood-floored festival tent things, tables, a PA system, whatnot
<q3k> you're all welcome :)
<q3k> cc G33KatWork
<marcan> lol, just realized something... the congress overlaps with winter comiket (which is why I've never gone to that... until this year)... and the camp seems to also overlap with summer comiket.
<marcan> not that I wouldn't prefer going to winter comiket over summer comiket, but lol, talk about bad timing
pie__ has joined ##openfpga
pie_ has quit [Read error: Connection reset by peer]
s1dev has quit [Ping timeout: 264 seconds]
pie___ has joined ##openfpga
pie__ has quit [Ping timeout: 268 seconds]
X-Scale has quit [Ping timeout: 240 seconds]
X-Scale has joined ##openfpga
Bike has joined ##openfpga
<G33KatWork> uh yeah, we'll be camping like the kings :>
Miyu has joined ##openfpga
<travis-ci> whitequark/Glasgow#19 (master - 66aa942 : whitequark): The build passed.
<felix_> the cccamp is awesome. (and i don't even like camping in general) you should realy bring a space blanket to put on your tent against the sun and earplugs though
<q3k> space blanked _and_ tons of cyber^Wduct tape
<q3k> because if you don't affix it to your tent properly, it will a) fly away b) before it does that, wake you up in the middle of the night because of the noise it makes when the wind picks up
<rvense> wake your neighbours up too
cpresser_ has joined ##openfpga
<felix_> yep. and don't use the cheap tape; in the sun the cheap stuff melts and gets nasty
cpresser has quit [Ping timeout: 256 seconds]
brizzo has quit [Ping timeout: 269 seconds]
brizzo has joined ##openfpga
clifford has quit [Read error: Connection reset by peer]
cpresser_ has quit [Quit: Reconnecting]
cpresser has joined ##openfpga
Ekho has quit [Quit: An alternate universe was just created where I didn't leave. But here, I left you. I'm sorry.]
clifford has joined ##openfpga
tinyfpga has quit [Ping timeout: 240 seconds]
tinyfpga has joined ##openfpga
Ekho has joined ##openfpga
ZipCPU has quit [Ping timeout: 240 seconds]
<daveshah> Shame they stopped at 1MHz, we may never know whether or not building microwave circuits on a cork substrate is the way forward
<kc8apf> seems cork can be anti-static or insulative depending on natural, composite, or agglomerate, temperature, and humidity as % of weight
<kc8apf> so, uh, keep a spray bottle next to your cork workbench mat
<daveshah> Might reduce the fire risk. But also create loads of debugging problems if high impedance exposed pins touch it
<awygle> humidity doesn't surprise me, wood is the same
ZipCPU has joined ##openfpga
m_t has joined ##openfpga
rohitksingh has quit [Quit: Leaving.]
digshadow has joined ##openfpga
fseidel has quit [Remote host closed the connection]
X-Scale has quit [Ping timeout: 240 seconds]
pie___ has quit [Read error: Connection reset by peer]
pie_ has joined ##openfpga
Miyu has quit [Ping timeout: 264 seconds]
pie__ has joined ##openfpga
pie_ has quit [Read error: Connection reset by peer]
pie__ has quit [Remote host closed the connection]