egg|nomz|egg changed the topic of #kspacademia to: https://gist.github.com/pdn4kd/164b9b85435d87afbec0c3a7e69d3e6d | Dogs are cats. Spiders are cat interferometers. | Космизм сегодня! | Document well, for tomorrow you may get mauled by a ネコバス. | <UmbralRaptor> egg|nomz|egg: generally if your eyes are dewing over, that's not the weather. | <ferram4> I shall beat my problems to death with an engineer.
awang has quit [Ping timeout: 200 seconds]
pizzaoverhead_ has joined #kspacademia
pizzaoverhead has quit [Ping timeout: 200 seconds]
<kmath>
<jonnysun> i fixed the extremely long leg in the new ready player one poster https://t.co/28VuR1ZzoV
awang has joined #kspacademia
e_14159 has quit [Ping timeout: 200 seconds]
Alaura has quit [Ping timeout: 186 seconds]
e_14159 has joined #kspacademia
* UmbralRaptor
? things
<UmbralRaptor>
So, I noticed that I lost a debit/atm card. Got it cancelled, new one sent out. Annoying given the weekend, but eh.
<UmbralRaptor>
I found it sitting in my car today.
egg|zzz|egg has quit [Read error: Connection reset by peer]
<Ellied>
that's the second time I've heard someone tell that story in, er, I think about a week
<UmbralRaptor>
o_O
<Iskierka>
... so the electric tingle off a battery is *way* stronger on a bit of your lip where you've bitten off dry skin
<Iskierka>
like weirdly strong for what's a dead 3V battery
<Iskierka>
or possibly anywhere on lips
<Ellied>
that's why I never test 9V batteries that way, my lips are always cracked
<Iskierka>
if this is what dead 3V does I'd probably jump from 9V
<Iskierka>
though I've only ever done those on tongue which is oddly not as strong
<Iskierka>
tongue just lets you taste the electrons
<bofh>
Iskierka: so it all depends on conductivity, and in such a spot on your lips I think the current path from lip surface to whatever branch of the maxillary branch of the trigeminal nerve innervates your lips.
<Ellied>
[no notifications for two days] "Micah Scott retweeted your Tweet:" [30 notifications in 5 minutes]
<Iskierka>
I mean yes that makes sense but it was still a bit of a "woah" moment when I got stabbed by a button cell
<Ellied>
when I was little I licked the jacket of a 12VAC barrel plug power supply. Turns out, on that particular supply, the jacket was *not* tied to ground.
<bofh>
A) ouch, B) Casio?
<Ellied>
don't recall. I think it came with a lego set or something, but I don't know what set because I got the adapter on its own secondhand from whoever originally bought it
<Iskierka>
... I would be very confused at lego having AC supply
<Ellied>
yeah, me too
<Iskierka>
I mean, power users: motors, simple low-power ones, so DC? And mindstorms, so electronics, so DC
<Ellied>
yeah, I have no idea
<Ellied>
I almost wonder if it was something really janky like putting the rectifier in the lego end instead of the wall plug end
<Iskierka>
I don't understand why you'd do that but ... maybe, I guess?
<Ellied>
this was the same power supply that I went on to vaporize my first pencil graphites with, so it definitely supplied a handsome amount of current too
<Iskierka>
fun
<UmbralRaptor>
Eek
<Ellied>
doing that is really fun if you have a ton of current available to draw. It ends up being kind of like a slow-motion flashbulb, a reddish glow gradually increasing until it's briefly bright enough to light an entire room, then quickly dwindling away after it gets too thin to support its own weight and snaps. If you put a big inductor in the loop, you'll get a nice angry arc flash when that
<Ellied>
happens.
<SnoopJeDi>
bofh, our departmental Christmas party was tonight and I ended up recounting the story of you inadvertently discovering 1.5 GB of quality orchestral music on an FTP server for probe (?) data
<bofh>
SOHO probe data raws, yes.
<whitequark>
Iskierka: lego has a series with trains
<whitequark>
or rather it had until like 2003
<whitequark>
I'm still sad about it :(
<SnoopJeDi>
(context: one of our professors plays in a jazz band and performs at this shindig)
<whitequark>
that used an AC 9V supply
<bofh>
hey it was *very* good choices of Mozart performances, & the Haydn selection wasn't bad at all either.
<SnoopJeDi>
mostly I stumbled onto the subject because I think it's kind of funny to try and explain my internet friendships in meatspace
<bofh>
whitequark: damn, that sounded awesome, pity they don't make it any longer
<SnoopJeDi>
"uh so I know this person and we both have *some feelings* about space probes..."
<bofh>
accurate
<SnoopJeDi>
the band started playing "The Girl from Ipanema" which drew me back into the foyer because it's familiar :D
<whitequark>
bofh: yeah it was p good. me and my father even automated switches
<SnoopJeDi>
Using part of the kit, or with some homebrew fixins?
<Iskierka>
they still do trains, or possibly restarted, but I'd've thought they're DC and surprised that previous ones weren't?
<whitequark>
homebrew fixins
<whitequark>
Iskierka: restarted
<Iskierka>
particularly as it post-dates scale model railways that are DC
<whitequark>
the old ones were very NIH-y
<whitequark>
it rectified the 9V input right away anyway
<Iskierka>
NIH?
icefire has quit [Read error: -0x1: UNKNOWN ERROR CODE (0001)]
<e_14159>
So, I was wondering how other fields are doing on the whole reproducibility and experimental rigorousness?
APlayer has joined #kspacademia
APlayer has quit [Ping timeout: 207 seconds]
<UmbralRaptor>
IIRC, big problems in psychology, more modest ones in medicine.
<UmbralRaptor>
No idea with CS?
<e_14159>
UmbralRaptor: With machine learning at least, it often feels as if the answer to "how reproducible are your experiments" is "HAHAHA".
<e_14159>
Most papers don't publish code. Those that do sometimes mention one set of hyperparameters in the paper, another set in the code, the authors (if they answer) apparently used yet a third set for the results in the paper and if you're using that you *still* don't get their results.
<e_14159>
Probably because they only report the best out of dozens of experiments.
<BPlayer>
There are a good few hundred km to the Alps (If that's how you call them in English)
<BPlayer>
Also, this much snow is definitely not normal, especially so early into the winter. We have a ski lift here (Imagine a big field that goes down a slope on the local town hill, and there is this T-bar sort of lift there), but the lift is only open a day or two per year, on average
BPlayer is now known as APlayer
icefire has quit [Read error: -0x1: UNKNOWN ERROR CODE (0001)]
<soundnfury>
heh. Well, any snow at all in December is not normal for here
<soundnfury>
Cambridge usually gets its snow in February (and sometimes at Easter)
<e_14159>
APlayer: Ten centimetres? I envy you.
<APlayer>
I guess, I was not outside yet. But I will go and clean it in 10 min or so
<APlayer>
Looks like 10 cm judging by the footsteps
<e_14159>
We have one, maybe two here. Which means it's going to be mush by tomorrow.
<APlayer>
Oh, that's true here anytime
<APlayer>
Uless it's -5 deg or so, you'll get mush by tomorrow
<e_14159>
+0.6°C
BPlayer has joined #kspacademia
<BPlayer>
Grr this router sucks
APlayer has quit [Ping timeout: 207 seconds]
<BPlayer>
Okay, I've got to go now. See you guys!
<egg>
e_14159: yeh it's raining over the snow here too :-\
<e_14159>
egg: Y'know, I remember when we still had *real* winters.
<egg>
might snow again later in the week?
<egg>
<soundnfury> Cambridge usually gets its snow in February (and sometimes at Easter)
<soundnfury>
e_14159: did you have to walk uphill in the snow both ways to your septic tank?
BPlayer has quit [Ping timeout: 207 seconds]
<e_14159>
soundnfury: Nah, to my universal constant factory.
<e_14159>
egg: Well, at least we have less rain.
<UmbralRaptor>
!wpn egg
* Qboid
gives egg a bear
* egg
pets the bear
<egg>
UmbralRaptor: I wonder whether I should try that game that Fiora and Atlas have been tweeting about
<UmbralRaptor>
Opus Magnum?
<egg>
yes
<UmbralRaptor>
Yes. ?
<egg>
!tell awang can you test https://github.com/eggrobin/Principia/tree/use-the-current-time with realism overhaul to see if it fixes your krash crash, and whether it introduces weirdness in RSS launches (try using procedural SRBs to check that #1421 isn't brought back by this)
<Qboid>
egg: I'll redirect this as soon as they are around.
<Qboid>
[#1421] title: Child transforms move on load flight state with RSS | This (a) may be a duplicate of the Proc SRB Bell issue, if that's been reported, and (b) may be an issue, albeit smaller, with stock size systems (why would I touch stock, so I don't know).... | https://github.com/mockingbirdnest/principia/issues/1421
<egg>
UmbralRaptor: hmm
<awang>
egg: Sure, I'll get that loaded after my next KSP crash
<egg>
:D
<awang>
Which might be coming up soon, since I've been spending a while in the SPH
<egg>
:D :D
<awang>
idk if I can test proc SRBs though
<awang>
Haven't unlocked them, and haven't used them at all, so I'm not familiar with the broken behavior
<egg>
awang: well if the bell is off to the side of the booster and spins things up on ignition it's broken
<egg>
awang: also sandbox is your friend for those tests
<egg>
awang: note that there are C# side changes so you need to build the plugin and the adapter (but make release does that iirc)
<awang>
I'll give that a shot then
<awang>
Is it just "place proc SRB on vessel, launch, laugh at brokenness"?
<awang>
Yeah, `make release` builds the C# part
<egg>
awang: yeah
<egg>
!wpn whitequark
* Qboid
gives whitequark an eta group
<Ellied>
I remember when I put a procedural SRB at KSC that was 10m across, about 100m tall, and had the thrust set to the lowest available. It would have had a burn time of about a month in-game, discounting timewarp and times when you're out of physics range of KSC (during which it stops)
<egg>
Ellied: what did you use it for? lighthouse?
<Ellied>
basically, yes
<Ellied>
I had the fiery end pointing upwards, so it looked about like a very smoky lighthouse.
<Ellied>
maybe a lighthouse with the wrong parts on fire.
<kmath>
<plaidfinch> some.............. ..............BODY ONCE TOLD ME THE PLUMS WERE IN THE ICEBOX THE ONE THAT YOU KEEP IN THE SHED… https://t.co/SASgbOA8UE
<SnoopJeDi>
farmstead rofl
awang has joined #kspacademia
icefire has joined #kspacademia
<awang>
egg: I get a "no viable overloaded '+=' when trying to compile
<kmath>
<bofh453> Nice example of a JFET⇔Triode direct correspondence! https://t.co/asBVvKJ7nd
<egg>
UmbralRaptor: ?
<UmbralRaptor>
egg: reviewing problems for the Classical final tomorrow.
<Ellied>
Also, this once again reminds me why the doping scandal^W^Wtendency for N-channel devices to be more common than P-channel is a thing.
<Ellied>
'cause it turns out you can't get holes to boil off a heated electrode and fly through a vacuum, or something.
<egg>
Ellied: what if you use positrons,
<Ellied>
I think finding a material to make wires out of that allows positrons and electrons to flow opposite directions with the same freedom, without annihilating, would be very tricky indeed
<bofh>
let me just find my spare batch of antitungsten...
<Ellied>
although likely very interesting if it were to exist
<Ellied>
also if you made a junction between that material and a normal material you could effectively have an x-ray-emitting diode, although only at one wavelength.
<egg>
bofh: well the nice thing with that is that if you misplace it it brightly reminds you of that
<egg>
!wpn Ellied
* Qboid
gives Ellied an irritating bus
<egg>
!wpn bofh
* Qboid
gives bofh a thermionic imaginary Brand New Photoneutronic Engine
<egg>
!wpn -add:wpn positron
<Qboid>
egg: Weapon added!
<egg>
!wpn -add:wpn electron
<Qboid>
egg: Weapon already added!
<egg>
!wpn -add:wpn gluon
<Qboid>
egg: Weapon already added!
<egg>
!wpn -add:wpn muon
<Qboid>
egg: Weapon added!
<egg>
!wpn -add:wpn pion
<Qboid>
egg: Weapon added!
<egg>
UmbralRaptor: welp I installed opus magnum
<UmbralRaptor>
:D
<egg>
!u ?
<Qboid>
U+1F701 ALCHEMICAL SYMBOL FOR AIR (?)
<egg>
wait why is that one astral
<egg>
!u ????
<Qboid>
U+1F701 ALCHEMICAL SYMBOL FOR AIR (?)
<Qboid>
U+1F703 ALCHEMICAL SYMBOL FOR EARTH (?)
<Qboid>
U+1F702 ALCHEMICAL SYMBOL FOR FIRE (?)
<Qboid>
U+1F704 ALCHEMICAL SYMBOL FOR WATER (?)
<UmbralRaptor>
bofh, Ellied: literally what dilithium is supposed to do,
<egg>
!u ♄
<Qboid>
U+2644 SATURN (♄)
<egg>
!u ?
<Qboid>
U+1F714 ALCHEMICAL SYMBOL FOR SALT (?)
<UmbralRaptor>
egg: ah, the triangles were all part of the same batch?
<Qboid>
U+1F706 ALCHEMICAL SYMBOL FOR AQUA REGIA (?)
<egg>
UmbralRaptor: something something nomal
* UmbralRaptor
feels anxious.
<Ellied>
wow, JFET amplifiers have impressive bandwidth
APlayer has joined #kspacademia
<Ellied>
2N5485 has no apparent amplitude drop between 0 and 3MHz.
<Ellied>
and it, uh, faithfully reproduces all the noise and distortion this shitty function generator puts out.
<bofh>
that's partly why we use them on op-amp inputs (the other reason is they *basically* have infinite input impedance at the gate, so are perfectly suited for an op-amp inut stage).
<Ellied>
aye, I at least knew about that
<Ellied>
but their frequency characteristics are still a bit foggy for me because datasheets never seem to want to make a single mention of them and I don't quite conceptually understand what the primary limiting factor is
<Ellied>
okay, this is weird. Now I'm getting an output that looks *less* distorted than the input at first glance. Yes, I did double-check that I've got the channels of the scope the right way around.
<APlayer>
I have no idea what you're talking about, but did some noise cancel out? :P
<bofh>
I feel like there's some capacitative coupling mismatch happening on the input, it really looks like the triangle wave is sagging.
<bofh>
Like this makes *sense* to me intuitively but I'm not sure *why* it happens.
<Ellied>
yeah, probably, I didn't make any attempt to match impedances
<Ellied>
so is that distortion between the input and the scope, then?
<bofh>
try adding a decoupling cap between "input" & scope input (1uF, 10uF, 100uF, not sure which is optimal) & see if that changes anything?
<Ellied>
all the coupling caps I've got now are 10nF, could try larger ones
<Ellied>
ohhhh I'm silly. I had the function generator plugged into the amplifier's output; I was seeing the result of feeding it through in reverse on the input. >_<
<Ellied>
no wonder I was seeing such weirdly good frequency response >_> <_< >_>
<bofh>
*glare*
<Ellied>
that's what I get for putting the input and output on directly adjacent breadboard rows.
<Ellied>
probably should've tipped me off that the "output" wasn't inverted.
<bofh>
it's like a common-gate amplifier topology, but much shittier
<bofh>
heh, same
<SnoopJeDi>
was that glare because of being robbed of a good puzzle, bofh?
<Ellied>
well, I've still got a puzzle, it's "where the fuck is all this noise coming from"
<Ellied>
it's not the function generator after all, it's there even when it's off
<Ellied>
must be the AC power noise goblins again, same ones that keep getting into my apartment and/or phone chargers and making my touchscreen go nuts
<egg>
argh least common multiples
<APlayer>
Quick question: So I have an initial vector and a perpendicular "target" vector. Is there a way to calculate a vector so and so many degrees off from the initial vector towards the target vector? I'd do it with the cross product to get an axis and the usual rotation formula, but is there a better way?
<Ellied>
rgh, there are a bunch of promising-looking low-cost op amps on Mouser with great advertised bandwidth, but... they're all from Maxim. >:I
<Ellied>
given that comparably-priced devices from TI consistently have about half the bandwidth, there must be something wrong with them.
<bofh>
What's wrong with Maxim?
<Ellied>
whitequark warned me to avoid them because their products tend to be half-assed ASICs that their customers designed and then they sold without adjusting the design, IIRC, so it's easy to get poorly-designed parts from them.
<Ellied>
I don't have any firsthand experience of this, but they did make that one part that was a watchdog timer and/or three bytes of RAM and/or whatever the fuck else