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.
<egg>
UmbralRaptor: on a more serious note, crows?
<Iskierka>
... so it was a few days ago but I'm very confused at the guy telling Ellied that SRAM loses data immediately on losing power. SRAM holds data CONSIDERABLY longer than DRAM
<egg>
UmbralRaptor: in my first semester at ETHZ I took an astronomy course taught by Schmid and Schmutz
<UmbralRaptor>
egg: maybe. Somewhat difficult to go back to galaxies. >_>;;
<Iskierka>
Yeah. I don't recall exact design details that could explain why it persists to much, but my dad worked with SRAM systems some while ago, and via assuming what values would be initialised to (zero) he found that there would be very large amounts of data remaining after more than a minute of no power
<UmbralRaptor>
Hm.
<Iskierka>
meanwhile DRAM is refreshing like 15 times per second to ensure integrity
<Ellied>
EPICS, aaaaaa
<UmbralRaptor>
EPICS?
<Qboid>
UmbralRaptor: [EPICS] => Experimental Physics and Industrial Control System
<Ellied>
we could have named categories, like "astronomy" "aerospace" "math" etc, so if there are multiple you can say "astronomy EPICS?" and get that, or "physics EPICS?" and get the one that's there now.
<egg>
wait why is he in the physics dept and not the astronomy one Ꙩ_ꙩ
<Ellied>
if the acronym exists but doesn't have any explanations for that category, it'll say what categories do exist.
<egg>
Ellied: and then you notice the guy is in a physics department doing astronomy :D
<Ellied>
my friends today were surprised when I told them that some unis actually have an astronomy department instead of just having that folded into physics
<egg>
very german english page: "Mr Prof. Dr. Günther Dissertori"
<egg>
(he taught my first semester physics course)
<UmbralRaptor>
Ellied: my advisor does lots with detectors, so I could ask him for suggestions?
<egg>
yeah so astrophysics share an institute with particle physics within d-phys apparently
egg has quit [Read error: Connection reset by peer]
APlayer has quit [Ping timeout: 383 seconds]
<BPlayer>
Okay, so I just found this random function while trying to get a rather specific curve... And what the heck is this? f(x) = x^(1/x)
egg|laptop|egg has quit [Ping timeout: 180 seconds]
<BPlayer>
Seems like it has a global maximum at x = e?
BPlayer is now known as APlayer
<APlayer>
Does this kind of function even have a name?
egg has joined #kspacademia
<Iskierka>
this is a fun graph
<Iskierka>
also twitter is being dumb. I've been unfollowed from at least two accounts and haven't been using an interface where unfollow is an easy misclick
Iskierka has left #kspacademia [Leaving]
Iskierka has joined #kspacademia
<Iskierka>
also hexchat is still dumb
<egg>
yeah
hattivat has joined #kspacademia
hattivat_ has joined #kspacademia
hattivat_ has quit [Read error: -0x7880: SSL - The peer notified us that the connection is going to be closed]
<egg>
!wpn hattivat
* Qboid
gives hattivat a spectrophotometer/hydrochlorofluorocarbon hybrid
<Qboid>
[#1640] title: Detect collisions in the force computation and thread errors through the integrators and the ephemeris | This is a step towards solving #1628. | https://github.com/mockingbirdnest/Principia/issues/1640
<Qboid>
5d 0h 0m 0s left to event #17: et al./Falcon 9 v1.2 [at 2017-12-08 18:20:00]. Say '!kountdown 17' for details
ferram4 has joined #kspacademia
APlayer has quit [Ping timeout: 207 seconds]
<egg>
!wpn bofh
* Qboid
gives bofh an extensive stellarator/conflict hybrid
<kmath>
<lindsey> @palvaro @bridgetkromhout I should do one of those "latency numbers you should know!" things, except instead of "L1… https://t.co/nQc4xKL0eG
BPlayer has joined #kspacademia
BPlayer is now known as APlayer
<soundnfury>
egg: L1 latency is 2 * (1.5Gm / c) = 10.006 seconds RTT.
<UmbralRaptor>
Uh
<soundnfury>
I, uh, might be talking about a different L1 ;)
<soundnfury>
(also I gave an unjustified number of sig figs, but hey)
<kmath>
<TetZoo> Breaking maniraptoran news: the Haarlem Archaeopteryx is not an Archaeopteryx, but a new taxon of anchiornithid nam… https://t.co/GqDJSYb52g
<UmbralRaptor>
Iskierka: exploiting this is a key feature of VLBI (Very Long Barkline Interferometry)
<kmath>
<goodfellow_ian> One of my favorite samples from the Progressive GANs paper is this one from the "cat" category. Apparently some of… https://t.co/fRSS0F7Jsg
<egg>
argh, whenever I see talk of significant figures I wonder whether it's about @sigfig
<egg>
(tbh it's probably best to say sig. dec. or sig. bits anyway)
<kmath>
<stephentyrone> @csferrie @aemylt Hot take: Anyone who uses `+` for inclusive OR or string concatenation is a monster.
<UmbralRaptor>
Not sure I agree, but using + for XOR or NAND *is* monsterous.
<soundnfury>
UmbralRaptor: nah, + for XOR is exactly right in F₂. Also, × or · for AND.
<APlayer>
I'd use * for AND and + for inclusive OR or string concatenation?
* UmbralRaptor
generally agrees with APlayer here
<egg>
UmbralRaptor: wat? + *is* XOR
<APlayer>
This may be the first time someone just agreed with me on this channel :P
<APlayer>
egg: Uh, what about inclusive OR?
* egg
slaps UmbralRaptor with an ichtyosaur
<APlayer>
Also, is there some curve that starts steeply at 0 and gets flatter as x increases? I am looking at roots and logarithms here, but neither curve fit my need, and I have so far not managed to modify them properly. Any alternatives?
<APlayer>
In particular, I need something flatter than square roots for big x, but steeper than logarithms for small x
<APlayer>
And it would be nice if I did not have to use log(x + 1) to prevent it from getting negative for positive x, but that's doable
egg is now known as egg|tea|egg
ferram4 has quit [Read error: Connection reset by peer]
ferram4 has joined #kspacademia
<UmbralRaptor>
ow
<hattivat>
APlayer: you mean someting like 2*ln(x)?
<APlayer>
I tried using a factor k for k * ln(x), but couldn't find a good value, as the flat part would grow too large before I was satisfied with the steep part
<hattivat>
you seem to have really narrow constraints
<APlayer>
I do. Landing a booster in KSP Falcon 9 style, and trying to pinpoint the landing using lift guidance. I currently try to apply a single all-around formula for the whole descent, but the conditions vary wildly between free-fall/high distance to the landing site, max Q/medium distance up to terminal velocity/low distance
<APlayer>
The curve I need it the amount of correction I need to apply given the impact error distance, which should output around 30 for values of 2000 and more (using a Min() function here), and around 5 for values smaller then 300. Around 500 it should go rather steeply up towards 20 - 30.
<APlayer>
I currently use 3 * sqrt(x), but that causes wild oscillations for smaller inputs, as it grows really quickly
<soundnfury>
APlayer: maybe try fitting a ratio of polynomials? even something as simple as y=\frac{x+a}{x+b} is surprisingly versatile
<APlayer>
Woah, that looks nice!
<APlayer>
Okay, this is plain and simple amazing! Thanks a lot, soundnfury!
<APlayer>
I think I found a curve that suits my needs!
<APlayer>
Doesn't even need the Min() stuff now! :D
Alaura has joined #kspacademia
<APlayer>
Okay, I've got to go now. See you tomorrow! Thanks again for that formula!
egg|tea|egg is now known as egg
<egg>
!wpn UmbralRaptor
* Qboid
gives UmbralRaptor an infinite Maclaurin cathedral
<egg>
!wpn -add:adj Laurent
<Qboid>
egg: Adjective added!
APlayer has quit [Ping timeout: 383 seconds]
<egg>
!wpn bofh
* Qboid
gives bofh a multipole bolt with a snek attachment
* UmbralRaptor
? things, and tries to set up the integrals for a cone.
* egg
pokes UmbralRaptor with kozai
<soundnfury>
!wpn egg
* Qboid
gives egg a Ляпунов Van Allen wrench
egg|phone|egg has joined #kspacademia
<egg|phone|egg>
Moo
<soundnfury>
OoM
<egg|phone|egg>
UmbralRaptor: Why a cone
* UmbralRaptor
hands soundnfury a mana potion.
<UmbralRaptor>
egg|phone|egg: Homework problem involves the moments of inertia of a cone.
<egg|phone|egg>
The only thing there is to life is triangles?
<UmbralRaptor>
The triangles aren't so bad. The circles, though.
<egg|phone|egg>
Ah yes they're poor triangles
<UmbralRaptor>
Stabbing someone with a circle is hard.
<soundnfury>
UmbralRaptor: ?
<UmbralRaptor>
soundnfury: random memory from when I played MMOs: OoM == Out of Mana
<soundnfury>
I see.
<soundnfury>
Out of Memory.
<soundnfury>
as in the fabled "oom-killer"
<UmbralRaptor>
Why you sometimes open a web page and your music player and IRC client die.
* UmbralRaptor
?
<soundnfury>
well, that's more because browsers are fsckin' hogs…
<egg>
UmbralRaptor: well then shrug, x1 = r cos φ, x2 = r sin φ, dV = r dφ dr dx3, sum for x3 ranging from 0 to h, for r ranging from 0 to R x3/h, for θ ranging from 0 to 2π of what you said
<egg>
um φ=θ because I can't decide what to call the angle
<egg>
ϡ
<UmbralRaptor>
hm
<egg>
(like I would have started by saying the principal axes are obvious, find the centre of mass, compute the principal moments, and then do the stuff about the apex by parallel axis, but shrug)
<egg>
I guess if you like integrating more than geometrying that works :-p
<UmbralRaptor>
I got inconsistent results with geometrying the triangle. =\
<egg>
UmbralRaptor: uh of course your formula has an r but that's not the same as my r
<UmbralRaptor>
Yeah
<egg>
UmbralRaptor: your r is r^2+x3^2 (quite obviously, also immediate calculation with the sum of the squared sting of death and costing of death being 1)