« first day (2588 days earlier)      last day (374 days later) » 

12:01 AM
*lecturers
:-)
 
Hi, Can someone provide me some self reading material for Condensed matter theory? I've done QFT previously for which I could happily read Peskin supplemented with David Tong. Can you please suggest some references along those lines? Thanks
 
@skullpatrol The second one was in my MSc and covered considerably less than my first and (I felt) didn't do it in any particularly great way, so distinctly average. The third was pretty decent - I liked the way he did things and was essentially a more mathematically detailed version of the first :)
 
12:23 AM
@Mithrandir24601 Who was it?
Let's see. You're a Brit, so...Hughston or Tod?
Felice/de Clarke?
 
12:39 AM
Sounds like the first guy was a real gem.
 
1:51 AM
@heather^
 
Anonymous
2:29 AM
@SaiKrishnaDeep Shankar's new book has a few chapters on CMT. Heard it's good but haven't used it.
 
@SaiKrishnaDeep i can
still there?
 
Anonymous
@ooolb Just type it. He can read your message later :) I'm interested in this too
 
well i need to know what level he is at
or what he wants to learn
 
Anonymous
He said he knows QFT...
 
Anonymous
@ooolb Ya, that is true
 
Anonymous
2:41 AM
CMT is too huge a field
 
Anonymous
*broad
 
@Blue not enough information
 
Anonymous
@ooolb Agreed.
 
Blue acting like he's a physician
 
Anonymous
@0celo7 "physician" ?!
 
2:48 AM
someone who does physics
 
Anonymous
...
 
@Blue yes?
 
Anonymous
Yes?
 
not sure that word means what you think it means
 
I agree @enumaris
@Blue mb look it up
 
2:56 AM
I turned my neural net into a 37million parameter beast...takes forever to train it now
 
Hi physics
well, see you
 
3:36 AM
see you later space cowboy
 
@ooolb are you at perimeter?
 
right now?
no
 
If so, I need you to find Carl Bender and interrogate him at gunpoint about matched asymptotics
 
oh i know that dude
he wrote that book
on approximations
 
ok go find him
 
3:40 AM
what's in it for me?
 
my love and admiration
 
i already have that.
 
you wish
 
also is he at perimeter right now? i'm pretty sure he was just visiting back in 2013.
 
oh, was he?
ah
he's at St. Louis
 
3:42 AM
ooo the h bar is quite empty right now
we're basically alone
kinky
 
I'm leaving now
 
😢
 
oooooooooooooooooooooo
I have two perturbation terms
there's two boundary layers
do your homework, kids
 
if you do your homework at school is it schoolwork or homework?
 
homework
@ooolb do you want an exercise?
 
3:47 AM
send on discord.
save me the public embarrassment thanks
 
Last night dream has a lot of crazy scifi stuff:
1. A strange warp drive geometry and JohnRennie's comment on it
 
@ooolb it's there
 
2. A weird particle or state that is made of a superposition of a torus region with clockwise momentum and anticlockwise momentum, resulting in one that has no momentum along the major circumference of the torus but still nonzero momentum in directions that are not pointing along the torus
 
@ooolb sorry now it's there
 
k
i never did LSD before but if I do I'll make sure to do it with @Secret
 
3:59 AM
I take no drugs. unless my kidney medicine cyclosoprine A has strange side effects of creating strange dreams
 
@ooolb so did you see it?
the computations involved are formidable
 
take drugs all the time, cause I like to lose my mind, every day, edge close, to the day you overdose.
 
As will be elaborated with some annotated pictures later, the particle like state mentioned in 2 seemed to behave like a gravitational analogue of anapoles
where imagine a region of spacetime that got simutaneously frame dragged in the clockwise and anticlockwise direction
Since gravity does not obey superposition, I am not even sure if the result of having two opposite directions of frame dragging will cancel each other out
The dream also mentioned of some GR simulator that can simulate any type of spacetime
 
@Secret I need this
Do you know how awesome that would be
one could test many conjectures
 
is it not on steam?
 
4:07 AM
Same thought as you, however I think the major challenge of such simulator is the computational cost. GR calculations with its highly nonlinear nature, might be more costy than a computation of a protein.

However I can see some ways approaching it. Recall how Slereah was building some kind of spaceitme database, that could be the first step. Next, one might be looking for machine learning techniques to help on the simulation by using the classifications of spacetimes as machines are known to perform very well on sign problems as a recent paper has shown
Since GR equations are ultimately a system of 10 nonlinear PDEs, it might be possible the solution strategy has some relation with the class of spacetime that is under consideration, thus that might help heavily reduce the parameters need to consider to simulate them
 
I once asked how hard it would be to do a GR sim of something with just discrete symmetry (cubical) and two people said it would be a very hard PhD thesis
 
actually 6 nonlinear PDEs right
 
we can sleep with secret tonight and figure out how the sim works in his dream.
free Phds
 
@enumaris don't forget the constraints
and the gauge conditions
and matter equations if you have them
 
that's what I mean tho, the 10 PDE's can be reduced to 6 due to "gauge symmetries" right
unless I remembered wrong
 
4:15 AM
@enumaris Yeah but you still have to solve the gauge constraints
It's a giant mess because those also depend on the solution
you need background metrics and the whole thing is pretty bad
But I don't know anything about the numerics. I should take a class in grad school
 
I just mean this: The EFE is a tensor equation relating a set of symmetric 4 × 4 tensors. Each tensor has 10 independent components. The four Bianchi identities reduce the number of independent equations from 10 to 6, leaving the metric with four gauge fixing degrees of freedom, which correspond to the freedom to choose a coordinate system.
there aren't actually 10 independent equations
 
@enumaris But choosing the right coordinates is a nontrivial task
You want wave gauge
So to pick the coordinates you have to solve more PDEs
 
@ooolb Even if that is really possible (I always can talk about things in a non joking perspective), the issue is that 1) Unlike other people, I cannot incubate my dreams for a certain topic due to Mechanism 1 (consicous desires have reduced probability of appearing in dreams), and 2) For 6 years, my dream still yet to show any sign of revisiting the exact same idea, and there are no known instance of either sequel dreams nor recurrence dreams
 
mmm
 
it's not impossible unless it's impossible.
where are you should I bring my pillow
 
4:20 AM
impossible is nothing
-adiddas
adidas*
I can't spell -.-
 
@enumaris we used to have a numerical GR guy who complained about the constraints being very hard to maintain throughout the solution
you have to trim in a clever manner to keep the constraints fulfilled without killing the solution
 
@0celo7 I felt this aspect can be helped by machine learning. You can train a neural network with some PDEs of a known class with some known constraints, and let it figure out the best solution for some new PDE after say training it on 1000 different PDEs
Actually that makes me wonder, are the space of all coordinate choices more than all possible moves of Go?
 
o.o
 
what?
 
hwat?
 
4:27 AM
hwta?
 
ok, let me ask in a different way. What are the possible number of gauges modulo isomorphism?
 
uncountably infinite?
x->x+a for any real constant a consitutes a valid coordinate transformation
well
constrain that to a neighborhood
 
what does he mean by isomorphism though
 
i dunno
 
say something like x->xsin^2x+xcos^2x is isomorphic to x->x
so the isomorphism here is some given function that can be represented as different expressions
 
4:33 AM
isomorphic = one to one and onto mapping? All coordinate transformations are diffeomorphisms of the manifold...as such...they are also isomorphisms...and so...1?
 
right, I see, so it's not a very useful notion in counting number of coordinates in the context of manifolds
 
I think the answer is either 1, or uncountably infinite depending on what you mean...
 
I am trying to find a good way to sort the gauges in equivalence class that are easily manageable but so far I can only think of the class of gauges related by linear or affine transformations
 
I can't imagine a way to preferentially pick out only a finite number or even a countable infinite set of coordinate transformations...
maybe
like...
uh....
There is a maximum number of symmetries of a manifold, i.e. the number of killing vector fields admitted by a manifold.
in 4 dimensions the maximum number for that is 10
given by the 10-dim Poincare group
 
5:01 AM
@Semiclassical :)
 
@0celo7 hyperbeam?
 
@ooolb fluid flow over a plate
 
kamehameha?
 
5:27 AM
The following is what the warp drive saw in the dream look like. Red denote regions where the spacetime is expanding and blue denote the region where the spacetime is contracting
and, don't ask me why it has those weird portals or tube like geometries, it is simply what it looks like in the dream
though given the original alcuberrie drive can actually host CTCs, it may be not that far fetched
 
If the universe can have CTC's it's bad news for 0celo!
 
@Secret I didn't know you ventured here
 
enumaris: From what I understood from the dream, the warp drive showed here may be some variation of the alcuberrie metric with a global topology that has 4 holes in it whereas the original alcuberrie drive, if I recall, don't have holes
orbit stabilizer: h bar is my home chat, because this is the first SE chat I joined. Maths chat is the 2nd one I joined, followed by periodic table, biosphere, factory floor and many others
 
5:44 AM
can't really do ADM GR without global hyperbolicity afaik
bad news for 0celo indeed
 
also another interesting thing is that Dream Johnrennie has a remark about the flat spacetime bubble in the middle:
He said that the bubble of that radius [of the order of a starship] cannot be stable with spherical geometry
As for the real Johnrennie, I am not sure what he will say, but from what I knew, I don't recall anything controlling the size of the flat spacetime region of any warp drive metrics
 
I don't know of any limit to the size of the central region, assuming you've no limits on your supply of exotic matter.
 
I see
 
6:12 AM
Btw, since gravity is nonlinear, do we expect if we have a region where spacetime is frame dragged in the clockwise direction being superimposed on a spacetime that is frame dragged in the anticlockwise direction will result in a spacetime with no frame drag? (one possible physical scenario that I can envision such can occur may be when two massive rotating objects with opposite angular velocity are on the course of merging)
 
6:23 AM
@JohnRennie any idea how to put a number in a circle in TeX?
 
Sorry, no :-(
 
7:12 AM
Good Morning (from Brazil).
Well, it is make sense to ask about some sort of Gravitational Double Slit experiment?
 
Hlw
what is difference between shape and orientation?
 
@JackClerk Hi Jack.
 
Hello, @JohnRennie. (By the way, I already deleted the question)
 
You'll have to clarify what you mean by a gravitational double slit experiment.
 
7:15 AM
so no the question is still open
@Fawad Consider two cones pointing in different directions in space. Both have the same shape but different orientation
 
@Secret thanks man
 
Gravitational waves would in principle be diffracted just like light waves or electrons, but in practice they interact so weakly with matter that it's hard to see what you could make the slits from.
 
Well. I'm a begginer in the study of General Relativity ok? My knowledge about the subject is based on books like Schutz, Hartle,Carroll and introductory papers.
About quantum mechanics I have a poor knowledge yet.

So, what I meant about "Gravitational Double slit experiment" is:
There's and gravitational analogue of the Double slit experiment, for gravitational waves?
@JohnRennie
 
Hmm, I am thinking of an array of LIGOs as the detector, and thus the slits can show up as the relative amount of the arms get stretched comparing one LIGO device to another...
but I guess the major challenge is how to find a single graviton source, if any
 
@JackClerk the double slits experiment is just interference of two coherent sources, where we get the two sources from a single light beam using the two slits. But gravitational waves interact so weakly with matter that it's hard to see how we could screen a gravitational wave to get two coherent GW sources.
But if we could figure out a way to do it then yes GWs would interfere just like light wave.
 
7:24 AM
Thank you @Secret and @JohnRennie . But for conclude the discussion, I want to put a "silly picture" here: Imagine a huge double slit plate in space close to a strong source of gravitational waves. Then like water waves, and light, we will see the pattern?
 
How will the nonlinear nature of gravitational waves play a role in Jack Clerk's scenario, Johnrennie, I suspect the interference will be a lot more complicated than its optical counterpart?
 
@JackClerk if your plate is made from normal matter the gravitational waves will just go straight through it so it won't act like double slits.
But if you could come up with some hypothetical stuff that does interact strongly with GWs then yes you'd get a diffraction pattern just like light.
 
I thought gravitational waves don't obey superposition due to the nonlinear nature of gravity?
 
@Secret
Pardon.
@JohnRennie @Secret.

So in the regions of destructive interference, what I would feel? kkkkk
 
@Secret the non-linearity is negligible unless the instensity of the waves is stupidly high.
 
7:29 AM
Some sort of flat geometry region?
 
Johnrennie: I see
Jack: Probably the same as if there is no gravitational waves passing through the region, i.e. things don't get stretched
 
@JackClerk gravitational waves produce tidal forces on any matter they pass through. As you move from a maximum to a minimum those tidal forces would become zero.
 
So, if the source (like a Black Hole binary) are sufficent away, then in the regions of destructive interference, space-time would have a flat geometry and then with we put a spherical object in this region the metric will become schwarzschild-like.
if**
Pardon, I just spend some naive-phylosophy time here with these discussions**
 
Yes.
Note that even when the GW intensity is non-zero spacetime is still flat on average i.e. the effect of the GW is to produce oscillations around a flat mean.
 
8:12 AM
Hi, everyone :-)
 
Morning :-)
Everyone = just the two of us right now. I guess everyone else is in exams or revising for exams ...
 
I have almost all of Physics left to revise! x'D (::She says. laughing it off with a slight quiver in her voice::)
 
I am always silently watching in the background, but yes I am kinda busy thus $||\langle me \vert h bar\rangle||^2 \to 0$
 
Coding all night
Not by choice lolz. Took a project in the afternoon. Have to deliver by morning
hehehe
 
@Kaumudi.H when is the exam?
 
8:17 AM
Tomorrow!
 
For those interested, this is what I am reading for my PhD in the background:
 
@Kaumudi.H Eek :-(
 
No, no, I will finish!
The situation was even more dire for Calculus and I managed!
This is a neat strategy I have found-revision becomes more bearable when I have The h Bar open on the side.
In all honesty, I actually prefer exam season! At all other times-as I have observed in this semester, at least-there is nothing exciting to do. This system of tortuous panic, followed by a reward is obviously very satisfying.
 
Sid
@JohnRennie Not me! My exams are done. I am just sleeping these days or preparing to organize an event
 
@Sid Hang out, then! I need company in this arduous feat :-P (Only from time to time, of course)
 
8:24 AM
@Kaumudi.H Kind of like beating your head against a wall? It feels wonderful when you stop :-)
 
I feel less panicked when I am in the company of friends, you see, that is the key to my strategy!
@JohnRennie Exactly! :-P
 
My opinion is that I need you kaumudi to decrease the probabilty of h bar having software system infrastructure conversations, which confuse me like hell and is why I take refugee in the maths chat a few weeks ago
 
You should plead to @JohnR for that, my friend!
 
I keep telling K she needs at least three laptops, but to no avail :-)
 
@JohnR: Are you going to leave soon?
 
8:26 AM
No, I'm here for hoooooooooooooooooooooooooours yet ...
 
lol
 
Once I get all my server checks out of the way I have to attempt to understand activity coefficients.
 
For chemical reactions?
 
I'm writing some software to calculate equilibria in reactor secondary cooling systems.
 
I see.
 
8:28 AM
(Not that I have questions to ask or anything; like I said, it is a little relieving to be with friends while I am panicked. I think it is possible to gauge how much of a social recluse I am from this, because I spend some of my free time hanging out with you lot, even though I am literally inside a hostel teeming with hundreds of my peers)
 
Seems an entirely sensible approach to me :-)
 
Sid
@Kaumudi.H meh. Puzzling is rather quiet these days and i don't really hang out much on Physics.SE. plus, I have got a bit of work to complete and some questions to write.
So, I don't hang out much around these parts
 
Ah, OK :-)
@JohnRennie :-)
Alright, then, back to revision!
 
Sid
@JohnRennie what's the best source to learn Data Structures for free?
 
8:30 AM
Being a strict ambivert, I have an unusual strategy of generating social safe zones in conference environments and other formal celebration events
 
@Sid Depends what you mean by data structures
As in a programming language?
 
Sid
Yeah
 
There's little to learn. Data structures are pretty simple.
Did you have a particular programming language in mind?
 
Sid
Python, obviously
@JohnRennie I have heard things to the contrary
 
Does Python have data structures that aren't classes?
I guess if you count lists and tuples ...
 
8:34 AM
Well yes, those indeed
and of course you can build every other data structures from this
 
in the end it's all pointers and arrays anyway :p
 
Google for example code to try out the various data types in anger ...
 
Never google in anger
4
 
Why not?
 
8:38 AM
If you are a python user and want to do some data analysis, you might want to check out pandas
My favourite database structure are multiindexed dataframes
 
I really don't understand why Python became some big data analysis thing
It's a weird idea
I'd rather use either some low level stuff like C or FORTRAN or some specialized thing like R
 
@0celo7 Mike Hobson - he's actually an astrophysicist
 
Well it has its pros and cons. Python, like matlab, can vectorise many operations and has some really simple API, but being to used to it will make you unable to deal with things under the hood
I don;t have enough R experience to comment about R, though
 
R is great for stats
 
I found python easy to use mainly because my way of thinking and coding is like a matrix and array: I read much faster when things are presented vertically
 
8:41 AM
The LHC software is all in C++
Or should I say
"""C++"""
It's that kind of C++
The one mostly made of static classes
ie C
 
One thing I don't like about C and C++ is all those semicolons syntax, but otherwise it is fine and you can write something as complicated as a game from it
in fact, a lot of python libraries are ultimately C++ codes really
 
Well, either you use semicolons or you have to deal with indentation
And indentation isn't fun due to the varying standards of indentation
 
What about line numbers?
 
Well it's not 1982 anymore, so no
 
that's true. though back in high school ,regardless of code, our teacher taught us to always indent your code to allow easy reading and troubleshooting. We are also taught the 4 spacebar indentation convention
 
8:46 AM
I have an Amstrad CPC
Coding with line number is kind of awful
And it only existed because old computers were very bad at text editing
 
this is better than indenting by tabs because tabs can be unpredictable at times, especially in python where it relies on tab to understand what you are saying
 
Ah, the tabs vs spaces wars ...
 
I use tabs because I'm lazy
 
My editor automatically inserts the spaces for me :-)
 
...and line numbers :-)
 
8:47 AM
Yeah that is the nice thing about good IDEs
Visual Studio does all the indentation
Old computers were kind of weird because BASIC was the operating system
 
I use VS for big projects, but for small apps I use Codewright.
 
not an ideal OS
 
\o @yuggib
 
@JohnRennie I wish I can just tab because I am also lazy, but sometimes tab insert 4 spaces while other times it inserts 5-6 spaces, thus screwing up a block of if then conditions in my code, which is why I had no choice
 
But then again only millionaires had hard drives
So you didn't really need anything to deal with file systems and such
 
8:50 AM
O/
 
and couldn't even dream of multitasking
monotasking was enough of a chore
 
How are you? @yuggib
 
just waiting for that bloody tape to load a program for 5 minutes
 
I'm good
 
I currently automate almost everything from job submission to data extraction, and later on, with the help of the machine learning group in my uni, we might be able to automate a GUI library search thingy
 
8:54 AM
I can do all tasks related to my work without leaving the text editor (of course, such text editor is emacs). The only inconvenience is that some websites don't render in a optimal way (but most of the work-related ones do)
 
having very old computer is fun in one way because it's like
You can see the computer THINKING
The algorithm very slowly unfolding on screen
Moving things pixel by pixel
I have a 1982 program with a paint program and using the fill bucket is an event
you can see the screen filling line by line
 
I have a quick question. If the displacement of a particle executing S.H.M is given by $x=a\cos \omega t+b\sin \omega t$, then how are we to find its amplitude?
 
then when it reaches an edge
going the other wa
 
Graph it?
 
@Kaumudi.H differentiate and set equal to zero to find the extrema ...
 
8:58 AM
Dang it! Of course! Thanks!
 
There's probably a quick way ...
... that I can't remember
 
This is quick enough. I got the correct answer, thanks!! :-)
 
Graph it, to check the answer :-)
 
Another one: Given the equation of a 3-D transverse wave $\psi (x,y,z,t)$, how are we to prove anything about the way that it moves?
 
Sid
@Kaumudi.H that is confusing..
 
9:13 AM
Give me a moment to upload a picture of the whole question...
 
Hi to all. Does anyone know where I could write matlab code online(for free)? Apparently another one of my institutions great inspirations is to have a matlab-oriented computational physics course without having matlab on the universities pcs. Thanks.
 
(Do ping me if you know)
 
9:33 AM
@Kaumudi.H Hacky way: 1st thing is that $\psi\left(x, y, z, t\right) = \psi\left(x, y, t\right)$, so no propagation in $z$-direction. Now, in '$1$ unit' of time, it travels $\frac{\sqrt{3}}{2}$ units in the $y$-direction and $\frac{1}{2}$ units in the $x$-direction. Use this to form a triangle and you'll get the answer with simple trig :)
 
Ah, nice, thanks! :-)
 
Anonymous
9:50 AM
@ConstantineBlack There is an online version of matlab. And on android store you'll get it for free.
 
It's pretty weird to think that physics didn't really use matrices until like
The 30's or so
 
@Blue: Hey! How was the exam?
 
Anonymous
@Kaumudi.H Ah, it was okayish. It was mostly memory based. Each small question was of 10-15 marks. No idea what they expect me to write for questions like "Describe acoustic and optic phonons" for 15 marks!! I only wrote two small paragraphs...meh. I don't like this subject much :P (physical electronics). Hope to do better in the upcoming tests so that there isn't a huge effect on the gpa.
 
10:06 AM
::rant about memory tests::
 
Anonymous
@Mithrandir24601 lol...i and my classmates have been ranting about it for the past 2 hours XD Just came back home after a group rant session!
 
@Blue Seriously, testing how much someone's remembered about a subject has exactly no bearing on how good they are at that subject! ::glares at examiners::
You're just testing how good their memory is! What use is that!!
2
 
@Blue Ok, thanks. I found a way by connecting to the servers of the university( the program isn't installed on the pcs on the computer room, but if I connect to the server of the university- which means running remotely another environment, i found an older version of matlab). But thanks again.
 
10:23 AM
"It is clear that such a limiting past light-cone will be a null hyperplane"
No it's not >:|
 
@Blue Ah, damn, that really sucks!
 
10:44 AM
@Mithrandir24601 are you saying memory has no use in understanding?
 
@Blue I found a fantastic book on dynamics if you're interested.
 
@user685252 No; I am saying that it has no bearing on how good you actually are at the subject - it has no bearing on how good you are at applying knowledge; it doesn't test problem solving skills; it doesn't take into account that, if I'm sitting in the office having forgotten the difference between different types of matrix decomposition or something, I can just search the internet (or a textbook), so it doesn't say how good someone is at research in that subject;
it doesn't test how good you are at deriving anything - someone can write down a definition without any understanding, while someone who can derive it, but has forgotten it probably won't have time in an exam situation. In short, testing memory is not the same as testing understanding
If you really want to test someone's understanding, give them a few problems in that area that they've never seen before and give them a reasonable amount of time to do it, with access to textbooks etc.
 
I am bad at memorizing so I have to pretty much reconstruct/reprove whatever from the start.
I did that with a lot of coordinate geometry on my high school exams
And trig, too
 

« first day (2588 days earlier)      last day (374 days later) »