The h Bar

12:00 AM

I think that's...hep-ex?

@0celo7 I did the modern kind of nuclear physics which is rather more about nucleons then the old kind. My advisor joked that nuclear physicist do what particle physicist did thirty years ago, only to a order of magnitude better precision.

Which isn't true in general but is reasonable true of JLAB physics.

0celo7

@ACuriousMind void.sw

I think this phenomenon is important enough to get its own section

Secret

lol, I have no idea, Macbook autocorrect love to make awkward autocorrects.
I see.

In that case, using the principle for the asymmetric rail case, the centre of mass will be rotating wrt two points of contact, so it means the $\omega$ have to satisfy

$$v_CM1=r_{eff1}\omega1$$
and
$$v_CM2=r_{eff2}\omega2$$

But the situation looks a bit more complicated. Do I just vector sum the $v_CM$s together to get the overall $v_CM$

I also noticed there's effectively a circle rolling inside another circle. I am not sure how to solve this case because I recall that rotational stuff don't obey superpos…

0celo7

@Secret my macbook has never autocorrected me on Stack

hence the shit-show that is my spelling

Secret

http://i.stack.imgur.com/r0HOK.png
@dmckee

@0celo7 lol then your macbook is well behaved to not cause troubles

0celo7

12:11 AM

my phone, on the other hand

dmckee

@Secret I'm not positive what you mean by the asymmetric rail case, but the axis of rotation won't be horizontal unless at least one side is slipping or the track is curved.

0celo7

>buy album in 2016

>contains song from 2014

wut

@dmckee Reading one of my advisor's papers, pretty sure it was typed in Word

dmckee

@0celo7 I wouldn't be surprised. The question of LaTeX versus word processor is anything but uniform across physics or even mathematics. You can find cases where one itty bitty subfield uses one and a next door itty bitty subfield that outsider would lump together uses the other.

So, you're allowed to have strong opinions, but you have to be able to behave yourself in professional setting.

Which is why I don't boo the jerks with their powerpoints.

Instead I think it very hard.

ACuriousMind

Am I allowed to spit into the drinks of people who don't use TeX?

0celo7

@ACuriousMind Wow

dmckee

12:23 AM

@ACuriousMind You have to package up the procedure and post it to CTAN first.

0celo7

You'd spit in a fellow UH person's drink?

ACuriousMind

@dmckee Heh.

0celo7

Any Windows people here use iTunes?

So I have this issue that if I plug in headphones AFTER I open iTunes, it will not play audio through the headphones.

This is really annoying.

Does anyone have a similar experience?

@Secret are you familiar with crystallographic symmetry groups

Secret

@0celo7 I have a course on that last semester, thus the memory should still be fresh

0celo7

@Secret give me the tl;dr on the difference between $Fd\bar 3m$ and $Fm\bar 3m$

I have a book on crystals that I won't be able to get to until the summer

(winter for you?)

guest

12:36 AM

Why waste your spit? @ACuriousMind ;)

ACuriousMind

@0celo7 more like get 40 rep :P

Secret

@dmckee Hmm, I am guessing you mean something like this, that the rotational axis will always orient such that the points of contact will be the same distance away from the rotational axis? (This is something new to me, btw)

0celo7

@ACuriousMind holy fuck people just upvote you for no reason

ACuriousMind

Why do I even bother writing technical answers? All I need to get rep are pithy answers to effortless questions.

2

guest

True dat^

ACuriousMind

12:40 AM

@0celo7 Funnily enough, this only happens for posts without much math. Hawking's "every formula halves your sales" is true as "every formula halves the upvotes" here.

0celo7

:/

guest

:\

ACuriousMind

Yeah, I don't like it, either

0celo7

so I was thinking (I know, shocking) and it occurred to me:

I talk with my physics advsior and I feel like I have much to learn

I talk with my math adviser and I feel like an idiot

advisor/adviser

I don't even know any more

ACuriousMind

Conclusion: Math is harder than physics ;)

Not even sure whether I'm joking there.

guest

12:43 AM

Talk to yourself first.

user54412

@0celo7 Sound -> Playback -> select the headphones as default device? That said windows (and I'm guessing other OS's) handles switching audio devices rather poorly, so the only surefire way is to set up the device you want and then run the software.

0celo7

Maybe I need a math adviser who wasn't reading Arnold/GR at age 16

user54412

Weren't you reading that at 16?

0celo7

Yes

I don't see how that matters

@ChrisWhite Sound?

Where is that

user54412

right click the speaker taskbar item and click sounds (on win 7)

0celo7

12:46 AM

it won't let me make them default

user54412

for more modern windows, I can only guess you have to enable a touchscreen, use some voice commands, and request microsoft take control of your machine and open the sound window for you

0celo7

lel

I'm using 8.1

ACuriousMind

lol

0celo7

pretty garbage

oh great we're in for some more programming "goodness"

user54412

@0celo7 then you have bigger issues I feel

0celo7

12:47 AM

Hmm I can bass boost even more

tpg2114

@dmckee Thanks for the tips -- I don't need to apply for funding yet, but I may be working in my current lab in a position where I can get funding. So we could be considered "established". I ended up trolling through grants.gov and sbir.gov and found some things to give me an idea to aim for

Secret

@0celo7 Generally speaking Fd3m and Fm3m differs in that Fd3m has a diagonal glide plane (d). This happens when you have some atoms or ions being placed in the empty spaces of the Fm3m (aka Face centre cubic) space group so that they form the vertices of an inscribed tetrahedron

Illustrations can be found here
http://www.uwgb.edu/dutchs/symmetry/3dspacegrps/3dspgr225-226.htm
http://www.uwgb.edu/dutchs/symmetry/3dspacegrps/3dspgr227-228.htm

0celo7

@Secret Right, my Fd3m crystal is missing an oxygen

ACuriousMind

Oh god crystal groups. I had to discuss those with a prof who had no idea what a mathematical group was. It was depressing.

I think the most advanced math we did in that lecture was proving the cross product is orthogonal to the two vectors it's made of.

0celo7

@ACuriousMind Pretty sure Lang is that way too.

dmckee

12:49 AM

So my department is "Chemical and Physical Science" which means we also have chemists, a geochemist and a geologist/geophysicist.

0celo7

But hey, UH gave him a PhD.

ACuriousMind

@0celo7 I have no idea who your Lang is, when I hear Lang I think of Serge Lang, the algebraist

0celo7

Maybe you guys need to raise your standards.

Secret

For me, rep is not my major concern. Whether the user found my answer (or everyones answer combined) can adequtely answer his question (and bonus points, for any questions that can be generated from his question) is my major concern.

That's why I rarely wrote an answer unless I knew I have know enough details to write a perfect answer

dmckee

The inorganic chemists are all about groups. The organic and biochemists look slightly green when you mention them.

Strange.

0celo7

12:50 AM

@ACuriousMind I know who Serge Lang is -.-

dmckee

@Secret Yes. That.

user54412

@0celo7 hard do know without benchmarking, but maybe 2-3 days

0celo7

@ChrisWhite Jesus!

That big huge thing!?

ACuriousMind

@Secret That wasn't a serious question. I know that I write those answers partly to help someone out and partly to convince myself I actually understand it.

tpg2114

Titan isn't even that big ;)

Or rather, 2-3 days isn't that long. It depends on how you look at it

0celo7

12:52 AM

@tpg2114 Biggest in the US last time I checked

imagine playing some vidya on that thing

I wonder if it can run TW3

user54412

Titan has 7.2 million core-hours per day. For perspective, our group just got a 50 million core-hour allocation, and the biggest criticism was they felt we didn't ask for enough.

dmckee

@ChrisWhite I'm always staggered by people who make productive (or even possibly productive) use of that kind of horsepower.

Makes me feel like some script kiddie.

tpg2114

@0celo7 2nd fastest, but not the biggest...

At least in terms of cores

Which depending on your benchmarking, cores /= performance

user54412

Also, our code isn't GPU-enabled, so we wouldn't be able to utilize Titan well (and we'd never get time on it)

tpg2114

^ That's where we are too

dmckee

12:54 AM

Someday I'm going to have to actually look at what GPUs offer. Just to have some idea what kinds of problem might run well on them. I mean, I know they're deeply pipline and they offer a specialized set of SIMD tools, but that it.

0celo7

I wonder if my school gets preferential treatment for Titan time because we maintain the Oak Ridge campus

dmckee

@0celo7 Having the guys that control it on hand means having access to real experts about what its good for and how to make best use of it. That kind of thing really helps.

tpg2114

@dmckee Compared to serial codes, I've seen huge improvements. Compared to already-optimized parallel codes, not so much. And if you need to communicate between nodes, it's even worse

user54412

@0celo7 There might be some fraction of the time allocated for that. I think Blue Waters gives time to UIUC.

tpg2114

I like the idea they are trying for, but I don't know if I buy into it yet. Intel is releasing their multi-core architecture as processor sockets instead of boards. Next year, they'll drop a 72-core/socket CPU

I think that's the better way to go for multi-core/multi-thread. No need to go through the PCI-e bus like with GPU's.

0celo7

12:57 AM

@ChrisWhite Blue Waters?

tpg2114

That said, GPU within a single node is really good. So pre/post-processing of data on a workstation will see big improvements.

user54412

@dmckee I follow my adviser's line of thought on this. He notes that (1) people often compare 1 GPU to 1 core, when really it's more like 1 GPU vs. 16 cores. (2) If people put as much effort into cleaning up their CPU code as they do in developing a GPU code, they'd often get comparable speedups.

tpg2114

@ChrisWhite That's my view/experience with it also

Secret

in that case @tpg2114's approach on the question I and him discussed earlier is the correct approach (that it is a rotated symmetric rail system)

And now for the (hopefully) last of my question related to rolling tracks:
I am trying to look for a general equation of the rolling track problem (it is ok if it has summations , integrals or other non closed form expressions) that can allow me to (in principle) calculate a tumbling irregular piece of rock (assume rigid body) down a track (it can be curved, straight, slanted, thick or thin, uneven etc.), because I planned to do some modelling of…

user54412

@0celo7 Blue Waters

0celo7

12:59 AM

@ChrisWhite I googled.

user54412

just thought I'd help you with that :p

ACuriousMind

@0celo7 I'm so proud of you.

His first (googling) steps.

0celo7

@ACuriousMind I'm crying right now :)

So happy

user54412

Aug 5 '15 at 17:10, by FenderLesPaul

sniff they grow up so fast

0celo7

@ChrisWhite Well Titan isn't ON our campus but a division of Univ. Tenn. has the Oak Ridge contract for X years

tpg2114

1:01 AM

top500.org/list/2015/11 Looks like I have access to 13, 18, 21, 28, 51... and I stopped looking

ACuriousMind

@ChrisWhite Wow. For a not really specific utterance, that's a remarkable memory

0celo7

Not the Y-12 part, that's some spook outfit

tpg2114

We had time on Titan for benchmarking awhile ago, but I wasn't involved in that

0celo7

What's SNL (the lab)

tpg2114

Sandia?

0celo7

1:03 AM

Checks out

tpg2114

They do some cool stuff out there

I just came across some experiments they ran that I will try to get done as a "capstone" part of my thesis. In addition to a bunch of the turbulent flames I need to run as validation

0celo7

@ACuriousMind Did you remember that one?

user54412

@tpg2114 Pleiades. Isn't that the one Donald Glover took over for several days to calculate a single orbital trajectory in The Martian?

tpg2114

@ChrisWhite Is it? I haven't seen the movie...

I read the book, but don't remember those kinds of details

0celo7

Who needs such a precise orbit for a movie

tpg2114

1:06 AM

@0celo7 One of the students who got his PhD in our lab is out in Hollywood working on special effects. His job is to make more realistic looking flames and explosions for movies

They pay good money for real-looking things

dmckee

Now that's cool.

I took a graphics class right at the beginning of grad school. Long story.

tpg2114

It is pretty awesome. We spend most of our time trying to hide the fact the answer is wrong by making it look cool. He gets to make things that look cool and doesn't care if it's right

0celo7

Lol

dmckee

But ever since I can't go to a movie with CG stuff in it without critiquing the CG. Drove the better half half nuts, until she decided just to mock me mercilessly about it.

tpg2114

Haha

I've actually borrowed a lot of algorithms from graphics and video games... those guys care about speed before anything else. They always have way better algorithms than the scientific community comes up with

0celo7

1:10 AM

one side effect of growing up that I don't like is that bad CGI is really bad.

@tpg2114 except for TW3

user54412

Or Crysis

0celo7

pretty sure they're calculating martian orbits while running the game

tpg2114

@0celo7 Even at 30fps it's way faster than what we can do in the scientific computing arena

0celo7

30 fps ;_;

I wish

plays at 20fps @ 720p lowest settings

dmckee

Favorites that jump to mind: Mulan, Brave, Ice Age (doesn't try to hard but took the time to get what they did really right). Shit list: Anastasia, XXX and something I watched the other night that really sucked ... oh yeah some episode of The Murdoch Mysteries with a skyline shot of early twentisth century Toronto that wasn't even up to video game standards.

0celo7

1:14 AM

How does one make sub-bullets in ppt?

I've completely forgotten how

dmckee

Who are you expecting to know that?

tpg2114

What do you mean sub-bullets?

0celo7

someone

tpg2114

Just hit tab

It should move stuff over

ACuriousMind

5 hours ago, by Chris White

@0celo7 You use beamer :p

tpg2114

1:14 AM

If that doesn't work... use beamer.

0celo7

Ah!

@tpg2114 Engineers use TeX?

tpg2114

The only time I use PPT/Excel/Word is when some government bureaucrat makes me out of some contractual obligration

Which is sadly more often than I like. But otherwise? LaTeX literally everything. I did every single homework in grad school in LaTeX

0celo7

Hmm, it seems like the branch of solid state physics I'm working in does not use LaTeX

tpg2114

Although for the record, I'm more on the science side than engineering side. My work is entirely fundamental and not design oriented.

ACuriousMind

@0celo7 Then revolutionize the field and make them use it

dmckee

1:17 AM

@tpg2114 How did you do the figures? I'm trying to move to that for my teaching material, but the figures are a sticking point.

0celo7

@ACuriousMind I don't really want to work in this field long enough to revolutionize it :P

I'm getting paiiiiiiid

(in a week or so)

tpg2114

@dmckee Tikz primarily. That, and matplotlib in python if it's actual data

dmckee

I use tikz slowly and badly, and sometime I sketch something on my wacom, but I'm not good enough yet to be very neat or precise.

0celo7

@dmckee I know how you can get better

tpg2114

I'm not great either. I spent a lot of time picking parts of what I want from the gallery

dmckee

1:18 AM

Fair enough. I use root for plotting data because ... well, because I already know that horrible beast.

tpg2114

And asking over on TeX.SE. Those guys are wizards

PGFplots is supposedly good for plotting math functions in LaTeX. I haven't messed with it much though

0celo7

@dmckee go to the library and get a book called "Morse Theory" by J. Milnor

draw all of the pictures

then send it to me

by the end of it, you'll be a master

tpg2114

Morse theory is really interesting. I've been working on discrete Morse-Smalle complexes

ACuriousMind

@dmckee What kind of figures do you want to do?

dmckee

@tpg2114 I've used that a couple of time to do graphs for tests. It's pretty nice, but it will clutter up your directory with extra files.

0celo7

1:20 AM

@tpg2114 that's used in engineering?

dmckee

@ACuriousMind Right now, mostly the ones you draw for intro physics, classical mechanics and modern.

tpg2114

@dmckee Have you tried Inkscape for drawing? I don't use it personally, but I know a lot who do

0celo7

/fluids

tpg2114

@0celo7 It's a very, very new idea which is why I want to get in on it

dmckee

@tpg2114 I looked at it, couldn't understand or find the right tutorial and tabled it. Need to look again.

0celo7

1:21 AM

ok I have to leave and get work done

user54412

@0celo7 abstrusegoose.com/105

tpg2114

But yes. Topological descriptions of data are immensely useful. Instead of dumping terabytes of data from 200,000 cores, if we can get a topological description of the data and store that? Way better

0celo7

you people are the worst influence

dmckee

I understand that it works well with drawing tablets.

tpg2114

I'm all for the free side of things, but I have been told it's good also.

@ChrisWhite I have to send that one to my brother...

He's getting his PhD in disgustingly theoretical math.

dmckee

1:23 AM

Speaking of Bloom County, today's strip is just brilliant.

0celo7

@tpg2114 how are you learning Morse theory without (presumably) a background in algebraic topology

tpg2114

The longer we go on, the less helpful we are to each other when we talk.

0celo7

I'm considering doing a reading course on morse theory over the summer while I'm working

ACuriousMind

@ChrisWhite Not even sure if I'm the guy holding the gun or the one sucking up the bait there...

tpg2114

@0celo7 Because I can read journal papers and work out the things I need to know from it :p

Just because I wasn't training in super theoretical math or physics doesn't mean I'm helpless in understanding topics

dmckee

1:24 AM

As far as I can tell the author usual identifies with the head with the glasses on.

ACuriousMind

@dmckee No, the author is the guy with the blue cap, if present

0celo7

@ChrisWhite it annoys me that "gauge theory" is presented as useful

dmckee

Well, who is the glasses dude then? I mean, other than me?

0celo7

@tpg2114 you can "figure out" algebraic topology from journal articles oO

dmckee

abstrusegoose.com/156

Secret

1:25 AM

@dmckee what is the name of the equation that will allow one to calculate the most general case of the rolling track problem?

0celo7

I should just go clean toilets

although I probably can't do that either

tpg2114

@0celo7 To the extent I need to know it to apply it to my work? Yes. So long as I can understand the assumptions and verify my case fits within them, then I can apply the algorithms

dmckee

@Secret Not a clue. Is there one? The stuff we've been talking about relies on lots of simplifying assumptions.

tpg2114

For an example of them in my field: citeseerx.ist.psu.edu/viewdoc/…

ACuriousMind

@dmckee Hmmm...perhaps his brother? I don't know. But I'm pretty sure that the blue cap is the symbol for abstruse goose's author surrogate

tpg2114

1:27 AM

I'm not looking to invent brand new theories in algebraic topology. I'm just trying to apply what exists to what I need

dmckee

Now that you mention it I think you might be right. He shows up in strips with captions that say "I" a lot.

Spin and so on.

0celo7

> Any host material for HLW must be able to withstand the sands of time

ACuriousMind

@dmckee There's one about "my girlfriend" that makes it pretty clear

0celo7

Pretty sure my prof would execute me if I wrote that into the talk

POWERPOINT WHY DO YOU SUCK

::mashes keys::

tpg2114

Cause it's not beamer.

0celo7

1:34 AM

the fuck I can't have a small bullet and then a large bullet

I HAVE TO WRITE THE THING BACKWARDS

Secret

This is what I am planning to model

I felt like it has to be a differential equation of some sort, but I am not sure how to mathematically write the next points of contact for each step

@dmckee I also felt like given the ubiquity of physics engines out there in the gaming industry, the equation (or something similar) should have already been derived by someone, except I cannot self study because I am not aware of any names (if any)

0celo7

satan

why did you make powerpoint

ACuriousMind

chill, man

0celo7

@ACuriousMind no

@ACuriousMind oh god void swelling

I have to talk about void swelling

ACuriousMind

@Secret I think you "just" write something that calculates the forces at each timestep, and then timesteps f=ma forwards. It's not that you write down one big equation and then solve it.

0celo7

1:40 AM

wtf is void swelling?????

dmckee

Well, you can treat your object as an inverted physical pendulum for each partial tumble (and the parallel axis theorem will be your savior here), but there is no reason to imagine that this problem has a well known name.

I don't even know that there is a name for the "rolling on rails" case we were talking about earlier.

> "Students ask about formulas, professors ask about principles".

@0celo7 This is the point at which a real saint wouldn't say "I told you so."

I told you so.

2

0celo7

i hate you

wtf is loop nucleation

Secret

Well typically I am concerned about principles more than formulae, but for general cases if a formula already exists it is often good to find them as a check on how well one understand a principle (especially for people like me who constantly made conceptual mistakes in relatively unfamilar topics in physics without knowing they have made mistakes)

@dmckee @Acuriousmind I see, I should try it and see how it goes

dmckee

@0celo7 These are questions to ask that grad student overlord you were talking about. He (or she) needs practice explaining it anyway.

0celo7

@dmckee I don't think it's that relevant anyway

ACuriousMind

1:45 AM

@0celo7 Then why are you stressing out about it?

0celo7

@ACuriousMind I've read the paper 3 or so times and it never occurred to me I have no clue what it is

dmckee

I so hate it when that happens.

On the up side, the local Wallgreens drug store is selling pens with both a LED light and a diode laser in them 2 for $5.

They are merely adequate as pens, but they have laser in them and they're dirt cheap!

0celo7

why did you specify drug store

is there any other kind of Walgreen's

dmckee

@0celo7 Because there are a few parts of the country that don't have that chain.

So some people might not know.

0celo7

Really!?

ACuriousMind

1:50 AM

@0celo7 I have never heard "Walgreen's" before.

0celo7

@ACuriousMind but you know what a drug store is?

ACuriousMind

yes?

0celo7

what is it

that's like saying you know what a tissue is but don't know Kleenex

ACuriousMind

@0celo7 A Drogerie

dmckee

The drug store market isn't as fragmented as grocery stores, but it used to be and (almost?) no chain has complete coverage of the US.

user54412

1:51 AM

You know what we need? The local corner druggist.

0celo7

@dmckee Food stores I know: Aldi, Food Lion, Harris Teeter, Whole Foods

Shoppers

Costco (does that count?)

dmckee

HEB, Kroger, Harps, Pricecutter, ...

0celo7

HEB is a Texas thing

Kroger...hmm...I don't think that's common in VA.

ACuriousMind

@0celo7 Y'know, we call them Tempos here, after our dominant brand.

0celo7

Never heard of the other two

@ACuriousMind I know what they're called in German

dmckee

1:53 AM

(some of those have the same corperate overlords now, but they used to all be different)

0celo7

I still remember SOME words

dmckee

@0celo7 Lower midwest.

user54412

I once read a great book on making your own telescope, written in the 30s or thereabouts. It told the reader to simply go to the local drug store and ask the guy for some silver nitrate and other fun compounds.

ACuriousMind

@0celo7 So it's better to say "tissue" than "Kleenex", and better to say "drug store" than "Walgreen's".

dmckee

@ChrisWhite Those are fun. And eye opening in this age of tight oversight.

0celo7

1:53 AM

@ACuriousMind I don't see what point you're making