The h Bar

Slereah

18:02

Why does spinor GR exist

The spinor Riemann tensor is ridiculous

Shashaank

20000 in advanced. But they just test how quickly you solve a complex mathematical problem ? They don't promote any any curious thinking or why or what or how types of things about any physical question. That i think is not Physics @Yashas

0celouvskyopoulo7

18:16

@Shashaank Give it up

people like yahas are so embedded in the program they can't suddently admit it's BS

@ACuriousMind ahahahahah

My lattice energy was wrong

time to start over :)

ACuriousMind

Heh, that happens

Shashaank

@0celouvskyopoulo7 Sorry , I didn't get you ! What do you refer to by program and BS

0celouvskyopoulo7

JEE

Shashaank

18:35

@0celouvskyopoulo7 if you say Bullshit , then I like your choice. There is hardly any good physicist in India who went to an IIT . They give Physics too . Focus is engineering. But even their engineering course is pathetic. They judge machine..... BTW if you are familiar with the type of jee questions , I would like to ask whether Einstein n all would be pretty easy at it( n Jackson , Griffiths type questions) or not ??

Sid

Einstein would ace Physics and fail in Chemistry.

There you go. He would have failed JEE. :P

0celouvskyopoulo7

@Shashaank I am a big JEE hater.

There is a lot of controversy surrounding it in this chat

Sid

Yes, @Shashaank it's better if you move that conversation to the room where it belongs.

Physics Meta

0

Q: Can anyone show me any mathematical equation in duplicate

This is my question Mathematically, What is contradiction between quantum physics and General relativity theory? This is what it called duplicate Can anyone show me any mathematical equation in [duplicate]!? this is an example of how this site suppress certain users for some unknown reasons...

Shashaank

@0celouvskyopoulo7 I don't know where it belongs to....but surely it doesn't belong to the jee room.... controversy as in ....

@Sid I don't think he would have passed Physics too....

Mithrandir24601

18:40

@Shashaank I know one person who went to IIT, and he seems at the very least decent at physics (I don't know him well enough to say whether or not he's really good, but he's certainly not bad)

Shashaank

@0celouvskyopoulo7 I am a bigger jee hater than you :)

@Mithrandir24601 I know that 10000 people go to an IIT . Not even one of those know the difference between " what they WANT to do " and "what they have to do"

ArtOfCode

maybe let's not compare academic assessment procedures to religious wars?

0celouvskyopoulo7

@Shashaank No

I got someone to delete his account because he was a JEE supporter

ArtOfCode

that... isn't something to be proud of

0celouvskyopoulo7

@ArtOfCode Actually, @ACuriousMind has ruled JEEhad to be acceptable.

Emilio Pisanty

18:46

@0celouvskyopoulo7 @Shashaank There is strong experimental evidence to back that up.

ArtOfCode

@0celouvskyopoulo7 and now I'm gonna say it's not. Avoid it, please.

0celouvskyopoulo7

@ACuriousMind ?

I said I don't see a reason why it should be bad.

Someone decided to delete it, not sure why.

ArtOfCode

Because you're comparing a process that plenty of people go through and plenty of people support to religious war. That's not respectful, so you should avoid doing it.

0celouvskyopoulo7

@ArtOfCode I don't care if it's respectful. The JEE test is not a person, it's a thing.

I can say apples are assholes and that's perfectly acceptable.

Danu

Hi @ArtOfCode long time no see

ArtOfCode

18:48

o/

1 min ago, by ArtOfCode

Because you're comparing a process that plenty of people go through and plenty of people support to religious war. That's not respectful, so you should avoid doing it.

Avoid it, please.

0celouvskyopoulo7

@ArtOfCode I am not comparing any process to anything. I am making a play on words, where an object is the aim. Not people, not a process.

The piece of paper.

Felipe Lopes

Hi, one of my answers was taken down because it was a solution to a homework problem. I only answered because the asker seemed to have shown some effort towards solving the problem. I am new to the site, could you please link me to the policy on homework problems so that I can avoid these problems in the future? Thanks

Shashaank

@0celouvskyopoulo7 spot on !!

0celouvskyopoulo7

Now, I do feel like the process is ridiculous and is like a religion.

But since I probably cannot insult religions (even though we all know which one I'm talking about), I guess I won't say anything more.

ArtOfCode

@0celouvskyopoulo7 Intentionally, perhaps not. Unintentionally, certainly. More to the point, your use of it was flagged by someone who indicated they felt it was offensive. The fact that it's not respectful of other people's support for the qualification, and that someone finds it offensive, should be more than enough to persuade you to stop.

Emilio Pisanty

18:51

73

Q: How do I ask homework questions on Physics Stack Exchange?

What is the policy on asking homework questions on Physics Stack Exchange? What kinds of questions are considered homework questions? Are homework questions allowed? What should I include in a homework question? Why don't you provide a complete answer to homework questions?

discussion faq homework

some aspects under long-term discussion

0celouvskyopoulo7

@ArtOfCode Meh, people always flag it. Let them.

Emilio Pisanty

latest installment

1

Q: Let's get things rolling again with the homework policy

Okay, so the last time we had a focused discussion of the homework policy was August 2016. Blech. So, let's start thinking about this again. And before you collectively groan, this first post isn't that painful because there's only one thing in here for debate. Let's rename the homework policy....

discussion homework

0celouvskyopoulo7

Flags don't reflect anything.

If I get the 30 minutes, I go do something else.

ArtOfCode

@0celouvskyopoulo7 Yes, they do. Especially when there's a mod acting on them and asking you to stop doing something.

Shashaank

@EmilioPisanty I don't know much neither do I say that I am very great in my level Physics , but I feel ( and I think it's absolute ) that jee is a big rubbish .....even for engineering.

Felipe Lopes

18:52

Thanks

0celouvskyopoulo7

@ArtOfCode I would like word from our room's mods, thank you.

ArtOfCode

@0celouvskyopoulo7 All moderators have equal power to moderate in chat.

Emilio Pisanty

0

Q: How to provide working for the question if we don't even know how to proceed in a question?

Most of the time when people ask Homework Related questions they are put on hold.They ask for our work on the question. Sometimes we all face the situation that we don't even know how to proceed further in a question and its impossible to show working in such cases. Why most of the people don't u...

discussion questions homework close-reasons

also relevant

0celouvskyopoulo7

@ArtOfCode Respectfully, that's ridiculous. How can you say that with a straight face?

Maybe your face isn't straight, I can't see it.

Danu

@0celouvskyopoulo7 What do you mean?

ArtOfCode

18:54

@0celouvskyopoulo7 Because it's patently true. Every moderator is given the powers and responsibilities for moderating chat.

Bernardo Meurer

@ArtOfCode Yes, they have equal power, but that doesn't mean they should exercise it when the moderators respective to the site are around. Chill.

ArtOfCode

@BernardoMeurer They're not.

0celouvskyopoulo7

@Danu It doesn't make sense that one mod should be able to come to another chat and dictate what people can and can't say.

ArtOfCode

@0celouvskyopoulo7 all mods are enforcing the same policy

Danu

@0celouvskyopoulo7 That's just wrong. Every moderator is a chat moderator.

Bernardo Meurer

18:55

@ArtOfCode I see ACM's watchful head

ArtOfCode

@BernardoMeurer the fact that he's in the room has no bearing. He's not available as of this moment.

Bernardo Meurer

All this brouhaha because of JEE[INSERT_PUN] ?

0celouvskyopoulo7

@Danu I know that, and it's stupid.

ArtOfCode

@0celouvskyopoulo7 It's also how it is. You don't get to ignore it because you don't like it.

Mithrandir24601

@0celouvskyopoulo7 I, for one, am a non-diamond who disagrees with that comment

Danu

18:56

@BernardoMeurer No, not really. The point is the subsequent response to being told not to do that.

0celouvskyopoulo7

@ArtOfCode I get to ignore what I want. Whatever happens afterwards, happens.

@ArtOfCode I don't believe that for one second. Every moderator has their own variations on various things.

There are degrees of leniency, and I don't think foreign mods should come and "set room mods straight" or whatever it is you're trying to do.

ArtOfCode

Look, someone stated in a custom moderator flag that they found your 'play on words' offensive. You should take that in good faith, and Not Do That (tm).

0celouvskyopoulo7

Sorry to whoever that was, but it's not offensive.

Perhaps I should leave before I get banned.

enderland

352

Q: If you're gonna talk Politics, you must respect those who disagree

This is sort of a follow-up to two past discussions: Toward a philosophy of Chat Does the Be Nice policy require SE users to "be nice" to people who are not SE users (e.g. public figures)? Over the past year, there's been an uptick in discussions of politics in chat. JUST LOOK AT THIS CHART! ...

discussion chat etiquette

0celouvskyopoulo7

Politics?

Who the hell is talking politics?

Mithrandir24601

18:59

@0celouvskyopoulo7 Just because something isn't offensive to you, doesn't mean it's not offensive to everyone else

enderland

> Ultimately, this all boils down to something that's come up repeatedly in past discussions of chat: respect. You can't effectively talk about anything in chat if you despise the folks you're talking to, and politics by its nature tends to bring out topics where people find it impossible to respect opposing views... or the people who hold them.

0celouvskyopoulo7

@Mithrandir24601 Protip, the whole notion of offense is made-up.

ArtOfCode

maybe it was just about okay two months ago when leniency was greater and nobody said they found it offensive. Now that we're trying to be more consistent in enforcement and someone has directly said they find it offensive, it's no longer okay, and you need to stop using it

0celouvskyopoulo7

If offensiveness is a criterion for language being disallowed, everything should be banned.

ArtOfCode

@0celouvskyopoulo7 Yeah, you don't get to decide that.

0celouvskyopoulo7

19:00

@ArtOfCode Then why does that other person fucking get to.

I think this whole thing is offensive.

ArtOfCode

someone has in good faith said they found it offensive. You should in good faith assume that they do, and have some respect and stop using it.

0celouvskyopoulo7

I don't respect someone who thinks that.

Shashaank

Perhaps this jee thing has gone a long way now. I just referred to yashas that jee is difficult ( but I believe it is useless ) . Probably we should leave jee to IIT. Anybody got annoyed by remark on jee , I am sorry . But that is how I feel about jee and and I have the right to have my own thinking about jee. Anybody hurt of my comment , sorry !

ArtOfCode

@Shashaank this, @0celouvskyopoulo7, is an appropriate response. You are welcome to your own opinion, but avoiding shouting about it once you know someone finds it offensive is just common respect.

enderland

@0celouvskyopoulo7 SE isn't a place where people can say anything they want, regardless of whether they respect other people - respect is a fundamental piece of how people interact with others here, as codified in many meta posts and the "Be Nice" policy (which does apply to chat too)

Bernardo Meurer

19:02

Woah

Who deleted that

It was constructive!

ArtOfCode

Me. I've told you to stop using it. Stop.

0celouvskyopoulo7

Is any mention of jihad disallowed in any context? Can someone please tell why?

Mithrandir

Once someone has said something offends them, then yeah, it's not allowed.

ArtOfCode

@0celouvskyopoulo7 No, as long as you're having a constructive discussion about it. It was the comparison that was found offensive.

@BernardoMeurer calling people ignorant isn't nice either.

0celouvskyopoulo7

@ArtOfCode Am I allowed to say JEEhad if I mean it in a nice way?

ArtOfCode

19:05

@0celouvskyopoulo7 no.

0celouvskyopoulo7

Well why not?

"the spiritual struggle within oneself against sin."

ArtOfCode

I've told you to stop. Stop trying to find ways around that.

0celouvskyopoulo7

So context doesn't matter?

Why are so many mods here?

Mithrandir

@0celouvskyopoulo7 Because someone found it offensive.

0celouvskyopoulo7

@Mithrandir So why can't I find donuts offensive and flag it?

ArtOfCode

19:06

@0celouvskyopoulo7 because that's pretty clearly not in good faith

@BernardoMeurer That's not what I've said. I said someone found the comparison offensive, no matter how it was meant, so you stop using it

Bernardo Meurer

Why exactly are you banning it? I mean, just the fact that you're prohibiting us from using the word is very offensive, and borderline disrespectful on your part

ArtOfCode

Because someone found it offensive. Have I not made that abundantly clear by now?

Mithrandir24601

:37495923 Earlier, you said: "Protip, the whole notion of offense is made-up." Therefore, there's no way you could be offended by doughnuts, or you're downright lying

Sha Vuklia

Mithrandir

If you found donuts offensive, then it would be polite to stop... But there's a difference between something that is entirely harmless to most of the world and something that has lots of negative connotations, no matter how unfounded they might be.

Sha Vuklia

19:10

Guys, we know that $d\mathbf l\times\mathbf r=dl\,dr\sin\phi$, where $\phi$ is the angle between $d\mathbf l$ and $\mathbf r$ (my $\mathbf r$ is supposed to be this curly r in the picture). Griffiths seems to be saying that $d\mathbf l\times\mathbf r=dl\,r\cos\theta$. So I'm guessing $\sin\phi=\cos\theta$. I can't really see this tho. Is there a handy way to draw this to make it clear?

Shashaank

Woh! This chat is going places. Any Muslim or religion out there got offended , sorry ! Was just saying jee not good....

ArtOfCode

@Shashaank No worries, but thank you for apologising :)

Shashaank

@ArtOfCode yeah I felt that its better to apologise if someone got that to their religion , but you know I never said anything about you know what , and you know about what I said. Even though i feel nothing for any religion ( as it is mankind is destroying the biggest religion ) , I thought it be better to apologise so take the discussion back to Physics rather than all that ......( P.S- all the way I meant only "joint entrance....

Sha Vuklia

oh, it's solved!

Anonjohn

@ShaVuklia He isn't saying dl\times r is dl r cos \theta

Sha Vuklia

19:19

I should be just looking at the perpendicular part

Anonjohn

No, no

notice dl and r are perpendicular

Sha Vuklia

yea okay

no they're not!

Anonjohn

right.

Sha Vuklia

only the cos component is

I just realized

Anonjohn

the sin theta component is along z axis

its perpendicular to the plane contain dl

Sha Vuklia

19:21

yea, but I was talking about $\sin\phi$

I know it's not depicted

Anonjohn

so thats perpendicular as well

Sha Vuklia

but $\phi$ would have been the angle between the two vectors

but they don't consider that, because you might as well take the perpendicular component right away

Anonjohn

$\phi$ is 90 degrees

Sha Vuklia

oh that's true

huh

Anonjohn

He is taking the cos theta because he looks at only the z component of B

Sha Vuklia

19:22

ohhhh

Anonjohn

Since all other components cancel by symmetry.

Sha Vuklia

that's what you meant:P

right right

Anonjohn

Yea!

Sha Vuklia

ok, I did not realise that. thanks

Anonjohn

cool. No trouble :)

Sha Vuklia

19:28

oh lol, they literally explain this right after giving the formula

I should adopt the habit to continue reading for at least a couple of sentences when I don't understand something :P

@SirCumference btw, I can't unread "circumference" anymore, thanks to you:P I now read it as "Sir Cumference" in my head XD

Sid

@Kaumudi.H How was the exam?

ArtOfCode

20:45

→ 5 messages moved to Physics Meta

DanielSank

Hi, everybody.

Slereah

Hm

How to call the point of a fiber bundle

this paper calls it $m$

DanielSank

So call it $m$.

Slereah

21:00

Just checking if it's the most common notation

DanielSank

::shrugs::

Whatsa fiber bundle?

Slereah

the most famous one is the tangent bundle

The set of all tangent spaces

DanielSank

I've heard of that.

Long ago there were differential forms in my life.

Sir Cumference

@ShaVuklia My pronunciation is better tho

Slereah

So @DanielSank

when will we get quantum doom

DanielSank

21:06

@Slereah Probably, like, I dunno coupla weeks.

Slereah

I read a paper about quantum C, once

It was mostly like regular C

But with a few more bitwise operators

Vertical and horizontal bundles

In mathematics, the vertical bundle and the horizontal bundle are two subbundles of the tangent bundle of a smooth fiber bundle, forming complementary subspaces at each point of the fibre bundle. The vertical bundle consists of all vectors that are tangent to the fibers, while the horizontal bundle is then a particular choice of a subbundle of tangent bundle which is complementary to vertical bundle. More precisely, if π : E → M is a smooth fiber bundle over a smooth manifold M and e ∈ E with π(e) = x ∈ M, then the vertical space VeE at e is the tangent space Te(Ex) to the fiber Ex containing e...

wikipedia does call the point $e$

I'm having troubles picturing the vertical space

How many dimension is the vertical space?

wew

rob

21:52

→ 1 message moved to Physics Meta

Physics Meta

1

Q: Link to meta page for guide to MathJax commands under text fields

Wouldn't it be helpful to provide a link under the text box, shown while asking questions or writing answers, to the meta page dealing with MathJax commands: https://math.meta.stackexchange.com/questions/5020/mathjax-basic-tutorial-and-quick-reference It is not only difficult but also tiresome t...

discussion feature-request mathjax

Slereah

Oh no

they just keep multiplying

dmckee

Anyone know of either a source for (a) 7-12 HR diagrams for clusters of different ages all in the same format or (b) a suitable data set from which such a collection could be produced in a reasonable amount of time.

I'm teaching a gen-ed level astronomy course again in the fall, and I have a guided inquiry lab I tried out last year, but having my HR diagrams in slightly different formats confused and misled a non-trivial fraction of the class.

I'd like to reduce that as much as possible.

I actually have several things I want to improve labs and assignments for that class, but I've got to start somewhere.

heather

22:17

is there any significance to the term "flux vector"?

(this is in a paper about continuous chemical reaction networks, if that changes anything)

dmckee

@heather It's means the direction of flow of [what the flux is of].

rob

@dmckee We actually just made some of these for exactly this purpose. I have an email with raw data (magnitudes and color indices) in spreadsheets that I could forward to you in thirty seconds, or I can get the pretty final versions from my colleague who made them.

dmckee

@rob Ooooooh. I'm no gluten for punishment. I'll take the finished product if your colleague doesn't mind.

Do you have my contact info?

rob

@dmckee I've found your contact info before. I'll reach out tomorrow if I can't find it again.

dmckee

Great. I appreciate it.

Sir Cumference

22:23

Uh, I might be crazy, but were these buttons on our profiles always this black?

heather

hmm...direction of flow...not sure if that makes sense in this context.

it defines the flux vector as $\mathbf{u}\in\mathbb{R}^R_{\geq 0}$

where $R$ is a finite set of reactions over $\Lambda$, which is a finite set of species (chemicals/states)

which as far as I can tell means that the flux vector is in the set of functions from the reaction to the set of all real numbers such that the real number is greater than or equal to zero. Which I'm not really sure what it means, that's just the symbols.

dmckee

Yeah. I see your concern. I wouldn't call that a flux, but then I'm a physicist.

I think we can deuce the intended meaning, however.

Consider that the state of the system at any given time encodes a point in this (abstract space).

And if you know the point you can work out the current composition of the system.

heather

right, because conservation of mass.

dmckee

As chemistry occurs the composition of the system changes, meaning that the point representing the system moves.

heather

oh, wait.

dmckee

22:33

So they are talking about the direction of 'motion'.

heather

so when you said "And if you know the point you can work out the current composition of the system" were you assuming no reactions have happened in the meantime?

dmckee

@heather Yeah. If you know the instantaneous point, then you know the composition at that time.

heather

okay.

dmckee

And the 'flux vector' tells you how things are changing.

heather

so basically just the movement between one state to another? which is represented by a vector taken in a roundabout way from the set of all possible reactions?

that makes sense.

dmckee

22:36

Something like that. I'm basically guessing from the exceprt you offered. But try it out as a fist understanding and see if it makes sense.

heather

it does.

thank you.

nbro

Suppose you have two three-dimensional gyroscopes: one in the forearm and the other in the upper arm. Now, if you keep your elbow frozen, would this imply that, even if you move for example the shoulder, the two gyroscopes would detect the same angular velocity, right?

Bernardo Meurer

@dmckee I need your algo expertise :P

dmckee

@BernardoMeurer You make that sound very impressive. I'll give it a try.

Bernardo Meurer

@dmckee github.com/bemeurer/prog

Do you have access to that?

dmckee

22:48

Never a good sign when we start with a github link. I might have to think!

@BernardoMeurer 404.

Bernardo Meurer

Hehehe

@dmckee Do you have a github account?

nbro

If that's true, it should imply that, no matter where we place the gyroscopes on the forearm or upperarm, the angular velocities should not be affected by translations, since they are vectors...

dmckee

@BernardoMeurer I think so. And my computer might even know my credentials. Give me a minute.

Bernardo Meurer

@dmckee Sorry for the hassle, I need to keep the repo private to stop my classmates from snagging my code

dmckee

OK. I'm signed in, but I am still getting 404 on your link.

Is that what you need to authorize me to look?

Bernardo Meurer

22:51

@dmckee I gotta add you, yeah, one moment

github.com/bemeurer/prog/invitations

@dmckee ^

dmckee

I'm at a friend's place in the country and internet slowish and high latency here.

OK. I can see the repository.

Bernardo Meurer

Check utils.c

dmckee

... and project2 has been recently modified. Drilling down.

Bernardo Meurer

Inside project02, yeha

In the bottom there are two sorting functions for doubly linked lists

They are merge sort implementations, I figured it was best

Could you take a look and see if my implementation is reasonable?

(They are identical so only look at one)

dmckee

I'll try. You've gotta know how sensitive routines on linked structures are.

Bernardo Meurer

22:55

Yeah, it's a PITA that they make me use linked lists for this

dmckee

@BernardoMeurer OK. I don't understand how your divide step is working (I'm only looking at the station code right now). Nor do I see a recurrance.

Merge source is a recurrent strategy, and for it to be maximally efficient the division should be roughly into two even parts at every stage.

If you divide into 1 and N-1 at each state it's O(n^2).

heather

if I want to do actual work with chemical reactions (not just retrieve values) what's a good python package (must be compatible with python 3)

dmckee

So, first question, can you write a merge-sort on an array of integers?

Bernardo Meurer

@dmckee Eh, I think so, I didn't try really

I just wanted to get this over with so I found a pseudo code and some old C implementation and hacked it together

dmckee

On a block scale a merge sort looks like ...

Bernardo Meurer

23:05

Because I figured merge sort was the best algo for the job

dmckee

1. Divide the range into lists A and B

2. mergesort sublist A.

3. mergesort sublist B.

4. Recombine the lists maintaing order as you do so.

Bernardo Meurer

Okay, that makes sense

but on a linked list

that would mean that I need to go over the whole list every time!

That's not practical

dmckee

@BernardoMeurer My guess is that it is a good choice provided you can devise an efficient division scheme. With a double linked structure perhaps working both ways around the list until you meet yourself and then partition from head to the meeting place.

Bernardo Meurer

Yeah, I can do that but then I do O(n/2) every time I find a half

That will be no good because I need to find many lists

dmckee

@BernardoMeurer You will have to walk the list each time, but that still only adds up to O(n) so it doesn't effect the limiting order.

Remember that 1 + 1/2 + 1/4 + 1/8 ... = 2. That kind of divising and count scheme is O(n).

Bernardo Meurer

23:09

Ah, true

nbro

Well, the usual implementation always (not just "roughly") divides the list in half. In the case of odd size, you arbitrarily choose which sub-list to make bigger.

dmckee

^ That.

Bernardo Meurer

Hmmm

Is merge sort even the right approach?

I just want to quickly sort this damned list

nbro

Merge sort isn't the easiest sorting algorithm to implement in C since you need to care about new memory allocation. Quicksort is a lot easier to implement.

dmckee

I don't actually have a brain-file marked "good sorting algos for linked lists".

nbro

23:11

But it depends on your knowledge of the algorithm.

dmckee

Look on Stack Overflow.

Bernardo Meurer

@nbro You cannot quicksort a linked list, can you?

@dmckee Shame on you :P

DanielSank

Why not spend one run through the linked list building up a hash table of all the elements?

Bernardo Meurer

I think the Linux kernel does heap sort

THINK

DanielSank

Then you can treat it like an array.

:-P

nbro

23:12

Btw, there also exists an in-place merge sort algorithm, but you may not want to implement it... but if you're curious you can have a look at the source code of the PintOS. There you find an implementation of it.

Bernardo Meurer

@DanielSank I might just end up doing that

dmckee

Heh ...

1

Q: Merging two Sorted Circular Singly Linked Lists

I am practicing some algorithm problems before an exam in a C language course and and I got stuck (for at least 3 or even 4 hours) at this question which I don't know how to answer: You have two circular singly linked lists that are already sorted, you have to merge them and return the head of t...

c algorithm singly-linked-list circular-list

@BernardoMeurer Same problem as with merge sort. You have to do the partition.

DanielSank

@dmckee I've seen that problem before.

dmckee

But you don't do swaps like you do in an array, you move them to sublists.

nbro

@BernardoMeurer What made you think that you can't quicksort a linked list?

Bernardo Meurer

23:14

@nbro I said it, and then I realized it was stupid

I do that a lot, don't worry

DanielSank

^

Bernardo Meurer

@DanielSank Ay, screw you :P

DanielSank

lolol

dmckee

@DanielSank We don't. We do it too.

DanielSank

@dmckee We don't?

Don't what?

dmckee

23:15

Sorry. I clicked the wrong line. That was aimed at @BernardoMeurer. We don't worry.

Because we have quickly correct dumb thoughts, too.

With experience you learn to wait for the correction before saying them out loud, of course.

Bernardo Meurer

@dmckee I want to make a merge sort in OpenCL

With arbitrary precision arithmetic

Fuck the rest of the project :P

nbro

lol

dmckee

Then run it with a potato. We could have guessed.

nbro

you can't, I think, OpenCL does not allow recursion...

but you may want to implement it another way

GPU programming usually does not allow recursion

Bernardo Meurer

@dmckee I don't currently own a potato computer :(

nbro

23:18

15

Q: Why is recursion forbidden in OpenCL?

I'd like to use OpenCL to accelerate rendering of raytraced images, but I notice that the Wikipedia page claims that recursion is forbidden in Open CL. Is this true? As I make extensive use of recursion when raytracing, this will require a considerable amount of redesign in order to benefit from ...

raytracing opencl gpgpu

Bernardo Meurer

I have my FPGA with a custom CPU that runs linux though :^)

DanielSank

\begin{rant}

I have spent almost a week trying to get a Hello, World website to run on Google App Engine.

No success.

nbro

I must say, though, GPU programming is a lot of fun

DanielSank

\end{rant}

Bernardo Meurer

@nbro I was completely joking, but I appreciate the resource, that as nice to read about

Jaime Gallego

23:20

@BernardoMeurer I wonder if gcc would be smart enough to remove that facade linked list and use an array directly :P

Bernardo Meurer

I want to try and implement schonhage-strassen in OpenCL with GMP

I just need to not be lazy

@JaimeGallego Lol, maybe

GCC probably not

dmckee

My expeience with [powerthing]s that won't [run simple thing] despite lots of trying is that there is usually some assumption about the underlying model that the programmer of [powerthing] made and I haven't.

Hate that.

Bernardo Meurer

ICC would solve the halting problem AND change the list into an array

dmckee

^ And automatically vectorize the code, too.

nbro

lol

dmckee

23:24

Heavy rain here. Internet going to pot.

Bernardo Meurer

@dmckee How do I make fgets() not read those trailing \ns in stdin and "skip" user input?

Should I just use scanf()?

I just hate scanf

dmckee

The best way is by not using fgets.

nbro

It also started raining here...

dmckee

But I imagine that is not an option for you.

Bernardo Meurer

@dmckee It is

I chose fgets() because I like how it avoids buffer overruns

What else should I use?

dmckee

23:29

@BernardoMeurer Well, worst case you can scan the input yourself. One character at a time and define the behavior you want. Pain in the butt.

Bernardo Meurer

@dmckee I'm not a masochist

:P

dmckee

I usually use getline, but you've indicated before that you're not allow that, and besides it reads the newlines.

If I had to write a scanner that wasn't very simple I might actually use flex, but again I fear you are not going to be allowed that.

Bernardo Meurer

Ah, I am!

Here I am

Aha!

Nice reminder

Transcript for