What's more difficult, programming or theoretical physics

Programming, and I mean this unironically.
Ever heard an engineer say something like "Our theoretical models of physics just suck. The margin of error is too damn high and it costs us millions!"

Programming is close to impossible.

Attached: Selection_109.png (842x206, 159K)

wew lad

OP pic is chemistry, so for OP it's definitely physics.

Programming is for brainlets

Theoretical physics.

p00j33t can program just fine.

Pic not related, is just high school chemistry

>implying programming is anything remotely like

shut the fuck up m8

Definitely theoretical physics.
CompSci is the brainlet STEM subject.

Developing a system that doesn't crap itself is seriously difficult, type theory is just a brainlet filter

Attached: boV2nqN.jpg (433x432, 35K)

This.

/thread

That has literally nothing to do with the fact that the majority of programmers aren't dealing with shit like that pic.

Ok

You can import programming libraries but you can't import physics knowledge.

>libraries spontaneously sprout into existence and are not written by someone

finding a bf

Based and redpilled

I thought that was common knowledge.

isn't this chemistry more than physics?

No.

Theoretical physics is hard math. Computer science is simple math. Programming is glorified legos.

Theoretical physics, of course.

Theoretical programming (aka computer science) is probably equally difficult to theoretical physics.

The kind of Lego where you have sets with 10 trillion pieces and you make some of them yourself and everything is held up mostly by sheer luck

Programmers can be replaced by 70 IQ pajeets from the third world. Theoretical physics can't. CS brainlets have the most inflated sense of self worth out of any profession. It's on par in difficulty with accounting.

> Theoretical physics is hard math
Can be, but you're lucky to usually focus on a single/small set of problems.

YMMV if you're currently fixing up a "Theory of everything" that works.
There are hints that that's about as hard as doing code that is objectively proper, sound, and provable to the maximum extent possible.

> Computer science is simple math.
No. It's all maths, really. With a bit more of an intent to struggle to actually use it than the usual theoretical maths has.

Oh really?

Theoretical physicists can be replaced by a rock or drop of water, because no one actually needs them to do or fix anything right now.

I just want to kms doing dumb mindnumbing grunt work but love solving hard shit to the point of happily noliving over it. Both physics and coding have both type of work so it depends what you do more than what field you are.

I talked once to a guy working in CERN who complained about wasting all his time going through the data trying to find something interesting to write about with no result for several years now(tried to get doctor). At that point his life was no better than doing dumb webdev shit.

How exactly does that prove that programming is easier than theoretical physics? That's an entirely different argument - you can talk about the economics of the professions until your face goes blue but that won't prove or disprove difficulty. I wouldn't expect much more from a programmer though.

There is a significant bias towards thinking computer science is simpler, because the simple computer science tasks are actually useful, while the simple physics tasks are not. "Simple" physics is known as engineering, "Simple" computer science is known as programming.

If you do research in either field they will be equally hard. But with computer science you are likely to actual find useful results and be able to code something up yourself without huge funding. With physics, unless you discover something like the laser your results will not be that useful until years later integrated into a larger theory. That is why I find computer science much more fulfilling than physics. But no field is harder or easier than the other, the state of the art for each field always gets harder.

Try theoretical mathematics...

Attached: 1543795008775.jpg (450x701, 61K)

You can import physics libraries. A shit ton of physics is done via programming so the original question is stupid, physicists should know both.

>state of the art computer science is the same as "programming"
lmao

It is, you mongoloid

I did not say or imply that.

A good 90% of theoretical physics is just mental masturbation. Programming is actually useful. It's a tool that's used to develop real world solutions to real world problems. Figuring out what inside a black hole or the center of the sun matters fuckall.

Programming is harder because there are real life consequences for being wrong.

Attached: 1543705224967.jpg (750x559, 33K)

please say on topic of the thread. if you recall:
>What's more difficult, programming or theoretical physics
This thread isn't about CS research

>"Simple" computer science is known as programming.
You mean, "Simple" Electrical Engineering is known as Computer science?

>Programming is harder because there are real life consequences for being wrong.
The risk of getting fired from your McJob for messing up a burger must make being a fry chef really hard in your mind.

I remember saying this when I saw it at my dads. hearty keks how dumb this show is.

More like theoretically useless lmao

OK, theoretical physics is more difficult than programming. Its common knowledge.

>What is Electrical Engineering.

This is an example of a computer science publication. larsblackmore.com/iee_tcst13.pdf And this isn't even an academia publication, its a SpaceX one.

>Mechanical Engineering the paper
Very computer science of you

So many physicsniggers that think complex analysis and differential equations are hard math. Much of the pure math involved in fields of CS makes that math look like elementary school shit.

>larsblackmore.com/iee_tcst13.pdf
>Optimal aerospace controls problems are an example of a CS publication
Imagine being this pathetic. This is why aerospace engineers et. al laugh at your typical CS person who doesn't know how to stay in their own lane.

Fucking kek. Any theoretical physicist will have to know how to program, in large part to get numerical solutions to the stuff they're doing. Barely any problem can be solved completely analytically, so they have to know their way around a computer.

Basically, physicists know physics AND programming. In fact, most theoretical physics students at my uni end up in jobs working as programmers. By comparison, how many programmers do you think could do theoretical physics?

>not doing any of them, correcting those who do both: the post

A theoretical physicist uses programing as a single tool in their toolbox. Programmers profession revolve around a single tool. It's like a chisel subject matter expert saying that since they know a lot about chisels, they're Michelangelo

If you are a mechanical engineer, I can understand how you might confuse some of the problems addressed in the paper with physics or dynamics, but the entire point of the paper is how to convert a certain type of optimization problem int a convex one. This belongs strongly in the field of computer science, and specifically in the sub field of optimization.

Specifically, rocket thrust cannot be throttled below a certain level, and a lower bound like that cannot fit into the regular Second Order Cone Program model. Regular SOCP constraints have the form norm(Vx)

Nigger, I do robust H control. It's an aerospace controls paper through and through. Stop pontificating about shit that you don't understand

Optimisation is indeed not part of CS, it's part of all Engineering, Physics and Math What your paper has problem with is F=ma and whole paper revolves around that problem, making it Mechanical Engineering and all of it's derivatives problem.

>be theoretical physicist
>get free money
>fuck about for a bit
>"I believe X because Y"
>another guy comes along
>"no X is like Z because A"
>oh well, time to try again later

>be programmer
>work on embedded systems
>build aircraft control program
>it malfunctions mid flight
>plane has to make emergency landing
>people could be hurt or killed
>plane is fucked
>all the passengers sue or take settlements
>millions of dollars lost

Yeah, it's real fuckin hard to be some egghead faggot working in a university.

Attached: 272.jpg (496x599, 36K)

>hiring a cs brainlet to do aerospace controls
there's your problem.

Then we're both right. It's a computer science publication about an aerospace control problem.

>Optimisation is indeed not part of CS
False. From the Wikipedia page on optimization:
en.wikipedia.org/wiki/Mathematical_optimization
"In mathematics, computer science and operations research, mathematical optimization or mathematical programming, alternatively spelled optimisation, is the selection of a best element (with regard to some criterion) from some set of available alternatives."

>What your paper has problems with is F=ma
No, F=ma is not the main problem. It does pose a problem for regular SOCPs (which the paper discusses) because of the varying mass over time, but this is addressed. The main problem, as I said before, is conversion from non-convex to convex. As the title of the paper states:

"Lossless Convexification of Nonconvex Control
Bound and Pointing Constraints of the Soft
Landing Optimal Control Problem"

>theoretical physics is often useless
>I've placed a number on how much of it is
People vastly underestimate how important basic research is. It's not just the layman like you user. But professionals in fields.
There should be some sort of agency that receives fundamental industry problem definitions and suggests solutions. It sounds a bit daft but incidents like how quaternions weren't used in 3D graphics for a long ass time because people hadn't realized their usefulness in that field (I believe it was at least a decade of dealing with stupid problems like gimbal lock). While mathematicians knew full well their utility in that sphere. There shouldn't be such a major lag there. If something that fundamental is left unnoticed for such a long time you have to wonder just what treasures lies in basic research that the right people haven't heard about.

A patron uses an artist/etc as but a single means to beautifying his estates. It's like Sargent saying that since he knows a lot about painting, he has the intellect to go around paying other people to paint things for him.
Pulling rhetorically-dishonest analogies out of your ass like they mean anything is fun, right? So much for the intellect of the Jow Forums theoretical physicist. ZZZzzzz

CS is just applied math. Talking this or that doesn't belong to CS is silly. There's no limit to what people teach in a cs course because if it runs on a computer somebody will consider it as CS and run with it. The same shit thought in physics / math department is thought in CS. The average codemonkey will never use it but it's there and some people do.

Intellectual niggerfaggots who have never gotten their hands dirty don't know what risk is. The world would probably be a better place without them.

>False. From the Wikipedia page on optimization
Yea, I realised my bad form after I posted, I wanted to say "not ONLY part of CS"
>because of the varying mass over time
How is this negative F=ma? If anything it proves my point.
>is conversion from non-convex to convex
Yea. conversation of non-convex F=ma to convex F=ma

That's not how theoretical physics is. And it's utility goes beyond being right or wrong eventually. It's a huge time save for experimental physics. With just experimental physics you'd try to brute force the universe. Good luck.

Differential geometry is considerably harder than diff eq.

There is some truth to this, as obviously the average computer science student will know more programming than the average theoretical physics student. But to say physicists regard programming as just "a single tool" is just horribly wrong, and I have to wonder what exactly your idea is of what physicists do with computers. You might be surprised by how much emphasis there is on the subject in a physics curriculum, and most students choose to take even more programming courses.

Compare programming with mathematics, which is also often described as nothing more than a tool of physicists. By contrast, a mathematician would have an easier time moving to the field of physics than most other professions, and a physicist could with a similar amount of effort make a move to mathematics. Can the same thing be said for a programmer?

This is a dumb question. Physicists program, saying what is harder makes no sense. If you are trying to compare theoretical physics to CS, then it depends on the area. There is a lot of bullshit research in each field.

Are you only talking to first year students?

Except the artist and theoretical physicist are both the critical actor in the scenario. A physicist can program, a programmer can't do theoretical physics. An Michelangelo can create art without a patron, a patron cannot reproduce Michelangelo. Nice try though. Try to not let your emotions get in the way next time.

>Yea. conversation of non-convex F=ma to convex F=ma
Not exactly. Since I am doing research in this area I can post some code for it. This code is done with Julia and the NLopt solver. The first block is the constraint for physics. The second block is the constraint for thrust. As you can see, I am allowed to use a norm(...)

There have been so few technological advancements in the last two decades or so that it's disgusting. And the advancements that are made just allow for retards to create worse and worse regressions. For example, faster processors and more RAM just gets used for more JavaScript and autoplaying videos embedded in web pages.

And half the new tech that's being developed isn't even used for anything good. It's all squandered on shit like cleaning water in Africa so that the nigger populations can explode and cause more problems. These people are weak links with bad ideas that will end up killing this entire fucking world, or throwing us back to the stone age to start all over.

Attached: ede2b35ab00e4365be8e199112a17567.jpg (363x550, 21K)

Also, m is not mass, m is the model. This code does not account for decreasing mass over time, but it can be done and kept linear with much effort.

>just F=ma
Nobody said that, F=ma is never "just" F=ma it's one of the most important things in whole physics.
What you are describing in this problem is infact F=ma if you cannot by the looking of your own variable realise that, then it speaks volumes about your understand of physics.

Interesting point. Although it's easier to get 'close enough' or to get started with finding the result you're looking for than with physics. There's less building blocks that are required to get you going with programming.

You said
>What your paper has problem with is F=ma and whole paper revolves around that problem
Which is untrue.
>making it Mechanical Engineering and all of it's derivatives problem
Which is true, it is an engineering problem but it was solved using the the theory of convex optimization, which is not an engineering problem. It is a mathematics and computer science problem.

Programming theoretical physics.

I wanna learn all math shit, i want to do it for the rest of my life.
But this shit doesnt give me money, and i hate my job as a codemonkey for engineers that get all the credit.

I cant find my way out of this and is fucking stressing me because life fucking sucks. Any advice?

Attached: anigif_enhanced-buzz-22769-1379695964-2.gif (478x369, 263K)

Yeah it is. I fucking love logos and love programming

>Which is untrue.
Wrong
>onvex optimization, which is not an engineering problem
Wrong

Attached: 1483194534635s.jpg (125x43, 1K)

Quit your job.

mitrikitti.fi/opthist.html
History of optimization, no dirty engineers in sight.

this

CS undergrads are the dumb-dumbs among the STEM students. Ironically, they always seem to act like they're the smartest.

Trying to claim that using convex optimization to solve a problem turns an optimal aerospace controls problem into a CS problem just because "teehee we use convex optimization too" and "but but the paper was posted on IEEE" is the most autistic straw grab I've ever read on Jow Forums.

>t. someone who literally does robust controls in aerospace

>Programming is glorified legos.
Sounds about right.

Amazing. How do you think this proves any of your statement I don't know.

By "literally does" do you mean plugging your problem into a solver written by computer scientists? Categorically different from proving a convex problem solves a non-convex one, as Blackmore does.

>now check this history of arithmetic, no dirty engineers in sight. checkmate

If it was an engineering problem, I would expect engineers would be the ones pushing the field forward, not mathematicians and computer scientists.

Do physicists ever have to deal with undecidability? Are there some results that are unprovable? Is there an equivalent of Godel's theorems in physics that will eternally cuck the field, like some result that proves 'there is no possible unifying mathematical model of physics'? I am genuinely curious.

Whole of quantum mechanics?

yeah, because most problems are automatically in the right form that you can plug them into a solver, right?
>reee someone else in the CS field also uses LMIs and LFTs this just doubly proves it is still a CS problem! I read a spaceX paper posted on IEEE so by the transitive property I'm a rocket scientist!
The hubris of CS students is astounding.

Point me where I claimed I was a rocket scientist. You two are literally trying to prove the field of convex optimization falls under the category of engineering just because you use it. Mathematicians invented optimization, and later computer scientists found ways to solve ever more broad categories of it efficiently. Now, it is used in finance, engineering, sales, manufacturing, shipping, and control software like you use. That doesn't make optimization part of those fields, it just means they use it.

>You two are literally trying to prove the field of convex optimization falls under the category of engineering just because you use it.
No, you're trying to claim that optimal aerospace control is CS because it CS people also occasionally use convex optimization. We're both correctly calling you a faggot and informing you that it is an optimal controls problem.

>hurrr hurrr hurrr I didn't say that
>If you are a mechanical engineer, I can understand how you might confuse some of the problems addressed in the paper with physics or dynamics, but the entire point of the paper is how to convert a certain type of optimization problem int a convex one. This belongs strongly in the field of computer science, and specifically in the sub field of optimization.
It's an optimal controls problem and controls engineers deal with these kind of problems and they convert non convex problems into convex representations. Stop trying to associate yourself with aerospace engineers in an attempt to win a troll debate.

Attached: Lj2XHPl.gif (1114x174, 220K)

Yeah you're right, it is more of an engineering paper than a CS paper.

Reall though, CS people don't "occasionally use" convex optimization, they literally are the field. The only reason you can solve your control problems is because CSists proved important properties and made the solvers. Mathematicians too but not engineers.