The h Bar

@0celo7 Well, why not? This way, money stays a meaningful resources to worry about all the time - it actually irks me that it becomes so inconsequential in most other games

0celo7

2:05 PM

in TW3, enemy levels are fixed

ACuriousMind

@0celo7 I'm curious to learn which way you think that is.

0celo7

@ACuriousMind We fundamentally disagree on this issue then

Slereah

Gold Diggers of 1933 - "We're in the Money"

►

0celo7

@ACuriousMind Sword and Quen. Alt-doge, power attack. Quen for if you get hit. Invest points in Quen so you can stun enemies.

Slereah

Unfortunately a lot of indy sandbox RPGs suffer from shitty final level syndrome

aka the shinfles

that's because they usually have a lot of good ideas but then they run out of money

for instance Arcanum or Vampire Bloodlines

RPGs with a great first half but then the game runs out of steam

0celo7

2:07 PM

I also find that having a sword with a high critical damage boost works best.

DPS through the roof

And invest points in the critical hit chance perks

ACuriousMind

@0celo7 You'd be amazed how easy the game is if you get Igni up to 100% ignite, then. The first few levels Quen is kind of a must, but after that I find that both leveling swordplay and leveling pretty much any sign is viable - what's not good is spreading your points around or leveling alchemy too much :P

0celo7

@ACuriousMind I don't even know why alchemy is in the game

@ACuriousMind Ignite is which level of the perk?

The only potion I use is Swallow

Slereah

@0celo7 Because other games do it

The Great Powers produce steel, China must produce steel

ACuriousMind

@0celo7 No perk - it always has a chance to ignite enemies, which means they'll stand around screaming and taking damage instead of attacking you. I think some perk might increase that explicitly, but it's mostly about increasing sign intensity

0celo7

It's like why was Destruction magic in Skyrim?

There was literally no viable way to use it

Slereah

2:09 PM

Yeah it was pretty shit

ACuriousMind

@0celo7 Because the potions are an integral part of witcher lore

Slereah

Although I don't think destruction magic was ever really that good in the Elder Scrolls

ACuriousMind

Most potions feel very weak, I agree there

Slereah

I've played a lot of it and it's always kind of shit

0celo7

When I do my next playthrough of Skyrim, I will try a mage

ACuriousMind

2:10 PM

Decoctions are actually pretty sueful, though, at least some of them

Slereah

I never played a guy who was a destruction wizard

0celo7

Going to hunt down magic mods on the Nexus

@Slereah my first playthrough I started off as magic, but switched to warlock, then to warrior

the game was too hard with pure magic

and not in a good way, it just became impossible to kill enemies

especially the Draugr which have infinite HP

ACuriousMind

@Slereah It was really good in Morrowind - if you had the Destruction skill and mana to use it often enough, which without alchemy shenanigans was out of reach for most characters :P

Slereah

also you level up magic pretty slow

@ACuriousMind I mostly used magic for teleport and levitation

that was the good stuff

0celo7

@Slereah See that's the thing. You could never level magic damage, only use more powerful spells.

Slereah

2:12 PM

@0celo7 The Apple model

0celo7

But with swords you got both better swords and higher damage on top of it

Slereah

Gotta buy a new spell every year

0celo7

Sadly the Skyrim Special Edition broke many of my favorite mods

I should wait until they have an SKSE for that, then mod it

Slereah

The one skyrim mod I usually get is "infinite carrying capacity"

I kinda hate the inventory system

"Sorting my items" is not a fun activity

0celo7

that's true

And in Skyrim you could find all of your crafting components in the wild

Things like Dimeritium ingots are impossibly rare in TW3

So you end up spending 4,000 gold

Slereah

2:18 PM

really in Skyrim I could just defeat the Empire and the Stormcloak just by getting a monopoly on blacksmithing

I just take over all blacksmithing and devaluate their currency

You just go raid a dwarven ruin every once in a while

Make a million dwarven bows

and buy out the empire

There should be a game like that really

you can either defeat the great evil

or just buy him out

0celo7

@Slereah civilization?

Slereah

Civilization isn't the hero's journey, though

I'm trying to think of an example where you can do it

Oh I know

Guild 2

that one is all about ancient capitalism

store.steampowered.com/app/39650/The_Guild_II

I remember a game that was also like

Running a ganster empire

and you could just buy out big chunks of the city

it was a fun game

0celo7

2:37 PM

@ACuriousMind I need physics help

Slereah

then why do you ask a math guy

0celo7

but he is a physician, no?

@ACuriousMind I want a uniform damage profile for my ions

Does that mean I want a heavier ion?

For Ni the Bragg peak is pretty pronounced

Sir Cumference

@0celo7 Shouldn't I take diff equations and linear algebra first?

0celo7

@SirCumference For the GR you want, I don't think you need differential equations at all

If you want to learn what I do, then you'd want a whole lot more math

there are levels

not necessarily in difficulty, just in amount of math needed

ACuriousMind

@0celo7 I'm afraid heavy ion physics is not exactly my forte :P

Sir Cumference

2:46 PM

@0celo7 What kind of stuff do you do?

Slereah

For basic GR just learn tensor calculus

that already covers most of it

0celo7

@SirCumference Existence and uniqueness of solutions of the Einstein equations

Sir Cumference

That sounds important.

Slereah

It's not

He doesn't contribute to the GDP at all

0celo7

@SirCumference Nothing any of us do is important

Sir Cumference

2:47 PM

I mean, if I want to be a cosmologist, I probably need a strong grasp of the subject

0celo7

Except maybe @JohnRennie

Sir Cumference

Gtg more class

John Rennie

@0celo7 all I do is fix servers these days.

Secret

Fun fact: There exist one thing that makes my anger explode, and said thing is much worse than computer error troubleshooting

->is a fancy way of saying that computer errors are very bad for my psyche

0celo7

@JohnRennie Are you responsible for someone making money?

Slereah

2:49 PM

For a good GR grasp learn topology, differential geometry, group theory, analysis and linear algebra

0celo7

@Slereah group theory?

John Rennie

@0celo7 the company I contract for makes money as a result of my work, if that's what you mean.

0celo7

@JohnRennie good enough

Slereah

@0celo7 Lorentz group and whatnot

ACuriousMind

@SirCumference For comparison, perhaps note that the field of fluid dynamics has worked pretty well so far but the existence and uniqueness of solutions to its fundamental equations - Navier-Stokes - is still an open problem. It's "important" in the sense that it'd be very nice to know, but (perhaps somewhat surprisingly) the knowledge of the existence of solutions in a general case is not that important for physics

0celo7

2:50 PM

congratulations, you are worth more than us

Slereah

@JohnRennie is the true GDP man

Everyone needs toothpaste

He contributes more to society than any of us

John Rennie

I suspect you're worth more than me. Young organs fetch more on the dark web than old ones do.

0celo7

Crap

I left a monte-carlo simulation running all night

how does my laptop do calculations when it's closed

ACuriousMind

@JohnRennie Annnnd...you know that how? ;)

Slereah

My organs are full of caffeine

0celo7

2:52 PM

@ACuriousMind less abuse

it's basic science

John Rennie

@ACuriousMind it's the only way I've stayed alive this long despite a lifetime of dissipation and self abuse.

0celo7

That's way too uneven

gold is slighly better

John Rennie

@0celo7 Wow, that's a pretty classic Bragg curve!

Slereah

>using windows for data analysis

Disgusting

0celo7

@JohnRennie Yeah, a 74,000 ion simulation.

AccidentalFourierTransform

2:55 PM

>data analysis

Disgusting

ACuriousMind

>quoting things like this

Disgusting

TROLL SLAYER

>

John Rennie

@0celo7 ah, that's simulation not an experiment. OK, I'm less impressed :-)

0celo7

@ACuriousMind do you even know what it means

@JohnRennie I am doing the experiment next week, but I need to select my ion source

ACuriousMind

@0celo7 what what means?

0celo7

2:57 PM

@ACuriousMind >

@JohnRennie Let's try U :)

Wow

How do I get a U source

ACuriousMind

@0celo7 "greater than" :P

But I guess you want to hear that it'S the typical style of quotations on imageboards, in particular 4chan :P

TROLL SLAYER

"is greater than"

0celo7

How did that s get capitalized

Do you need to hit shift for '?

ACuriousMind

@0celo7 Yes

Balarka Sen

@0celo7 like thiS

0celo7

2:59 PM

o.o

ACuriousMind

German keyboard - we usually have little use for the apostrophe :P

AccidentalFourierTransform

0celo7

@ACuriousMind I think he is suggesting I am retarded

AccidentalFourierTransform

I ThiNk He IS sUgGesTinG I Am reTardED

ACuriousMind

@AccidentalFourierTransform Stop with the trolling, please :P

TROLL SLAYER

3:01 PM

please

AccidentalFourierTransform

fiiiiine

TROLL SLAYER

thank you

:-)

hi @heather

0celo7

@Slereah I'm doing it on a mac, to boot!

Apple laptop plugged into a Dell monitor using Windows

ACuriousMind

Steve Jobs rotates in his grave

heather

hello

TROLL SLAYER

3:06 PM

how are you?

heather

good, yourself?

TROLL SLAYER

fine thanks

0celo7

@ACuriousMind Are our usernames allowed to threaten people?

Like, "Mod Slayer"

0

Q: Mind in worlds with number of time directions not equal to one

In our universe "mind" can be roughly defined as development over time. But what is life and/or mind in a universe with the number of time directions not equal to one? Is it development in each of time directions? in at least one of them? Well, what's about universes having only space dimensio...

@Slereah something for you

heather

Is there a length limit to usernames? Like, the chemical name of titin?

0celo7

@heather that would probably break the site

heather

3:09 PM

exactly.

0celo7

I doubt they have multiple KB for the usernames

ACuriousMind

@0celo7 No. It's debatable whether "mod slayer" constitutes a threat, though - usernames are, like your profile, the area where we are most reluctant to meddle

@heather I think it's 36, but I can't find anything on meta on it

If you want to really find out, just try it ;)

Secret

If you only have space dimensions and no time... How can anything MOVE? And having life without movement, change, etc. Well, it would not be life as we define it. And for multiple time dimensions, you would first have to find a way to make cause and effect work with them... — Florian Schaetz 5 mins ago

I think we have discussed before on the impossibility of movement without a time variable, but I don't remember the exact location in the chat transcript

AccidentalFourierTransform

2

Secret

will just pull the lever out of boredom

heather

3:13 PM

@ACuriousMind it's 30

you can't even have supercalifragilisticexpealidocius =/

TROLL SLAYER

supercalifragilisticexpealidoc is close enough

Balarka Sen

bababadalgharaghtakamminarronnkonnbronntonnerronntuonnthunntrovarrhounawnskawnto‌ohoohoordenenthurnuk

I want that as my username

TROLL SLAYER

8 mins ago, by heather

Is there a length limit to usernames? Like, the chemical name of titin?

Balarka Sen

I'll fight the authority for it

custom username is an individual right

TROLL SLAYER

ask on meta

good luck

AccidentalFourierTransform

3:20 PM

Uvuvwevwevwe Onyetenyevwe Ugwemuhwem Osas

►

heather

Technical Communication proposal on area 51

2

(if anyone wants to support)

John Rennie

Re this: if you have more than one time like dimension how do you calculate the evolution of a system? Is there an equivalent of a Langrangian that you can extremise?

ACuriousMind

@JohnRennie Point is: You don't. Equations "of motion" in more than one time dimension tend to be rather ill-behaved: They're ultrahyperbolic, and specifying initial conditions would no longer uniquely determine the future state.

John Rennie

@ACuriousMind Thanks. I wonder if there is a simple example that illustrates this type of behaviour ...

ACuriousMind

@JohnRennie Well, the example is the wave equation in such a signature

It's not hard to see that if I have $\Delta_s \phi - \Delta_t \phi = 0$ with $\Delta_s,\Delta_t$ the Laplacians for the spatial and temporal dimensions, respectively, that specifying spatial initial data no longer determines a unique solution if the number of temporal dimensions is larger than 1

0celo7

3:31 PM

This seems like something awfully specific

What class did you see it in?

ACuriousMind

I'm pretty sure I read it in one of Moretti's answers, actually

Currently searching

Hm, can't find it

0celo7

@ACuriousMind I must be missing something, is it supposed to be obvious that the Cauchy problem is ill-posed?

Slereah

@JohnRennie There's CTCs through every point in a ultrahyperbolic spacetime

0celo7

I don't think time travel has the same meaning when there are two time directions

Slereah

Well if you only allow particles to move in a timelike direction, particles can have worldlines that are closed

that's pretty important

John Rennie

3:45 PM

@Slereah is there a simple way to explain why that is?

Slereah

@JohnRennie take a small neighbourhood around a point of the spacetime

0celo7

@JohnRennie you can imagine having a 2-dimensional timelike slice

then do a loop in the slice

Slereah

Consider it to be $\approx \Bbb R^n$

Take the plane spanned by the timelike vectors

Draw a closed curve in that plane

0celo7

you can do a loop in a 2-dimensional slice

Emilio Pisanty

@Slereah wait, what's a timelike curve in that setting, exactly?

John Rennie

3:47 PM

But doesn't that mean at some point you are moving in the negative $t_1$ or $t_2$ directions?

Slereah

that's pretty bad behaviour for a spacetime

Emilio Pisanty

negative $v^2$?

Slereah

@EmilioPisanty a curve with a timelike tangent vector

@JohnRennie Well sure, but that's not forbidden

at no point do you take any discontinuous turn

0celo7

You mean non-$C^1$ turn

John Rennie

@Slereah then why can't I oscillate between positive and negative time displacements with one time dimension?

0celo7

3:48 PM

@JohnRennie Because that's not $C^1$.

Emilio Pisanty

@JohnRennie I guess the point is that there's no such thing as "negative $t_1$" anymore

Slereah

@JohnRennie That would require a discontinuous change in the tangent vector

0celo7

That's literally what I said.

Slereah

You can show that locally, it's impossible in a spacetime to change lightcone continuously for a timelike curve

John Rennie

@Slereah @0celo7 ah, OK, yes I see. Thanks.

Emilio Pisanty

3:49 PM

in (1+N)D you get future and past light cones

Slereah

even worse by the way, ultrahyperbolic spacetimes admit arbitrarily small closed timelike curves

Even pathological spacetimes aren't that bad

Emilio Pisanty

in (2+N)D those presumably form part of a single connected component

0celo7

Who says 1+N instead of N+1?

ACuriousMind

@0celo7 Mh, I think it is obvious that you can't get a unique solution in d dimensions by specifying data on a d-2 submanifold. What's not so obvious is that you also get an ill-posed problem if you try to pose the problem as "initial" conditions on a d-1 hypersurface in this space, but it's also true: Well-posed initial value problems are hyperbolic, cf. e.g. math.stackexchange.com/a/14663/143136 (and thus apaprently Hormander)

Emilio Pisanty

@0celo7 I do.

I'll choose any signature that will rile up people who get riled up about signatures.

Slereah

3:50 PM

There's actually TONS of problems with ultrahyperbolic spacetimes

It's quite fun to read about

How terrible they are

Emilio Pisanty

@Slereah examples?

0celo7

@ACuriousMind It's not obvious to me because I don't have an exact solution in mind.

Emilio Pisanty

oh, as in, ultrahyperbolic examples have problems

Slereah

yes

0celo7

I can believe it because the Laplacian in two dimensions is much weaker than in one.

Emilio Pisanty

3:51 PM

not as in there's problems where the spacetime is ultrahyperbolic

0celo7

So how are you seeing it?

Slereah

I've seen some proposals for ultrahyperbolic spacetimes, but they're pretty obscure and not too interesting

ACuriousMind

@0celo7 Well I can get a solution $g(t) = g(t_1)$ that only depends on one of the two time coordinates by solving the corresponding well-posed initial value problem in one dimension lower. But of course I can also do that for $t_2$, so I already have at least two distinct possible solutions.

0celo7

@ACuriousMind Ah, freezing one time coordinate.

That's what I thought.

Slereah

Max Tegmark did a fun paper on the topic, by the way

A paper on the anthropic principle and the number of spacetime dimensions

from the point of view of a single particle, multiple time dimensions don't change things too much, since it just follows its own worldline, but when you get interactions things get unpleasant

John Rennie

4:02 PM

Just to clarify my thoughts on this: if I have a single time dimension then I can solve a hyperbolic equation in $\mathbf x$ and $t$ if I know the initial conditions $\mathbf x$ and $d\mathbf x/dt$. Is that correct so far?

0celo7

@JohnRennie Perhaps

The technical conditions are not nice

You need conditions on the ellpticity of the spatial part

Slereah

You need a whole bunch of conditions

John Rennie

@0celo7 assuming the sorts of system physicists usually deal with on a daily basis ...

Slereah

But basically if your field is nice and your spacetime is nice

then yes you can

(up to gauge :p)

John Rennie

What I was getting at, is if I have two time dimensions and I know $\mathbf x$ and $d\mathbf x/dt_1$ and $d\mathbf x/dt_2$ then this isn't sufficient. Is that still correct?

0celo7

4:05 PM

Why are you using $\mathbf x$

Use $u$

John Rennie

to mean a vector

Emilio Pisanty

^ yeah

also, what are $d\mathbf x/dt$?

weren't we talking about wave equations?

0celo7

@JohnRennie As a partial differential equationist, use $u$ for PDE solutions

John Rennie

OK, OK, but back to my question ...

Emilio Pisanty

@Slereah See, that's just mathematician-speak. Just switch from a debit POV on rigour to a credit POV. Not "how nice are my hypothesis, and what kind of shiny results do I get?" but "Assume I have infinitely nice hypothesis, what's the shiniest results you got? You can just bill me for rigour later"

John Rennie

4:07 PM

If my initial conditions are $u$ and both time derivatives of $u$ then is it correct that this is no longer sufficient to predict the evolution of the system?

Secret

Do we also not get a unique solution for ultrahyperbolic PDEs even if we specified TWO initial conditions, one for each time variable?

Slereah

I think free particles may work fine?

Sir Cumference

@JohnRennie @Jim @ACuriousMind Well say I want to research the earliest periods of the universe. Wouldn't I need a very strong grasp of math, beyond what is required for GR?

Emilio Pisanty

@JohnRennie Well, if you fix $u=\frac{\partial u}{\partial t_1}=\frac{\partial u}{\partial t_2}=0$ at some $t_1=t_2=0$ hypersurface, you can presumably still get nontrivial dynamics

i.e. just assume the spatial part is constant

and solve the Laplace equation on the time part

Slereah

Free particles are usually okay even in pathological spacetimes

0celo7

4:09 PM

@ACuriousMind I can't actually find that theorem in Hormander...

Slereah

Although some spacetimes do have closed timelike geodesics, too

But I think free particles in $\Bbb R^{2,n}$ are fine

They'll just be straight lines intersecting the spacelike hypersurface

Secret

I just have this naive guess that the difference between nice spacetimes and spacetimes filled with CTCs is that you have a lot more constraints on the dynamics since consistency will mean that the dynamics will depend on a time interval instead of a point in time

If you have CTCs, then you must have some form of retrocausality, and since the simplest model on dynamics of this is to ensure events evolve consistently at all point, then in some sense you can still have dynamics, but it is a lot more restricted since it has to obey not only constraints imposed in the past, but also the future…

So, you can theoretically still have interactions, but I have no idea how will one account for all the constraints imposed due to worldlines looping back

Slereah

The constraints can be strong enough to forbid all fields

Secret

That's also what I am suspecting, but I am not 100% certain

Slereah

I am

There are examples

There are no scalar field of mass $m$ on the timelike cylinder unless there's a relation between the mass and the charactertistic dimension of the spacetime

It's not too hard to show, really

Same proof as wavelengths on the torus

@JohnRennie you can check this without too much difficulty, by the way

Pick the spacetime $\Bbb R^{2,1}$, with 2 timelike dimensions and 1 spacelike

Then you can mostly use basic geometry

Emilio Pisanty

4:19 PM

:38732902 My thinking is that saying $u=\partial u/\partial t=0$ at $t=0$ and then $\nabla^2=0$, in 1D, forces $u$ to be zero

Secret

quantisation (?) of a scalar field of mass m in terms of the characteristic dimension does not sound too bad. So I am guessing is when they are brought into interaction, then either they cannot interact, or there is huge inconsistency problems?

Emilio Pisanty

but that homogeneous part of the problem probably does have nontrivial solutions in more dimensions

Slereah

@Secret Use the classical pool game example

Or the GUN

I think Kriele has the gun experiment

0celo7

Hmm. I think it's Theorem 12.3.1

Willie reworded it significantly.

Secret

books.google.com.au/…

Well, if we demand consistency (like how the physics community handle retrocausality), the contradiction cannot happen. I think the more important thing is in the subset of dynamics which is consistent with a CTC topology, is that subset empty or all the allowed dynamics not interesting

Slereah

4:24 PM

I think Kriele might be the only GR book to use a gun for an experimental setting

the classical experiment is the pool game, though

It's in... Klinkhammer I think?

It's in Friedman, anyway

"J. Friedman, al - Cauchy problem in spacetimes with closed timelike curves"

Apparently the pool problem is from Polchinski

I wonder if it's string theory Polchinski

John Rennie

@Slereah Marcus Kriele's book Spacetime: Foundations of General Relativity and Differential Geometry ?

Slereah

yes

Secret

arxiv.org/abs/1609.01496 (Why do they love to put quantum computing stuff in it)

John Rennie

@Slereah Check the description on the UK Amazon site ...

Slereah

Secret

4:30 PM

Yeah I get what you mean. If consistency is obeyed in retrocausal spacetime, such dynamics simply cannot happen to preserve consistency

or they have to happen in a specific way to self fullfill it

e.g. ball A enters the wormhle time machine because its past self pushes it (At this point, I stop worry about freewill)

I am not sure how will QFT prevent such dynamics to happen though, as the above diagram is an insane number of classical trajectories need to be forbidden or somehow altered to ensure consistency

(My path integral skills is still nonexistent to address that yet)

uh, Nature, are you sure what you are talking about, entanglement in CTC spacetimes...?

> What happens, however, if causality is not strictly broken? In this context, Pienaar et al. considered a special case of Deutschian CTCs known as open timelike curves13 (OTCs). Consider a particle that travels back in time with respect to a chronology-respecting observer, but is completely isolated from anything that can affect its own causal past during the time-traveling process (See Figure 1). While the time-traveling particle has the potential to break causality, its complete isolation ensures that causality never actually breaks.

but from the look of it, OTC seemed to be basically the same as the non interacting case, or is there more subtleties, hmm...

Ok,guess not, though some of the weird properties are reproduced
---
https://arxiv.org/abs/1005.2219 post selection...

0celo7

4:46 PM

@JohnRennie what's up with it

@JohnRennie oh

that seems bad

Slereah

Then again, I don't think most people will buy Kriele because they read the Amazon page

It's a pretty niche product

0celo7

"Banach's theorem"

Gee, way to be precise

Slereah

At least it's not Riemann's theorem

Secret

I cannot seemed to google any nice papers that talked about interacting quantum fields solution in CTCs, most papers I came across are quantum computing, which is not even relativistic, I think I need some help in locating good papers in that topic

Slereah

There's no paper on the topic.

Except maybe the papers on unitarity in spacetimes with CTCs

Transcript for