The h Bar

12:19 AM

@bdegnan Maybe the digital ones were not enough exact? What if the digital circuits use signal levels of 10 digit precision? What if they use 20 digits?

enumaris

3:01 AM

@bdegnan hey, I had to drop off and go home, but it was very interesting reading the chat here! :)

I've only just been exposed to these techniques/methods so I'm interested in learning more

vzn

3:54 AM

@bdegnan checks out! :) (reserving judgement on your other wild story) :P Temperature robust programmable subthreshold circuits through a balanced force approach smartech.gatech.edu/handle/1853/47548

John Rennie

7:11 AM

It's hot in the UK.

2

Slereah

we will all die

John Rennie

And forecast to get hotter

undefined

yeah hot in Germany too. Tomorrow shall be the hottest day

John Rennie

Indeed. Tomorrow sounds like a good day to hide from the sun and do ... well ... as little as possible :-)

undefined

but I'm in the north of Germany. So it wont get as hot as in the south-west. There its said they gonna break the 40 C (104F)

I filled vacation request form today for tomorrow and the day after tomorrow. It will get too hot in the office (no AC or such)

anyway, I was thinking about Can a photon loose "energy" or decay or something like that? I was thinking about if its wave-like, will the wave be "flattened" over time and space until its undetectable? Because I was thinking about what would happen to a photon if it's placed into a tiny sealed box which is reflective on the inside? Would it just bounce forever in the box?

John Rennie

7:33 AM

@undefined if the mirrors in the box were perfectly reflecting then the photon would bounce forever. In real life mirrors are more like 95% reflective so it doesn't take many microseconds for the intensity to fall to zero.

undefined

7:45 AM

@JohnRennie how does it fall to zero and what exactly does fall?

John Rennie

@undefined suppose we have a metal mirror. The light has an oscillating electric field and that makes the conduction electrons in the mirror oscillate. Those oscillating conduction electrons reradiate light and the incident and reradiated light interfere. The result is a wave going back away from the surface i.e. we get reflection.

But this reradiation is never 100% efficient because the oscillating conduction electrons can scatter off the metal lattice and transfer energy to it i.e. some of the light energy is converted to heart and lost.

undefined

is it the same photon or is the original photon lost the electrons emit/reradiate a new one?

John Rennie

It's very tempting to regard light as an array of photons. Kind of like little balls moving in a group. But the photon is a great deal more complicated than this.

undefined

yes

John Rennie

Light is a quantum field and photons and waves are just different ways the quantum field can appear to interact.

Slereah

7:52 AM

All photons are the same photon

One photon universe

Novalium Company

Good morning

John Rennie

@undefined So asking if it's the same photon isn't that useful a question because for phenomena like reflection the particle approach to light, i.e. photons, is not an approprate approximation to use. We would use a wave approximation instead.

@Slereah Fock that

@undefined you might be interested to have a read through:

45

Q: Do photons truly exist in a physical sense or are they just a useful concept like $i = \sqrt{-1}$?

Reading about photons I hear different explanations like "elementary particle", "probability cloud", "energy quanta" and so forth. Since probably no one has ever seen a photon (if "seen" it supposedly - and rather conveniently - ceases to exist) but many experiments seem to verify its properties ...

undefined

yes but as far as I know and understand we are able to create single photons. And if this photon bounce off a mirror, is it still the same we created? Maybe its a stupid or not useful question but for me somewhat still interesting. But if I think longer about it, I'm not sure if its even answerable

I read that question yesterday already many times :)

John Rennie

@undefined what we call a photon is actually something called a coherent state that I have to admit I don't really understand.

A coherent state is more like a wavepacket than a particle.

undefined

if the word photon is bad in that question then one could also ask if its the same wave

John Rennie

7:57 AM

@undefined well the reflected wave has a different momentum. If you insist that a wave is a specific momentum state then incident and reflected waves have to be different waves because they have different momenta.

Whether this is really a useful statement is debatable.

undefined

@JohnRennie you are one of the most smart people I know. If even you don't understand it, I don't think I'll ever understand it :/

John Rennie

Coherent states falls into the area of quantum optics, and though I've never studied this we have lots of people who have. Emilio Pisante for example is something of an expert in this area.

undefined

@JohnRennie so it could be the same wave but with different momentum? Or could it also be possible that the electrons of the reflective surface create/emit a new wave?

John Rennie

This is largely just terminology. Whether it's the same wave or a different wave depends on how you define the wave and it isn't a useful distinction anyway.

undefined

my thoughts behind all those questions was, how/if it the answer would change if we have entangled photons

John Rennie

8:08 AM

It's outside my area of expertise, but I believe that reflection does not change any entanglement of the original light i.e. you can generate pairs of entangled photons and after reflectin you still have a pair of entangled photons.

Whether it's the original photons forming an entangled pair or new photons forming an entangled pair doesn't make any difference.

undefined

ow for me it does make a difference. I'm not sure why but I feel like its interesting

John Rennie

If you ever find yourself working in experimental quantum optics you'll get over it :-)

undefined

but only after I know the answer :)

because I don't like it if I don't know something. And if I have the opportunity I'm always putting all my effort in finding an answer

my boss often calls me a "stickler for details"

he dont like it much, tho

Slereah

Apparently the Bohm people like to call standard QM "Orthodox Quantum Mechanics"

to make it sound more sinister, I presume

undefined

8:24 AM

is there some kind of formula to calculate the deference in momentum from reflected waves?

John Rennie

@undefined yes, the total momentum is $p = h\lambda$ and if the incident wave has momentum $(p_x, p_y, p_z)$ and we take the mirror to be in the $xy$ plane then the reflected wave has momentum $(p_x, p_y, -p_z)$.

It's no different to the elastic collision of a ball with a surface.

Sir Cumference

When I went home for the summer I didn't expect it'd actually be hotter

undefined

@JohnRennie thank you. Does this formula have a name, so that I can read more about it?

John Rennie

@undefined I'm not sure if it has a name. It's just the result of a collision with an infinitely heavy object. When you conserve momentum and energy you get this result.

(we assume the mirror is infinitely heavy so it doesn't move in the collision)

undefined

@JohnRennie I see. I would just like to read about what exactly the h and mean λ

Slereah

8:39 AM

don't buy an infinitely heavy mirror

it's hard to move up the stairs

undefined

lol

John Rennie

@undefined $h$ is Planck's constant and $\lambda$ is the wavelength of the light.

The equation $p = h/\lambda$ works for all particles, where for massive particles $\lambda$ is the de Broglie wavelength.

undefined

thank you

Pandya

8:54 AM

This seems copy pasted from quora — Pandya 41 secs ago

Slereah

9:11 AM

vixra.org/pdf/1510.0502v3.pdf

:)

Slereah

9:30 AM

You know, I know a bunch of linguists and such

and how glad am I that it is always possible to find a physics paper online

it is much harder in the field of humanities

you may have to ask libraries

order books and such

Slereah

9:47 AM

"Seeliger's papers, especially his (1895a), contain the most detailed and general development of the problem. The price is that his exposition is the most cumbersome of an expositions; he alone resorts to infinite series expansions in Legendre polynomials and includes tidal forces in the analysis. all later commentators managed to reduce the exposition of the core to one or two lines of formulae."

Bloody Seeliger

Slereah

9:59 AM

reading historical physics is always a bit weird

like you realize how many problems of the days are never really commonly discussed anymore

Gyromagnetic

10:44 AM

Hallo hbärs!

Anyone familiar with Bayesian optimization? I have a few questions :p

Slereah

11:26 AM

"Kelvin addressed the question of whether was imponderable, that is, has no weight."

Emilio Pisanty

11:54 AM

... or, as has been said previously, "Your brain is continuously and egregiously lying to you about what your eyeballs have actually detected". — Emilio Pisanty 11 secs ago

cc @dmckee

undefined

12:20 PM

@JohnRennie hola @EmilioPisanty Do you maybe have an answer to this?

ow wrong @ I'm sorry

Jake Rose

12:35 PM

@JohnRennie “nobody has ever seen a photon” tickled me

@EmilioPisanty do you know a guy called Tijmen Euser?

Slereah

I have seen a photon

I keep it in my pocket

undefined

you lucky :)

Emilio Pisanty

@undefined what is "this"?

@undefined it it's this, then John has already answered it

@undefined if it's this, I'm utterly uninterested in that type of metaphysics

Slereah

"Herr Seeliger sees in the fact that, according to Newton's law, an infinitely long matter-filled cone would also exert an infinitely great attraction on a point on its axis, an objection against the general validity of the law itself;
the author of a cosmogonic hypothesis can only draw the converse conclusion, that that mass distribution, purely imagined by him and inaccessible to observation, could have been possible at no time, for it cannot be explained by the forces abstracted from the world of experience accessible to us." (Wilsing 1895a: 388)

Snap!

undefined

12:54 PM

@EmilioPisanty no, more like the question if a reflection affects the entanglement in anyway

and I dont want to discuss but I would be glad if you could elaborate why you call that metaphysics? Because it's not important/meaningless?

Semiclassical

@Slereah historical math is strange too. I've been trying to trace the origin of a certain stats concept and it's been surprisingly tricky despite the fact that I only care about the story from roughly 1890 onwards

JMac

@undefined What's the physical definition of "same proton"?

Semiclassical

(namely, the notion of a "hilbert space of random variables". it's simple enough to verify that that makes sense, but figuring out who first had that insight is not so easy)

Ryan Unger

@Semiclassical Kolmogorov?

Semiclassical

That seems the most likely, yeah.

For instance, there's a 1941 paper of his on stationary sequences in Hilbert space which addresses stochastic processes

undefined

1:08 PM

@JMac Photon, not Proton. I would like to know/understand/learn that too

Semiclassical

the 1941 paper is found here on google books

JMac

@undefined I meant photons

Slereah

just before Pearl Harbor

Semiclassical

actually, the clearest page in that book is from another 1941 Kolmogorov paper (same topic area): books.google.com/…

Slereah

"Is Octonionic Quantum Gravity in the Pre-Planckian Regime of Space Time Violated? Input from a Worm Hole Bridge from Another Universe Considered as a Model for Examining This Issue."

Oh Vixra

Semiclassical

1:18 PM

Interesting that Kolmogorov references his book in that section. I've looked in it and haven't seen any sign of Hilbert space geometry in there

(might be implicit somewhere of course)

Slereah

Apparently Newtonian gravity is totally unfit to deal with homogeneous mass

quite unfortunate

Semiclassical

riiiiight

Slereah

It's troo

Has been known since the 19th century

you just get divergent potential all over the place

Semiclassical

ah

Slereah

and you can't do any regularization that works in a consistent way

Semiclassical

1:23 PM

ew

Slereah

basically Newtonian mechanics was a shambling corpse by the end of the 19th century

it was full of holes and it's quite fortunate that QM and GR appeared

Ryan Unger

@Semiclassical Kolmogorov is credited with introducing measures in probability theory

Semiclassical

@RyanUnger yeah, or at least for making clear how to properly axiomatize using measure theory

(Borel had tried to introduce measure theory to probability I think but he didn't succeed?)

@Slereah I imagine that's related to the issues with classical electrodynamics e.g. self-energy

Slereah

@Semiclassical Well not really, although that was certainly part of the picture

although of course

Me saying that it was done is only with hindsight

I'm sure most people thought those issues had solutions within the framework

Semiclassical

@RyanUnger The interesting thing is that there's a 1936 paper of de Finetti where he also gives a geometrical interpretation of random variables as vectors, along with distance/inner product/angle between them

(He actually gives two such interpretations, the second being the familiar one and the first being a quotient of such. but I don't think that's so essential)

Slereah

1:30 PM

Hm

Apparently Newtonian gravity's issue comes from the lack of unique solution to the Laplace equation if you drop boundary conditions

Semiclassical

see pages 8-11 here (brunodefinetti.it/Opere/AboutCorrelations.pdf) for the de Finetti paper

@Slereah i mean, classical electrodynamics and newtonian gravity both involve inverse square laws so I imagine there's some necessary similarity between their issues. big difference I guess is that gravity is purely attractive

that and, y'know, only one scalar potential for newtonian gravity

Slereah

@Semiclassical Oh yes

The same problem arise if you consider homogeneous charge (that isn't zero)

but of course that's less of a problem in EM

Ryan Unger

I'm having to write the graduate school about my transcript

If I have to send it again I'll be pissed

b degnan

1:45 PM

@peterh The issue with "bits" is actually that noise has value in the biological systems as it makes hysteresis. Analog is basically infinite bits. The C. elegans models match the analog behavior very well, and the VLSI implementation that is digital does not. This is basically due to the "hum" of the analog circuits.

Ryan Unger

@Slereah arxiv.org/pdf/1907.09922.pdf

peterh

@bdegnan Yes, but doubling the worthy bits of the signals roughly doubles the computing effort, while it squares the signal-to-noise ratio.

Slereah

@RyanUnger is that quartic scalar stuff

b degnan

@peterh I just wanted to mention that I had no idea in the "bit equivalent" side of these systems. :)

peterh

2:20 PM

@bdegnan n-bit signal means a signal, where the values can have $2^n$ different, value-quantized, discrete values.

@bdegnan For example, if we have a 3-bit signal, between 0V potential and 1V potential, then the allowed values are 0V, 0.125V, 0.25V, 0.375V, ... 0.875V

@bdegnan Most calculations in digital circuits depends roughly linearly with the bit size. This is a very raw estimation but roughly so is it.

b degnan

@peterh I'm aware of that, but in the context of how it changes the neural network behavior. I made the ADC/DACs for programming the floating-gates on the neural networks. I have just never looked into the digital neural network models nuances. I only know that the FPGAs 1) didn't follow our models, and 2) used a lot of power.

Emilio Pisanty

@undefined It doesn't.

@undefined I find the question "but is it the same photon?" to be about as interesting as "but have you ever really looked at your hand?" coming from someone that's been staring at their hand for the past two hours while smoking weed.

I don't think it's interesting as a question and I'm uninterested in discussing it.

peterh

2:42 PM

@bdegnan That I agree! Also I've always tought that neural networks should work with analog circuits. They exist, there are EPAC (eletronically programmable analog circuit). Essentially, you can "upload" the analog circuit diagram into the chip, and then you will have the analog chip.

@bdegnan FPGA is good for massively parallel tasks, but it is very bad for general computing. Probably your FPGAs were not enough well-developed.

@bdegnan I suspect, human brain could be already simulated even with the todays technology, with a network of EPACs. We need around 10billion neurons with around 20Hz. Next step will be to analyze the circuit of a human brain with image recognition algorithms, and then create the equivalent network with EPACs. If it would work, the result would be... maybe the heaven, maybe the hell. Somewhere deeply I really hope, there is some inherent, unknown thing in the universe what avoids it.

theDoctor

Ha, you don't need better imagery, you need epistemology.

peterh

Imagine that you get a dead man. Extract his brain and freeze it in liquid oxygen. Then grind it in 0.1$\mu$m slices, and you take a photograph, also with 0.1$\mu$m precision, from all the slices.

You save the whole thing into a big database. Then, you eat it with image recognition algorithms. So you will have the circuitry of the whole, once lived human. You create the equivalent analog circuity, and upload it into a network of EPACs.

theDoctor

You won't have enough information.

peterh

At this point I say, heaven would be better

theDoctor

Besides, someone's figured it all out with pure research (no biology).

peterh

2:50 PM

@theDoctor What information would be missing?

@theDoctor With what research? I think, todays technology could already do it, but even 20 years old tech wasn't yet enough well-developed for that.

theDoctor

As I said it didn't require more biology or neural imaging, but more epistemology.

peterh

@theDoctor Can you upload a human soul into a computer with epistemology?

b degnan

3:19 PM

@peterh we have neuron ICs. I have ganged up 10s of these: hasler.ece.gatech.edu/Published_papers/Neuron_papers/…

I don' t know the max the lab did. I went off to make cool stuff on 14nm.

Physics Meta

4:19 PM

0

Q: Cases in which edits are an unethical mean to generate attention

I made 6 minor edits to my latest answers. The last edit (addition of a link to a lecture on the topic) I did 2 days after I gave the answer. I noticed that the question received a lot of new attention after the edit. It seems with each edit the question is bumped up on the main site again. Don'...

Transcript for