The h Bar

12:00 AM

I mean generally speaking, is there such a thing in electromagnetism as, amplitude dependent interference? Somehow trying to find an explanation for how router can tweak this :)

ACuriousMind

@user929304 1. I don't actually know anything about how WiFi routers work. 2. I would assume that WiFi routers generally try to conserve energy, i.e. transmit below their maximal signal strength. If the noise is too high, simply turning up the signal will improve the signal to noise ratio. 3. I would also imagine that the routers can tweak their frequencies to channels that are less noisy.

Jake Rose

@user929304 Im not sure you can 'change' it in the settings

@ACuriousMind that sounds pretty reasonable

user929304

E.g In the SNR impact performance paragraph here wireless-nets.com/resources/tutorials/define_SNR_values.html may be relevant, suggesting that lowering SNR will translate to more frequent transmissions and less noise

Jake Rose

@user929304 that is about optimising the signal, and emphasises you want a good SNR

user929304

@JakeRose you can, e.g. service.avm.de/help/en/FRITZ-Box-Fon-WLAN-7490/012/…

Jake Rose

12:06 AM

It uses SNR as a number you can use to say if the signal is good or bad

ACuriousMind

@user929304 You may be misunderstanding the post. It is not saying that you actively lower some "SNR" setting in the router. It is saying that if you are in a region with low SNR - meaning that there is much noise compared to signal - then the router will on average need more transmission of a package to get it through to the receiver.

Jake Rose

I expect it changes the value by moving the router, or changing paramters about the environment

@user929304 I dont see anything in that second link about changing some setting on the router etc

@user929304 its just a useful number

user929304

@ACuriousMind in that post it s true, but Fritzbox router allow you to change snr!

Jake Rose

@user929304 where does it say that?

user929304

Hence my confusion

Jake Rose

12:09 AM

user929304

See step 4. Here en.avm.de/service/fritzbox/fritzbox-3490/knowledge-base/…

Jake Rose

Pay attention to the last line

ACuriousMind

@user929304 Yeah, for that, see my first message. the router is probably simply changing frequencies and signal strength settings until it hits one where it detects so little noise that it predicts the SNR at an "average reception position" to be equal to what you set.

user929304

@ACuriousMind right i see, but is the noise level itself, that the transmitted signal is susceptible to, itself dependent on the amplitude of the signal?

ACuriousMind

Well, the stronger the signal, the higher the signal-to-noise ($\frac{\text{signal}}{\text{noise}}$) ratio :P

user929304

12:12 AM

An alternative explanation may be that increasing snr required more frequent transmissions between client and router, and as a result it can more quickly error-correct sent data packages?

ACuriousMind

@user929304 Since the SNR is measured in dB, this seems purely about strength of the signal and not about successful transmission

user929304

Ye i guess makes sense. But does it then explain why increasing snr reduces throughput?

ACuriousMind

Hmm, that would indicate Jake is right and it has to do with re-transmission of packages

user929304

Man im confused, what the heck are these routers doing? :PP

ACuriousMind

Which makes little sense to me because I thought the policy was generally "re-transmit any package until the client acknowleges reception or some time limit is exceeded", but again, I don't really know how any of these things work

user929304

12:17 AM

Could re-transmissions intefere constructively and therefore improve signal quality? :)

ACuriousMind

I have a feeling there is an SE site where this sort of question would be very much on-topic but I can't put my finger on it

Mo_

@ACuriousMind DSP.SE

and EE.SE

@user929304 retransmissions do not take place at the same time. How can they interfere?

user929304

@ACuriousMind such a wierd diagram here community.arubanetworks.com/t5/Controller-Based-WLANs/… see last figure. It seems snr and data-rate are related non-linearly!!

@Mo_ true, trying to make sense of how different data rate leads to different SNR :(

ACuriousMind

@user929304 That's because dB is a logarithmic unit. The diagram is just saying that the better your signal reception is, the more packages you will successfully receive on average

Mo_

There is a scheme called MIMO where multiple antennas are used to transmit data to compensate for adverse effects that may have affected signals arriving from various paths

@user929304 each SNR level leads to a single bit rate, which depends on the performance of the coding method used to transmit data

user929304

12:27 AM

@Mo_ is data rate different here than signal strength?

Mo_

@user929304 yes. Data rate is an indicator of the number of bits per time you can transmit

signal strength is the power of the (analog) signal used to transmit the data

ACuriousMind

It depends very much on the protocol. Imagine a terrible protocol that sends all data as a single blob. A single bit error will invalidate the package, needing re-transmission of the entire data. You need to strike a balance between the effort needed to partition the data into meaningful chunks and the effort needed to retransmit a failed chunk

user929304

@Mo_ ye sure, but does that mean lowering data rate can potentially increase SNR? We re trying to figure out how router let user tweak the snr

enumaris

oh boy, there's so much logistics lol

Mo_

@user929304 SNR depends on the physical environment (antenna, temperature, etc.)

@user929304 That probably means lowering the transmission power which leads to lowering SNR

and it is useful when the current SNR level is much higher than what you need to achieve the expected data rate (say 100Mbps), and thus you can lower transmission power for other reasons (such as security)

user929304

12:34 AM

@ACuriousMind true, so i guess that makes rly hard to relate throughput loss to SNR increase

ACuriousMind

@enumaris Low-level communication protocols are really sophisticated to achieve reasonable speeds while not sacrificing accuracy. Checksums such as CRCs are a crucial part of it.

user929304

@Mo_ indeed. I wonder, how any of this can explain that by increasing SNR, throughput (dl and upload rates) is lowered? Maybe as acuriousmind says jt s a matter of used protocol

enumaris

@ACuriousMind oh, I was talking about onboarding logistics and documents I had to sign lol

I have not been following your conversation XD

ACuriousMind

@user929304 Where do you get that from? The diagram you linked has throughput monotonously increasing with SNR

@enumaris If you're gonna interject into the middle of a conversation, I'm gonna assume you're commenting on it :P

enumaris

@ACuriousMind ah yeah, fair assumption...but I mumble random thoughts in here often XD

user929304

12:38 AM

@ACuriousMind the fritzbox link i sent you, it warns about that, namely if snr is changed data rate changes

ACuriousMind

@enumaris hmmm

enumaris

It's not this chat, it's me

ACuriousMind

Also, did you expect a German company to not be bureaucratic? :P

danielunderwood

I'm not sure if it's related to that SNR option, but transmission power can also be configurable in order to have a piece of hardware work with different regulations. Sometimes firmware will let you have free choice of your transmission power with a disclaimer

Mo_

@user929304 By increasing SNR, throughput is either increased or left constant. You can't have worse communication quality when you increase the signal level!

enumaris

12:41 AM

@ACuriousMind XD

danielunderwood

@enumaris you should tell them that you want to work on COBOL too

and ML...using COBOL

enumaris

nothx

user929304

@Mo_ yeah makes sense. I guess i m mis-interpreting the whole thing. Confusion arose from point 4. Here en.avm.de/service/fritzbox/fritzbox-3490/knowledge-base/…

danielunderwood

or SNOBOL or SPITBOL

because early computer scientists had a sense of humor

user929304

Where somehow interference resistance can be tweaked in router settings but at cost of lowered throughput.

ACuriousMind

12:44 AM

@enumaris Really, a German company that automates business processes. Turning individual workflows into standardized logistics is what we do

enumaris

I'm joining an innovation center

So I may not have the best high level overview of SAP :P

danielunderwood

You should automate automating business processes

enumaris

Automate SAP

sounds legit

unfortunately, I have to stay within the innovation center's charter

ACuriousMind

We...might be doing that :P

enumaris

which is on smart hospitals or smart healthcare

Mo_

12:46 AM

Assume I have a gas of two kinds of particles (A and B) in thermal equlibrium. I can calculate the statistical average of the number of A and B particles in a region around the origin. Now, does anything change if A and B particles are + and - charged (like a plasma)?

enumaris

maybe in a year I can influence them to be broader in their directions

ACuriousMind

In the end, I'm happy that I have nothing to do whatsoever with specific businesses :)

enumaris

one of the reasons I chose this role is that I felt it had a high(ish) degree of influence within the innovation center :D

user929304

@ACuriousMind fritzbox manual calls it " increasing interference resistance", does that more likely relate to what you were saying with tweaking freq instead of strength?

enumaris

@ACuriousMind are you like...in an IT lab, isolated from the rest of SAP?

do you use COBOL to warm you up during the winters?

ACuriousMind

12:49 AM

@enumaris No, I work on source code analysis (of code written in ABAP), which is mostly independent of specific businesses. Garbage code is garbage code regardless of what process it's supposed to model :P

danielunderwood

hmmm I wonder if we may be your competitor. I share a floor with an IoT group, but I'm not really sure what industry they're in

enumaris

is there a central HR directory at SAP...could I in theory find you in the employee section somewhere...if I spent enough time digging through it...

danielunderwood

well @enumaris...I don't think they're working on code analysis IoT...I hope

ACuriousMind

I'm the opposite of isolated - our work is part of every ABAP containing product we ship.

enumaris

SAP is probably your competitor in some domain...they have a ton of businesses...

ACuriousMind

12:50 AM

@enumaris Yes.

And you don't even need to dig - just enter my surname and you'll find me, it's unique at SAP

enumaris

Sounds like I got something to do for when I'm bored LOL

oh

that's interesting

unique among 95,000 people is a feat

danielunderwood

I'm not even the only Daniel Underwood at Cisco

enumaris

although not as impressive as if it were unique within facebook

ACuriousMind

Rare German name with an umlaut :P

danielunderwood

as people found out when they tried to email me

enumaris

12:52 AM

so you're at Cisco? lol

you didn't reveal this before

Yeah I'm pretty sure SAP and Cisco compete

although the primary competitor in the US would probably be Oracle

Cisco has more of an emphasis on telecoms right...

danielunderwood

I work for the anonymous company C***o

enumaris

sounds legit

ACuriousMind

@danielunderwood Every SAP employee has a username based on their surname, and, if this is not sufficient, on their first name. For common German names like "Müller" and common first names they've had to assign numbers to people with duplicate names, complicated by the fact that the usernames are limited to 12 characters for historical reasons

enumaris

Cisco has government contracts so I guess that was why you worried if it was confidential? o.o

I'm apparently number 168

cus I got a 0168 number

appended

danielunderwood

Well what I do is a bit confidential. I don't suppose the fact that I work there is

And wow

I was impressed when I saw someone with a 5

ACuriousMind

12:55 AM

So there's people with usernames like MUSTERMANN14

enumaris

yeah my surname is pretty common if you romanize it

it's unique if you use the Chinese character

and by unique I mean afaik I'm literally the only person with that surname

not even my father/grandfather have my surname lol

ACuriousMind

But my username is just my surname. I'm special :)

danielunderwood

Our username system is quite weird though. Like I have half of my first name and half of my last and the other with my name has each offset by a letter

enumaris

My dad disowned me before I was born so I have a different surname than him

(jk about the disowning part)

ACuriousMind

I'm also special because both my names contain umlauts and it's never consistent across all tools how they handle it :P

enumaris

12:57 AM

so far I have not logged into an employee portal

I've only logged into some account to sign onboarding documents

time to play some more overwatch :D

danielunderwood

I sat there for an hour on my first day thinking I was waiting for my password change to propagate and that's the reason I couldn't log into anything...I had actually typed in my username incorrectly

enumaris

1:13 AM

lol

enumaris

1:45 AM

hmmm

anyone here use wechat? I'm not receiving new message notifications from it for some reason...

-.-

danielunderwood

2:08 AM

Use uschat

or youchat

or ichat

Sir Cumference

@danielunderwood Huh, someone else remembers ichat

danielunderwood

lol nope

Sir Cumference

:(

danielunderwood

The cool people use AIM

I also used MSN Messenger at some point...or whatever it was called

Hey it was MSN Messenger. At least my memory isn't terrible

Sir Cumference

man i miss the old internet

Huh you can still download an AOL browser

danielunderwood

2:24 AM

uhh

> Windows 7, 8 or 10, 266 MHz or faster computer processor

From discover.aol.com/products-and-services/… feel terribly sorry for anyone on those OSes with a 266MHz processor

Sir Cumference

@danielunderwood Aw never mind then

How can apple not support aol after all these years

danielunderwood

Did apple ever support AOL?

Sir Cumference

yeah when i was like 5 i think

now i'm getting flashbacks of old computer disk games and dial up

danielunderwood

2:41 AM

I'm pretty sure my first game was Carmageddon, but then my parents only let me have E rated stuff and I played through the Harry Potter games as they came out

I also remember the assorted color plastic computers Apple had

enumaris

3:13 AM

apparently Arena of Valor is the same game as 王者荣耀。。。

but not quite the same...my main is slightly modified...

not as strong it feels like

Dude156

3:46 AM

Hey guys, if you din't mind me asking

how is velocity ever measure? like in problems in physics when we have a ball being launched at 20m/s, how this actually haopoen? Dont we need to measure the final and initial positions divided by time to calculate velocity?

Sir Cumference

@Dude156 That would give you average velocity. Here, there is a notion of instantaneous velocity: intuitively, we obtain at the velocity at launch by measuring the smallest amount of time possible around it

This time is actually taken infinitely small

Dude156

4:19 AM

but still, how would we ever reach that number? i was thinking maybe f

fdt=mdv

John Rennie

7:52 AM

@Dude156 you are quite correct that we can never measure instantaneous velocity because all measurements take a time greater than zero. However we can make the error in the velocity measurement as small as necessary by making our measurement fast.

Slereah

8:06 AM

Morning

Avnish Kabaj

10:46 AM

GodotMisogi

12:38 PM

-1

A: What would happen if the Earth was tidally locked with the Sun?

In my oppinion if the earth were to stop rotating (only once a year), still the Moon could unlock it. The Earth used to spin faster, but the Moon slowed it down (and is still slowing it), so it makes sense for the Moon to create the opposite effect. Although coriolis force would be 364 times les...

Sir Cumference

1:57 PM

-7

A: Why aren't neutron stars full of dark matter?

Why aren't neutron stars full of dark matter? Because dark matter doesn't consist of particles. There's something of a myth that it does, which I think comes from particle physicists who have never actually read Einstein's original material. I also think science is something of a competitiv...

please stop banning JD. he comes over to astro in the meantime.

Swapnil Das

2:12 PM

@Blue Can you write a C++ code for me? :P

pZombie

When people say that an object cannot heat up another object above its own temperature because of the 2nd law of TM, they mean it only in regards of using mirrors/lenses etc, right? I am pretty sure i could use solar panels and batteries to collect energy from the sun and then use some process to heat up an object way beyond the sun's temperature

Slereah

2:33 PM

@SirCumference Ban him network-wide

pZombie

what if he reincarnates?

Slereah

I'm not worried about baby Duffield

4

pZombie

so can a cold body heat up a hotter body after all? Solar panels + batteries + some process seem to allow for such a feat

maybe not at a high efficiency, but it can be done, or so it appears to me

John Rennie

3:42 PM

@pZombie no you cannot use lenses or mirrors to get a temperature higher than the source. See:

14

Q: What temperature can you attain with a solar furnace?

A solar furnace is a device that concentrates the sun's light on a small point to heat it up to high temperature. One can imagine that in the limit of being completely surrounded by mirrors, your entire $4\pi$ solid angle will look like the surface of the sun, at about 6000K. The target will then...

Anonymous

@SwapnilDas I can write in C. Haven't used C++ in a while.

Anonymous

Anyway, I could try. What's the question?

Anonymous

@SirCumference Hmmm, I was expecting that.

Anonymous

I don't use Astro SE, but this is something you really need to pick up with your mods.

Emilio Pisanty

3:59 PM

@PhysicsMeta wow.

This is just getting weirder and weirder.

Transcript for