« first day (2225 days earlier)      last day (2812 days later) » 

2:13 PM
@Bob permanently off to keepass?
tempted to get 1password
 
Bob
@HackToHell ?
 
The blue icon is keepass no?
iirc you used lastpass right
 
on asset registers would you register the kb/mouse?
 
I never did
Though when recording the computer I'd make a note of the type of keyboard and mouse
 
Ah, running is fun
 
2:20 PM
In the past: nope.
These days: yes
Just you can point to the person who goes though a dozen keyboards per year.
 
@Rahul2001 Oh god, talking about inexperienced bikers not wearing helmets as if it's a good thing
 
Why isn;t it.
I do not think a bicycle helmet will help you vs 10 ton crushing trucks.
And normal cycling is something you learn around ago 5 or lower.
 
Because it's illegal in half of Europe? :-P
 
2:31 PM
It is?
I never wore a helmet. But I only cycled in NL and DE
 
i'll be honest i thought it was the law in NL to wear a cycle helmet
 
Nope. You rarely even see cycle helmets here.
Some racing groups often wear one (and go at stupid speeds)
 
Most countries in Europe require it by law for children, Spain requires it for adults as well.
 
no you need to give them to pedestrians for when cyclists run them down
 
But regular people just going to work or going shopping? No
 
2:33 PM
Personally I never do, but they say it does drastically reduce your chances of head injury in a crash.
 
when i was in amsterdam i got knocked down by atleast two cyclists
 
I've never knocked down anybody, though I've twice run into idiots who stepped out in front of me without looking when the pedestrian light was red.
 
In the last 30 years I have been hit twice by a bike. One overtaking to close. On turning left without looking when I was overtaking.
 
the first time i was looking the wrong way when crossing the cycle path
 
I've also hit two stationary vehicles, once cause I was on my phone and in my defence, the second one was black and illegally parked at night. Although being on my phone probably contributed to that one too
 
2:35 PM
And people do tend to look to the cycle lanes before stepping off the pavement to the street.
 
the second i was drunk and just wandering aimlessly
 
A lot of whether people look or not depends on local cycle culture, and you also have to bear in mind in touristy cities like mine, a lot of people are not local and are oblivious.
 
I just do not see the advantage for most casual riders over here.
Good separation from normal traffic, usually not that high speeds (do not want to arrive sweaty at work).
Decent snow free cycle paths.
 
hence amsterdam
 
Amsterdam is a mess. It is annoying when walking there or by car. Never tried a bicycle there
 
2:38 PM
hence getting run over
i like you dutch, you were all nice and kind and funny when i was there
 
If you are happy with a hat which help when you get in a soft colision, do wear it
I just fear fewer people will cycle if it is mandatory.
And it does not help with the #1 cause of cyclist deaths (heavy trucks)
 
Funny how a passive M.2/PCIe port converter costs twice as much as an active SATA => IDE converter.
 
Newer, more sexy, higher price ?
Or naybe, people who buy relative expensive new stuff will also pay more for adators from regular PCI-e to M2 format then
people who need something for old hardware?
 
Urgh, cheapest PCIe SSD that can match a SM951 M.2 SSD in IOPS is £275 :-(
 
It might stay a bit cooler though.
 
2:51 PM
Hmm, Samsung 850 Evo SSDs seem to have gone for 1/1024th the flash size in RAM - the 4TB version has a 4GB DDR3 cache
 
3:04 PM
how weird... getting a 404 even though the file exists
 
Bob
@HackToHell I've never used lastpass.
 
@Burgi Did you forget to do populateWebsite()?
 
@kerbalspacecat my guess would be they just go for 1-4 since its common
 
@JourneymanGeek Huh?
 
3:13 PM
oh, the ram cache?
 
Oh
Well the 500GB version has 0.5GB of RAM, and the 250GB version has 0.25GB etc. etc.
 
No wait, the 250GB has 512MB and the 120GB has 256MB, buut apart from those two it's all 1/1024
4GB of cache though... I still have hard drives with less than 4MB of cache
 
@kerbalspacecat faster storage = more buffer
 
@JourneymanGeek Obviously not linearly, if a 60MB/sec hard drive has 2MB of buffer and a 600MB/sec SSD has 4096MB...
 
3:17 PM
@kerbalspacecat probably takes into account overall cost, handyness of the ram (for example for speeding up FDE...)
 
None of the drives I've mentioned have FDE, and handyness? When is it ever handy to buffer 8 seconds of max speed reads or writes?
The latter would even be utterly dangerous
 
BUFFER ALL ZEE THINGS!
 
@JourneymanGeek O_O There is left handed and right handed RAM? ;p
 
@DavidPostill Of course!
 
@JourneymanGeek BTW do you know a unix.stackexchange.com mod? Offensive user name with no posts to flag unix.stackexchange.com/users/188945/nope-fuckers
 
3:28 PM
In cases like that, you can flag any post and link to the user
 
@BenN I'll flag your chat post then :)
 
@DavidPostill brought it to the attention of a mod
 
@JourneymanGeek Thanks
 
I could have sworn @terdon comes here
 
It's been too long for a normal ping
 
3:30 PM
he's a U&L and AU mod
ya. Grabbed him elsewhere
Its actually pretty handy that the trilogy + UL+ AU mods know each other to an extent XD
 
@JourneymanGeek Yeah, well you have a secret room to hang out in :)
 
Pingbot FTW
 
Bob
What secret room? There's no secret room.
Nope, not at all.
Never existed.
 
only zuuuuuul
@DavidPostill that's not really all tho
there's some sites where I don't really know the mods at all
 
@Bob So all these rumors of a "Teacher's Lounge" are totally unfounded? :)
 
3:42 PM
I don't know what you are talking about.
 
@jokerdino lol
 
But I like the idea. It would be cool if we could have a sekrit room.
 
or many seekrit rooms
 
@JourneymanGeek I only know a handful of SU mods and one U&L mod who also happens to be one AU mod.
Your ties with other mods are probably stronger and wider.
 
@jokerdino You can run but you can't hide :)
 
3:48 PM
Don't know what you are talking about but that was not me either way.
 
@jokerdino I have reasonably good contacts in SF, and one of the SO mods pings me for cross site spam
oh, and SR ;p
 
I'm still really low in the ranks of the secret mod society.
I already managed to achieve the first step, being a RO
And step two, get an accomplice to impersonate other mods and confuse everyone
But that step is a secret, can't let the mods know
 
@ThatBrazilianHeadlessHorse you should flag your message and get a mod to delete it so they won't kn... oh
 
do what I
did...
 
Then the confused mods will ban a mod while trying to ban someone else
 
3:53 PM
eliminate the competition, by letting them win the election
then sneak in and win the next one.
Worked for me!
 
there may not be another election for SU
 
And then when trying to restore that mod, they'll instead make ME a mod! bWA HA HA
 
@jokerdino I was referring to this
yesterday, by jokerdino
@DavidPostill re: your spam flag on Ask Ubuntu, any explanations?
but never mind
 
It was not me talking to Andy.
 
@JourneymanGeek I heard the pay is not that great...
 
3:54 PM
You get a SU and SE T shirt ;p
 
@jokerdino OK. That's why I said never mind.
 
@allquicatic we'll have one when needed
 
> NAND memory bit cost reduction will continue at a steep rate via scaling 3D NAND architecture.
Yay
 
@ThatBrazilianHeadlessHorse Didn't someone already try this? ;p
 
@DavidPostill I don't have multiple important conversations at the same time usually.
 
3:56 PM
@allquicatic Really?
 
@DavidPostill Maaaaaaybe. I'll admit nothing.
 
@ThatBrazilianHeadlessHorse lol
 
@kerbalspacecat hope that "bit cost reduction" is passed onto consumers instead of the 850 Pro sitting pretty at $450 while they reap increasing profits
2 mins ago, by Journeyman Geek
@allquicatic we'll have one when needed
 
IIRC SE pokes each site's mods around each anniversary of the last election to see if they want another
 
@allquicatic Hopefully when the competition actually start shipping finished commercial products...
 
3:58 PM
@BenN other way or something to that extent
 
> 3D NAND with 48 layers is closer to the 2D NAND price-per-bit curve than 3D NAND with 32 layers. We expect that 3D NAND will reach price parity with 2D NAND at 64-layer chips. Going to 64L, 96L or even 128L 3D NAND would be possible in a few years
 
I imagine the mods can request another at any time though (speculating)
 
pretty much
@BenN we don't really get poked
 
@kerbalspacecat how long until the chips are visibly thick in the width dimension (along the "Z" axis if you're looking at the board sitting flat on a table)?
 
Ah, OK. I was thinking of this MSE answer
> Most elections are scheduled as a result of moderators reaching out to Stack Exchange to ask for additional help.
 
4:00 PM
@allquicatic Hah. No idea. I suspect they won't go beyond 128-layers though
There's already diminishing returns in increased cost/complexity/etc. and less proportional benefit per layer.
From the sounds of things even 64->96 will be hard
 
I wish I knew what "hard" means when they say certain IC fab things will be "hard".. are they discovering entire new fields of science and immediately applying them to practice? are they just developing new fab machines based on known principles but making them "better"? are they using entire new chemical elements?
 
@allquicatic a good chunk of it is yield
 
I know they did start using entire new chemical elements with Hafnium, Gallium and Arsenic in the recent past which they didn't use before
 
For doping?
in this case tho I suspect the tricky thing is building up layers on a wafer
 
@Bob oh it was allquixotic then I thin
 
4:29 PM
@allquicatic Yield mostly. They go into some detail in the article though not too deep
> For 1y nm 2D planar NAND such as 16 nm or 15 nm MLC/TLC NAND devices, for example, the process integration on the memory cell array and peripheral region comprises well/active/isolation (SA-STI, self-aligned STI) formation; cell FG/CG and peripheral gate formation; and contact and interconnection (metals and vias) formation. Of course, patterning approaches such as DPT (double patterning technology) or QPT (quadruple patterning technology) and an air-gap process for active, wordline and bitline patterns on memory cell array region should also be included for manufacturing 2D planar NAND p
> Compared with the 2nd generation 32L V-NAND, the 3rd generation 48L V-NAND cell structure, of course, has a higher number of cell gates, which means the requirements and controllability for process integration are more challenging.
etc.
 
 
1 hour later…
5:34 PM
I bought some of these apple airpods in the pub. Sound quality is really poor, but my ears are now clean.
 
@Hennes how much does the blue rings cost? 100usd?
 
Half that since they do not have the apple logo on them.
 
@Hennes but won't that cause the earphones to brick while updating them?
 
@Hennes There speaks a man who doesn't go to the pub because the beer is cheaper at home :)
 
I got two Canon (kanon) here, and bakes minced meat, beans and pre-made 'greek dinner'
If you drink that on am empty stomach then you learn why it is called Kanon
 
5:54 PM
@Hennes Mine came with three different coloured rings in the box... free...
 
0
A: Under what conditions will a group-regulated PSU damage a modern computer?

bwDracoThe issue is specific to certain Haswell-based system configurations. Haswell introduced new, extremely low-power states called C6 and C7. The processor virtually shuts down in these power states, placing loads as low as 0.05V on the +12V rails. Since it's usually only high-power devices such as...

Longest answer I've written in a while.
 
Can anyone recommend a cheap and easy way to snoop and drop DHCP packets with a router or some kind of device? Basically I have a network cable coming from a switch where there's a DHCP server on the network and I don't want those DHCP packets coming down over the wire. Routers and switch on both ends don't support DHCP snooping and I don't want to buy a new switch/router just for this, rather buy a small device and plug the cable into that and then to its destination.
Perhaps a cheaper router/switch
 
Do they support port mirroring?
 
@bwDraco awesome answer, but "group regulation can cause the voltage on the +12V rail can be overcompensated" - grammar ;p "to be" overcompensated
 
Cisco RV082 routers on each end and the switches are unmanaged
 
5:59 PM
Fixing.
...done.
 
TIL @bwDraco
wasn't aware of that particular issue
good question, good answer
@qroberts you'll need a system with a stateful firewall; basically any Linux box should be able to do this... depending on how hacky / reliable / performant you want it to be, you could go with a Raspberry Pi, or a compute stick, or an old laptop, there are loads of options
 
@qroberts or just replace it
 
I feel the sudden inexplicable urge to run chkdsk
But I don't have anything I legitimately need to chkdsk.
Gotta love what EARFCN stands for -_-
 
6:21 PM
@kerbalspacecat fsck
 
E-ARFCN
E-UTRA-ARFCN
E-UMTS-RA-ARFCN
Three acronyms nested inside each other
Acronymception!
 
I like fsck more as it supports ext3 etc too
 
@kerbalspacecat A!
 
Dammit missed the edit window by 5 seconds. I actually missed a T
E-UMTS-TRA-ARFCN
From one of my Facebook posts:
UARFCN is an acronym that stands for UTRA Absolute Radio Frequency Channel Number, where UTRA stands for UMTS Terrestrial Radio Access where UMTS stands for Universal Mobile Telephony System. UMTSTRAARFCN.
Good to see the progress with simplyfying acronyms, we've now moved on to EARFCN.
@allquicatic \\\\\\\\\\\\\\\o
Oh you mean I should expand the A?
Technically if we were to use 3G in space or on a plane, it wouldn't be UMTS anymore
 
Universal mobile telephny system terrestial radio access absolute radio freqency channel number
it's just a collection of words! D:
 
6:32 PM
@djsmiley2k At the end of the day it's just an arbitrary and roundabout method to define "frequency"
The only advantage is it gets rid of a dot, that's about it.
 
@kerbalspacecat no, I mean you should acronym Acronymception to just A
@kerbalspacecat imho they should drop all the pretense and just say "PHOTONS" for the current state of the art; then, if they find something better that still uses photons, call it "PHOTONS 2", etc.
just "P", "P2", "P3", etc. for shorthand being displayed next to your signal strength bars on the status bar
 
@allquicatic Oh
@allquicatic lol
We're nearly there already, with it displaying "4G" or "4G+" next will be "4G++" and "4G+++"...
 
soon your entire status bar will be UMTTRAARFCNANAGLEREPLFOOBARGSQFHNSA
no room for other icons
 
We should be able to hit 1Gbps (the original definition of 4G before the Americans came along and messed it all up) by the time we get 4G+++
 
@kerbalspacecat I'm sure Verizon and AT&T will be up to 9G by then
9G marketing-wise, but still 4G
 
6:42 PM
@allquicatic I'm sure some providers have already tried calling "4G+" 4.5G
 
the funny thing is, since photons have zero mass by definition, the Earth's gee forces don't even impinge on them except at relativistic distances
so calling it 4G is silly when it's technically 0G
 
or 0.000000000000000000000000000000000000000000000000001G
 
@allquicatic Actually I do wonder, how do we calculate/define the effect of gravity on a photon
 
@allquicatic Would pfsense do the trick?
 
6:44 PM
@kerbalspacecat think you'd have to learn either special or general relativity to understand that, but I do wonder that also
 
Given mass-energy equivalence, photons do have mass, and also, I don't see why Earth wouldn't apply 1G of acceleration/force to a photon like it does for baryonic matter
 
I believe it depends hugely on the distance traveled, because even the Sun wouldn't bend light if it's just traveling between planets in our solar system
you'd have to pass a beam of light past the sun for a significant fraction of a lightyear for us to be able to even measure the curvature
 
It's just 9.8m/s/s would be a relatively miniscule deflection on something travelling 300,000,000m/s
But really, as I understand relativity, it would just be 1G, at least for the fraction of a second it's at/near the earth's surface
 
@kerbalspacecat I don't think mass-energy equivalence means that anything that has energy also "has" mass; IIRC it's a fundamental transformation to cease becoming energy and start becoming a tiny speck of mass
otherwise we could build bridges by shining a light at the side of a hill for a really long time
 
> Mass-energy equivalence states that mass is concentrated energy.
 
6:47 PM
@kerbalspacecat things get weird at the speed of light though because it also changes time
 
Relevant Physics.SE: physics.stackexchange.com/q/2229
 
@BenN /q/? you're just trying to get a badge for getting people to click your link! CLICKBAIT! :P
 
Ah-ha!
> Einstein also introduced the concept of relativistic mass (and the related mass-energy equivalence) in the same paper; we can then write

mc2=pc
mc2=pc
where mm is the relativistic mass here, hence

m=p/c
m=p/c
In other words, a photon does have relativistic mass proportional to its momentum.
 
There's no user portion in that link; I don't even have an account there :p
I just used the share link because it was closer to my mouse than the address bar
 
69
A: If photons have no mass, how can they have momentum?

NoldorinThere are two important concepts here that explain the influence of gravity on light (photons). The theory of Special Relativity, proved in 1905 (or rather the 2nd paper of that year on the subject) gives an equation for the relativistic energy of a particle; $$E^2 = (m_0 c^2)^2 + p^2 c^2$$ w...

@BenN No, it's good, thanks
 
6:51 PM
> The theory of Special Relativity, proved in 1905
 
@allquicatic I think you underestimate the immense amount of light you would require
 
I wish all human knowledge were just a SE chatroom
Einstein could reply to my question with a link that said "116 years ago"
like the "2 hours later..." messages you get
 
Sounds like we should register at Resurrection.SE
 
human knowledge propagation is laughably slow
I learned about math the Greeks discovered in B.C. less than 10 years ago, and I'm learning about physics we discovered 116 years ago just now
 
@allquicatic So you want to replace humans with machine-learning AI I guess
 
6:54 PM
> And here we have it: photons have 'mass' inversely proportional to their wavelength!
fascinating
 
Anyway, the mass of a photon would be so miniscule I don't think Windows calculator could actually calculate it
@allquicatic Indeed, since wavelength is inversely proportional to frequency, as frequency increases the energy of a photon increases, and therefore so does its mass
 
@kerbalspacecat If we had a ridiculously enormous source of energy, like the Sun, and were harvesting it with a Dyson Sphere, couldn't we have energy-to-mass reactors that would create chemical elements of our choosing, then we could just resort to perfectly ordinary fabrication / 3d printing / subtractive manufacturing methods to make objects?
 
Mass-energy-equivalence, as I naively misunderstood it! Hazzah!
 
@kerbalspacecat I knew that higher frequency photons are more "energetic" but I didn't understand that that energy could be viewed as a momentum and also as a "relativistic mass"
where relativistic mass is a completely different thing or concept than classical mass, where the mass of a photon is always 0
 
@allquicatic Ah, see there's the thing. I presume you knew why it's impossible to accelerate anything with mass to the speed of light - as its speed (kinetic energy) increases, so does its mass, requiring more and more to accelerate it as it gets faster. AIUI the energy put into it, effectively becomes the relativistic mass of the object.
 
6:58 PM
I guess it makes intuitive sense because light couldn't be affected by heavy objects like stars unless it had some mass
 
Though I'm probably wrong now that you mention it
@allquicatic Well there is that, though some would argue gravitation as a force doesn't work quite the same as other forces that act on mass. Also we still don't really understand how gravitation works at all.
 
the LHC disagrees ;)
not even the people that run it; the literal LHC, the thing, disagrees
it has very strong opinions on these matters
no pun intended
@kerbalspacecat isn't "kinetic energy" just photons affecting baryons to make them more energetic? O.o
 
One interpretation is that gravity is acceleration. Another is that gravity is the physical bending of spacetime so that the light isn't being "affected" by any force but rather is travelling in a straight line through a locally curved space-time
@allquicatic Well, no, because photons aren't the only source of kinetic energy. In fact they're just about the worst source possible
 
> This question has an open bounty worth +100 reputation from allquicatic ending in 7 days.

One or more of the answers is exemplary and worthy of an additional bounty.
5
Q: Under what conditions will a group-regulated PSU damage a modern computer?

AshThese days group regulated power supply design is considered obsolete and is ideally avoided -- primarily due to the poor regulation under crossload. For example, Shilka pulls no punches here, and while the Wolf is more subdued in this review it's hardly complimentary. Others around the place are...

stupid thing making me wait 23 hours
remind me in 23 hours to go back and reward it to bwDraco
*kicks SE for implementing that dumb rule when you select "reward an existing answer"*
need to implement reminders in Cavil
@kerbalspacecat are you sure?
I know, I know, IANAP
 
u are not a potato?
 
7:11 PM
Except it completely misinterprets why Haswell has power supply compatibility issues...
 
@kerbalspacecat thats cuz gravity isnt a force. sigh
i asiked physics about this
they kibda confirmed
 
@allquicatic Relevant MSE: meta.stackexchange.com/q/116072/295684 (now that's a badge-producing link)
 
@kerbalspacecat Then write a better answer in the next 23 hours :D
 
It's just a myth propagated by people who don't understand the difference between a C7 sleep state and a S3 sleep state.
In the S3 state, where most people are complaining about their Haswell+cheap PSU systems "crashing", neither the 5v or 12v rails are even on
 
omfg my ear is driving me nuts
 
7:17 PM
@allquicatic Hey I didn't say I knew anything about crossloading PSUs or the actual question, just that Haswell's C7 states are a red herring and totally irrelevant.
 
7:32 PM
My Lenovo laptop got here faster than I thought it would, and I have to get Linux ready to teach a class starting Monday. Default choice right now is Ubuntu 16.04, but I was wanting to go with Debian or Fedora so I can better learn how my OS actually works as opposed to pushing GUI buttons.
Any suggestions? Anyone know how to make sure I can keep using my Windows license in a VM?
 
@AaronHall Use Ubuntu without a GUI then?
Debian and Fedora both come with GUIs too you know
 
Yeah, I want a GUI for e.g. web and VMs and stuff like that, but otherwise, I don't, really.
I'm sure Fedora on Lenovos are a no-brainer for a lot of people...
Multiboot?
Anybody do this a lot?
grub on lenovo, I guess is what I need to search for?
 
@allquicatic lol
@kerbalspacecat: can you explain what is wrong here? I said nothing about the S3 state. In the C7 state, the machine is running and idle, just not doing anything. The system is not asleep.
@allquicatic: Can you relay this to @kerbalspacecat? I'm ignored and would like to discuss this in detail.
 
Should I try to create a boot thumbdrive from Windows before I do the reformat to MBR? (I've started checking out the directions here: forums.lenovo.com/t5/Linux-Discussion/…)
 
> (and remember that office PCs are often idle much of the time, meaning that this cross-loading condition can persist for extended periods of time)
I never said asleep. I said idle.
C6/C7 are just ultra-low-power idle states. The machine is not asleep or off in either of these states.
In S3, whatever is still on is driven by the +5VSB rail. This has nothing to do with the cross-loading issue.
 
7:54 PM
Is it trivial or at least doable to install Ubuntu 16.04 and then later add Fedora and Debian boot options? On an encrypted drive?
Maybe I should take this to ask ubuntu...
 
@AaronHall if you use UEFI you can just use one install of GRUB2 and a separate root/boot partition for each Linux OS and use the GRUB boot menu
 
thats how you do it on old bios toooo
 
@fixer1234: Thanks for the fix. However, read the above for some context behind the downvote.
 
43 mins ago, by kerbalspacecat
@allquicatic Hey I didn't say I knew anything about crossloading PSUs or the actual question, just that Haswell's C7 states are a red herring and totally irrelevant.
it seems like he doesn't actually know something that would directly suggest your answer is incorrect
 
Hmm...
 
8:05 PM
ok.
gotta run...
 
8:26 PM
@bwDraco, I suspect you're right about the source. That's pretty lame when somebody admits they don't know anything about it. But the bounty should balance things out. :-)
 
> Wow, we didn't see that coming! It looks like someone over at Phison forgot to run the run the power optimization script and integrate it into the firmware.
lol
 
8:50 PM
Okay. Something completely off-topic here. I've run into a TV commercial on OTA HDTV signals which is significantly louder than the rest of the content. However, the FCC requires that all content pass through ATSC A/85 post-processing to ensure uniform volume.
Out of curiosity, does the standard as written contain any loopholes that would allow an advertiser to "beat" A/85 and make their commercials sound significantly louder?
The standard document is here (PDF), if you're technically inclined.
The downvote was because this answer is not only very wrong, but also very misguided. This issue is NOT specific to Haswell - that C6/C7 "problem" has nothing to do with group-regulated PSUs (other than being an example of very minor crossload). C6/C7 problems were down to issues with underloaded 12v2 rails. Group reg PSUs don't have 12V2 rails. Group-reg PSUs do NOT regulate all rails together - only 12v/5v. ATX12V allows ±10% on 12V under load. All ATX2.31 compliant PSUs must handle Haswell-like crossloads >5A on 3.3/5v&<=0.1A on 12v. Haswell C7 was never a real issue in practice. — Potato Cat 9 mins ago
Can we discuss this in greater detail?
 
@bwDraco I get that all the damn time >_<
the extra-loud TV commercials, that is
 

« first day (2225 days earlier)      last day (2812 days later) »