« first day (2873 days earlier)      last day (1963 days later) » 
00:00 - 16:0016:00 - 00:00

4:00 PM
For all I know, the 2019 model will be thinner, lighter, and more shatter-resistant.
4
 
technically microbits, being at the bottom of the Hierarchy. does not have to be hollow. merely compact
 
More expensive is virtually guaranteed though.
 
@Geobits yeah
 
@quartata ahh heck
Wht else to use for seed then?
 
You could have the user roll some dice and then run a PRNG with the results to get a seed.
 
4:04 PM
@Downgoat /dev/random
 
@Geobits Given "various fractures" you'd better hope it's more shatter-resistant
 
Eh, the usual kid stuff. Monkey bars and various jungle gym equipment
 
read(random, &s, 16);
 
@quartata Didn't Downgoat reject that yesterday for portability reasons
 
he hasnt cared about Windows so far
he can add that later, i showed him the function for that
/dev/random works everywhere else
 
4:07 PM
@Geobits but will the 2019 edition be waterproof and support fast charging?
 
geez, I hope it supports slow charging
 
Water resistant I think, not waterproof. Fast charging is likely, but I'm pretty sure it will be wireless either way, so that's good.
 
I know a guy who supports fast charging IRL
He legit gets 3 hours of sleep a night
And I can tell this is consistently true because of Steam in-game tracking
 
Or he bots lol
Just sayin
 
lol
 
4:11 PM
No like, participating in and (occasionally) winning ranked Dota 2 games
 
Yep, sounds like a bot. Doesn't really sleep, competes online in games, haven't seen his face... failed the turing test
 
Well I've seen his face
We go to the same school
 
Yeah, but you could be his cover-bot
 
Good point, good point.
But being fine off very little sleep is actually a well know genetic condition: ncbi.nlm.nih.gov/pmc/articles/PMC2884988
 
@Geobits is microbits also your son/daughter or is it minibits' son/daughter
 
4:15 PM
.oO(How old is minibits?)
 
@Pavel The "Sleepless Elite" as they're known
 
@Pavel That condition shows an average of 6.25 hours, vs 8.06 without, according to that paper. 3 hours is a bit of a stretch over long periods.
@Poke Mine
@Pavel Not old enough for that shit
 
@Geobits I couldn't find a better source since I don't know what the condition is actually called, but yeah it's not that helpful :/
 
0
A: Sandbox for Proposed Challenges

simonalexander2005Where am I? Given an input folder structure (tree), and a pointer to a directory (node) within that folder structure, the program should output the full path to the directory the pointer is pointing at. Alternatively, an actual file system could be used. If this was linux, the rules would be "y...

 
There's a lot more research on disjointed sleep periods iirc, or smaller periods, but more frequently
 
4:17 PM
@Geobits now that you've answered I realize that the question doesn't affect my response. Good job
does this mean you're going to frequent chat less frequently than you already do
;)
 
It will either a) make me less frequent in chat b) make me more frequent in chat c) make me chat the same amount d) all of the above
 
ooh ooh i think it's d
 
@Geobits now that you've answered I realize that you were never going to have a definitive answer
i'm going to go back to work now
also hi
 
0
Q: Code Golf: Letter-Ception

Nicolas BrauerThe Challenge: For an input of one letter X (upper or lower case from A to Z) and one digit N (0-9) print the corresponding letter X made of N * X. The letter has to be from this list: AAA BBBB CCCC DDDD EEEEE FFFFF GGG H H A A B B C D ...

 
Hi!
 
4:21 PM
@Geobits Congratumalations!
 
Thanks!
 
4:33 PM
@Pavel if he gets 3 hours of sleep he should be winning way more often
 
Well it's probably roughly half the time right
Assuming matchmaking works
 
Well. Half isn't occasionally is it :P
 
IDK, is it? ¯\_(ツ)_/¯
 
Ok Google, define occasionally
> Occasionally: adverb -- at infrequent or irregular intervals; now and then
Not very helpful, Google
> I'm sorry
 
> infrequent
> irregular
 
4:37 PM
Infrequent may be helpful, but irregular isn't much help at all
 
> now and then
 
Yeah
 
If I win both now and then, that means I win a lot, right?
 
@Pavel now and then is relative
 
yes
 
4:39 PM
Is this cosmological now and then or bacterial division cycle span now and then :P
 
It's most likely correlated to the amount of sleep the guy gets
Sleep deprivation isn't known to be good for the hand-eye coordination or reflex speed
 
length-of-a-dota-game now and then
@Geobits Assuming it is the afformentioned genetic condition, there are no harmful sleep deprivation effects
 
That also works in favor of my bot argument
 
@Pavel ~45 minutes
 
@Pavel obviously. Otherwise it's not a genetic condition, it's just insomnia
 
4:41 PM
wait dont you play dota
or was that
nathan
 
No I play Dota
 
I don't play dota, doat, dato, daot, dtoa, or dtao
 
I'm just saying, if you main techies your average game length is gonna be longer than 45 minutes
 
@quartata I see how you have the highest number of messages posted now
 
(Not that I know if he does or not)
 
4:48 PM
@Pavel actually it averages out
because either you instantly win or it goes on forever
 
I suppose if you take into account your teammates throwing because you picked techies
 
5:24 PM
0
Q: Find index of largest number

Justin CaseFind the index of the largest number in N(6<=N<=100), if you're given the value of index of 3rd and 4th largest numbers on every input query of 5 distinct indices. Example: For an ascending sequence Output: Input: 6(Total number of indices) 1 2 3 4 5 ...

 
 
1 hour later…
6:44 PM
Wooh, finally k8s on DigitalOcean.
 
Fizz`Buzz
┌────┬────┐
│Fizz│Buzz│
└────┴────┘
@H.PWiz
 
Intersting...
 
7:43 PM
@Geobits ditto
@Geobits Congratulations :)
I assume this means you'll be needing a micro fedora: sophomoredev.com/tag/micro-fedora
 
Anonymous
@Geobits Congratulations! Good to see you back here!
 
0
Q: Recursive Multiplication

JakeRivisheroTwo kids stand over a board of 100 sweets, each with a number from 1-100 drawn on it. They try to eat as many of them as they can and keeping a track of the running total. For example if a kid eats 5,7 and 10 then their running total is 5*7*10 = 350. The winner is the kid with the highest tota...

 
8:02 PM
@trichoplax Contrary to popular opinion, that old pic was one of maybe two times I've ever worn a fedora. And yes, I regret it as much as I should.
 
I never realised you wore a whole fedora...
 
@Geobits Do you also regret the man bun
 
Can't tell if serious
 
@NewMainPosts possibly decent challenge, but missing winning criterion
 
@trichoplax Completely serious. Personally I hate that hair style ;)
 
8:04 PM
@Poke I mean that isn't Geobits in the photo...
 
Oh. He was serious
 
Same question on Stack Overflow suggests this is an Olympiad Question. — Laikoni 19 secs ago
 
@trichoplax hmm. unlucky
i suppose i misread your initial comment
though i do consider that fedora a minifedora
rather than a micro
i'll see myself back to work now
 
My comment makes very little sense if you weren't on PPCG a few years back, and only slightly more sense if you were
 
i was but i guess i also assume that coders own fedoras
and neck beards
granted i own neither
i guess i'm a hypocrite
 
8:09 PM
wait its not downscaled enough
 
@NewMainPosts This contest seems odd, to say the least. I need more info. How are the kids eating candies? Is it alternating turns? If so, player one has a huge advantage. If it's a free-for-all grab-and-smash, the faster kid wins. If the point is to make a kid sick, kudos. If it's to have anything to do with math.... that's unfortunate.
 
well... the mathematical aspect really doesn't account for that
poor kids...
 
@Poke Do you at least have a plaid button-up short sleeve shirt that you wear open over your sarcastically funny and/or retro game t-shirts?
 
@Geobits Also the example given "194 is only possible with 97 and 2, but the other kid must have eaten 2" clearly must depend on picking order somehow.
 
better
much clearer
it looks like a spilled bottle now
 
8:11 PM
oh yeah I see that
 
It does!
 
@Geobits negative. i think i might be a bad coder
 
@Skidsdev Yeah, and the kid to call out a number last has an advantage also, if you assume they don't eat all the candy and he can factor a number
3
 
@Geobits also when does the game end? When all candy has been eaten?
 
okay perfect
 
8:12 PM
> eat as many of them as they can
Terrible end condition
 
So they stop when they can't eat anymore?
Has the person writing this never met a child?
 
I think they've both never met a child and also never invented a game/contest
CMC: How many candies does a kid have to eat to guarantee victory, assuming both are truthful and he gets to pick all his candies first?
 
oooh
so how few numbers in the range 1-100 does one have to pick to ensure that the product of the chosen numbers is larger than the product of all remaining numbers?
 
Yeah. I'd assume you'd just start at 100 and work back from there
So basically, where is the split mark
 
alternatively: What is the largest n below 100 such that n! is smaller than the product of all integers in the range [n..100]
of course that would technically give the wrong answer, but 100 - n would give you the right one
 
8:16 PM
um, you mean (n..100]?
 
right
 
i.e. the smallest n such as (100 - n)! < 100! / (100 - n)!
 
8:30 PM
@Geobits based on my naive implementation, you would need to eat the 42 largest candies to guarantee victory
ie the product of [59...100] is greater than 58!
 
My naive implementation got the same answer, so there is no possible way either of us is wrong
It simply wouldn't be reasonable
 
I'm going to confirm it with a naive implementation of erik's formula
yep, that agrees
 
It stands to reason that the answer is 42
I mean, this is one of the all-important questions
 
of course a better challenge would be: Given an upper limit x, what is the smallest n such that (x - n)! < x! / (x - n)!
 
8:41 PM
oooo
 
I think putting on it would just end up being the same as a fastest code for the factorial function
 
yeah probably
division and lt comparison aren't likely to be optimized
 
Lookup table for the win :D
 
@Geobits good luck encoding a lookup table when 0 < x <= 2^32 - 1
 
Yeah, but you can eliminate factorial computation by making a table on runtime, assuming you have the memory for it.
 
8:47 PM
aah
maximum 1kb of memory
 
Are you confident you can fit (2^32-1)! in 1kb? :P
 
yes
it can fit into 4 bytes
 
Not 2^32-1, (2^32-1)!
 
oh right yeah
hmm
time to ask python
 
after an eternity or so
 
8:52 PM
I suspect a while loop will be faster than recursion due to function call overhead
and also I'm too lazy to google how to change the max recursion deph
I'll let you know when it terminates, I just started it
 
Well, Geobits is back ... it seems
 
@Zacharý While not very confident, you still sound more confident than you should
 
...
 
@Zacharý he's back... today
 
@Poke I assume that it is just now that he has an opening in his schedule?
 
9:01 PM
@Zacharý Using the last few months as any indicator I'd say so.
he also wanted to share that his wife is perfnant
 
@Geobits when one of your kids grows up, will they become macrobits?
 
Depends on the size I guess
 
Oh.
 
Although there I go assuming again. All he said was that there was a microbits on the way
 
@Geobits what's your SO? Aerobits?
 
9:02 PM
Could be adoption or something
 
Details, details. Just assume it's good news and all will be well :D
 
Maybe Geobits is just Angelina Jolie's internet persona
 
How did you know????
I thought I hid that rather well :/
 
@Geobits I studied the nude scene in Wanted
 
My questions are leading up to the obvious: When do we get a Geobyte?
 
9:04 PM
Ah, right. Damn
@Skidsdev Eight? You must want me dead
 
6 your wife and you count as well
 
Well to be fair, you, minibits, and microbits are all plural
you could already be a geobyte
 
true
 
Or a geoword or geodword
 
I can tell Geobits is back because half the starboard says by Geobits
 
9:07 PM
ahahaha
 
Poke and Geobits in TNB. What is this, 2016?
 
9:45 PM
who wants to hear a joke
 
@mınxomaτ I like to lurk in here even if I'm not talking regularly :]
 
10:11 PM
Hit me with it @LeakyNun
 
@Veskah youtube rewind 2018
 
Ayyyyy
 
10:30 PM
@Geobits you made quite some progress in rep since then.
Ah you probably just un-did some of your down votes.
 
10:52 PM
@El'endiaStarman It seems some of those permalinks on TNBDE are more like templinks
do you know what might have happened/is this on purpose?
 
11:16 PM
@LeakyNun and 2017
 
11:34 PM
@quartata hm ok
also @quartata apparently weird things happen when casting an unsigned integer >2^31-1 to a double, do you know why this happen :| like this Double::(UInt32::1 << UInt32::31) emits 0x41E0000000000000 which is completely different value and bit format
 
Off the cuff guess with not much merit: Conversion process treats unsigned int as a signed int while converting to a double?
 
11:52 PM
@flawr Hrm. Two possibilities I can think of: 1) I changed the permalink URL format and this still broke despite my attempt to provide backwards compatibility (or maybe that was just for VarLife) or 2) the text files corresponding to that permalink got deleted for whatever reason at some point.
Looks like the permalink format is the same so it's probably the second.
Actually no, the text files are still there as determined by SSHing to it. Wot.
 
00:00 - 16:0016:00 - 00:00

« first day (2873 days earlier)      last day (1963 days later) »