The Nineteenth Byte

Anonymous

12:01 AM

@DJMcMayhem That's not a good idea, because the new posts would inherit the votes of the old post.

I got a new computer, did I mess up paying $1,299 for this: acer.com/ac/en/US/content/predator-helios300-series (1060GTX, i7 (7th gen), 512GB SSD, 16GB RAM, 1 TB HDD)

It seemed like a great deal, considering NewEgg had it at $1,199 and the place I got it at let me pick it up in-store.

Destructible Lemon

@MagicOctopusUrn how is paying 100 more than another source prices it a great deal?

Downsheep

@MagicOctopusUrn It was a couple months ago so idk if fair but I was able to get my rig for $1.1K with gtx1070 1tb ssd 16gb ram and i7-7k

not sure how much being a laptop would increase price though

Magic Octopus Urn

@DestructibleLemon I smashed my laptop on the ground in the morning, had the extra $100, and paid for the convenience of not not having a PC for 7 days.

Destructible Lemon

@MagicOctopusUrn ok but i wouldn't say great tbh

like, acceptable would be my adjective of choice

Magic Octopus Urn

12:08 AM

@Downsheep The same laptop was 1,399 at my microcenter (same specs).

Anonymous

@MagicOctopusUrn I don't think so. That's a fair price for those specs.

Magic Octopus Urn

@Mego Microcenter... The store that exists in random places but doesn't rip you off, yay.

@DestructibleLemon After playing a buncha games on it compared to my 960M/8GB/i7(5th) the biggest difference is the graphics (obviously) and the heat that was killing my future children lol.

960M could still run all the games at 1/2 the FPS of my new one... So sad I knocked it off the counter in my kitchen.

Pheo

Thats pretty sad.

Magic Octopus Urn

@Pheo amazon.com/Acer-VN7-591G-792U-Generation-GeForce-Windows/dp/… This is the bud I killed in a moment of clumsy.

Pheo

@MagicOctopusUrn Wanna know something strange?

Magic Octopus Urn

12:13 AM

The corner struck the ground, and shattered the motherboard from bottom left to top right... All the way through. If it had hit any other corner it'd be fine.

@Pheo Also, sure, I love strange.

Pheo

It looks exactly like the one I am using right now.

Magic Octopus Urn

@Pheo Whatcha using?

Pheo

Chromebook acer

Magic Octopus Urn

@Pheo I have a chromebit

Pheo

Ack! The XKCD is filling my screen.

Magic Octopus Urn

12:14 AM

I personally have always hated ACER for forcing horrid bulky install packages on stupid people... But then I realized they sell the same parts at a lower cost. All I have to do is reinstall windows for an ASUS at a lower price...

@Pheo LOL, I can't tell if it's my browser limiting the options or it's just the end of the option tree.

Pheo

Umm.... Save is nowhere in the options.

My favorite: File -> Find -> who -> 's on first

Magic Octopus Urn

There was a laptop at micro center with 2 1080 TITAN gfx cards, 64GB ram, 2TB SSD, i7 (last gen) for $8,999.00 and I just laughed when the dude showed me it (while I was wearing a hoodie and cargo shorts) "If yer lookin' fer powuh luk no furtha'".

@Pheo WHO?!?!

bader

hi

Magic Octopus Urn

@bader 'allo

ASCII-only

12:37 AM

@MagicOctopusUrn there are recursive menus

DJMcMayhem

@Mego Votes are a terrible indicator of challenge readiness

I think inheriting old comments is a bigger issue

Riker

12:53 AM

code jam hype

ASCII-only

wait. is code jam normally this late in the year

Christopher

:/ "Additional library" just counts as "another language"... — user202729 yesterday

feedbook?

@Dennis dude I keep staring you messages and then reading the context. :P

that is a sandbox link but comment pop

quartata

nah there's a comment hash at the end

Christopher

w/e

ASCII-only

1:09 AM

@Christopher is that not valid...

Christopher

@ASCII-only i ment to post sandbox link without hash for comment :P

ASCII-only

@Christopher :|||||||||| how did you manage to do that 0/10

Nathan Merrill

1:25 AM

@cairdcoinheringaahing I have a significant contribution to that number. I would not want the sandbox to be cleared, nor do I want to post them yet

Posting a challenge is a significant time commitment to me.

Regardless, the size of the backlog really isn't the problem. With the exception of loading times on mobile, it doesn't really cause any issues. (Deleted posts appearing and people not reviewing it is far more of an issue than the size)

user202729

@quartata "comment hash"?

Nathan Merrill

@Mego that post was obviously planned: Why did you choose to have you answer it, and DJ post it?

Anonymous

@NathanMerrill The question was from the moderation team. We wanted one of us to answer it to give our perspective, so I wrote an answer that reflects my opinions.

Nathan Merrill

it just seems easier to have the same person post the question and the answer, but w/e :P

not a criticism of the post, I think its a great question

Anonymous

@NathanMerrill Having the same person post the question and the answer makes it seem like a unilateral decision being made by the mods.

Anonymous

1:32 AM

Rather than a proposal with a request for opinions, accompanied by an opinion

Nathan Merrill

I guess? I can't think of a time I've ever thought that on a meta post, but I also can't think of a post where a mod both posted a question and an answer at the same time

New Meta Posts

Can we raise the privilege levels?

11

totallyhuman

Hmm.

I have a feeling most people were more interested in the design than the privileges.

Nathan Merrill

The design is cool, but the privileges are important

Anonymous

@NathanMerrill We try to avoid it because it's not hard to get the wrong impression

Anonymous

1:37 AM

Considering that we made our own design with a userscript, the privileges are really the main thing

totallyhuman

No, I definitely agree.

Nathan Merrill

@Mego it seems as if you talk out of experience?

I personally don't see it, but I also don't know the moderation world

totallyhuman

(It's not like I'm losing any important privileges anyways.)

Nathan Merrill

I vastly prefer the low-responsibility world :)

Anonymous

@NathanMerrill I don't have any specific experiences in mind. I just see how someone could get the wrong impression. Because people (especially new users) tend to take everything moderators say as statements of fact or law, we have to be extra careful about what we say.

Nathan Merrill

1:39 AM

Yeah, I absolutely agree with that. I do that. It's hard not to do that.

Leaky Nun

"Yeah, I absolutely agree with that" sounds sarcastic

user202729

> I'm using "healthy" to mean "we actually upvote posts" (as opposed to the voting culture on Stack Overflow).

Nathan Merrill

not sarcastic :)

user202729

Actually... my experience is quite the opposite (for SO).

ASCII-only

@Pheo +1 for the link at the end

user202729

1:41 AM

(normal posts get a lot of upvotes, sometimes bad posts are upvoted for unknown reasons)

totallyhuman

@totallyhuman Wait, crap, I'll lose VTD abilities.

user202729

Mego has less than 25k too and so is PhiNotPi

quartata

Re privileges: I'm about 80% sure the CMs actually asked us if we wanted to have our privileges raised immediately or at design and the answer was at design

Anonymous

@user202729 It actually doesn't matter for me :P

user202729

(I don't know what to do with the site analytics anyway)

totallyhuman

1:42 AM

I need 7k rep, stat.

user202729

@Mego Uhm... right.

Anonymous

@quartata That was back when we thought the design would come somewhat soon after graduation

quartata

I know, just pointing out that a. This isn't new and b. This is feasible

user202729

(anyway what do people do with the site analytics?)

quartata

they're pretty nice

Anonymous

1:44 AM

@user202729 I constantly see posts getting downvoted very quickly after being posted on SO. Granted, quite a few of them should be downvoted, but not all of them. Also, good, detailed answers will get maybe 5 upvotes (Jon Skeet and Eric Leppert notwithstanding). We vote up a lot more, considering our vastly smaller userbase.

user202729

>>>What<<<? Really? 1 rep to vote up?

Anonymous

@user202729 They're nice for monitoring the traffic.

Anonymous

@user202729 Yes. Literally anybody can vote up, so long as they register an account.

totallyhuman

You need 15, IIRC.

Anonymous

It's 15 for designed sites

totallyhuman

1:45 AM

codegolf.stackexchange.com/help/privileges/vote-up

Anonymous

Oh man I was looking at the wrong column

totallyhuman

stackoverflow.com/help/privileges/vote-up

Doesn't change, apparently.

Anonymous

Yeah I was looking at private beta

Anonymous

Oops, egg on my face

totallyhuman

Can I have a link to the chart you seem to be looking at?

quartata

1:49 AM

If you needed 15 rep to upvote in private beta no one would get rep

Anonymous

@totallyhuman Last link in the meta post

totallyhuman

Going to the help center is tiring...

Anonymous

50

A: Reputation requirements compared

Sites There are four main types of privilege thresholds (sorted from high to low): Stack Overflow and graduated sites with a design (example) public betas and graduated sites without a design (example) Ask Patents Private Betas and this Harvard University site. Except for Ask Patents, these ...

totallyhuman

@Mego Oh, thanks.

user202729

@quartata Accepting answers and accepted answers.

(which was very common on PPCG)

Zacharý

1:58 AM

@Mego Is this a meme or something that I'm unaware of?

Anonymous

@Zacharý It's a common saying in the US - it's like "shame on me"

Nathan Merrill

...is there a name for the system where: "If you get 1st place, you get N points, if you get 2nd, you get M points, etc"?

totallyhuman

Damn, SE still hasn't fixed the 1000k bug.

user202729

@totallyhuman That's intentional.

Anonymous

10

Q: Origin of "egg on my face"

Where does the phrase "egg on my face" come from, and what is its meaning?

user202729

2:00 AM

I have 7759 rep and it isn't displayed as 8k.

10000k would be displayed as 10m.

totallyhuman

@user202729 That doesn't make sense, considering how 1.0m is partially implemented.

stackoverflow.com/…

vs.

https://stackoverflow.com/users/22656/jon-skeet

user202729

I don't see any 1.0m.

Zacharý

^

user202729

(no onebox please)

And most importantly Jon Skeet don't want that.

totallyhuman

@user202729 Look at his Communities list.

user202729

2:05 AM

So what? My communities list display 7.8k for PPCG.

That's by design.

totallyhuman

@user202729 Everywhere your rep is shortened, it's displayed as 7.8k. It should be the same for 1.0m, but it's not. Not to mention 1000k is ridiculous.

user202729

My non-shortened rep should be 7759×1 then. (or whatever unit used for 1 rep)

totallyhuman

Your non-shortened rep is and should be 7759, just as Skeet's is and should be 1,017,768.

user202729

half-shortened or whatever it's called

Destructible Lemon

2:23 AM

@user202729 I see 1.0m as well though

mbomb007

3:04 AM

@user56656 I was using deduction for a couple of lemmas.

So I had an assumption, reached a conclusion, and from that I proved the implication.

user56656

Oh ok, so you weren't using sequent calculus or anything like that.

mbomb007

No

It was just hard for me to piece it together Hilbert-style

It's been 7 years since I took that class, too

Riker

3:33 AM

@ros @mods mind if the starred link is edited to "Can we raise the privilege levels?" (with a link)? kinda hard to tell what's being starred as-is (wihtout clicking through on mobile)

Anonymous

4:09 AM

@Riker eleven'd

Downsheep

5:25 AM

btw @Mego could you change my username back to Downgoat

Magic Octopus Urn

5:42 AM

c

betseg

6:09 AM

github.com/michael-lazar/rtv thats one nice client

Cows quack

6:36 AM

@mods could you please unfreeze chat.stackexchange.com/rooms/56779/code-jam-stuff

DJMcMayhem

in code jam stuff, 39 secs ago, by Feeds

status-completed

ASCII-only

6:48 AM

anyone here good with .NET/CIL?

ASCII-only

7:55 AM

nvm

caird coinheringaahing

8:14 AM

@NathanMerrill I'm definitely not suggesting that we go through the Sandbox, post every challenge to can be posted and delete the rest. It'd work, but it'd also annoy plenty of users, such as yourself. Instead, I'm suggesting removing the obviously finished posts, whether they be ready to post, or just terrible ideas. They're what's clogging up the Sandbox, not the continually-maintained proposals such as this

Posts such as this, which haven't been touched since 2014 and where the original poster was Last seen Nov 13 '17 at 10:10, go stale, gathering one or two upvotes and never get posted. That is what I'm aiming to remove, because there are literally thousands of such posts

ASCII-only

1

A: Sandbox for Proposed Challenges

Play Your Cards Right code-golf Game Rules Play your card right is a game played with a pack of 52 cards (no jokers). At the start the deck is shuffled and the dealer will lay a single card on the table. The player then has to decide if the next card is higher or lower than the card on the ta...

> controller coming tomorrow

caird coinheringaahing

Exactly

ASCII-only

the one after that is a terrible image-manip

flag one after that is super hard

caird coinheringaahing

This finally explains how Martin found the time to make the Hexagony quine :P

FrownyFrog

8:31 AM

on his github hexagony page it say "pointy-topped grid"

but it’s flat-topped

tut tut

ASCII-only

@cairdcoinheringaahing :|||||

note to self: ask people here how bad on a scale of 1 to 10 how bad a lisp-based golflang would be

also: how many people here actually use Esoteric IDE? Enough to make Charcoal worth porting to C#? (sadly IronPython 3 isn't out yet :/)

caird coinheringaahing

Q: Anyone have a main-ready Sandbox post I can use as an example?

TFW you get more upvotes in the Sandbox than on main :/

ASCII-only

O_o

caird coinheringaahing

9:01 AM

Any mods around?

Ok then. Posts a ton of spam :P

Leaky Nun

@cairdcoinheringaahing do you python?

caird coinheringaahing

@LeakyNun Yes, somewhat. Why?

Leaky Nun

can you help me golf python?

caird coinheringaahing

Probably not, I'm not that good at Python golfing, sorry :/

Leaky Nun

ok

caird coinheringaahing

9:12 AM

There's this (frozen) chat room for Python golfing, but I'm not sure if that would be of any help

@LeakyNun I might as well have a look at it though :P

Leaky Nun

2

A: Integers in Base Pi

Python 3, 471 317 bytes Surely there are golfs that I missed. Feel free to point them out in the comments. def h(Q,a=0,b=4,c=1,d=1,o=3,s=1,P=9,C=6): while P^C: a,b,c,d,s,o,P,A,B=b,s*a+o*b,d,s*c+o*d,s+o,o+2,C,0,1 for I in Q:B*=c;A=A*a+I*B C=A>0 return P def f(n,p): Q=[-n];R="" while ...

go ahead

caird coinheringaahing

@LeakyNun -3 bytes by switching to Python 2 and using ` instead of str.

Yeah, I'm really not good at golfing Python

Leaky Nun

lol

caird coinheringaahing

-6 bytes by changing indentation :P

-3 bytes by moving the assignment in h

ASCII-only

9:57 AM

@LeakyNun O_o 160 byte golf

Leaky Nun

lol

ASCII-only

:/ Charcoal doesn't have RealDigits yet

New Main Posts

5

Q: Output a binary path from a number

For an integer n that satisfies n > 0, write its value as a right-descending path based on its binary representation. Rules The first (most significant) set bit is always in the top-left corner. When the next bit is set (a 1), draw a character ("filled") on the next line in the same column as ...

code-golf number printable-ascii

ASCII-only

10:34 AM

@Neil you've probably seen this already but you haven't responded so i'm not sure, but: here and here you can make one of the Jump arguments a string since it auto-casts anyway, removing the need for a separator

Neil

yeah, I remember you mentioning

ASCII-only

also here

oh @Neil it seems inclusiverange isn't needed anymore since incremented is a thing, right?

Neil

I guess so

although what about InclusiveRange("0", "9")?

ASCII-only

10:49 AM

oh, that's a good point

offtopic: i rediscovered one of the many things i intended to fix but never did

Neil

11:18 AM

@ASCII-only would you mind reminding me, does Charcoal have either or both of a) list concatenation b) pairwise list addition

ASCII-only

both lists/both strings fallthrough to the last statement here: github.com/somebody1234/Charcoal/blob/master/charcoal.py#L3667

so no pairwise addition

unless i forgot something

Neil

OK np

ASCII-only

hmm. i wonder how useful some kind of multifor/multimap would be (map over several iterables at once). also wonder about elementwise operations too

oh also i wonder what operations would be useful for lambdas? e.g. composition? Doubled to make a lambda run twice? (should probably just double result) (note: lambdas probably work really really terribly)

Neil

I think lambdas broke default: { } as it reads it as case { }: instead

ASCII-only

oh :/ oops

caird coinheringaahing

11:29 AM

Something's not right here...

CMC: What's not right in that photo?

ASCII-only

@cairdcoinheringaahing suffers from link rot

caird coinheringaahing

No, there's actually a discrepancy in the picture :P Might be difficult to find though :D

ASCII-only

@cairdcoinheringaahing harry potter tab

caird coinheringaahing

@ASCII-only Nope, that's a fanfiction I'm reading

ASCII-only

@cairdcoinheringaahing post is cut off :|

caird coinheringaahing

11:34 AM

It's a long post :P

A hint: Something doesn't add up here...

ASCII-only

where

caird coinheringaahing

In the picture

user56656

less views than votes

caird coinheringaahing

@user56656 ✓ Nice :P

ASCII-only

oh. >_> didn't even think of checking views

@Neil Would you mind giving a testcase?

Kevin Cruijssen

12:02 PM

@Neil Can you explain a way to loop over a list but skipping the first item? I tried something like Assign(InputString(), a); For(Slice(a, 1, Length(a)){...}, but the Slice doesn't seem to work and gives an error. What would be the best approach to loop without the first item?

@Neil In Charcoal I mean**

Neil

@ASCII-only ^^^ I was having problems with Slice earlier, meant to mention it

Kevin Cruijssen

@Neil Is there a workaround for now, until Slice gets fixed?

Neil

for a string, Reverse(Mold(Reverse(a), Decremented(Length(a)))) might work

also, InputString(a); instead of Assign

caird coinheringaahing

Would a moderator mind unfreezing this room please?

Kevin Cruijssen

@Neil Thanks, that indeed seems to work.

New Meta Posts

12:14 PM

1

Q: It's time for some Spring Cleaning

I'll cut straight to the point here: We must do something about the Sandbox, and soon The Sandbox recently hit 2000 non-deleted answers (5700 total). That's over 2000 potential challenges festering in some corner of the Internet, most likely not going to be posted for years, if ever. I'm propo...

discussion sandbox

caird coinheringaahing

12:45 PM

Anyone up for another picture quiz? :P I just finished a new one, which I think is much harder :P

Leaky Nun

someone help me golf my answer pl0x

moonheart08

12:57 PM

please never bring "pl0x" into this chatroom. it reminds me too much of ROBLOX (Altho, i can say that it's earlier years were great. Stuff like survival 303 was awesome)

Anonymous

→ 2 messages moved to Trash

mınxomaτ

→ 1 message moved to Trash

Leaky Nun

sorry

Anonymous

@Downgoat Done

Nathan Merrill

1:16 PM

@cairdcoinheringaahing I'm really not convinced that we need to go back and remove poor posts from the sandbox. The size of the sandbox doesn't really affect anything. I do think it would be useful for us to say "Hey, if you are having trouble coming up with ideas, perhaps you should peruse some old posts in the sandbox"

Anonymous

@NathanMerrill I agree. I think the main problem is that people don't look at the Sandbox.

ASCII-only

@Neil deverbosifier? that should be fixed when Dennis pulls. I think. (wait, Dennis has already pulled. you might want to test again)

user202729

Even the newer posts in the Sandbox receives little care. Assume everyone sort by active, the sandbox being huge is not a problem.

moonheart08

1:37 PM

aww, i'll lose access to review once the priv levels are raised. Oh well, i wanted to start golfing again anyways lol

caird coinheringaahing

Here's how I see it: there are a multitude of problems with the Sandbox, but people view it as 'too important' to shut it down or do something that can fix those problems. One way I can see that fixes the 'feedback' issue, and make it smaller (whether that's important to you or not), is to have a Samdbox with a fixed number of answers, say 100, where new users get priority. But we won't do that, for various reasons.

Nathan Merrill

We need motivation for people to review in the sandbox.

Perhaps a fixed limit would provide that, but I'm not sure

user202729

Locking the sandbox doesn't really help getting people to review it...

caird coinheringaahing

If you want to post a challenge, but need another challenge to be posted in order to do so, you'd be more inclined to give feedback to the existing challenge.

Laikoni

@cairdcoinheringaahing People would just stop using the sandbox in that case.

Nathan Merrill

1:42 PM

right, that'd be the primary source of motivation, but it'd be rather ineffective at actually moving it out. I'd have to post a bunch of comments, and then wait a couple of hours before finally getting my challenge in

Poke

@NathanMerrill would result in low quality comments, probably

annoying flow for the user

Nathan Merrill

yeah. "Are you going to post or delete this" comments

I'd love it if we could provide rep for upvoted comments in the sandbox

...Do you guys think that a sandbox.stackexchange site would be closed?

like, its a standard SE site, but the entire goal is to provide feedback on posts for other sites

Poke

@NathanMerrill that would be a mess

Nathan Merrill

would it?

Poke

i imagine so

Nathan Merrill

1:46 PM

It's like codereview, but "postreview"

Poke

i don't want to see sandbox posts for IPS

etc

Nathan Merrill

you'd just have tags for the different sites

Poke

to downscale that a little bit

why don't we have a meta post per challenge

a sandbox meta post

like your idea you could have a tag for sandbox posts

Nathan Merrill

we've talked about that. I'll try to find the meta post

I'm having trouble finding it. IIRC, the biggest issue was the flood it would cause, drowning out more important issues.

Poke

maybe we should ask SE to add a "not" for tags in searches

not: [code-golf]

Nathan Merrill

1:53 PM

I believe that exists, but either way: We want way more emphasis on the actual meta questions, and having the average joe filtering those out is impractical

Poke

the average joe doesn't post on meta imo

Nathan Merrill

61

Q: Filtering questions by multiple tags

I know you can select a tag and then click on unanswered to view the unanswered questions for that tag. But is there a way to view questions only for a certain set of tags > 1 tag? Mainly I'd like the union of tag1 with tag2 with tag3 with tag4. But also the intersection of tags would be nice a...

support faq tags search

The average meta joe :)

the easiest path is going to be the one most taken. It's why default settings are so important

Poke

challenges are what drive the site. if they're flooding out actual meta discussions maybe that's not a terrible thing?

i assume the problem we're trying to solve is the number of items in a single post (the sandbox)

@NathanMerrill is also why a sandbox SE probably won't be used

user202729

There are prefix - but there is a bug...

that prevents it to work without something else.

Nathan Merrill

Possibly. I mean, I don't know why people participate on codereview either

I think it solves the motivation issue though: Now you are getting virtual points for reviewing

Poke

1:58 PM

@NathanMerrill i imagine the result of a codereview post doesn't end up on another SE site

Nathan Merrill

...ok? why is that an issue?

Poke

it's not an issue

i just don't see it working the same way

Nathan Merrill

Yeah, perhaps it wouldn't.

Poke

you could make a meta post about it :P

Nathan Merrill

Perhaps, but I don't think I'm really the person to lead the crusade. I enjoy reviewing posts, but I'm not super active about it and I don't think that would change with a new site

which is why you may be totally right: we very well might have the same motivation issue

caird coinheringaahing

2:11 PM

@Poke I beg to differ

Poke

@cairdcoinheringaahing i'm sure this is the general case :|

moonheart08

2:48 PM

cmake somehow confuses me when it appears to be the simplest thing ever

>_<

i have no idea how to quickly include every .cpp in my source dir, of all things lol

Pavel

I just let Clion do that for me

moonheart08

i prefer scons anyways. whips up a quick sconstruct script instead

C++ is confusing. even with clang's better error messages.

> error: reference to type 'const int' could not bind to an lvalue of type 'int const[4]'

i mean

i see the issue (& vs *)

but still

and seriously, why is it reversing the order of 'const int'

Pavel

I make my MAKEFILEs by hand when not using Clion

moonheart08

i don't like makefiles :c

Pavel

Just run g++ **/*.cpp :P

It all becomes hell with large project anyway

moonheart08

3:03 PM

heh yea

(It's even more hellish for The Powder Toy, some parts of the codebase are still in C land, using C style code, while pretending to be C++11)

Pavel

NetHack has a script to generate a configure script to generate a MAKEFILE and it's almost guaranteed to fail without manually modifying these build scripts for some reason or the other even when targeting Linux.

moonheart08

lol

Pavel

I had to edit one of the .c files because one of the includes was pointing to a file in the wrong directory

moonheart08

~~rewrite nethack in rust~~

Pavel

I can only imagine what compiling it for Windows is like

moonheart08

3:06 PM

probably soft, velvet covered rainbows... that secretly want to kill you.

Compiling anything for windows is awful unless it's a VS project >_>

and thats without trying to add even a single library dependency

Pavel

No way it'll build with the visual compiler

You need MinGW

moonheart08

Obviously. MSVC has issues anyways

Pavel

And NetHack has dependencies on Yacc, Flex, and a few other things

moonheart08

(It's really not that good at optimizing compared to GCC and Clang. And GCC now has experimental automatic threading of all things)

Pavel

It doesn't take into account that no one uses Yacc.

moonheart08

3:08 PM

> -floop-parallelize-all -ftree-parallelize-loops=8

Pavel

I had to edit one of the build scripts to replace yacc with bison -y

DJMcMayhem

Aww, and I thought we'd finally get our design on November 1st. — Erik the Outgolfer 2 mins ago

@EriktheOutgolfer Of which year? :P

moonheart08

i hope our design is dark themed. dark > all

Pavel

Except for the dark gradscript theme

Erik the Outgolfer

@DJMcMayhem at least you got the reference...

the year is intentionally omitted

Pavel

3:14 PM

in PPCG v2, Nov 5 '17 at 9:20, by Mego

@Sherlock9 November 5, 2020. Mark your calendars.

Erik the Outgolfer

unrelated to the design

Pavel

It's at least somewhat related

J. Sallé

@moonheart08 yes

caird coinheringaahing

3:33 PM

A better answer to this would be greatly appreciated. I don't think the existing answer covers the issue very well

totallyhuman

3:49 PM

reddit.com/r/Python/comments/8a8jb8/…

caird coinheringaahing

@totallyhuman Obligatory Comic Sans is terrible comment

moonheart08

4:43 PM

I try to find a solution to a libGL issue. I find a stackoverflow issue on it. It has no answers. ):

betseg

lol they have tumbleweed for that question

if you have the same problem upvote the question

moonheart08

Also, OpenCL is going to drive me insane

:v

Lynn

golf.shinh.org/p.rb?FizzBuzz#Haskell how in the...

I think henkma found a 81B Haskell FizzBuzz that uses some of the new Prelude (because Haskell got upgraded on golf.shinh)

I really wanna know what's up with it!! I can't get below 85B. I posted the 85B somewhere on here

I think shorter solutions use gcd 15n in some super clever way but I can't figure it out ~.~;

12

A: 1, 2, Fizz, 4, Buzz

Haskell, 85 bytes main=mapM(putStrLn.f)[1..100];f n|d<-drop.(*4).mod n=max(show n)$d 3"Fizz"++d 5"Buzz" Getting close to henkma's 81 bytes, but not quite there yet. d = drop.(*4).mod n is key here: d 3 "Fizz" is drop (n`mod`3 * 4) "Fizz". This is "Fizz" when n `mod` 3 is 0 and "" otherwise.

I wish I could forget this solution and reset my brain...

moonheart08

NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) No platform
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) No platform
clCreateContext(NULL, ...) [default] No platform
clCreateContext(NULL, ...) [other] Success [AMD]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) No platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) No platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No platform

H.PWiz

I tried for multiple hours a while ago (fruitlessly)

moonheart08

4:52 PM

WTF amd

all i wanted was working OpenCL

:C

mınxomaτ

> working OpenCL

Haha, good one.

Lynn

The trick is probably to operate on "FizzBuzz" right?! I thought of like, "FizzBuzz"!3!5 where ! is something clever... but no dice

moonheart08

@mınxomaτ now try and make it work on one of AMD's integrated Radeon graphics units. (APU)

Lynn

I hope it's a magical formula I could never find, so I can not beat myself up so hard about not finding it.

moonheart08

i mean, this APU has 6 compute cores dedicated to graphics. i'd love to actually be able to use them :v

J. Sallé

4:55 PM

@Lynn might also be something incredibly simple that no one would ever think would work

New Main Posts

0

Q: Expand Sine and Cosine

Trigonometry has LOTS of identities. So many that you can expand most functions into sines and cosines of a few values. The task here is to do that in the fewest bytes possible. Identity list Well, the ones we're using here. sin(-x)=-sin(x) sin(π)=0 cos(-x)=cos(x) cos(π)=-1 sin(a+b)=sin(a)*cos...

code-golf trigonometry

Transcript for