>>107833429
kek @ likely indian saying Linus' work is unacceptable
>but he didn't want to learn, and that's not okay!
one only has the time to learn a finite number of things

>>107834403
I'm suggesting it failed to follow the instructions in its CoT and then tried to retroactively shoehorn the result back to "no arithmetic" in a nonsensical fashion. You also failed to follow the instruction in your CoT but stopped at that.

File: 1761341221164679.jpg (83 KB, 1024x1024)

83 KB JPG

>>107831372
What's the copium faggots will use to tell you AI is just a bubble?

>>107831512
NOOOO TRUST THE EXPERTS!!! Castrated brain.

>>107834471
So it failed to follow the no arithmetic rule for adding the evens and odds? Okay, I guess that counts for arithmetic. What's the correct solution then? You can't propose a new sequence without knowing its sum. You can only deduce that although Jane's meets the criteria Bob then must be thinking of a different set of primes.

>>107833317
when m5 max though saar

>>107832992
i dont need, but i want

got a m3 airbook with 16/256, but that's a bit to small. so i want more disk and ram

>>107833690
No argument, if you want linux ARM machine, just buy the upcoming snapdragon x2 and pray fedora or someone manages to support it.
Using macOS and never trying to learn the key binds or how the environment works is waste of time.
Also if you want USB to work on both Mac/Windows just format is as exFAT.

I wouldn't do it. I went for the mini and it has gone unused for two months now (I have a windows laptop and a linux desktop as well).
At least apple ads don't work on me anymore, because when I think about getting one I remember I already do and never boot it up for anything.

>>107814412
The M4 chip delivers outstanding performance and battery life. macOS stability is version-dependent. While some users report no issues with Tahoe, widespread criticism cites compatibility problems, UI bugs (especially with Liquid Glass), and instability that's led many to advise waiting or staying on older, stable versions like Sonoma or Sequoia for critical work.

File: Z-00061-Short thin tail s(...).png (979 KB, 1152x768)

979 KB PNG

Morning anons

haven't seen him since killing LTT has he got a taste of that drama money and gone down the inevitable self appointed town sheriff?

>>107830239
?? But white people took over Venezuela in like 3 hours just recently? Lol

File: 1767398333305536.png (82 KB, 904x828)

82 KB PNG

>>107830928
It's trolling of course. Like pic related.
But as all good trolls usually have, there's a kernel of truth in it. Majority of anti ai posters are also usually aligned with other lefty causes and are pro tranny. The ones that screech the loudest anyways. It's a bit tempting to poke fun at them.

>>107830649
Kek I love people whining about ai. That alone justifies the trillions of dollars thrown at it

>>107834584
If by being pro gamer you mean a pedophile then hes the biggest gamer of them all, kys tranny

>>107831217
>$400
>for a worse keyboard than the Danger Hiptop
Let me know when someone makes a Danger Hiptop smartphone clone, then I'll be interested.

Every other phone keyboard fucking sucks.

>>107834607
The only people who say Danger Hiptop are zoomers who watched a YouTube video. It was the Side Kick you fucking weirdos who read a production name and then called it that belong in the ganges

Considering doing so, but i need to know if the SoC isn't shit like in the Kosherphone.

Nah. Shit SoC and too risky in general for an unproven product by an unproven company.

>>107831217
>Ctrl and alt keys
I'd consider it just for that so I can use the android version of Emacs and not have it be a fucking pain.

>>107834864
Yeah you totally should

And all this
While he literally runs out of money
...
;]

new thread
>>107834867
>>107834867
>>107834867
migrate

is ranfaggot esl? what do the rentries have to do with the topic of the thread?

>>107834874
let me guess you're also not ani you're just his biggest whiteknight (despite him being a discordfag)

>>107824313
Q2 26.

>>107824355
yeah naw, they already tried that, that is basically what rdna was, turns out focusing on gaming only is really fucking stupid, because you cant repurpose it for other workloads, case and point, the radeon vii can run fsr4 better than rdna 2, , which is ridiculous, basically the reason amd is so far behind on the ml upscaling and frame gen , is because they opted to gut all the advanced capabilities in favor of raw raster performance, while nvidia can sell its ewaste silicon to consumers, while selling the highest bins as quadros and compute for >$10k a pop.

>>107832283
they saw novideo make bank selling cheap garbage like kepler, maxwell, pascal and figured they can cut their shit down too. just when nvidia switched to more compute

>>107824286
The GPU are selling at MSRP

>>107824369
>the ceos are cousins
They are not.
>amd had already said they were withdrawing from flagship gpus
For this gen. They obviously want to have a high end competitor but couldn't make it work with RDNA. But we'll have to wait for UDNA next year.

Finally, a browser thread. Alternative to Firefox that's not a fork of it or Chromium? Also, light on RAM consumption.

>>107834378
>Microsoft should have just quietly switched Internet Explorer to Chromium instead of wasting time with Edge and having two browsers running at the same time.
EdgeHTML was the fastest browser engine at the time of its death. The only reason they dropped it was Google fucking with all their sites to constantly break it.
We'll never have scrolling as smooth as it was in old Edge.

>>107834446
K-Meleon

>>107834446
What your asking doesnt exist

>>107834538
But Dillo comes close. Too bad my distro doesn't have the pkg... First was XFE now not even Dillo? Aw, shucks.

File: 1761599508588733 (1).jpg (70 KB, 564x563)

70 KB JPG

>>107830366
Shovels?

>>107830409
>tl;dr
Line must go up.

>>107830414
/thread

>>107831288
It's a simulation, dumbfuck.

File: MV5BYTc4MDJmNGEtZTkzYy00N(...).jpg (84 KB, 850x478)

84 KB JPG

>>107831093
There will be jail if they lose all their investor's money and sell the company in the cheap to Google or some other company with lots of money.
They will rush towards an IPO to avoid this, and they likely have time to get there.
The ends justifying the means have less to do with it than just simply stupid boomers in DC who can barely operate a phone not understanding the tech and falling for Jewish tricks, or being Jews themselves and always looking to take more goyem resources and give them to Israel. There is no finish line with current (LLM-based) AI tech. It would be like saying if we just pursue phone tech far enough, it will allow us to fly. It just doesn't make any sense.

>>107834739
if you can get the datacenters to be the millions of monkeys with typewriters, then maybe they'll be able to convince all the sad rich people that they aren't sad anymore

>>107833884
Kind of, yeah. Because to this day you'll find retards telling you up and down not to use _utf8_off and _utf8_on.

>>107832962
sorry, my bad, when I said: >>107831557 , I meant how did you come up with this whole thing, it's called "short circuit evaluation"(I found about it just now)

in short: I meant how did you discovered this even if it existed? You guys read blog or source code of OSS?

>>107833842
I'll try to explain...

Perl string are treated as a sequence of characters. If the UTF8 flag is off, each character correspond to a byte that can have any value in 0x00..0xff. If the flag is on, characters can have an ordinal value > 0xff and each character can be represented by several bytes internally. I think the internal representation is similar to UTF8 but is not quite UTF8. The intended purpose is that a character represents a unicode codepoint but it can be almost anything. The UTF8 flag is on if any character is > 0xff, otherwise it's off.

perl -E 'use Devel::Peek; Dump chr(0xff)'
perl -E 'use Devel::Peek; Dump chr(0x100)'
perl -E 'use Devel::Peek; Dump chr(0xffffffff)' # ok
perl -E 'use Devel::Peek; Dump chr(0x7fffffffffffffff)' # ok
perl -E 'use Devel::Peek; Dump chr(0xffffffffffffffff)' # error

>>107834672
>>107833842
>checked for the internal UTF-8 flag of a scalar, *and converted it to ISO-8859-1* if it was set - and if it couldn't do that because ISO-8859-1 contains only so many characters, then it would shit into your log about the failed conversion.
They did it wrong.
You can either manipulate strings as bytes in which case you want the UTF8 flag turned off, or you can treat them as codepoints and you want the flag on. But when concatenating string you want both string to be bytes or to be codepoints. If one string is codepoints, the new string will be treated as codepoints, the bytes of the byte string will treated as codepoint and if it was utf8 where one character was > 0x7f, when you'll encode the string in order to have bytes when printing, those characters will be encoded twice and will be garbage.
Libraries like JSON::XS or HMTL::Entites can return strings with UTF8 on, so you either need to encode their output to UTF-8 or decode your other string from UTF-8.

The problem is that if the input bytes are not in utf8 (because they use a different character set or because they have been erroneously converted (converted twice, etc..)) and contains bytes > 0x7f, decode("UTF-8", $input) will return garbage. What I do is check if the bytes are valid utf8 and if so decode "UTF-8" them to have codepoints and if not decode "iso-8859-1". The advantage is that when the string will be encoded back to "UTF-8", the bytes decoded with "iso-8859-1" will be garbage but the transformation is completely reversible so you can still get the original bytes back without loss.

>>107834681
>when concatenating string
We're not talking about concatenating strings. We're talking about data that's being sent to a client. That module - I think it was FCGI, but honestly, so many other modules do the same shit - checks if the flag is set (which it always properly was because we were loading data from a UTF-8 database), and converts it to ISO-1 if it is (and I mean actually _convert_), because Perl has an unhealthy relationship with ISO-1 to this very day.

I'd notice this because we received proper UTF-8 characters if I included some Chinese characters in our output as comments (not present in ISO-1), but if they were missing we'd receive ISO-1 text (despite the header and everything clearly being UTF-8).

>>107834305
'tard, it is the name of the library we should base our own code on.

Although I second the name desutorrents. Doesn't immediately give it away.

>>107834272
who's pip?

t. neon

>>107834591
the python package manager

>>107834601
why are (You) saying hi to a python package manager?

File: a98964f782e3a4da0aa17310f(...).jpg (316 KB, 2768x2320)

316 KB JPG

>>107834613
what's the problem with that?

File: cloudflare-be-like-v0-zoy(...).jpg (88 KB, 640x503)

88 KB JPG

>>107834072

>>107834072
>The fact that 99% of sites would go down if Cloudflare went down is crazy.
You're right—it's a massive single point of failure. A significant Cloudflare outage would cripple DNS, CDN, and security for a huge portion of the web, exposing the dangerous centralization of critical infrastructure for convenience and cost savings.

it actually happened a couple of weeks ago.

adopting Cloudflare was a very silly decision on the internet's part. there are dark net markets that attackers have every reason to DDoS, and they do. but then those sites have enough drugs money to set up mirrors. all while (presumably) banning the DDoSers.

invent a problem, sell the solution. we're seeing the same with 'verify your ID' services.

>anon think 99% of sites are behind cloudflare
peak schizo

>>107834419
>PR2, depending on the revision, doesn't need EQ.
Fellow r2retards. not like this...

>>107834419
>there's more people that love the pos than hate it.
applicable to all pos if you really think about it

>>107834459
r2r will never join my signal chain. don't need it.

>>107834565
I thought you meant r2r. not revision 2.

;p

>>107834565
Yeah r2r is a meme

>>107833906

"Think different."

reality: Don't think, just keep buying!

>>107834461
stock must go up forever.

>>107834333
Airplane mode doesn't disable Bluetooth.

Airplane mode exists to stop passengers from spamming the 4G/5G towers on the ground, it has nothing to do with safety.

>>107833605
You do you but big over ear headphones make it clear to everyone you can't hear them so it makes social interactions easier.
People will either leave you alone or use hand signals to grab your attention before trying to speak to you.

>>107834610
I'm supposed to carry headphones around 24/7? That sounds far more socially inept than having earbuds in your pocket. Probably why everyone does the latter.

File: 1734117759811112.jpg (138 KB, 1280x720)

138 KB JPG

>W10 Pro
I've run into a really annoying problem: the font that I have installed is not appearing anywhere but the font menu in the new settings app. I have tried the following
>installing for all users
>converting it from ttf to otf
>deleting any traces of it in registry (in both places)
>reinstalling that entire font family
>done all of the above in safe mode as well
but it just does not work. At first I thought that the font was maybe broken in some way, but I threw it onto my craptop and it installed itself perfectly with no issues. Is there anything else I could do?

>>107814454
faulty ram?

I love Windows!

File: 1721825636719025.jpg (118 KB, 1280x720)

118 KB JPG

>>107832018
I managed to fix(?) it and it's probably the most retarded windows shit I have ever had to witness
>get font editor program
>change font family and weight name
>export
>all suddenly works
>AE doesn't even see anything wrong
My only assumption for it being fixed is that I already had a font of the same family with the same weight name (here it was Bold but my font was Condensed Bold).

>>107831832
>why is the new context menu so fucking big?
Because most of the UI elements are bigger now that's really why.