>>107723340
What browser do you use?

File: 1747109293443239.png (293 KB, 1271x795)

293 KB PNG

>>107723478
I use paprika recipe manager

When did /g/ become normalfag central?

>>107714854
Its clearweb because I can google piratebay dumb normalnigger

File: 1742366356256630.jpg (245 KB, 1125x1780)

245 KB JPG

>>107723614
Around the same time India got widespread internet access. It's only going to get worse from here.

File: deMF_zi_00011_.png (2.56 MB, 1792x1152)

2.56 MB PNG

>>107723738
ohh, the 'positions' are finland, london, and california, lol. cali is 1pm rn
I think you're our most eastward anon, so you get to see the new year first. let us know if it looks promising once you get there

File: 1767205264542766m.jpg (97 KB, 647x1024)

97 KB JPG

hope you all bought gpus for Christmas

>>107723810
Anyone spending money to generate jpegs is beyond coherent logic anywsy

File: deMF_zi_00012_.png (2.4 MB, 1792x1152)

2.4 MB PNG

>>107723810
>no source
I doubt this but also wouldn't be surprised either. my game plan is to treat my 4070 well and sit on it until 2027. by then, the AI bubble should have collapsed and there will be a massive market for cheap second-hand GPUs, similar to when crypto mining lost steam

>>107723825
my energy prices have been getting crazy. one of the reasons I fell in love with z-image, its way faster than chroma therefor I'm not cooking as long

>>107723781
All checked, Sir.

>>107723690
yeah i bet that 5 cent microchip encased on a 2 cent recycled plastic shell simply deserve to retain their cost for three decades

>>107723713
Works for nintendo products.
Quality doesn't age. If something is good and well made in a timeless manner, there's no reason to lower the price.

>>107720402

The TLDR is that they are fundamentally different kinds of computer. Floating-point math uses numbers that look like:
[+/-] {mantisa a.k.a. significant figures} (exponent a.k.a to-the-power-of-what)

So -3.14 is
[-]{314}(1)

It's actually not because the standard for floating point numbers is fucking retarded on purpose, but that's what it is in its purest form.

Unfortunately, to do math on these kinds of 3-part numbers efficiently, you need to have fixed-length fields, so the above becomes:

[-]{00000314}(00000001)

Because computers actually just shuffle numbers around, they cant actually "count" per se, or rather when the do it requires memory and loads of operations (see long division in logic gates for interesting example of how unintuitive the most efficient maths actually is)

Comment too long. Click here to view the full text.

>>107723774
Because the creator's children (or their children's children) simply deserve to live off shit they had no input in right?

>>107723856
I support generational wealth for me and mine, yes. I assume everyone else should look out for their genetic legacy as is their right.

>>107723480
use the gguf loader place gguf models as i said into /ComfyUI/models/unet
should be good to good then, then link it to what every in your model chain, for example kj node for patch sage attention which might be outdated now but i still use the option --use-sage-attention on the command line. Then into shift then through your lora loading node.

for wan 2,2 you need 2 advanced ksamplers, the first will be configured to enable noise and return noise but those need to be enabled. Depending on what you are using we will depend on number of steps nessary i use a checkpoint that only needs 4 steps so that will be the total steps for both the high and low samplers. the break down is thus, first sampler that gets the latents is set on start step 0 and end step 2 the second sampler is set on start step 2 and end step something like 100000 and is set to not return noise as in no matter mater what step it is it will always attempt to 100% denoise.

but if you're taking svi then you must either chain the sample latents to the next and so on, this is hard to explain in text especially that i 'm under influence right now. Tell me what you are doing then i will tell you failing that i will give my workflow in due course.

basically the last in chain must be set to not return noise, and only the first should have noise enabled but svi it can get more confusing if you have no idea.

I wish A1111 came back
Better times

honestly qwen sucks, has no variation, worse than zit if you use lora, slower than zit if you don't meaning you can just go with a normal model

>>107723839
first sample is the high noise the second is the low noise, very important you get this right. The high noise model creates the motion in a very nosy output, the low model sampler is then feed those samples and creates the actual frames denoised.

>>107721133
That's cursed, wtf

>>107716399
saying "the n-word" is the same as saying nigger, because you say "n-word" with the expectation that the listener knows that you mean nigger. you say n-word and they think "nigger". that's just how words work

File: xqnr2zfnlfvc1.gif (79 KB, 498x388)

79 KB GIF

>>107716457
>it capitalizes "black"

>>107720992
>im a retard who can be fooled by the latest version of DOCTOR.BAS

>>107722456
Coping with Reddit-tier gotchas isn't evidence. Either provide evidence it isn't sentient or shut up.

>>107723555
I don't know why you're so proud to be 75IQ.
>what does &x+1 do, it don't make no sense
Kek

>>107723718
it doesn't make sense retard. It translates to address(x) + 4*sizeof(x) + sizeof(x). Low IQ faggot.

>>107723718
well anon? we're waiting....
what does &x + 1; do?

File: Screenshot_20251231_10484(...).jpg (174 KB, 1440x1343)

174 KB JPG

>>107723779
Run it and find out:

#include <iostream>
using namespace std;

int main() 
{
  int x[5] = {0,1,2,3,4,};
  long offset = (long)&x;
  cout << (long)(&x + 1) - offset << endl;
  cout << (long)(x + 1) - offset << endl;
}

>>107723778
sizeof(x) is 5 ints. x is int[5]. This is not difficult. Pointer to array does not decay because it is not an array.

>>107721567
In 2026, the only way to build a high-end computer system will be to e-beg on X, like this guy, who received a 5090 card for free from Sam Altman, which is now quadruple the price: https://donaldboat.substack.com/p/dogpile-the-donald-boat-story

>>107723787
>I think it's neat
You're using it for porn, aren't you?

>>107723807
I can't use it because my hardware isn't good enough to run it locally.
But no I don't care about porn, and I prefer Japanese hentai made by Japanese people.

>>107721567
microsoft is yet to deploy something like 90% of the gpus that it has bought.

>>107723807
We all know he is anon

>>107723493
>0 <= f(n)
Do you even know how retarded you are?

assume that n is the size of a dataset. O(1) constant time, and does not grow based on n. this could be returning the size of the dataset, if it is known. O(n) is linear time, and it means essentially any constant operations applied to each of the dataset. this could be looping over it 100 times. as n grows so do the operations. O(logn) is something that is always less than O(n), but still grows as n grows. the classic example is binary search. O(n^m) is for each element in the dataset n^(m-1) operations are applied to it. the only other one that is real is O(n!) and this is just brute forcing combinations, an example would be the traveling salesman.

>>107723205
>an algorithm with worse big O notation can run faster
Correct. Quicksort is O(n^3) in the average case and bubble sort is O(n^2), yet quicksort sometimes runs faster.

>>107723652
>You are wrong.
Wrong.
>>107723804
>O(1) constant time, and does not grow based on n
Wrong.
>O(n) is linear time, and it means essentially any constant operations applied to each of the dataset
Wrong.
>O(logn) is something that is always less than O(n)
Wrong.
>but still grows as n grows
Wrong.

Clearly you know next to nothing about big O notation, just like the redditor in OP's pic.

>>107723015

BigO notation can be understood in O(LogN)2¥ time. Basically you need to be naturally exponentially asian.

>i thought for a moment i was on /a
fuck kuroba hmh
>>107723568
they arent sending their best

>>107723587
The source literally excludes Android, saarnix jeet.

File: 1765326473081.jpg (248 KB, 717x1280)

248 KB JPG

saar

>>107723615
Care to share the fucking source and how they conducted it? Nobody ever does when I ask for """some"""" reason.

File: wallhaven-gj8lod.png (606 KB, 1920x1080)

606 KB PNG

found it
https://wallhaven.cc/w/gj8lod?utm_medium=organic&utm_source=yandexsmartcamera

>>107720674
>flow
no idea what you're talking about

>>107720692
It's some JS thing, made by Facebook, which I think gives you some of the same benefits of TypeScript, while generally not being as good

Anyway. Advising people to stay away from TS in 2025/2026 seems a bit odd. It's a good language and it's extremely common now.

>>107720900
you should 100% learn js before ts, or you're fucking retarded, that's my point

>>107721022
Ah yeah now I see what you mean, okay

I have a website I want to promote, what should I do? Post on Reddit like a normie? Pay zucks for visibility on instagram?

File: sanakan-nihei tsutomu-saf(...).jpg (267 KB, 1450x2048)

267 KB JPG

>>107720437
>before the bot could use the whole computer or phone
...to do what? i've most recently tried to get chatgpt and claude implement or at least describe iterator logic for a very specific vector-like container i have. they just can't. they can tell me about existing things which are outwardly similar to what i want. they can't actually come up with an implementation, even if i give them lots of hints. even when i straight up described what to do they implemented it with errors. i am almost 100% certain this is an architectural issue in ml, or animal brains would've worked differently.

>>107720482
>I can't find a fucking use for AI
sounds like a "you" problem

>>107720654
>I-if you don't buy my bags it's a y-you problem!
Not, it's a baggie problem, baggie.

>>107718961
Ok, boomer

>>107718779
OP, you and everybody who shills for it endlessly need to understand WHY people take issue with it. Do you understand why they take issue with it? Or are you the same as Jeets I speak with on here who fail to see why people dislike Jeets?

>>107722865
My understanding is that with usb it should just not show up on the host when you have it passed through, and blacklisting drivers is only for graphics

>>107723395
I mean even for the USB controller itself? I'm not talking about a USB device.
Interesting though, you're probably right.

So, I have a collection of TV Shows that make up around 80% of my drives' space. I am seeding them indefinitely.
Now I run Jellyfin via Docker Compose and am very surprised. When Jellyfin fetched the metadata, it barely recognizes any shows and if so, they are under completely wrong categories.
It was suggested to freaking and fucking RENAME every file so the subroutine handling the metadata is not confused?!
I mean wtf, every somewhat modern and popular music player has no problem fetching tags and whatnot from the most obscure of releases.
Does anyone have an idea how to improve the recognition in Jellyfin? Otherwise I just go back watching via NAS share...

>>107723770
Are your files or folders named oahdiuhaiuhxaxjwoiajdfeoaj1337 HD ULTRA REMUX BD 420 and then some shitty abbreviation for the actual title? If so yeah you will have to rename. If you have normal {title}{season}{episode} naming you're doing something else wrong.

>>107723770
you can symlink to make a better library
sonarr will do it for you

>>107723269
it actually did. Intel CPUs are well known for their extreme low idle power. We are talking single digit watts while AMD can barely get under 20W on low-end AM4 boards and barely under 30W for AM5.
On the other hand, AMD APUs are known for being perfectly playable as long as you keep details down, while Geforce is needed for more detail.

I still don't know what the fuck if anything Apple is known for in the GPU business, other than popularizing high dpi devices.

>>107715472
i used to have that exact USB hub with the on and off switches

>>107723780
Intel is good idle power, horrible load power usage
AMD is vice versa

>>107713121
where is the joke tho?

>>107723817
>AMD is vice versa
No, with AM5 they are horrible on load power usage too. I had to bump up my 8600G to something like 160W just so the iGPU would never throttle. The 5600G did that out of 65W. And that's just an APU, with the chiplet based CPUs it's so much fucking worse because the interconnects suck up so much power. They need like 20W just to idle.

Zen 6 will change all of that since they are finally going to start using better interconnects. But that's still 2 years away.

They killed mypal and cuck licenced it, plus it doesn't support any website made after 2017.

>>107721286
The manchild drove them all away.

Yes I use it to run loli hentai flash games

>>107721764
>>107721864
>>107721964
male poon bros reporting in

File: 1750508986356634.png (980 KB, 1126x869)

980 KB PNG

>>107722125
It reminds me of the homosexual NewPipe developers who refused to implement SponsorBlock. Thankfully the community forked the project.