>Le
Reddit is that way --->

>LeCunt

File: 1703280117558384.png (422 KB, 746x900)

422 KB PNG

>>107742949
LeCun is really fucking based, he is one of the few that have power and that have the technical knowledge to say out loud that LLMs are a dead end.
Tis a shame that he is still a neural network cuck unable to recognize that neural networks themselves are a dead end, just like symbolic AI was at its time.
https://www.youtube.com/watch?v=4__gg83s_Do
https://www.youtube.com/watch?v=N09C6oUQX5M

>>107743448
If neural nets are a dead end, what do you think researchers should be looking into instead?

>>107743505
They should stop masturbating about stateless mathematical models (because both symbolic AI and neural networks are about logic and math not computing or programming) and start focusing on stateful programs that are actually agentic (in the original sense of the word not in the llm corrupted marketing sense) by having internal state, internal goals and a real model of the world (LeCun is right about this part), all of this not through mathematical models but through programming. If our current languages and databases are not enough then we should make progress there, for example by researching graph databases and new languages that make expressing complex and dynamic rules about the real world possible.
But corpos only want to sell their datacenter services, to them its a feature how inefficient LLMs are, and most researchers are not programmers they are stuck in their bubble of academia writing their papers in their ivory towers.

File: 00000021-1050944210032138(...).jpg (1.02 MB, 2880x2160)

1.02 MB JPG

>>107743226
nice

>>107743451
Yea but juniors are getting replaced with Indians.

>>107743415
Were they really "insane" or just hard? They're supposed to be hard, and 90 minutes for 2 of them sounds like ample time to me. Especially for 3 YOE, people are expected to really know their shit after 3 whole years in the field

>>107743533
Leetcode is a matter of exercise, not one of experience.

>>107743415
>Was I supposed to cheat?
Yes
Forced cheating is how they filter out whites

>>107743533
I've solved 200 leetcode problems, 90 of which were mediums and 25 hards.
I asked chatgpt and claude to find me the most comparable problems and both were hards.
I don't think these problems are in leetcodes database, and they were even more complicated than the comparable hards.
Problem 1 was an array question with a heap / sorting requirement and convoluted premise, problem 2 was a dynamic programming question with even more complicated shit that took me 10 minutes to even understand.
P1 was a medium- hard (closer to hard), P2 was a very hard.

>>107743116
What do you mean by that?

>>107742049
absolute fucking retard

>>107743246
Artificial intelligence implies natural intelligence. Artificial general intelligence implies artificial particular intelligence. AGI is a chimera built upon a chimera.

>>107742144
>farmers
dogshit is actually horrible for farming because its too acidic.

>>107742475
yeah when China develops AGI the West will twist it that way that they had to invade to stop autonomous killerbots from killing the human race
perhaps hollywood makes a DUNC prequel set during the butlerian jihad in time

File: wddDagpTstIHsyC5.jpg (1.36 MB, 2528x1696)

1.36 MB JPG

>>107743295

File: 20260102_1317_Surreal Sna(...).png (2.05 MB, 1536x1024)

2.05 MB PNG

the snail surface reminds me of something
something i can't possibly recall

File: NO.mp4 (533 KB, 464x666)

533 KB MP4

>>107743307
This is not what I asked for.

File: saar.jpg (219 KB, 1024x1024)

219 KB JPG

/v/ Dall-e is dead

File: _a518ab62-3fb0-4bfc-b909-(...).jpg (158 KB, 1536x1024)

158 KB JPG

>>107743522
¯\_(ツ)_/¯

>>107741996
The "you can't criticize Wayland because it's a protocol" cope is getting old. If there isn't a protocol supported by a strong majority of compositors that provides a given feature, the feature should be considered unsupported. If X11 had theoretical provisions for some feature in the protocol, but that feature wasn't actually available in X.org, we'd consider it unsupported too. Anything else would be dishonest.

>>107742077
It's pure security theater, you're not meaningfully more protected. There's a ton of ways to exfiltrate data, like reading your home directory or using LD_PRELOAD tricks. Wayland's "security" is simultaneously too lax and too strict. Too lax because it doesn't provide meaningful protection; too strict because security is merely an excuse for designing things that Wayland devs don't like out of the protocol. Configurable access control is a thing that exists, so there's no need to have a gimped protocol to achieve security.

>>107742943
>The "you can't criticize Wayland because it's a protocol" cope is getting old. If there isn't a protocol supported by a strong majority of compositors that provides a given feature, the feature should be considered unsupported. If X11 had theoretical provisions for some feature in the protocol, but that feature wasn't actually available in X.org, we'd consider it unsupported too. Anything else would be dishonest.
But the majority of the things labelled "Not Supported" here actually are supported in the majority of compositors, it's not theoretical. There's no loss of functionality in any real implementation, but because it's not required as part of the protocol they're claiming it's unsupported. It's very dishonest.

>>107743027
Maybe I have outdated information then. Is there now a cross-compositor
1. equivalent of xte,
2. way for programs to position their own windows,
3. way for programs to register global hotkeys,
4. equivalent of xrandr,
5. equivalent of setxkbmap?

Wayland is so bad that Valve put all their time and money into it!

> X11 has screen tearing
>no such thing in Wayland
Good enough for me, the rest of them will be integrated soon.

>>107741731
tried playing rdr2 on the projector. impossible.

im trans, what should i get from aliexpress

>>107738961
that wont work
it was 2+4+7 and I barely reached 7
it would not be worth it to split it to get just 2+4 and never reach 7

>>107742313
https://aliexpress.com/item/1005010081820095.html
https://aliexpress.com/item/1005009880901568.html

File: 20260102_194719.jpg (62 KB, 1200x1602)

62 KB JPG

>>107741435
Btw, just copied my building's entrance fob with the shittiest chink cloner and app (pcr532, picrel)

>>107743061
it doesn't matter
what matters is that i need to touch my dick to the output and the output needs to be of satisfactory quality for me

>>107741748
image is not a good metaphor, lossy compression can achieve > 100x size reduction with nearly imperceptible differences.

>107743438
Don't listen to this shill >107743465

>>107742235
The new score, for whatever it's worth, is 76,2 which is 1/10th of a point below gtp5.1.

Depending on your skillset and interests, a benchmaxxed leetcode model might serve a purpose. If you do the higher level architecture and function specifications yourself, the model could do all the function implementations.
If you don't have the skills, you need to use some other model to help you out.
You could even roll your own mixture of experts this way, an AoR, or assembly of retards.

>>107743438
QK_6 of a mistral small 24B finetune, qwq snowdrop, or a cope quant of GLM 4.5 depending on your system RAM. Models 24B+ are just enough to follow instructions and maintain a character well, but they'll still sometimes encounter the same problems as smaller retarded models.
Honestly, the powergap in local models is massive. We're all peasants. Anons on 8GB GPUs are in some cases running the same models as you guys on 24GB lol. Once you get to 200GB+ of RAM you've got some better options at non-cope quants.

>>107741973
how is it censored if its pumping out harcore porn videos non stop

>>107742282
Not hardcore enough.

>>107741617
TBF anon, I used to play TCGs at hobby shop. The very epitomy of "virgin nerds" and there was this couple there. FUGLY people but super nice. They got together, fell in love and had a daughter... No joke their kid is named SAKURA. This was around 15 years ago I think?

But dude, put yourself out there! You WILL find someone if you look!

>>107741973
>censored
No it's not I was literally having it do scat RP with me. Unless you're doing loli stuff with it, then sure yeah, then go to like polybuzz or janitor for it.

>>107742446
I bet this faggot likes handholding.

>>107743264
What about chub or agnai?

>>107743264
The cards on poly are poorly curated and look like ass

>>107742928
>well? how did she react?
well, how would you react when someone stole your printer and buried it in tinfoil? Given that he stabbed her to death few weeks later, I guess we can assume the stasi printer's Final Solution remained otherwise unresolved.

>>107738775
My mother was violent and abusive, physically and verbally. She would snap between moods like the flick of a switch. But that's my fault because I was born?
>incel hates his mom
I wasn't an incel, I was a child.

>>107740346
Chatgpt didn't talk him into killing itself, it was the sudden bolt-of-lightning realization that he was getting wasted on hard cider and treating a web browser application as a best friend causing him to spontaneously die of cringe.

In case Chatgpt suggests such things to you, stay away from harming your parents!
Goddammit. Instead you could research the surveillance methods better ...

>>107743504
totally not a glownigger post that prevents you from reaching zeus

>>107741103
imo the potential for customizable integrated development enviromments is huge. Emacs blessing/curse is the 40 years of baggage/design patterns, plus the need to have backwards compatibility for a shitload of really niche use cases.

Having the possiblity of rolling different frontends COULD(if someone does it) allow for embedding a shitload of stuff into the editor, like rolling your own gamedev tools or data processing shit, completely under your control as in emacs, with the benefit of being able to use any CL library to expand it.

BUT, the counterpoint is that the community's really small so you'll have to make your own dog food for the foreseeable future, no new packages, no information on how to solve your obscure bug from a mailing list exchange 15 years ago.

>>107741103
>>107741425
On that matter: https://kyo.iroiro.party/en/posts/100-percent-lisp/

>>107741103
I really was looking forward to using lem regularly eventually, but I think this is a deal breaker for me >>107732089
Maybe I want someone to convince me that I shouldn't care about it... But I don't think that's possible.

>>107740957
That's a clever use of the TAB character.

>>107727170
how do you handle user scripts? are they free to do whatever or is there some sort of interface they're obliged to use (i.e. they can't use system IO with disregard)? I'm not that familiar with CL so I don't know how user script integration works

>>107743324
You reply saying "based" and that's it. You are now a channer!

That's why we need lossless video with proper chroma encoding.
>t. Hi-Fi Video enjoyer

>>107741945
>neither scaling is integer
2160/720=3
3840/1280=3

>>107742155
>I have some archived 720p movies with high bitrate from over 15 years ago and they still look fine
Movies on DVD are 480i and that tends to look better than online 720p

File: Untitled.png (160 KB, 640x400)

160 KB PNG

>>107741289
when I was a kid, I thought picrel was photorealistic. it's almost as if your standards can change with time.

>>107743200
Hope they'll find a cure, anon.

if guidezo spent less time shilling his guide and more time improving it maybe people would actually read

>>107743181
>sliders are set at wrong frequencies
>can't make IEMs sound good :(

fyi cuckbox got ported to the hiby r3pro 2 so you can ditch your strange chinky hissblasters from no name manufacturers

>>107743426
Trvke

>>107742276
is this a thing with revanced? I haven't used youtube outside of a chromium web browser in a long time.

>>107742118
you either, use them, or learn your lesson the hard way (eventually).
t. was caseless for years, but not anymore.

>>107742118
>>107742886
caseless for more than 10 years but only had cheap phones
when i buy something expensive it'll be cased

>>107741660
>Why don't they make phones like XYZ any more?????
>That's too expensive I'm not buying it
Every time lol

File: Screenshot.jpg (269 KB, 1004x1114)

269 KB JPG

OneUI 8.5 on my Fold 7 which to be honest is not much different to 8.

How can you get smooth playback with mpv on a 240 Hz display? "video-sync=audio" has microstutters, and the "display-*" options consume far too many resources and cause frame drops. Is there no "video-sync=video"? "video-sync=desync" is smooth and uses few resources, but the audio gradually goes out of sync. There has to be a way, right?

I have ripped a tv series from blu-ray with makemkv. each episode is 6.3 gb (23 minutes, so ~4.5 MB/s), 11 episodes total. ffprobe output for the video stream:

Video: h264 (High), yuv420p(progressive), 1920x1080 [SAR 1:1
 DAR 16:9], 23.98 fps, 23.98 tbr, 1k tbn

I think 70 GB is too much for one tv series, so I'm thinking of re-encoding this; should I and if yes which format?

>>107743382
if you're still getting microstutters with profile=fast hwdec=no buy a better GPU that doesn't have horridly broken drivers. Alternatively if you're on Linux maybe try Zen/Cachy kernels.

>>107743382
These are your options:
>video-sync=audio with VRR
>display-* with hacks like https://github.com/mpv-player/mpv/issues/11122#issuecomment-2041618717
>switch to madVR

>>107743444
I am on Windows with an NVIDIA GPU.
I use shaders, but with "video-sync=audio" the GPU usage is only around 20-30 percent. With any of the "display-*" options, GPU usage is much higher and I get dropped frames.

>>107743487
Apparently, VRR does not work with "hwdec=d3d11va", and the hack does not fix the microstutters.

Is there a way to use something like "video-sync=desync" and automatically offset the audio to compensate for the desynchronization?