@lisicamica Beko semzamenjal takoj po garamciji z Boschem. Ker sta mi tako pomivalc kot pralni stroj crkovala. Bosch ima modele tudi razdeljene po razredih. Cenovnih in kvalitetnih.
Ne gledat na garancijo. Garancija je danes zavarovalniski biznis in skoraj ni povezana s kvaliteto.
Stanovanje v Trnovem najemata Robert Golob in Tina Gaber pri podjetju, ki je od države potegnilo že skoraj 30 mio €.
Isti krog je povezan s spornim dopustovanjem para Gaber Golob na Ugljanu.
➡️ Kdo nadzira Golobovo intimo in milijone javnih sredstev? https://t.co/sx5AKZBR0s
The scariest number here: 3.61% of CPUs in one large-scale study were found to cause silent data corruptions. Not “a few bad chips.” Nearly 4 out of every 100 processors doing math wrong, silently, with no error log.
Google coined the term “mercurial cores” in 2021 after their production teams kept blaming software for data corruption. They’d debug for weeks, find nothing wrong with the code, swap the machine, problem gone. The actual cause: manufacturing defects at sub-7nm that pass every factory test, then degrade unpredictably months or years after deployment.
Facebook confirmed the same thing independently. Hundreds of affected CPUs across hundreds of thousands of machines. The defect doesn’t crash your system. It just gives you 5 instead of 6 when you multiply 2x3, under specific microarchitectural conditions, with zero indication anything went wrong.
Now think about what this means for AI training. A single corrupted GPU or CPU in a distributed training cluster doesn’t just produce one bad output. It feeds corrupted gradients into a synchronization step that gets averaged across every accelerator in the cluster. One bad chip can silently poison an entire training run. NVIDIA published a whitepaper on exactly this problem. Loss spikes during LLM training that nobody could explain traced back to silent hardware corruption.
The part that keeps infrastructure engineers up at night: traditional defenses don’t work. ECC memory can’t catch this because the corruption happens during computation, not storage. Checksums like CRC heavily use vector operations, which are themselves one of the most vulnerable instruction types. The tools designed to detect corruption are running on the same flawed silicon.
Google’s current detection method? Roughly half human-driven, half automated. And of the machines humans flag as suspicious, only about 50% are actually confirmed mercurial on deeper investigation. We’re debugging trillion-parameter models on hardware where we can’t reliably tell which chips are lying to us.
Moore’s Law gave us more transistors. It also gave us transistors we can’t fully verify.
🤣🤣🤣🤣🎣 A young guy from Texas moves to California and goes to a big departmentstore looking for a job.
The manager says, "Do you have any sales experience?"
The kid says, "Yeah, I was a salesman back home in Texas."
Well, the boss liked the kid, so he gave him the job. "You start tomorrow.I'll come down after we close and see how you did."
His first day on the job was rough but he got through it. After the store was locked up, the boss came down.
"How many sales did you make today?"
The kid says, "One."
The boss says, "Just one? Our sales people average 20 or 30 sales a day. How much was the sale for?"
Kid says, "$101,237.64."
Boss says, "$101,237.64? What did you sell him?"
Kid says, "First I sold him a small fish hook. Then I sold him a medium fish hook. Then I sold him a larger fish hook. Then I sold him a new fishing rod. Then I asked him where he was going fishing, and he said down at the coast, so I told him he was gonna need a boat, so we went down to the boat department, and I sold him that twin engine Chris Craft. Then he said he didn't think his Honda Civic would pull it, so I took him down to the automotive department and sold him that 4X4 Blazer."
The boss said, "A guy came in here to buy a fish hook and you sold him a boat and truck?"
Kid says, "No, he came in here to buy a box of tampons for his wife, and I said, 'Well, since your weekend's shot, you might as well go fishing.'"🎣🤣🤣🤣🤣🤣
I just finished shoveling my driveway. Took me 90 minutes. My back is killing me, and I have to drive to work now.
My neighbor has a snowblower. He did his driveway in 10 minutes, then offered to do mine.
I told him no thanks, I like the exercise.
Real reason? If he does my driveway once, I owe him. Next thing you know, he's asking me to "take a quick look" at his computer because "it's running slow."
I don't do free IT work. Not for family, not for neighbors, not for anyone.
Last year, this same neighbor asked me to help him set up his printer. I told him I'm a "network architect" and I don't work with consumer hardware.
He looked confused. I said, "It's like asking a heart surgeon to check your kid's scraped knee. Different skill set."
He hasn't asked me for tech help since.
So yeah, I'll shovel my own driveway. A sore back is temporary. Becoming the neighborhood Help Desk is forever.
Boundaries aren't rude. They're strategic.
Stigne Muji poruka od švalerke:
- Dođi brzo, sama sam kući!
Razmišlja on kako da se izvuče pa reče Fati:
- Jao ženo, umro tetak! Evo sad javili. Ostani ti i kopaj, a ja idem da vidim treba li šta pomoći.
Ode Mujo i ostavi Fatu na njivi da kopa. Posle nekog vremena vraća se on kući i kaže:
- Nećeš verovati. Ja otišao do tetka, kad on živ i zdrav. Rekoh, kad sam već tu, sednem sa njim, popijem kafu, upitam ga za zdravlje i shvatim da me neko lepo namagarčio. Ne mogu da verujem da se ljudi šale na takav način!
Na to će Fata:
- Ovaj svet je totalno poludeo. Evo sad kad si ti otišao, naiđoše dvojica pijanih drumom i napadoše na mene. Kažu, ili da im pevam, ili da se seksamo!
- I, šta si uradila? - upita je Mujo, vidno zabrinut.
A Fata će:
- Pa, kako da pevam kad je tetak umro?