Colarsis

Développeur et joueur de Lol, Hearthstone (un petit peu quand même) et Minecraft. Songe a une carrière de chanteur #lol

[Michael Shanks @MichaelShanks]

I’m gonna simply say this: if you are at all interested in a Stargate show with ANY of the original creators/performers involved, now is the time to say something. Otherwise it really will be the end of that chapter forever. Let them know you are THERE

[Colarsis @colarsis]

@Banksystre @Seb_Frit @adam_bros Mais justement, le film n'a aucune hype et c'est très bien comme ça
Pour être plus précis, il en as certainement dans une bulle d'information autour de Seb, mais ça en reste là
On est pas sur une promo dans toutes les intros du BDG. Laissons les gens vivre et expérimenter

[Aakash Gupta @aakashgupta]

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.

[Squirtle @squirtle_says]

the world you remember no longer exists

[Zyluh @Zyluhk]

@shpostx Y'as pas de pénurie d'ingénieur, y'as trop de cadres, pas assez de technicien, et des ingénieurs qui font pas des tâches d'ingénieurs

[John Carter @martianwyrdlord]

The neural queue never made sense to me. Why would animal life on Pandora evolve an open, cross-species connection that enables one organism to invade the central nervous system of another? What possible advantage does this confer? The downsides - infection pathway, loss of autonomy to predatory or parasitic organisms - are obvious.

Now that we know that it's because Eywa is a planetary-scale fungal parasite it makes a lot more sense.

It also points towards an obvious endgame for the series. Jake Sully and his merry band of alien bioterrorists hijack a starship and return to Earth, carrying Eywa spores. They smuggle these into the solar system and past the planetary defenses, landing on the planet to plant the seeds that will infect Earth with a new Mother. The spores take root, humans and other life forms begin to mutate as the infection takes hold, and a new Gaian Golden Age of ecological harmony takes hold as humanity is enslaved by the alien parasite, which asks only that the species give up its autonomy and renounce high technology, to be trapped for the rest of time in a state of hallucinatory primitivism.

It's possible that this is the Eywa parasite's entire purpose: a biotechnological Inhibitor that infects biospheres in order to prevent technological civilization from spreading beyond its home planet, designed as a weapon by some ancient race of engineers who didn't feel like sharing the Galaxy with competitors.

[LDLC @LDLC]

🎁 CONCOURS 🎁
Un PC Gigabyte AERO X16 à gagner ! 🍀

Pour participer :
▪ FOLLOW @LDLC
▪ RT CE TWEET

💻 Le PC en détails
👉 https://t.co/u6FHn61MB9

📢 Joue aussi sur Instagram pour multiplier tes chances. Bonne chance !

🥳TAS à partir du 05/01/2026

[Colarsis @colarsis]

@TopAchat #PetitPapaTopAchat

[TopAchat @TopAchat]

Concours 🎁 Petit Papa TopAchat 🎄
🔥 + 34 000 € de matos à gagner !

Ca continue avec le Setup 7 de 3439€ !

Pour participer :
🔁 RT ce tweet
➡ Follow @TopAchat
💬 Commente avec #PetitPapaTopAchat

Participe aussi ici ⬇
https://t.co/P9SiyHUTTx
🍀 TAS le 18/12 à 12h
@CORSAIRFRA

[TOUTCRAMER @ToutcramerV2]

Que ça fait du bien que ces vérités soient évoqués.

[TopAchat @TopAchat]

Concours 🎁 Petit Papa TopAchat 🎄
🔥 + 34 000 € de matos à gagner !

On enchaine avec le Setup 6 de 3634€ !

Pour participer :
🔁 RT ce tweet
➡ Follow @TopAchat
💬 Commente avec #PetitPapaTopAchat

Participe aussi ici ⬇
https://t.co/P9SiyHUTTx
🍀 TAS le 15/12 à 12h
@IntelFrance

[Colarsis @colarsis]

@TopAchat #PetitPapaTopAchat

[TopAchat @TopAchat]

Concours 🎁 Petit Papa TopAchat 🎄
🔥 + 34 000 € de matos à gagner !

C'est reparti avec le Setup 5 de 3994€ !

Pour participer :
🔁 RT ce tweet
➡ Follow @TopAchat
💬 Commente avec #PetitPapaTopAchat

Participe aussi ici ⬇
https://t.co/P9SiyHUm3Z
🍀 TAS le 11/12 à 12h
@Seagate

[HSVSphere @HSVSphere]

This produces more real value than your ChatGPT wrapper, btw. Good, honest application of ML.

[LCP @LCP]

.@denis_Masseglia défend un amendement pour créer "une taxe comportementale qu'on pourrait appeler la contribution à la modération fiscale", et qui taxerait "les députés qui déposent des amendements qui proposent des taxes". #DirectAN #Budget2026

[ksa 🏴‍☠️ @kosa12m]

okay, this is getting chaotic

[Corti (Cortiste) @cortisquared]

PTDR taxe sur la taxe quite literally

[Towa 🇫🇷 @TowaYowaIowa]

Vous ne devinerez jamais de qui ces articles parlent.

[Professeur Carré @Geometriquement]

Je découvre une liste d'abattements fiscaux, pour certaines professions particulières.

Qu'est-ce qui les justifie ?

https://t.co/3M3dcDh1St

[HotelDon @HotelDon]

Every 5 years or so someone in Valve's accounting department realizes they're one mistake away from accidentally becoming a bank so they do something really funny