# This Tiny Chip Could Make Google's Quantum Computer 1,000× Better ## Метаданные - **Канал:** Dr Ben Miles - **YouTube:** https://www.youtube.com/watch?v=SyLrbgjbuVk - **Дата:** 21.05.2026 - **Длительность:** 2:20 - **Просмотры:** 2,749,927 ## Описание Princeton scientists developed a quantum chip with coherence times over 1,000 microseconds, 15 times longer than the current industry standard. They achieved this by using tantalum on silicon substrates, combining cleaner interfaces with scalable manufacturing. Source Article: https://www.sciencedaily.com/releases/2025/11/251116105622.htm #technology #breakthrough #quantumcomputing #discovery #shorts My Patreon:🚀 http://patreon.com/DrBenMiles My Instagram: https://www.instagram.com/drbenmiles My TikTok: https://www.tiktok.com/@drbenmiles My Newsletter: https://drbenmiles.substack.com/ My Merch: https://www.rockstarscientist.org/ 🔗 Linktree: https://linktr.ee/drbenmiles MY GEAR 📷 Sony A7III https://amzn.to/3OWrmGd 🔎 Sigma 402965 16 mm F1.4 https://amzn.to/49BNJdq 🎤 Shure SM7B https://amzn.to/4sF3ngx 🎤 Zoom H4n Pro https://amzn.to/3OXsklB 🎤 Sennheiser AVX https://amzn.to/4geWnBi ## Содержание ### [0:00](https://www.youtube.com/watch?v=SyLrbgjbuVk) Segment 1 (00:00 - 02:00) Scientists at Princeton just built the world's first millisecond quantum chip that survives 15 times longer than the industry standard. The reason we don't have useful quantum computers isn't cubic count, it's coherence, the time that a qubit can hold its information before it leaks away. That leak comes from microscopic defects that trap the energy stored in the qubit and typically live in two places, the superconducting metal itself that the circuit is etched from or the substrate underneath, the flat wafer that holds the whole chip together. For a decade, the field has been stuck between two bad options for that substrate. Silicon, what the chip industry is built on, which is easy to manufacture and scale, but for years it lagged behind sapphire because the metal-silicon interface was a massive intermixed layers hosting energy-leaking defects. As a result, sapphire, a much cleaner insulator at the interface even though its fabrication is a nightmare, became the substrate of choice. Last year, I got to catch up with Andrew Houck and Nathalie de Leon's team at Princeton who published in Nature what they think is the next best bet. First, they switched the superconducting metal from niobium or aluminum to tantalum, which has dramatically fewer surface defects and unlike most metals, it can survive the harsh acid baths that clean contamination off a chip without losing its superconducting properties. Originally, the team tried building tantalum on sapphire qubits hoping it would allow them to remove even further interface defects and improve coherence times. And they got close to a world record, but they found that losses were now coming from the sapphire substrate below. So, they decided to switch to silicon. Tantalum is traditionally incredibly difficult to grow on silicon, but after a long process of refinement, they produced a device combining a material that can allow you to produce significantly cleaner interfaces with now a substrate that was both manufacturable and scalable. The result was a qubit with coherence times of over a thousand microseconds, 15 times better than the qubits in Google's Willow and three times better than the world record. Why this is so important is that errors in quantum computers compound exponentially with qubit count. So, a 3x per qubit improvement isn't a 3x better machine. If you scale that to a 1,000 qubit machine, that becomes about a billion times more effective at computation. Princeton design is already drop-in compatible with architectures that Google and IBM are already building on, and it is the single biggest coherence jump in over a decade. If you like science that's coherent for at least 1 millisecond, follow for more. --- *Источник: https://ekstraktznaniy.ru/video/52263*