Sebastian Sylwan

For @SinnersMovie, Michael Ralla and Pepe Valencia brought a steam train to their ideal filming location — without moving production across America. Using drone scouting, RealityScan, and Unreal Engine, they quickly iterated and shared previs on set. Watch the Fireside Chat: youtu.be/lUKR7lZoFYI

Producer Chris DeFaria on Creator Economy and Strategy Behind Startup Chronicle Studios: ‘We’re Trying to Invert the Development Process’ #FMX2025 variety.com/2025/artisans/… via @variety

@mundosdigitales @GRUPMEDIAPRO Pipeline o tuberia digital? 😆

Descubre el pipeline y el workflow utilizado para crear un largometraje 100% en Producción Virtual en un tiempo récord de 4 semanas. tinyurl.com/4efubsfb #MD25 #vfx #producciónvirtual #icvfx #pipeline #workflow @GRUPMEDIAPRO

Prime Focus Ltd. Expands with Strategic Merger! | MCap 3,034.37 Cr | PE 15.07 Prime Focus Ltd. completes a merger to boost AI in media! 🚀 🔑 Key Points: - Prime Focus Limited (PFL) announced the completion of a merger between its step-down subsidiary Brahma Merger Sub Inc. and Metaphysic Inc. on February 14, 2025. - Metaphysic Inc. has a turnover of $4 million for the fiscal year ending December 2023. - Brahma Merger Sub Inc. was incorporated on February 4, 2025, and has no turnover to date. - The merger aims to enhance Brahma's technology and AI capabilities in the media and entertainment industry. - Post-merger, Double Negative Holdings Limited will hold 90.91% and Metaphysic Inc. will hold 9.09% of shares in Brahma AI Holdings Limited. - There is no change in the shareholding pattern of Prime Focus Limited as a result of this merger. - The merger was completed as an all-share transaction. 🔹Company Profile: Name: PRIME FOCUS LTD Symbol: PFOCUS, BSE:532748 PE: 15.07 Market Cap: ₹3,034.37 Cr. == Approx. $0.36 B Sector: Media, Entertainment & Publication - Media & Entertainment What are your thoughts on this strategic move? Share your insights below! Follow us for more realtime updates! 🕒 This update of PRIME FOCUS LTD was recorded at February 18, 2025, 02:16 PM Disc: Information provided in this tweet can be inaccurate, verify through the source before making any investment decision.

(🧵NO ONE COULD HAVE PREDICTED THIS): To answer the question "What does the future hold for 𝑆𝐀𝑅𝑆-𝐂𝑂𝑉-𝟐?" it's worth examining how predictable its evolutionary trajectory has been so far. Evolution is stochastic, but stochastic processes can still yield predictions. (1/)

Introducing, the Paul Debevec long read, at @beforesmag. I'm sure many people know @debfx's incredible influence on VFX. At @SIGGRAPHAsia I got the chance to sit down with him to discuss a discreet portion of his work at Google & at Netflix. beforesandafters.com/2024/04/04/the…

@elonmusk The Roman empire did, the Moores did in Spain, Constantinople did. It's not empathy it's common sense! All those benefited (economically and in peace) from that help

Modern Western civilization has extraordinary empathy compared to its power. This is also arguably its greatest weakness. With the nuclear bomb, America could have subjugated every nation on Earth with ease. Hitler and Stalin would certainly have done so. Instead, America helped rebuild Germany and Japan! There is no historical precedent for a nation with so much power helping, rather than destroying, its defeated enemies.

Can we just make @ylecun president of AI and call it a day please?

@curiosityonx South island of New Zealand!

Which country is this as seen from space?

A series of scary graphs about how temperature anomalies for 2023 compare to other years. Yikes. 📈

@nickbaum @paulg Lego is what he's shooting. The app seems to be stop motion studio which is the best. My kids love it.

Sunday morning is for stop motion animation.

LOL 😂 #CalvinandHobbes

At DigiPro 2023, we’ll present Magenta Green Screen, a new compositing technique designed to record accurate alpha channels in an LED volume and to collect ground truth training data for natural image matting algorithms. See you there and @SIGGRAPH! dmsm.github.io/magenta-green-…

I hit a bug in the Attention formula that’s been overlooked for 8+ years. All Transformer models (GPT, LLaMA, etc) are affected. Researchers isolated the bug last month – but they missed a simple solution… Why LLM designers should stop using Softmax 👇 evanmiller.org/attention-is-o…

@andruyeung Venezia, Italia

Name a city more beautiful than this.

@IgorCarron Not wrong, there's a difference between spending 1B$ buying chips you'll amortize over 3 years and spending 1B$ to train a single LLM

Once again, even Nvidia's predictions were wrong. A customer is willing to spend $1Bn for LLMs now in 2023, not 2026. "it is entirely possible that in five years a company could invest one billion dollars in compute time to train a single language model" nextplatform.com/2021/02/11/the…

@ruima Source? Your math puts each "chip" at 10K$ which seems high.

This is a letter Feynman wrote to a former student who wrote congratulating him for the Nobel. I’ve posted it before but I really find it worth it to read especially as a student or early stage research person.

@debfx i remember us having a discussion about when this would become possible, circa 2005...

@kimusan No, sorry, it will have to be Unity! 😂

Looking forward to sooner rather than later getting my dojo and future virtual architecture into my Apple #VisionPro. It will be Epic! 🤩