# We Just Turned Down Millions of Dollars. Here Is Why. Source: https://www.youtube.com/watch?v=SWGBN1KvG6c Recap page: https://rapidrecap.app/video/SWGBN1KvG6c Generated: 2026-01-01T17:33:50.561+00:00 --- ## Quick Overview The speaker declined lucrative sponsorship offers, including several for dedicated videos costing thousands of dollars, because he values creating content independently and sharing research papers freely, which he believes is the best way to support the community, rather than pushing sponsored products. **Key Points:** - The speaker explicitly turned down paid partnerships for 5-8 minute dedicated videos, some costing thousands of dollars, because he prefers to maintain editorial independence. - He cites the need to share cutting-edge research papers freely with the audience, contrasting this with sponsors who often dictate the script and limit the content discussed. - The video showcases various complex physics and graphics simulations, including the breaking of bread (Ding et al. 2019), fluid dynamics (Barreiro et al. 2017), cloth simulation (Ando 2024), and character movement (Geijtenbeek et al. 2013), to illustrate the kind of high-quality content he produces. - The speaker mentions that for the last 10 videos, he hired someone to edit out stumbles and filler words, which accounted for only about 2% of the total work. - He highlights specific research achievements, such as AlphaFold 2 significantly advancing protein folding accuracy (Jumper et al. 2021) and new simulation methods offering massive performance gains (Gissler et al. 2019). - The speaker concludes by stating he will continue making videos for free, as his goal is to provide valuable, unbiased content, rather than making money from endorsements. ![Screenshot at 2:08: The speaker explicitly states his reason for rejecting sponsorships by showing complex, high-quality simulations of various scientific topics he covers, like the octopus interaction simulation, to illustrate the value he already provides.](https://ss.rapidrecap.app/screens/SWGBN1KvG6c/00-02-08.jpg) **Context:** The speaker, who hosts a channel often featuring advanced computer graphics and physics simulation research papers (like those from Google DeepMind, NVIDIA, and various academic institutions), addresses the frequent inquiries from companies wanting to pay for dedicated promotional videos. He uses a montage of complex simulations—ranging from food destruction and fluid dynamics to character animation and protein folding—to demonstrate the high value and technical nature of the content he already provides for free, explaining why he rejects sponsorship deals that compromise editorial integrity. ## Detailed Analysis The speaker explains that he refuses lucrative sponsorship offers, including those for dedicated videos costing thousands of dollars, because he prioritizes delivering unbiased, high-quality content derived from cutting-edge research papers. He shows numerous examples of the complex simulations he covers, such as the breaking of bread (Ding et al. 2019), realistic fluid dynamics (Barreiro et al. 2017), efficient cloth simulation (Ando 2024), and advanced locomotion learning (Geijtenbeek et al. 2013). He contrasts these visuals with the restrictive nature of sponsored content, noting that sponsors often demand specific wording or product promotion, which he refuses to do. He also shows the massive accuracy improvements in protein folding achieved by AlphaFold 2 (Jumper et al. 2021) and performance gains in physics simulation (Gissler et al. 2019) as examples of the valuable, free information he shares. The speaker emphasizes that his priority is supporting the community with knowledge, not generating revenue through endorsements, stating he would rather work tirelessly himself than compromise the channel's integrity for money. ### Sponsorship Rejection Rationale - The speaker rejects paid partnerships for dedicated videos, despite offers potentially reaching thousands of dollars, because sponsors dictate content and force specific messaging, which compromises his commitment to independent content. ### Demonstration of Content Quality (Various Sources) - The video showcases complex simulations including breaking bread (Ding et al. 2019), viscous fluid interaction (Barreiro et al. 2017), cloth simulation (Ando 2024), and locomotion learning (Geijtenbeek et al. 2013) to prove the high value already offered. ### Scientific Breakthroughs Highlighted - Examples include AlphaFold 2 achieving 93% match in the Global Distance Test (Jumper et al. 2021) and performance gains of 9.5x in physics simulation stability (Gissler et al. 2019). ### Channel Operations & Effort - The speaker mentions having an editor help with filler words for the last 10 videos, which was only about 2% of the total work, emphasizing that the core research presentation is done by him and his wife. ### Concluding Stance - The speaker reaffirms his dedication to providing content for free, stating he would never charge for content or accept restrictive sponsorships, concluding that this approach is what makes the channel valuable. ![Screenshot at 00:02: Visualization showing the simulation of bread breaking apart, illustrating complex material deformation research.](https://ss.rapidrecap.app/screens/SWGBN1KvG6c/00-00-02.jpg) ![Screenshot at 00:24: Comparison between Traditional MPM \(top\) and PFF-MPM \(bottom\) methods for material simulation, demonstrating the new method's ability to maintain material integrity when pulled apart.](https://ss.rapidrecap.app/screens/SWGBN1KvG6c/00-00-24.jpg) ![Screenshot at 00:53: A simulation showing the effect of energy \(E\) on material deformation, with E=150 resulting in a high splash compared to E=17 \(Han et al. 2019\).](https://ss.rapidrecap.app/screens/SWGBN1KvG6c/00-00-53.jpg) ![Screenshot at 01:41: Comparison showing isotropic versus orthotropic material failure simulation on a blue block, demonstrating anisotropic material modeling \(Wolper et al. 2020\).](https://ss.rapidrecap.app/screens/SWGBN1KvG6c/00-01-41.jpg) ![Screenshot at 04:38: A demonstration of fluid dynamics simulation where many small figures pour through a funnel, illustrating particle interaction simulation complexity \(Gissler et al. 2019\).](https://ss.rapidrecap.app/screens/SWGBN1KvG6c/00-04-38.jpg)