jongr.eth

@BadDhf34717 The X year yield is nothing but a predictor of the average yield from now till X years from now

My signal number 802 296 1959

@jukan05 Sounds like a fairly obvious optimization that was foregone so far to minimize complexity

Intel Raises Technology Barriers for Next-Gen Memory 'ZAM'… Preemptively Securing Patents Intel has been securing key patents for 'Z-Angle Memory (ZAM),' which is gaining attention as an alternative to High Bandwidth Memory (HBM). The technology is capable of dramatically reducing power consumption and heat generation in AI memory, and the move is seen as a strategy to preempt intellectual property rights and block fast followers. According to industry sources, Intel recently published patent 'Method and Apparatus for Pseudo-Split Die Memory Access (US20260017215A1)' in the United States. The patent pertains to the operation of ZAM, which Intel is co-developing with SoftBank. ZAM is a next-generation memory that, like HBM, is built by vertically stacking DRAM. What differentiates it from HBM is a significant reduction in the number of Through-Silicon Vias (TSVs) used for data transfer between DRAM dies. The architecture connects each DRAM die via diagonal (Z-shaped) circuit routing around a central TSV. The core of the newly published patent involves selectively activating only the necessary portions of each DRAM die comprising the ZAM stack, rather than powering all dies simultaneously. This concept aims to address heat generation and power consumption—the most critical vulnerabilities of stacked DRAM architectures. Additionally, by finely segmenting circuit routing, the design can also achieve higher bandwidth. This patent is a follow-up to Intel's previously disclosed 'Microelectronic Assembly Including Package Substrate with Multiple Die-to-Die Interconnect Structures.' While the earlier patent proposed the structural framework of ZAM, the latest one presents methods for actually operating it. Intel has also secured a ZAM process technology patent titled 'Self-Aligned Via Pattern Formation for Backside Interconnects.' The series of patents is interpreted as Intel's attempt to preempt ZAM technology rights and raise barriers to entry. This means that even if competitors develop next-generation memory with similar architectures, they would need to find ways to design around Intel's patents. Intel is developing ZAM in partnership with Japan's SoftBank, working alongside SoftBank subsidiary SyMemory. The goal is to reduce power consumption by 40–50% compared to conventional HBM. The Z-angle architecture is expected to simplify the manufacturing process to improve productivity while also increasing per-chip capacity. Intel is also developing a new semiconductor packaging technology called 'NGDB' to enable ZAM implementation. The commercialization target is 2029–2030. Intel unveiled a ZAM prototype for the first time at the 'Intel Connection Japan 2026' event held in Japan earlier this month. An industry insider noted, "The biggest technical hurdles for HBM are heat generation and power consumption," adding, "Intel is pursuing comprehensive innovation through ZAM—not only in DRAM stacking architecture but also in placement within AI semiconductors—so it will be worth watching whether this disrupts the HBM landscape."

@tobyliiiiiiiiii IPO hype

SpaceX & xAI released a render of their proposed mass driver on the Moon today, a machine that can launch spacecraft into space using electromagnetism. As part of their Moon Base ambitions, the company aims to use the mass driver to launch AI satellites into deep space.

@pickover 1 + 1 = 0 ??

Mathematics. The Woman, the Finger, and the Triangle: A Math Diagram That Will Make You Question Everything. Please enjoy this right-triangle for which a² + b² = c².

@DJSnM Should be in a landfill

Good thing Artemis is on the pad rather than in space

@maceskridge @KevRGordon If you're looking at a historical chart you can do actual forward pe

@KevRGordon Forward P/E uses imaginary earnings. This is the actual P/E over a longer time horizon than 5 years

Over the past 5 years, the S&P 500's forward 12m P/E is almost unchanged while the index is up by 83%

@ExploreCosmos_ Easy fix, just need a space sunshade

A recent study investigates how long Earth can sustain an oxygen-rich atmosphere under long-term planetary and stellar evolution. Using coupled models of climate, the carbon cycle, and biogeochemical processes, combined with stochastic simulations to explore uncertainty, the authors show that Earth’s atmospheric oxygen is not a permanent feature but a transient phase. As the Sun gradually brightens over geological timescales, increased surface temperatures enhance silicate weathering, which progressively removes carbon dioxide from the atmosphere. Once CO₂ concentrations fall below the level required to sustain oxygenic photosynthesis, biological oxygen production collapses. The models predict that this leads to a relatively rapid and irreversible drop in atmospheric oxygen, occurring on average about one billion years from now, well before the Sun enters its red giant phase. After this transition, Earth’s atmosphere would resemble pre-Great Oxidation conditions, dominated by nitrogen with very low oxygen and elevated methane, supporting only anaerobic microbial life. The study concludes that oxygenated atmospheres like Earth’s are likely short-lived in planetary terms and that oxygen should not be assumed to be a persistent or universal biosignature when searching for life on exoplanets. 👉 arxiv.org/abs/2103.02694

@ArcRaidersNow Inorganic. They use aggression based matchmaking, it's not a cultural phenomenon within the game

ARC Raiders players have formed a community called “Rescue Raiders” — squads dedicated to reviving and protecting strangers instead of hunting them. Extraction shooter… but with heart. ❤️ #ArcRaiders

@ExploreCosmos_ Only 30 to 40 times magnification?

In one of its most visually compelling and scientifically productive observations to date, the #JWST captured a remarkably well-defined Einstein ring formed by the distant galaxy SPT0418-47. This system lies ~ 12 billion light-years from Earth, so it is observed as it existed when the universe was only about 1.4 billion years old. The striking ring-like appearance is not a property of the galaxy itself but the result of strong gravitational lensing produced by a massive foreground galaxy positioned almost perfectly along our line of sight. The geometry of this alignment is unusually precise. The foreground galaxy appears as a compact blue-white point at the exact center of the image, while the light from SPT0418-47 is bent symmetrically around it into a luminous orange ring. This configuration occurs when the source, the lens, and the observer are nearly perfectly aligned, causing the background light to be deflected equally in all directions. The ring is therefore not merely adjacent to the foreground galaxy; it is geometrically centered on it, directly reflecting the curvature of spacetime induced by the lensing mass. Although the ring appears almost flawless at first glance, careful analysis reveals subtle asymmetries and brightness variations. These small deviations from perfect circularity are not observational defects but key scientific signals. They encode information about the distribution of dark matter within the foreground galaxy, since invisible mass slightly alters how strongly different parts of the background light are bent. By modeling these imperfections, we can reconstruct the dark matter profile of the lens with high precision, turning the ring into a diagnostic tool for mapping mass that cannot be seen directly. Gravitational lensing in this system provides an effective magnification of roughly 30 to 40 times, transforming JWST into a telescope observing through another natural cosmic telescope. This amplification allows us to resolve internal structures within SPT0418-47 that would otherwise be far below the telescope’s detection threshold. As a result, JWST can probe the physical and chemical conditions of a galaxy from an epoch when the universe was still in its early stages of assembly. Spectroscopic observations revealed the presence of polycyclic aromatic hydrocarbons, or PAHs, within SPT0418-47. These complex organic molecules are often referred to by astronomers as “cosmic smoke” because they are byproducts of stellar processes and trace chemically enriched environments. The detection of PAHs in this galaxy is particularly significant because it represents the most distant system in which such complex organic molecules have ever been identified. Their presence implies that multiple generations of stars had already formed and evolved rapidly enough to enrich the interstellar medium with heavy elements. Equally important is what JWST did not see. While PAHs were detected, the observations do not show uniformly strong signatures of hot dust across the same regions. This mismatch between “smoke” and “fire” suggests that star formation and chemical enrichment were not proceeding evenly throughout the galaxy. Instead, the data point toward a patchy or spatially differentiated evolutionary process, in which some regions had already become chemically mature while others lagged behind. This challenges simplified models in which early galaxies evolve in a uniform and well-mixed manner. Follow-up observations combining JWST data with measurements from the @almaobs revealed additional complexity. A companion galaxy, known as SPT0418-SE, was identified near the main system. Despite its similarly early cosmic age, this companion appears to be highly enriched in heavy elements, reinforcing the conclusion that chemical maturity can arise surprisingly early in galaxy evolution. The presence of such an enriched companion suggests interactions or shared evolutionary pathways that accelerate star formation and metal production in dense early environments. Taken together, these observations show that SPT0418-47 is not a primitive or chemically simple galaxy, despite existing so soon after the Big Bang. Instead, it exhibits organized structure, rapid star formation, and advanced chemical enrichment, all revealed through the combined power of gravitational lensing and JWST’s sensitivity. The Einstein ring is therefore not just an aesthetic curiosity but a powerful scientific instrument, allowing us to study the internal physics of early galaxies, trace the buildup of heavy elements, and map dark matter distributions in foreground lenses. This single system encapsulates how modern astrophysics extracts deep physical insight from the interplay between gravity, light, and cosmic time.

@Mrgunsngear Well for the same price as a US system there's 9 more that work fine

Let's check in on a Chinese MLRS system being used in the ongoing Cambodian Thai war and see how it's going. Not well. It's not supposed to do that... 👀💥 #war #cambodia #thai #MLRS #fail #FAFO #nope #chinese #rocket #rockets

@DefiIgnas Ethereum literally did that...

Gnosis is considering a hard fork to recover funds from the Balancer hack. After the hack, ~$9.4M was frozen on Gnosis Chain via a soft fork. Funds cannot move from the attacker's address. The funds are stuck. Returning them to victims requires a hard fork. I think Gnosis clearly sits in the low-risk DeFi category (ironically), where protecting users comes first. Because Gnosis is positioning itself for neofinance and enterprise-focused Web3 services, not for Ethereum-style immutability absolutism. In any case, the damage to 'immutability' is already done. The soft fork happened. Choosing not to recover the funds now isn’t neutrality. And tbh, we’ve already seen similar censorship interventions elsewhere: Berachain and Sonic after the same Balancer hack, and Sui after the $162M Cetus exploit. But this fork sets a big precedent: Do we hard fork for every hack? Only if losses > 5% of TVL? Why not 3%? Can app devs start assuming the chain will step in if they mess up, lowering security standards? The hard fork and the debate will end up setting hard-fork precedent rules for other chains to follow.

@SuhailKakar It hasn't really scaled L1 tho

ethereum has scaled so hard and reduced gas fees so much that i no longer use L2s gg ethereum

@BitBob88 @BrianRoemmele The brain is big and warm. Probably not quantum

@BrianRoemmele A quantum collapse of all probabilities in a scaled down section of eternity.

Visualization of what is inside of AI models. This represents the layers of interconnected neural networks. And yes patterns do develop and they can form a signature of how they think. The pattern can be seen as the thought process.

@VitalikButerin Of course there aren't producers of gas looking to hedge downside risk like there are for e.g. corn. So that side of the trade will have to be fulfilled by overcollateralized speculators with credit risk

We need a good trustless onchain gas futures market. (Like, a prediction market on the BASEFEE) I've heard people ask: "today fees are low, but what about in 2 years? You say they'll stay low because of increasing gaslimit from BAL + ePBS + later ZK-EVM, but do I believe you?" An onchain gas futures market would help solve this: people would get a clear signal of people's expectations of future gas fees, and would even be able to hedge against future gas prices, effectively prepaying for any specific quantity of gas in a specific time interval.

@charliebilello QE was the absolute most harmful dumb idea to ever be had. It does absolutely nothing and it took the place of actual stimulus that could actually have avoided the recession.

The Fed's balance sheet has shrunk 24% over the past 3 years while the S&P 500 has advanced 82%, dispelling the myth that the stock market is dependent on QE to rise. $SPX

@HarryStebbings @jonsidd Their mote is their network effect. Difficulty of building the software has never been a factor

SaaS is Over: " SaaS apps were built at a time when software was relatively hard to build.  Now many of these AI applications are easy to build on top of these LLMs. Most companies will start building custom software super easily." @jonsidd Love to hear your thoughts on this @destraynor @dharmesh @craigzLiszt @jasonlk @FabianHedin @spenserskates

@tonywestonuk @Kael2510 @Steve277Masters The area under the curve of 20M+ is not very big

@Kael2510 @Steve277Masters $1m is rich compared to most....But its a drop in the ocean compared to some of the megarich, who by buying up huge amounts of real assets, means less for the rest of us.. Its these people who need to be taxed heavily. finance.yahoo.com/news/bill-gate…

Taxing the rich is essential. The rich hoard assets, and less available assets means higher prices. This is economics, 101. Supply and f**ing demand . And MMT does say this when you look even skin deep.

@WilliamXnote @dmesg @ZackPolanski Taxes cause inflation? You have that backwards. The whole point of a tax is to lower inflation. To your point about investment, if the tax doesn't affect CPI then it's useless. But of course it will sooner or later since the end goal of all savings/investment is to buy things

@dmesg @ZackPolanski Taxing the rich taxes mostly investment; it doesn’t much alter their consumer spending and to extent it does doesnt have a huge macro impact Taxing ordinary people quickly goes to consumer price inflation

The Labour Government refused a 1% asset tax on the super rich, yet raised taxes on people already struggling. All whilst their "black hole" turned out to be utter nonsense. Active choices to refuse to fund our communities which they desperately need. Join.greenparty.org.uk

@paoloardoino 7/184 is only 4% overcollateralization. Since they are buying gold not treasuries tether holders are taking on the downside risk of their USD short position while they take on the upside potential

re: Tether FUD From latest attestation announcement (Q3 2025): "Tether will continue to maintain a multi-billion-dollar excess reserve buffer and an overall proprietary Group equity approaching $30 billion." Tether had (at end of Q3 2025) ~7B in excess equity (on top of the ~184.5B stablecoin reserves) + another ~23B in retained earnings as part of our Tether Group equity. Tether Group total assets: ~215B Stablecoin liabilities: ~184.5B S&P made the same mistake of not considering the additional Group Equity nor the ~500M in monthly base profits generated by U.S Treasury yields alone. Some influencers are either bad at math or have the incentive to push our competitors. Forever trusting who we are No, nothing else matters

@TeodoroFuente10 @JohnLeFevre You are literally freeloading off the medical technology R&D the American people pay for

@JohnLeFevre In the same way that the US has the most expensive healthcare system and most expensive universities in the world. And yet Americans are neither the healthiest nor the most educated people in the world.

why is it that vacationing in Europe is so amazing, and yet, the largest company in Europe is 10% the size of NVIDIA.