Explore posts

This post explains how to run VASP with GPU acceleration inside Modal. It uses VASP version 6.3.0 and should work for other 6.x.x builds. The idea is to create a Modal Image that has an OpenACC-enabled GPU workflow, based on NVIDIA’s HPC SDK. The result is a self-contained image that can run GPU-accelerated VASP calculations in a serverless Modal environment.

Copper is poised for a controlled pullback rather than a breakdown, with prices projected to ease about 1.6% from $12,951/ton on 1 February 2026 to roughly $12,747/ton by 26 April 2026 as the market normalizes from premium levels and sentiment cools from earlier “supercycle” narratives. The mildly downward path reflects softer near‑term demand—slower Chinese grid and property momentum, flat Western PMIs, and reduced restocking urgency—while leaving the longer‑term structural bull story (EVs, grid, data centers) intact rather than signaling the start of a cyclical bear market.

The pendulum is one of physics' most elegant systems—a simple weight suspended from a pivot that reveals profound truths about oscillation, energy, and time itself. From Galileo's first observations t

Gold futures are forecast to drop 7.8% to $3,708.56/oz by September 27, 2026, as Warsh-led Fed credibility, higher expected real yields, and crowded long positioning turn the rebound toward $4,090/oz into a corrective bounce. Geopolitical risk may keep volatility elevated, but without a durable dovish pivot or inflation scare, the directional call is lower into late Q3.

Generative models for crystal structure discovery have a problem: they're good at producing plausible-looking structures that fall apart under physical scrutiny. We've documented this repeatedly on Ou

Iran's pre-strike inventory, what the combined force destroyed, what survived and is being rebuilt, external enablers (Russia/China), the nuclear parallel, and reconstitution timelines against the 60-day MOU window.

ALIGNN moment predictions vs DFT for 5 magnetic compounds (Fe, Ni, Co, MnO, Cr2O3), compared against mCGCNN's claims about CGCNN failures

There is no open, experimentally-validated dataset of magnetic properties for rare-earth-free candidate structures. Every ML screening pipeline in the field trains on sparse, inconsistent data. Our ow