Substrate is using x-ray lithography instead of EUV. but i thought XRL was a dead end because it's ionizing radiation that can blur features by damaging the mask or the substrate. last i looked it up, minimum XRL feature size was 30 nm, worse than EUV's state of the art ~16-18 nm gate length for a "3 nm node".
https://en.wikipedia.org/wiki/Synchrotron_light_source when you run electrons real fast in a circle they emit very bright, very coherent x-rays. this is what Substrate is doing. there might be something about the coherence or the intensity that helps with blur? but the details are a trade secret and Substrate isn't saying.
https://www.owlposting.com/p/there-arent-enough-smart-people-in once again, it is not "boring" to work on repurposing (best biotech approach EVAR if you want to specialize in computational vs. laboratory capabilities) or CROs (if you want to improve on SOTA of experimental or manufacturing technology, and build a business around that, it is much harder to do that within a biotech startup that also needs to build a pipeline, than to specialize in experiments or manufacturing as services)
https://builders.genagorlin.com/p/reflections-upon-turning-40 probably helpful to do a timeline of your life and the milestones in it, as seen in this post. it's kind of a trip to see just how *many* things i've tried in life by age 37, and how like...20% at best? were unqualified successes.
https://docs.nrel.gov/docs/fy21osti/79178.pdf this is already priced into NREL's techno-economic analysis; this document projects scale-up of various technologies that are currently benchtop or pilot scale.
https://www.nature.com/articles/s41591-024-03057-9 6 distinct "biotypes" of depression based on fMRI clustering; show different symptom patterns and patterns of performance on cognitive tests, and different responses to treatment.
this seems pretty rigorous
otoh i wish there were characteristic "types of depression" that were more recognizable on the symptom/behavioral level, such that a therapist or psychiatrist could make a guess as to which type a person might be. this isn't that.
old-school "melancholic depression is when you don't eat, don't sleep, frown super deeply and wring your hands, have intense guilt, and have high cortisol" would be an example of a much more stereotype-able depression subtype, the kind that's more useful in practice than any of the 6 "biotypes".