It'll be interesting to see what happens when an AI-piloted F-16 flies for real against a human-piloted one. IIRC, the F-16 maximum maneuvering G-load is 9.5G, LBA (limited by airframe). Maybe that's the meatbag limit, not the airframe - possible to crank it up a bit? Certainly the various CCAs in current flight test will take much higher G loads.
Several years ago, DoD flew an AI against a human in a simulation. In basic fighter maneuvers (no BVR stuff like missiles, just a merge and fur ball), the AI killed the human - always. So G-limits may not matter. Pretty interesting.
YMMV a lot with Aliexpress. I bought a $700 chiller that arrived without a small attachment fitting for the return inlet line. I complained, received another chiller (with the fitting) free. Go figure.
eBay is very good for science-related goods, in my experience. Last week, I bought ~$12K of used gear for a project. All items work, would have cost ~2.5x that from the usual used instrument vendors, about 6x new.
YMMV. When we sold in 2021, our realtor connected us with an interest-free lender who put up $76K for repairs and staging. This, and other services, made the difference between our selling for $1.8M and the $2.4M we got. Maybe we could someday get similar service from a realty AI, but in our recent experience, the humans made all the difference.
The lighter centers could have been diffraction around the object and reconvergence. Or some kind of signal processing effect at the retinal level.
The retinal vessel network is a fun thing to inspect as you say. It works best for me when the sun is high in the sky. The bright, featureless blue brings out the branching network very well.
Reconsider the blood cell visualization thing. I've had bilateral PVD. Each was accompanied by an initial few minutes of seeing a few dozen small dark spots that had lighter grey centers. Most visible when I looked at the sky. They all disappeared after a few minutes. I think these were RBCs from minimal retinal blood vessel tearing at the PVD events.
Human RBCs are ~6µm - 8µm diameter. Human retinal light sensing cells range from ~0.5µm - 10µm diameter, depending on type and position. They're close packed.
Given the geometry, RBCs leaking onto retinal cells should cast shadows that could be resolved as images. And that's right where leakage is most likely to occur during a PVD event.
"Rocket Manual for Amateurs" was my favorite book after I found it in 8th grade. In high school I had a chem teacher who would give me chemicals so I could experiment with what I'd read. A great book for budding Raketenkinder.
I wonder if the results would have differed if LLM training data were biased to include a stronger correlation between use of nukes and subsequent collapse of technology that all LLMs require to run ("survive")?
I'm building a prototype chemical vapor deposition system in a space with strong Chinese interest and activity. I picked Prusa 3D machines over Bambu because of the potential for losing critical proprietary IP with Bambu. Can't take the chance.
Unknown at present (nobody sells diamond sandpaper yet that I know of), but not likely. Diamond is as bioinert as gold, fluorocarbons, and similar materials in not provoking inflammatory responses when implanted in tissue.
Your point about unexpected behavior of dust is a good one. AFAIK, the worst dust hazards (aside from outright chemical or radiological toxicity) are from spiky, acicular particles like asbestos. Diamond dust is either well-formed small crystals or blocky fragments, depending on how it's made. It doesn't exist in needle-like forms that trigger pathological responses from otherwise nontoxic materials.
One has to be trained to do it, the untrained tendency is to wait too long. There's a USAF film on Youtube titled "Ejection Decision" that discusses this and shows how little time there is to make that choice.
De Beers built and has abandoned a very large CVD gem diamond synthesis in Oregon. This was their LightBox operation. They've transferred it to Element Six to rid the parent (De Beers) of the money sink, probably as a touch-up before a sale.
There was a fire sale of LightBox diamonds a few months back. I picked up 3 diamonds, all brilliant cut: 3ct. white, 2ct. pink, 2ct. blue. Total for the three was $600.
De Beers sells devices that can distinguish between naturals and CVD synthetics. They're not cheap, but less than ~$80K, IIRC. They do a pretty good job, I've heard >90% success in identifying CVD stones.
Diamond grit for polishing and grinding is now a cheap commodity. In 10,000 ct. lots, I pay from 5¢/ct. to 30¢/ct. depending on specific grit properties. I haven't searched for it, but diamond sandpaper should be a thing at these prices.
Several years ago, DoD flew an AI against a human in a simulation. In basic fighter maneuvers (no BVR stuff like missiles, just a merge and fur ball), the AI killed the human - always. So G-limits may not matter. Pretty interesting.