cnschn

The thing is, it’s “only” a simulation, and with DFT stuff (the approach used in the simulation) there is usually quite a bit of wiggle room to get some interesting results. It’s still kind of a big deal, it says something like “hey these people’s results might not be complete nonsense”, but I’m still waiting for someone reproducing something experimentally.

And even if it were a sign of insanity, it would most certainly not be its definition.

I’ve used gandi for about 10 years as well, but they lost me to porkbun with their recent not-so-well communicated price increases (including dropping the small free email mailboxes)

I was a bit weirded out by this as well, but honestly - it’s pretty straightforward and has been working just fine for me. shrug

One more vote for Caddy, everything just works, simple things are simple but you have a lot of flexibility for more complex situations.

You could, but it probably wouldn’t help much. The overhead for each additional Postgres server is minimal, the RAM usage comes from each database. It doesn’t really matter if those are on the same postgres instance or on separate ones, in my experience.

Also survivorship bias. A few old cars lasted for a really long time, but you don’t see or think about the majority that didn’t.