Naive comment from a hobbyist with nothing close to $5M: I'm curious about the degree to which you build a "home lab" equivalent. I mean if "scaling" turned out to be just adding another Raspberry Pi to the rack (where is Mr. Geerling when you need him?) I could grow my mini-cloud month by month as spending money allowed.

(And it would be fun too.)

I paid 150€ for a Mini PC with an Intel N100, 16 GB of DDR5 memory, and a 500 GB SSD.

While I have no intention to scale up low spec hardware like this, it at least seems to beat the Azure VMs we use at work with "4 CPUs", which corresponds to two physical cores on an AMD EPYC CPU.

And that super slow machine I understand costs more than $100 per month, and that's without charges for disk space slower than the SSD, or network traffic.

Renting at Azure seems to be a terrible decision, particularly for desktop use.

It's hard to describe how slow a $150 / month azure VM really is. Holy heck are they limiting.

The degree is whatever you want to deal with. I had a rack at my last house (need to redesign the space for it at new house) with 3x Dell R620s in a Proxmox cluster, running K8s, serving Ceph from NVMe drives over Infiniband (for the mesh traffic), and 2x Supermicros running independent ZFS pools.

It was fun to build - especially Infiniband - but my next iteration is going to be a single beefy server, maybe with storage attached externally. What I had had outstanding uptime, but ultimately it was massively overkill, noisy, hot, and sucked power down.

You sure can. Pi are pretty underpowered you can get machines with more cores and memory and pcie lanes and networking out there and virtualize them

Interesting, there are a handful of PDFs in the drop that appear to be an email with a Base64 encoded attachment—inline.

OCR is so bad of course that decoding the Base64 seems futile without a lot of effort.

Example: https://www.justice.gov/epstein/files/DataSet%2011/EFTA02609...

(More mentioned here: https://old.reddit.com/r/Epstein/comments/1qu9az2/theres_unr...)

Would a few byte errors break a binary so much as to make it undecodable ?

I think it's more than a few bytes error. (I believe this because I spent about 15 minutes on the linked document and came up empty.)

I thought I had heard that the integrated RAM/VRAM makes local LLMs fairly quick on a RAM-maxxed Mac Mini.

Sarcasm much? Ha ha, you forgot: https://en.wikipedia.org/wiki/Brown_paper_bag_test

A half-empty kind of guy!

I'm a hobbyist and price, in the end, sold me on Bambu Labs.

(And I stayed once I saw the quality. Likely Prusa can match or exceed it, but not with what I was willing to lose from my wallet.)

Not criticizing your decision, but I went the opposite way, deciding that I was ok spending a certain extra amount initially in order to encourage a non-Chinese manufacturer. But I understand not everyone has this luxury.

I bought the Core One kit to understand better how the machine works, which reduced the price delta somewhat.

It remains to be seen over the long term which way is actually better financially, as Prusas have historically had long lives, while there is only limited data on the Bambu Lab side yet.

So far, I am quite happy with my decision. But competition is on. I am excited about the upcoming INDX system for the Core One: if it delivers on its promise, it will be fantastic!

In hindsight, I would have been happy to spend more if I knew the quality of what I was purchasing would be high.

My Ender that I had purchased years earlier sat in the closet gathering dust because of how much of a pain it was trying to dial in the bed to level, etc. I took a second chance at 3D printing on the Bambu, but might not have if it were costly.

If someone now tells me a Prusa (or whatever) is as good as and simple as the Bambu, I would not hesitate to spend even double.

Feels like this is about as unserious as flamethrowers, the Boring Company, and Hyperloop.

We're living in the Age of Distraction… amusing ourselves to death (as usual).

"Nothing in outer space treaty that enables first come / first serve squatting. Second mover can always park next door."

Antarctica then. (That's fine.)

In the same way Space Race 1.0 kicked the US into putting engineering at the forefront, I look forward to Space Race 2.0. Even if China kicks our (U.S.) ass, I'm be hoping for a sea-change in our attitudes (in fact, the US getting their asses handed to them might be the best medicine we need right now).

(Why do I use the word ass so often?)