My guess, 10,000x the cost on CL1. Even with the tech perfected, bio neurons fire much much slower than logic gates and electricity in a circuit board. If you have an ASIC (custom built board that isn’t really using a CPU), the ASIC would be much much faster for deterministic calculations at high speed with an active cooling system.

Bio neurons are great at self-organizing. If you already know how they need to be organized (e.g. a hashing algorithm), and you need max-speed output there’s no real advantage.

It’s not wrong to say bio neurons are power efficient, its just that power efficiency depends on what the activity is.

Don’t worry all neurons are cage free, grass fed, open range

For real though, where the neurons come from is as interesting/impressive as the computation itself. The guys at Cortical, at least in prototyping, give blood samples, revert blood cells into a stem cell state, and then (over the course of 6 months) they convert their stem cells into neurons before putting them into a dish. (To be clear, Cortical did not invent the stem cell tech at all. Apparently its standard practice and nobody in the bio engineering world cared to tell the rest of the world.)

Meaning… You could theoretically build a computer out of your own neurons and then program them.

The neurons in the machine (or at least the prototypes idk about every CL1) are neurons from the lab lead (Hans). And he has given consent 😁

I’ll ask them, I happen to see them in a zoom meeting occasionally.

I very very very much doubt its a new machine, however you might need to send your machine back to get it refilled as I imagine there is a precise integration between bio neurons and electrical hardware.

You actually legit feed it snacks haha. There’s a nutrient mixture/sludge to keep them alive.

Sounds good. Given its creeky all over the place, I’ll probably try mapping out the joists first. I’ll (hopefully) post an update after my attempt.

Thanks for the advice!

Blue heeler!

Yeah I’d say that’d be pretty tough with carpet

Nope, unfortunately

1. Grad student
2. Bad

For a post that sparks good answers that I’m happy to see, I’m sad to see the post itself have so many down votes.

Once you’re >25 this is just a flex

Glad I’m not alone on this

Usually you can just google tiktok downloader, paste in the url and download it. Its possible those services have died since I last checked though.

deleted by creator

I think it could be a great solution. I’ve never considered it before. That said there’s one sticking point for me:

Apportion payment to developers based on software use by paid users and the size of their contribution to that software.

That^ . That needs a lot more detail. If they provide solid details – details that most can agree on – then I will actually be on board with the solution.

Labeling datasets is costly process. When you dont opt out, you’re letting them build a labelled dataset on you-specifically for free.

Same for me: just say no, and they say OK. Effortless but the option is totally invisible.

The irony is, I’ve seen the staff stop using the face scanner for everyone halfway through the line to speed things up. So its not saving time, just costing money to increase surveillance.

No

Yeah, cookies, account logins, and other stuff make it hard too. Ex: randomly exploring gmail emails at different times of day, but not actually marking emails as read.