For some odd reason the 'Chia burns out hard drives!' is getting repeated as the fashionable fud. This is odd, because for the most part it's just plain wrong (thread)
If you plot with a plain old HD or an enterprise-class SSD then your drive will survive no problem. Plotting on hard drives, including the one which you'll probably leave your plot on, works fine. It's a bit slower and requires more head room, but works fine.
We've gone to great pains and put a lot of technical wizardry into getting plotting even on HDs to work completely acceptably. The amount of head room it needs isn't much more than one plot, and the time it takes isn't much more than double.
At the moment there's a huge premium on plotting quickly because (a) everything is new and (b) the netspace is growing, uh, a bit quickly
In this environment there's a huge premium on getting plotting done fast because the work difficulty is going up so quickly that the same amount of space now is worth a lot more than it will be in the future. Once work difficulty gets high enough this will settle down.
What point the S-curve will stabilize at can't be predicted with curve fitting during the growth phase. My guess based on running numbers is somewhere around 100 Exbibytes
Also plotting is much more prominent now because a plot is made once and stays around forever. You're going to see a lot of people with big plotting rigs with not much need for plotting in the future.
This has already happened to some extent with plotting as a service being available. Also some people planned ahead and got themselves a new gaming machine which first paid for itself by plotting
But back to the original point: Don't plot with consumer SSD! Or at least, only do a little bit of plotting with each consumer SSD
Also don't clean nonstick pans with steel wool, don't clean vegetables with soap, and don't use your phone as a doorstop. These aren't arguments against steel wool, soap, or phones, they're basic guidelines about using your tools properly.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
OpenAI's chatbot doesn't have much experience in the material world
Don't let OpenAI's chatbot take care of your pets
They seem to be updating it in real time so it gives better answers to specific questions than it did just a few hours ago but sometimes small variations can trip it up
Good morning everyone. Today's rant will be about analog audio and some truly maddening myths in that field (thread)
There's a persistent myth that as soon as any step in audio processing is digital the entire thing has been reduced to a lower digital quality and that's what you're stuck with. This is wrong.
The truth is that every step of processing, either analog or digital, reduces quality, and each specific process will have its own effects, but the limits of how little quality can be lost are much better when both the inputs and outputs are digital
There's a myth in the general public that Know Your Customer programs are important for banking. This is incorrect (thread)
KYC has nothing to do with bank auditing and solvency. It is a relatively recent program which is solely for helping law enforcement. To that end it is extremely expensive and ineffective
The thing which really keeps bank together is federal reserve auditing and FDIC insurance. You know, the things which keep them from turning into FTX. KYC didn't start until 1970. It makes up a significant amount of all banking costs with no benefit to bank customers whatsoever
Something which offends me about SBF is that he's a finance bro's notion of a charmingly autistic founder. He plays that role well because he's a finance bro himself (thread)
Here's the chart of SBF's corporate empire. Nobody sets up something like this without damn well knowing that they're up to something shady
Another obnoxious thing is his inability to tie his shoes supposedly being cute. An adult being incapable of tying their shoes is a red flag of someone who doesn't take pride in their work and can't be bothered to do much of anything for themselves
Someone who has not made major charitable contributions, either through money or deeds, can't claim to be an 'effective altruist' any more than they can claim to be a philanthropist
Someone who's starting out their career and plans to do it can claim that they aspire to be an effective altruist some day, or even have direct plans for it, and if they haven't yet made lots of money that's a reasonable claim
If someone had a plan of making lots of money and using it for charity in the end but their investments never panned out I'm even happy to give them the benefit of the doubt that they would have been very generous