16 comments

[ 5.4 ms ] story [ 63.5 ms ] thread
(Disclaimer, I generally am against cryptocurrency and I think it's at best pointless and at worst a scam)

This is overall an interesting idea. They have to burn the gas off, I'm assuming it's infeasible for them to send the electricity anywhere useful, so mining bitcoin let's them do something with the fuel they would otherwise just waste.

I wonder if the same idea could be improved upon to actually do something useful. For example, a portable GPU data centre, for ML model training? Maybe the bandwidth requirements are too high. Anyway, the concept is good (using waste fuel for on-site data processing), they just need to find a better application

does the assumption of infeasibility of sending electricity contradict the idea that wires are used to send the mined bitcoin?
I'd assume any data needed for the mining process is transmitted wirelessly. Sending electricity wirelessly is infeasible.
It's a profitability issue, not a logistics issue. It would not be profitable to convert that natural to electricity if it was used for anything besides Bitcoin.

And worry not, this is benefiting you even if you don't like cryptocurrency:

"The firm estimates that bitcoin mining allows carbon dioxide-equivalent emissions to be reduced by over 60% compared to routine flaring."

> It would not be profitable to convert that natural to electricity if it was used for anything besides Bitcoin.

My sibling comment talks about this a little more, but as someone running data centers on the same technology, I can assure you that it is possible to do profitable compute that is not crypto mining on this technology.

> "The firm estimates that bitcoin mining allows carbon dioxide-equivalent emissions to be reduced by over 60% compared to routine flaring."

https://www.crusoeenergy.com/digital-flare-mitigation has all the stats on Methane, CO/CO2, VOCs, NOx, etc.

For comparison, using one of our GPUs for a month offsets the equivalent of ~400 kg of CO2, which is about the same as a round trip flight from SFO to SEA.

Coming in a bit late to this (apologies for missing HN yesterday), but I'm the product lead for Crusoe Energy's cloud product--we're in an alpha of exactly this :)

We've got Nvidia A100s and A40s racked in data centers on site; ML training, rendering, simulation, etc. are all great applications for this type of data center.

If you're interested in learning more, crusoecloud.com has some info (plus a signup page), and I'm also available at mike at crusoecloud dot com.

This is pretty cool. Cryptocurrency mining is a great use for "stranded" energy like this. Rather than just being burned off, it secures the distributed network.
is the value rewarded by mining greater than the value of energy spent during mining?
Yes, because in this case the value of the energy spent is zero- it would have simply been burned off into the atmosphere.

Similar concept as mining next to underutilized hydroelectric dams.

It's a good way to use wasted gas. But I wish there was a better use for the burn.

Anyone know of any practical basic materials that require energy but can be created in a relatively small area?

I can foresee a mini factory that can be dropped close to a leaking well. It would use the gas to create the material instead of bitcoin. Same process different product...

Hydrolysis, carbon capture, folding genes etc...

Just not very profitable compared to Bitcoins

fertilizer. Hydrogen.