

PhysX has just been a CUDA application for a long time, there’s not been a dedicated PPU on any card in a very long time
PhysX has just been a CUDA application for a long time, there’s not been a dedicated PPU on any card in a very long time
Welp here’s hoping something pops up to replace gravity sync soon, can’t really upgrade until it does.
I just found https://github.com/mattwebbio/orbital-sync but it’s not apparent if it’s compatible with v6 yet
This is just one of many reasons, for example
The FAA head pushed out by Musk before all the planes started crashing had fairly recently fined spaceX for safety issues.
It’s just pettiness and entitlement. He’s going after everyone who has ever said no to him.
https://www.theregister.com/2025/02/14/youngsters_in_foss/
Read this a few days ago and it feels pretty relevant here.
I bet you can run 10base2 over it
Further evidence that a Republican government in the USA results in private organisations pushing the bar as far as they can.
In Reagan’s time it was Wall Street. Now it’s Silicon Valley.
You want private organisations working for your benefit and not that of their shareholders? You need a government that actually has the gumption to challenge them. The current US government is 4 years of a surrender flag flying on the white house.
Or we could bin off this fucking failed neoliberal experiment, but that’s apparently a bit controversial for far too many people
The problem is you’d need to split it down to an amount that people would be happy hosting and then host it multiple times in case any node goes offline.
Another comment in the thread says it’s likely over 100PB today (100,000 terabytes). I’d say 4 copies (spread over different time zones) is a relatively minimal level of redundancy (people may host on machines that aren’t powered all the time), and I reckon you’d get a network with the most participants, whilst still getting enough storage, at around the 150gb per node mark.
That comes to nearly 3 million participants needed just to cover today’s archive, new people will obviously need to join every day. Also given I imagine it would need to be open to all, the redundancy level could do with increasing to avoid malicious actors with a lot of resources taking on a lot of the network and forcing it all offline at once in an effort to cause data loss
Nothing here is insurmountable, but also not remotely easy
Techy people are a lot more likely to jump through a couple of hoops for something better, compared to your average Joe who isn’t even aware of the problem
Is there an extension of Betteridge’s law of headlines for tech blogs?