Proton dropped prices when their costs went down.

Gotta follow the 3-2-1 rule

I really want to know how they transferred the data at 650Gb/s

Isn’t Grok the worst?

Translating bash commands to powershell is actually one thing the Google AI really excels at.

Of course I don’t have it build my scripts, but it’s been useful if the Microsoft documentation doesn’t have enough information about the flags.

And I’m asking why you think that.

China’s models have gotten significantly less resource intensive based on updating the software.

This time is different because the chip fabs intentionally retooled for AI chips. It’s not an uncontrollable external pressure.

There’s no guarantee that they’ll switch back to consumer modules after the bubble pops. I’m willing to bet that Big Tech’s contingency plan to the pop is to utilize a dead consumer market to push subscription based virtual desktops and zero clients.

I can’t even buy the ram I need anymore, even if I wanted to pay out the ass for it.

I have to use it on a daily basis at work.

Powerful computers can act as space heaters in the winter.

They’d probably just confincate the drive if it’s obviously encrypted.

Why are you assuming that software improvements don’t include less resource utilization?

Integrated GPU

Testing with streaming at 1440p@144Hz from a cacyos host vm running sunshine to a cachyos physical client running Moonlight, the moonlight client used:

• 153.2MB RAM
• 2-3% of an i3-N305
• 24Mb/s (3MB/s)
• 15-20% of the iGPU

And none of that changed between running Autodesk Fusion with both a low and high polygon assembly, running a AAA game at 1440p@144Hz, or emulating a switch game at the same resolution and refresh rate.

The application being streamed is irrelevant to the resources used by the client.

Edit: Added GPU utilization.

deleted by creator

They seem to think criticism of the CCP means support for the US fascists.

It’s possible to hate both.

Some people have more money than sense