Show me another OS that you can undress to the kernel, a console, a file system and a disk driver and then build it all up again without missing a beat.
The kernel processes are actual processes so each of the drivers is fully sandboxed, an error in one bit of code can not cause any other processes to be affected unless you explicitly declare that it should be so (shared memory, for instance) and of course you don't do that.
The reduction of scope alone is worth at least 30 IQ points.
Absolutely rock solid. I built some specialized network devices using QnX and those things ran for a decade+ after first installation. Not a single reboot.
IBM 3380 Model K introduced in 1987, has 7.5 GB of storage and costed about $160000 to $170000, or adjusted for inflation it is $455000 in 2025 US dollars, that's $60666/GB. A Solidigm D5-P5336 drive that can store 128 TB costs about $16500 in 2025 US dollars, that is $0.129/GB. That's a 470279x price reduction in slightly less than 40 years. So what is likely going to happen to LLM pricing? No one knows and both your example as well as mine doesn't mean anything.
I think it really depends on the task at hand, and not that intuitive. At some point accessing the memory might be slower than repeating the computation, especially when the storage is slow.
have you considered using meshlets technique like Unreal Engine Nanite or Assassin's Creed? It could potentially open the door for better culling and more effective depth prepass.
It is important to not leave out the context that Google needed TPUs for their AI development. Now the question is what does Zoho email and office software gain from fabricating custom chips?
reply