At first I thought the goal of chip sanctions on China was to bring them to the negotiating table, negotiate out fair play and go on our way.
But if instead it keeps going the way it’s going, I wonder if 20 years from now we’re sitting in a split world, where the west has one set of ISAs, fabrication plants etc, and China has another which don’t communicate with each other, since China can’t use any new ARM extensions and eventually all of ARMv9 will be deprecated too.
I know it’s harder to get past about 5nm without EUV machines, but I wouldn’t really doubt that China putting a national effort into it could eventually get there, even if it puts them some years back for now.
Is…Is that split world even a goal? Or is that an accidental consequence of neither side getting what they want and a default path no one wanted?
Not really, translation layer/virtual machine is standard in compute these days. KVM, VMWare, Rosetta on Apple, or Kubernetes/Dockers. RISC-V is building on RISC-4 instruction sets, so compilers and all will be there. These are barriers thought up by politicos with limited tech knowledge, then get surprised by the speed of change.
At first I thought the goal of chip sanctions on China was to bring them to the negotiating table, negotiate out fair play and go on our way.
But if instead it keeps going the way it’s going, I wonder if 20 years from now we’re sitting in a split world, where the west has one set of ISAs, fabrication plants etc, and China has another which don’t communicate with each other, since China can’t use any new ARM extensions and eventually all of ARMv9 will be deprecated too.
I know it’s harder to get past about 5nm without EUV machines, but I wouldn’t really doubt that China putting a national effort into it could eventually get there, even if it puts them some years back for now.
Is…Is that split world even a goal? Or is that an accidental consequence of neither side getting what they want and a default path no one wanted?
Not really, translation layer/virtual machine is standard in compute these days. KVM, VMWare, Rosetta on Apple, or Kubernetes/Dockers. RISC-V is building on RISC-4 instruction sets, so compilers and all will be there. These are barriers thought up by politicos with limited tech knowledge, then get surprised by the speed of change.
Kubernetes / docker don’t do any translation at all. They’re running on the host kernel, namespaced away.
The speed of change of what exactly?
SMIC say it is possible to use DUV for 2nm. Yes there will be a split standard, US or China.