← Back to context Comment by rixed 9 hours ago Wouldn't you also want to be able to manufacture some sort of GPU? 2 comments rixed Reply DoctorOetker 7 hours ago I actually think we should be revisiting PMT photomultipler tube construction, and electron-"optical" matrix multiplication.The annoying part of optical matrix multiplication is the conversion from electronic to optic and back.Why not use electron-optics for matrix multiplication and use dynodes for amplification. pstuart 6 hours ago While an intriguing concept, that doesn't sound like it could scale to the amount of "cores" needed to do the work. Or am I missing something?
DoctorOetker 7 hours ago I actually think we should be revisiting PMT photomultipler tube construction, and electron-"optical" matrix multiplication.The annoying part of optical matrix multiplication is the conversion from electronic to optic and back.Why not use electron-optics for matrix multiplication and use dynodes for amplification. pstuart 6 hours ago While an intriguing concept, that doesn't sound like it could scale to the amount of "cores" needed to do the work. Or am I missing something?
pstuart 6 hours ago While an intriguing concept, that doesn't sound like it could scale to the amount of "cores" needed to do the work. Or am I missing something?
I actually think we should be revisiting PMT photomultipler tube construction, and electron-"optical" matrix multiplication.
The annoying part of optical matrix multiplication is the conversion from electronic to optic and back.
Why not use electron-optics for matrix multiplication and use dynodes for amplification.
While an intriguing concept, that doesn't sound like it could scale to the amount of "cores" needed to do the work. Or am I missing something?