The author explains how the performance of its Ring Buffer implementation went from 5.5M items/sec to 112M items/sec by looking at and addressing cache misses and so on.
I need to look into re-learning #thermodynamics. #Motherboard says the #CPU#core#temps are around 10C higher than what the #heatsinks are (I built my own #sensor set for this machine - because I wanted to.)
Granted, this is one of the old E5-2600v2 series #motherboards in the computer I'm typing on right now.. (different computer, not the server - waiting for #Amazon to deliver my new #frelling CPU coolers), an X9DRH-7TF by SuperMicro, so maybe that's #normal?
Yeah, I know, WAY overkill for a #workstation, but it's also a #pull from a previous #server iteration, so...
Yes, I'm #weird. I upgrade my #servers, then my #workstations get all the old #parts.
Usually in a single-core performance environment a desktop device is twice as fast as a mobile device, my results shows. Multi-core speed comparison is more irrelevant, may have more level of difference also, depends on core numbers and core efficiency though.
Usually in single-core performance environment a desktop device is twice as fast as a mobile device, my results shows. Multi-core speed comparison is more irrelevant, may have more level of difference also.
I am currently composing a new #PC for my father...
And I'm currently faced with the question "#AMD or #Intel" - It's not supposed to be a high-end system, but rather a well-performing computer that should also last for about a good decade. (He VERY rarely gets a new one) So I will use the latest of the latest.
As for #CPU selection, I find AMD simpler simply because there aren't as many models. (I'm also a bit of a AMD fanboy) What do you think?
Chinese Researchers Used AI to Design RISC-V CPU in Under 5 Hours (www.tomshardware.com)
The chip ran Linux benchmarks at about i486 speeds.