0
submitted 3 years ago* (last edited 3 years ago) by tronk@lemmy.ml to c/todayilearned@lemmy.ml

The speed and the efficiency of these chips are just insane:

"As just one example, a recent microprocessor design using RISC-V has a clock speed of 5 gigahertz, well above a recent, top-of-the-line Intel Xeon server chip, E7, running at 3.2 gigahertz. Yet the novel RISC-V chip burns just 1 watt of power at 1.1 volts, less than one percent of the power burned by the Intel Xeon." (Source)

Perhaps reading from the source is better? ๐Ÿ™‚ https://riscv.org/

Sorry if I'm preaching to the choir, given the type of Lemmy users, in terms of already knowing about RISC-V and in terms of assuming y'all know about the M1. Still, I just learned about RISC-V today, and I'm awed!

top 1 comments
sorted by: hot top controversial new old
[-] Jeffrey@lemmy.ml 1 points 3 years ago

It's worth noting that 1GHz x86-64 is not the same as 1GHz RISC-V or 1GHz ARM. Different CPU architectures have different instruction sets, so it might require more cycles to achieve the same result.

e.g. 1GHz x86-64 and a 2GHz ARM - The x86-64 chip has a desired instruction built in, so it is able to perform the desired calculation in 5 cycles. The ARM chip does not include the desired instruction, so it takes 50 cycles to achieve the same result. Even though the ARM chip is twice as fast, it will take ~5x longer to run the same task.

I've never used RISC-V, but I did some testing with a couple ARM CPUs and a couple x86-64 CPUs last year and the results were roughly: ARM took ~5x longer, and x86-64 used ~5x more electricity. For the chips I was using, and for the work I was doing, there wasn't any efficiency gain by switching fully to ARM.

I am super excited for RISC-V, though! I can't wait to have a RISC-V coreboot machine.

this post was submitted on 27 May 2021
0 points (NaN% liked)

Today I learned

7585 readers
1 users here now

founded 4 years ago
MODERATORS