Re:AMD laptops have a hidden 10-second performance delay
by srazzaque on Wednesday November 25, 2020 @04:53AM
(#25207547)
Great links. Whilst I'll admit a confirmation bias on my side, benchmarks aside, I'm pretty happy with my AMD X13. This is after literally 15 years of only purchasing Intel laptops - aside from thunderbolt, I dont feel like I've lost out on anything.
An interesting little excerpt from the first link:
> this tuning Intel does is in software and for Windows only. If you change OSes or even if the OS is upgraded, chances are this tuning will vanish.
Re:AMD laptops have a hidden 10-second performance delay
by tsujp on Wednesday November 25, 2020 @05:06AM
(#25207624)
My first ever PC, that was mine mind you, was some cheap AMD Athlon back in 2008, after that when the i7 came out I was on that train until very recently when I picked up a 3700X instead -- so essentially 11 years of Intel for me.
I'm not really attached to either brand just what's currently the best, can run Linux, and whose owning company/companies are behaving themselves (stability/support/development).
Re:AMD laptops have a hidden 10-second performance delay
by antpls on Wednesday November 25, 2020 @02:09AM
(#25206666)
I thought the "race-to-sleep" method was the most energy efficient one? Why adding such delay?
Re:AMD laptops have a hidden 10-second performance delay
by mlyle on Wednesday November 25, 2020 @02:24AM
(#25206730)
With really aggressive boost, the pendulum has swung the other way: the amount of performance per watt when heavily boosted nosedives as voltage to the part is cranked up and leakage increases.
Here this is debunked: https://semiaccurate.com/2020/11/23/intel-looks-at-laptop-po...
Another debunk: https://www.extremetech.com/extreme/317657-intel-is-spreadin...
AMDs own slide on performance tuning OEMs have access to: https://www.extremetech.com/wp-content/uploads/2020/11/AC-vs...