News for hackers. Stuff that matters.

A comment on AMD's Zen 6 Family: Feeding the Beast.

Re:AMD's Zen 6 Family: Feeding the Beast by joebonrichie on Tuesday November 10, 2026 @12:53PM (#50443687)
The instruction level parallelism in the 2-ahead branch prediction is super cool stuff, I know zen5 had this feature but it's kind of crazy that memory is so slow relative to the cores that this complexity is beneficial and worthwhile.
[ Parent | on HN ]
Re:AMD's Zen 6 Family: Feeding the Beast by WithinReason on Tuesday November 10, 2026 @03:38PM (#50445976)
[delayed]
[ Parent | on HN ]
Re:AMD's Zen 6 Family: Feeding the Beast by InsidiousWombat on Tuesday November 10, 2026 @04:22PM (#50446484)
SIMD stuff is also generally pretty big part of the chip, tho of course dwarfed by cache
[ Parent | on HN ]
Re:AMD's Zen 6 Family: Feeding the Beast by stingraycharles on Tuesday November 10, 2026 @11:04PM (#50449627)
It has always been like this, and it’s amazing how branch prediction and speculative execution work. CPUs are so much more complex than the simple mental model you first learn.
[ Parent | on HN ]
Re:AMD's Zen 6 Family: Feeding the Beast by magicalhippo on Wednesday November 11, 2026 @03:01AM (#50450843)
> It has always been like this

Nah. I recall putting multiplication tables in RAM because it was faster to read the tables than doing the multiplication.

RAM used to be relatively fast. These days it's dog slow unless you're streaming.

[ Parent | on HN ]