Announcement

Collapse
No announcement yet.

Intel Core i9 12900K vs. AMD Ryzen 9 5950X On Linux 6.0

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • torsionbar28
    replied
    Originally posted by r1348 View Post
    Well, the explicit purpose of the 12900K was to "one-up" AMD at the cost of horrible inefficiency
    Echoes of the P4 Prescott era. What's old is new again!

    Leave a comment:


  • yump
    replied
    Originally posted by hamishmb View Post

    Fair point. Do we know if Intel's idle or near-idle power usage is better though? It aught to be due to the E-cores, but would still be good to know.
    Almost certainly, but not because of the E-cores. AMD's problems are the I/O die and the mandatory discrete graphics.

    Leave a comment:


  • qarium
    replied
    Originally posted by hamishmb View Post
    So performance per watt is now very similar across both CPUs on the average, which is interesting. Still a bit worse on the Intel chip, but not by a huge amount.
    yes thats very interesting i always thought intels 10nm node is much worst but it looks like the different is small but amd still has a small advantage.

    but what we do not know is what is because of intel ISA war and what is because of the 10nm node.

    Leave a comment:


  • hamishmb
    replied
    Have they? According to the benchmark data here, the efficiency is only marginally worse. Do you have a source that paints a worse picture?

    Leave a comment:


  • r1348
    replied
    Well, the explicit purpose of the 12900K was to "one-up" AMD at the cost of horrible inefficiency, so I think Intel has succeeded there.

    Leave a comment:


  • hamishmb
    replied
    Originally posted by HEL88 View Post

    It's depends. Eg. programmers may fully loaded CPU only for a few minutes per day, but may play games eg. 1h so Intel becomes more effective .
    Fair point. Do we know if Intel's idle or near-idle power usage is better though? It aught to be due to the E-cores, but would still be good to know.

    Leave a comment:


  • otoomet
    replied
    Originally posted by HEL88 View Post

    ...programmers may fully loaded CPU only for a few minutes per day, but may play games eg. 1h so Intel becomes more effective .
    In that case maybe consider something like i5-12400 or ryzen 5600 instead?

    More seriously, I'd love to see the power efficiency data for more mainstream cpu-s. The flagship models are not what most people will get.

    Leave a comment:


  • HEL88
    replied
    Originally posted by hamishmb View Post
    So performance per watt is now very similar across both CPUs on the average, which is interesting. Still a bit worse on the Intel chip, but not by a huge amount.
    It's depends. Eg. programmers may fully loaded CPU only for a few minutes per day, but may play games eg. 1h so Intel becomes more effective .

    Leave a comment:


  • caligula
    replied
    Originally posted by milkylainen View Post
    Efficiency has a new meaning these days.
    Parts of Europe are easily hoovering around $0.5 per kWh on average.
    I doubt that's gonna be a problem. People want max perf and are willing to overclock which can double the cost.

    Leave a comment:


  • Michael
    replied
    Originally posted by JEBjames View Post
    Michael

    typo page 4:

    "to perform better whike​" should be "to perform better while​"

    grammar/typo page 7:

    "across this wide span of benchmarks and many covering many different workloads." should perhaps be "across this wide span of benchmarks and covering many different workloads." (no double many).


    Thanks,

    Leave a comment:

Working...
X