Announcement

Collapse
No announcement yet.

Running The Ryzen 7 1700 At 4.0GHz On Linux

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

  • #41
    What memory frequency was it running thru these tests? (in one picture i saw 2133MHz)
    Have you had any luck at higher memory frequency?
    https://www.gskill.com/en/press/view...-for-amd-ryzen
    Last edited by Nille_kungen; 06 March 2017, 05:16 AM.

    Comment


    • #42
      Originally posted by LeJimster View Post
      Until somebody figures out why some of the gaming performance is *much* worse than on windows with some titles and vulkan (that confuses me), I will be watching and waiting. Still leaning towards buying a Ryzen system as it hits the ball out of the park on everything other than gaming right now.
      ZOMG Ryzen has a few less FPS on a few games on Windows when running at settings NOBODY will run it at, despite it having the same framerate at real world settings, OMG It's the end of the world as we know it!!!!11!!1!!!!!!11!

      Seriously everyone needs to stop this nonsense. Ryzen isn't bad for gaming, and for 1080p/60fps gaming it works just fine, and in fact other than a few games released in the last 2/3 years my A8-3850 (Athlon II x4 2.9GHz equivalent) was more than adequate to run 1080p/60fps games. In my personal testing of a Ryzen 1700 at stock clocks, there's nothing that causes it problems in Windows gaming, and Doom using the Vulkan render showcases what the future holds. In non-particle-effect zones the in-game diagnostic was reporting average of 12 thread usage with 30-50% load per thread, in particle effect zones it would jump up to the full 16 threads but still around half or less utilization per thread. Over the next year or two most games are going to head there, and for current games Ryzen is adequate. Now... in some current games can the i7-7700k get more FPS than the Ryzen 1700-1800x? oh absolutely, and for 144Hz gamers there may be a point, but... that's a generalized argument for using modern quad cores instead of octocores (note that the Intel equivalents perform about as bad in the same situation), and the quad core Ryzen parts are coming, which not having to deal with being octocores should be able to clock up to that 4.5GHz range to compete with the i7-7700k, and if they can get them to turbo to 4.7-4.8 or so then they'll get the stock clocks gaming performance crown.

      Once again though, everything else is heading to where Doom is today (Which can take advantage of 12 Threads in some lighter zones, and 16 threads in the heavier particlely zones), so the glamour of quadcores is going to rapidly fade in the near future, just like the glamour of dual cores faded.

      Comment


      • #43
        Nice benchmarks. Performance per watt figures would have made this article a wet dream.

        Comment


        • #44
          Originally posted by LeJimster View Post
          Until somebody figures out why some of the gaming performance is *much* worse than on windows with some titles and vulkan (that confuses me), I will be watching and waiting. Still leaning towards buying a Ryzen system as it hits the ball out of the park on everything other than gaming right now.
          scheduling issue, Windows moving threads around on the cores, and the resulting data transfers thru cache transfer between the two cpu complexes, ...

          Comment


          • #45

            Ryzen 1700 at stock with R9 285, 3000MHz (2933MHz currently) DDR4 CAS 15 at 1080p at max settings (with Motion Blur and Depth of Field turned off)

            Comment


            • #46
              Michael, so I guess now it's time to test the 1800X again, this time with maximum overclocking?

              Comment


              • #47
                Originally posted by Azpegath View Post
                Michael, so I guess now it's time to test the 1800X again, this time with maximum overclocking?
                I really wouldn't recommend this when you don't have any runtime temperature monitoring so the only way to detect an overheating is when the screen goes blank and a plume of smoke rises from the motherboard.

                Comment


                • #48
                  Originally posted by L_A_G View Post
                  I really wouldn't recommend this when you don't have any runtime temperature monitoring so the only way to detect an overheating is when the screen goes blank and a plume of smoke rises from the motherboard.
                  Well, even though there is always a risk, I'm fairly certain that the motherboard and the CPU will switch off a while before that, based on hardware fail-safe switches.

                  Comment


                  • #49
                    Originally posted by Azpegath View Post
                    Well, even though there is always a risk, I'm fairly certain that the motherboard and the CPU will switch off a while before that, based on hardware fail-safe switches.
                    While these switches do exist, they're not completely failure safe and components can fail before they run up against the limits set by them. People who do overclocking as a sort of sport blow boards and chips all the time, specially when they do the hotter air cooled runs without stuff like liquid nitrogen or CO2 ice to keep things at freezing temperatures.

                    Comment


                    • #50
                      Originally posted by L_A_G View Post
                      While these switches do exist, they're not completely failure safe and components can fail before they run up against the limits set by them. People who do overclocking as a sort of sport blow boards and chips all the time, specially when they do the hotter air cooled runs without stuff like liquid nitrogen or CO2 ice to keep things at freezing temperatures.
                      You are right, people blow shit up sometimes

                      However, when I said Maximum Overclocking I meant "maximum safe overclocking". Since Michael could crank up the 1700 to 4 Ghz, I guess the 1800X should be able to go a bit higher. I could have been more clear in my statement.

                      Comment

                      Working...
                      X