Announcement

Collapse
No announcement yet.

DDR3-800MHz To DDR3-2133MHz Memory Testing With AMD's Kaveri

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

  • #11
    Originally posted by jakubo View Post
    what about timings? dont they have influence on the performance?
    seeing the results i really must wonder why AMD did choose not to go all the way and go for GDDR5 and put some pressure on Intel. Or at least triple-channel...
    The timings don't have much affect since the bottleneck is primarily a bandwidth one rather than a latency one. I remember that it was originally planned to have GDDR5 embedded on the chip but that was canned for the first gen Kaveri -- perhaps due to budget constraints. Instead of triple channel, I don't see why they couldn't just go straight for quad channel.

    Comment


    • #12
      Originally posted by jakubo View Post
      what about timings? dont they have influence on the performance?
      seeing the results i really must wonder why AMD did choose not to go all the way and go for GDDR5 and put some pressure on Intel. Or at least triple-channel...
      Expensive memory modules, I guess.

      Comment


      • #13
        Originally posted by mmstick View Post
        The timings don't have much affect since the bottleneck is primarily a bandwidth one rather than a latency one. I remember that it was originally planned to have GDDR5 embedded on the chip but that was canned for the first gen Kaveri -- perhaps due to budget constraints. Instead of triple channel, I don't see why they couldn't just go straight for quad channel.
        Documents indicate that kaveri may have support for quad-channel and GDDR5 memory, but that it has been disabled. I suppose a different package would be needed for quad-channel, and GDDR5 requires very short traces, so it would likely only work right with BGA systems where everything is soldered down.


        Wow, the GPU tests scale linearly with memory bandwidth available... that means the GPU is very bandwidth starved? Maybe they could have gotten much better performance by tuning the memory controller? Benchmarks indicate that the AMD memory controllers are less efficient than the Intel memory controllers.
        Last edited by grigi; 21 January 2014, 06:58 AM. Reason: wasn't finished

        Comment


        • #14
          Originally posted by Rexilion View Post
          I find it rather disturbing for the fact that the i7 4770k performs almost twice as fast (CPU wise) than the a10-7850. Especially since the i7 runs with 1600mhz memory while the a10-7850 can go much higher.

          It's been 7 years since I last build my computer (I was mostly guided in my choice by someone else tbh). I'm considering to buy an i7 3820 (Ivy-Bridge e, no iGPU ... yay!), but according to cpu-world.com this CPU does not even perform as well as i7 4770 which even has an iGPU. Confusing...
          You're comparing a 300US cpu to a 190US CPU... Gee Gorge can I pet the rabbits. The thing is Benchmarks can lie in Intels favor but the truth is the Kaveri (like other CPUS in it's bloodline) Is pretty memory starved and lack of a L3 hurts it's performance. Does that make it a bad dual module? No it makes it a pretty darn good one just with room for improvement...

          That being said if they make a 4 mod steamroller based FX I would buy that in a heartbeat.

          Comment


          • #15
            Originally posted by nightmarex View Post
            You're comparing a 300US cpu to a 190US CPU... Gee Gorge can I pet the rabbits. The thing is Benchmarks can lie in Intels favor but the truth is the Kaveri (like other CPUS in it's bloodline) Is pretty memory starved and lack of a L3 hurts it's performance. Does that make it a bad dual module? No it makes it a pretty darn good one just with room for improvement...

            That being said if they make a 4 mod steamroller based FX I would buy that in a heartbeat.
            The truth is, Intels are far better in fpu performance and their memory controller is more efficient. They also consume less power.

            AMD is just cheaper/price competitive on low end and with a better igpu, though it is quite bandwidth starved and you don't get to see its real performance.

            Also HSA and Mantle are vaporware, especially on Linux. By the time AMD brings the software that exploits them, Carizzo or its succesor will be available...

            Comment


            • #16
              Originally posted by UraniumDeer View Post
              It's pretty interresting how many of the performance results scales about 1:1 with the increase of memory speed.
              ...
              I did a few calculations on the first test, and while it doesn't remain 1:1, but drops more and more the closer you get to 2133Mhz, the numbers seems rather predictive. I think you could make a pretty accurate mathematical projection of the performance effect of an increase in memory speed, if one was so inclined.



              Next test speed / current test speed * current framerate result = rough estimate of next test
              actual result of next test / rough estimate of next test * 100 = actual performance compared to rough estimate of next test in percentage
              1066/800*17,53=23,36
              23,59/23,36*100=100,1%
              1333/1066*23,59=29,50
              29,19/29,50*100=98,95%
              1600/1333*29,19=35,04
              34,29/35,04*100=97,86%
              1866/1600*34,29=39,99
              39,18/39,99*100=97,97%
              2133/1866*39,18=44,77
              42,50/44,77*100=94,93%

              Comment


              • #17
                Originally posted by Article
                The timings on the system memory were maintained the same at all frequencies so it's solely a comparison of the memory bandwidth as it impacts the A10-7850K Kaveri APU at its stock frequencies.
                You mean you kept the timing artificialy high for the lower frequencies instead of using the profiles? That makes the benchmarks (especialy the CPU ones that care more for latency than bandwidth) complete bulshit.

                Comment


                • #18
                  Originally posted by TemplarGR View Post
                  The truth is, Intels are far better in fpu performance and their memory controller is more efficient. They also consume less power.

                  AMD is just cheaper/price competitive on low end and with a better igpu, though it is quite bandwidth starved and you don't get to see its real performance.

                  Also HSA and Mantle are vaporware, especially on Linux. By the time AMD brings the software that exploits them, Carizzo or its succesor will be available...
                  AMD's Bulldozer CPUs have half the number of FPU per core, so of course that Intel is twice better in floating point operations than AMD.
                  Carizzo will also support HSA and Mantle...

                  Comment


                  • #19
                    Originally posted by tomtomme View Post
                    learn to read the article before posting in the forum ... please! This is going to many forum-posters out there - not only you AJSB...
                    I usually do but was in a hurry going to do some work, so, i simply did a quick look to the charts and didn't read the text

                    Comment


                    • #20
                      if they need short ways for the gddr memory they could place the Memory on the backside of the mainboard. Cases would have to adapt different possibilies of removing memory and/or heat dissipation. You could for instance make some jelly-like pillows on the RAM modules to push against the case without having electrical contact to use the case as a massive heatsink. there are ways. and i think that AMD should make some innovative and risky jumps in order not to lose completely to Intel.

                      Comment

                      Working...
                      X