Announcement

Collapse
No announcement yet.

Benchmarking - how 'bout a new round...

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

  • #16
    R2J2, is yours 533 or 800? I wonder why mine is at 2997

    Eric Boer
    Dev

    Comment


    • #17
      mines 800, that is strange yours is only 2997 were did that last 2 go?

      Comment


      • #18
        =:-/
        Laurent

        Comment


        • #19
          ahhh geezus my time is pathetic... LOL
          Its a good thing im getting that upgrade

          Athlon 2700+

          Comment


          • #20
            2 x xeon3.6 + HT

            Comment


            • #21
              2 x Opteron 275 (2.2 ghz) on Win x64



              This isn't a bad time for my computer but I spent $ on these 275s because I was expecting gains similar to that graph in reRender's post. My speed gains aren't anywhere near that however... even with the 4GB on x64. I wonder if my machine is hamstrung in someway. Or Vray doesn't use the opterons' power in the same way as scanline. Any ideas?

              Comment


              • #22
                Why are R2J2 and reRenders times 10 minutes apart? Aren't they using the same setup? Is it different amount of cache on the chip or something?

                Comment


                • #23
                  Originally posted by davision
                  Why are R2J2 and reRenders times 10 minutes apart? Aren't they using the same setup? Is it different amount of cache on the chip or something?
                  are you sure they are the exact same processor model? also dont different cpu batches have slightly different yield performance?
                  ____________________________________

                  "Sometimes life leaves a hundred dollar bill on your dresser, and you don't realize until later that it's because it fu**ed you."

                  Comment


                  • #24
                    I don't know how much performance would differ within a specific cpu product line -- but 10 minutes is about 15% in this case. That's seems improbable.

                    Some specs must be different. The readout in the framestamp is quite different, but I don't know what it means.

                    Comment


                    • #25
                      Mine was running pretty hot, it doesn't have very good circulation and room was pretty warm, I'll stick it half outside tonight and try it again.

                      Interesting the 3.6's are different too, it could be like Percy said.
                      Eric Boer
                      Dev

                      Comment


                      • #26
                        Yeah, the 3.6s only seem split by about 1.5 minutes so it's not that much of a discrepancy. Interestingly, the one with less ram was faster. Perhaps he's using faster memory chips or a faster mobo or something.

                        Comment


                        • #27
                          good thing to remember. also im usually running the backburner manager on my machine when rendering animations using the renderfarm, and even though my machine is technically the fastest (dual 3.2 xeons w/ht as opposed to the render farms dual 2.8 xeons) i typically render slower due to cpu cycles being used up by the manager...who knows what other proggies running in memory do the same.
                          ____________________________________

                          "Sometimes life leaves a hundred dollar bill on your dresser, and you don't realize until later that it's because it fu**ed you."

                          Comment


                          • #28
                            R2 and Rerender - are either of you doing anything funny with your BIOS settings? You might also want to check to see if there's any BIOS updates from your mobo manufacturers... The last benchmark we did I discovered my RAM timings were off. A quick 'default' and I was back up to happier times.
                            LunarStudio Architectural Renderings
                            HDRSource HDR & sIBL Libraries
                            Lunarlog - LunarStudio and HDRSource Blog

                            Comment


                            • #29
                              ours are strait from the supplier, so no. the benchmark ran on a free machine while I was working, so there wasn't any other apps running (except the usual crap) that might have been the differance.

                              Comment


                              • #30
                                I kept it cooler and got a little closer. must have been stepping down, now it is at 3000MHz. There are really no settings to mess up in the bios, supermicro is very conservative. I don't have much running to get in the way Windows is idling at 164MB and 0%cpu

                                Eric Boer
                                Dev

                                Comment

                                Working...
                                X