zsero

Member
  • Content Count

    8
  • Joined

  • Last visited

Community Reputation

0 Neutral

About zsero

  • Rank
    Member
  1. But isn't there an actual test for this? I mean all the windows file servers usually do only 1 thing, which is to access multiple files at the same time. I think it must be on our side where we have the problem. And isn't there an actual test instead of this manual HD Speed x multiple instances? I mean there should be a file-server test that just does that. But isn't there an actual test for this? I mean all the windows file servers usually do only 1 thing, which is to access multiple files at the same time. I think it must be on our side where we have the problem. And isn't there an actual test instead of this manual HD Speed x multiple instances? I mean there should be a file-server test that just does that.
  2. Actually to say something positive, I have just tried Win7 for this particular reason, and in experience it was much smoother during intensive write-read tasks on the OS disk, than using XP 64. Using HD Speed it produced the same results, BTW. My actual problem is that under XP 64 my computer freezes for moments or seconds when I do a huge data copy from one hard drive to another. With big files it could get over 110 MB/s in Total Commander, and during this copy my computer is unusable! It freezes the mouse for seconds sometimes! That's why I all started finding why is that happening. Actually I tried a simple seek rate test during these copy operations and it became 1200 ms average, with 2500-3000 ms peeks! On XP 32-bit it was about 25-35 which is normal I think for heavy file copy, but not 2-3 seconds as on XP 64!
  3. update: I have tested the same Samsung hard drive in both Sil3114 and ICH9R. Operation system was different, but as I have seen, all Windows OS-es have the same performance for this test. Results Samsung Spinpont F1 1TB (HD103UJ) Sil3114 PCI card, Win2003: 1 instance: 87 MB/s 2 instance: 74 MB/s total 3 instance: 23.8 MB/s total 4 instance: 14 MB/s total Intel ICH9R AHCI mode, WinXP 64-bit: 1 instance: 112 MB/s 2 instance: 15.6 MB/s total 4 instance: 14 MB/s total
  4. I couldn't manage to screen shot from a remote desktop server, but beleive me, it's 84.7 MB/s single instance and 2x37=74 MB/s with two instances. It has a PCI bottleneck of course, but I could not reproduce any of these performance results under any Intel SATA controller. BTW, I have just tried Win7 on the Intel SATA and it was the same (total 14 MB/s) as XP 64-bit.
  5. I really start to think it is not about hard drives performance, what we are talking about is how buggy Intel SATA controllers are when handling multiple simulateous operation. I have just tried the read test on a $20 Sil3114 PCI controller card and it didn't slow down more than 5-10%! It was with a Samsung HD103UJ!
  6. How good that I found this thread, I am really puzzled about my drives multithreaded performance! I have a new Seagate 7200.12 1TB and an old Samsung HD501LJ. The Seagate has amazing single threaded performance, the Samsung a little less, but when I try just 2 reading threads they almost die. I mean both of them get an average of 7.8 mb/s (or 15.6 mb/s if you say double) and they really lock up your system when you write to the system drive. I am on a X38 chipset based motherboard with ICH9R. I tried all different Intel RAID and AHCI drivers but it's exactly the same. OS is XP 64-bit.
  7. I have just shared my experiences about 5 Seagate 7200.12 1TB drives over at SPCR. http://www.silentpcreview.com/forums/viewt...p=472121#472121 In short 2 out of 5 were bad, one drive had bad sectors as it arrived, an other one had very crazy write caching issues and with the remaining 3 the performance was _really_ different between the drives depending on the firmwares: CC34 firmware -> seek 14.5 ms (and AAM disabled) CC35 firmware -> seek 13.5 ms (and AAM Max Performance) CC44 firmware -> seek 18.5-19.5 ms (and AAM Balanced Performance (208)) I think it's really a nonsense to produce drives with such high failure rates and with so much difference between batches. Just imaging building a performance RAID array from the drives above (35% difference in seek times!) I think that's why the different review sites have different results about the same drive.