unregistered

Member
  • Content Count

    137
  • Joined

  • Last visited

Community Reputation

0 Neutral

About unregistered

  • Rank
    Member
  1. unregistered

    SSD drive in PIO 4 mode?

    Some of the older Intel chipsets have the option of SATA or IDE ot Legacy mode for the SATA controllers, make sure you select SATA (in the BIOS)
  2. unregistered

    OCZ Vertex 2 Pro Preview

    Unfortunately, most of the stuff is under non-disclosure. What I can tell you though is that in PCMark Vantage, the Vertex score lower in RAID (Level0) than in single formation because you get additional latencies from the arbitration. That is, if you have a card with a large on-board memory buffer, you can get better performance (some of the HighPoint controllers or higher end LSI versions). Typical scores with the Vertex were around 23k (single drive), 19-20k dual and 16-18k 4-drives (depending on the age and firmware). What is interesting also is that if you configure the cards to run as "RAID" cards and make them bootable OS, you lose the "windows soft cache" and the performance takes a "minor" hit regardless of which benchmrk you are running. I ran a few comparisons also with the Summit drives and they are slower in single drive configurations but marginally faster in multiple drive configurations. I am not sure where this comes from but it seems that the actual burst transfers of the Vertex have a small onset latency (which shows up as scrw-ball result in HDTach for example where the burst speed is lower than the sustained speed ) IPeak.. the problem is that I can't say where I got it from and it is not publicly avalable. I know that some of the original SR staff had a copy, that would probably be the easier way to approach this. I am also working with FutureMark (Oliver has been a good personal friend for over 10 years) and SiSoft (Adrian is pretty perceptive to suggestions) to get benchmarks more consistent and get rid of some of the old stuff that pops up in articles over and again and where the paradigms are so off that the results are not only useless but plain and simply wrong. This is BTW one of the things that I have high hopes for with the new SR. I hope I answered most of your questions, but that's about all I can disclose. BTW, congrats to your Hall of Fame placement. Feels good, does't it :-) I was in the top 10 graphics some 8 years ago with a hand-modded el-cheapo FIC RADEON.
  3. unregistered

    OCZ Vertex 2 Pro Preview

    Actually, I flashed with the RAID firmware, and yes, it can be used as bootable OS RAID in that configuration. I have also used IOMeter, Sandra, Everest, CrystalMark, WB and whatever other benchmarks are out there. I am not a fan of IOMeter because it doesn't reall correlate with real world performance, one of the reasons why Intel has stopped supporting it. PCMark Vantage is an excellent tool, and there is iPeak for selective trace playback as well as SCSI Toolbox.
  4. unregistered

    OCZ Vertex 2 Pro Preview

    Actually, that's not true, with the LSI 3801 and four Vertex drives I easily got over 1 GB/sec sequential reads. The problem was that for example Atto only returns "0" values once you hit 7 digit numbers. With 2 cards I came out at about 540 MB/s.
  5. unregistered

    OCZ Vertex 2 Pro Preview

    Depends on the interface, if you add protocol overhead you saturate the bus on a 4 lane card already with 2 drives and even an 8 lane card will not get you much further on the Core 2 platform because of the FSB bottleneck. PCIX maxes out at around 650 MBs. On the current AMD / Intel Core ix platforms soft-RAID is actually fastest because you use system memory as cache. To get anywhere near that you need pipe burst SRAM on he cards, SDRAM is getting too slow. It is not that easy because you throw off the wear leveling by hitting the drives sequentially instead of across the etire volume. Really? I didn't know
  6. unregistered

    OCZ Vertex 2 Pro Preview

    RAID may not be that important anymore, at least not when you talk about RAID Level 0. With the transfer rates achieved with those drive you pretty much saturate any Core2-based Intel system already because the FSB cannot keep up because of snooping. In the case of nVidia chipsets it is even worse because they don't do overlapped snooping of busmaster transfers so you are effectively limited to some 650 to 800 MB/sec, which is the sustained read transfer rate of 2-3 drives, no matter how wide your interface is (PCIe x8). With the newer systems (x58, P55 or any of the AMD systems) you have more headroom but if you look at those numbers and add arbitration latencies, you may actually negate the bnefits of SSDs unless you are going ultra high-end on the RAID controller (including a large cache).
  7. unregistered

    Storage Review Site Update

    Welcome back! We have done a bit of coverage of SSD technology fundamentals on LostCircuits but Iam glad to see SR back up and running. Let me know if there is anything I can do to help jumpstart this site again.
  8. nVidia appears to have a sizeable edge over ATI in this field, at least from any benchmarking that I have done. There are a few applications in 3dsmax where the ATI cards are faster, however, that is mostly caused by incorrect "ignoring" of e.g. the Omni lighting flag and other imperfections within the ATI drivers. Anything higher than a Quadro FX3000 (which is already hopelessly obsolete) will stomp the FireGL's. Alternatively, there is the 3DLabs series of cards but Creative Labs mutilated them so badly that there was only a vague hint of resemblance to the original 3DLabs specs after the acquisition left. I hear that the new ones are better again but I am not willing to deal with them anymore. Forgot to mention, for any CAD, high fidelity rendering, you will still need OpenGL and that is where the special nVidia drivers make the cards fly.
  9. unregistered

    Meshing

    Thanks everybody, that was helpful. Essentially what it comes down to is that even if you don't fill up a "block" of a RAID setup, a subsequent write would still have to go to a different disk. Almost by definition, this would cause mismatch in the filling level of the different drives and as Gilbo pointed out correctly, keeping track of what is where would be very complex but it is still possible if you think of daisy-chaining with the limitation that every transfer will cause a rotation of the disks in the array. On the other hand, it is possible to define virtual disks that are composed of corresponding blocks of different physical disks .... just rambling on here..
  10. unregistered

    Meshing

    it is sort of an academic question but I came across a description of such a system that sounded somewhat awkward to me but ... I am not the expert and that is why I am posting here to get the feedback of all of you'se
  11. unregistered

    Meshing

    mhhh, nobody ever heard of that? TIA M.-
  12. unregistered

    Meshing

    where meshing is defined similar as striping, however, is set apart from the latter by not using a defined block size but rather writes one complete transfer to one disk and then the next chunk to another disk on a rotating basis. In other words, like a rotating JBOD where every consecutive transfer hits a different physical disk. Regards M.-
  13. unregistered

    What is NCQ ?

    That pretty much hits the nail on the head, there is no performance penalty in a point to point protocol as the one used in SATA. The problem with Legacy Command Queuing was that mostly that the entire queuing scheme was led ad absurdum by the issue that the drive cannot set up its own DMA engine and that the memory addresses had to be supplied together with the data stream after a service command. In other words, the system wouldn't know what hit it until the data started flooding in and that caused enough contention downstream to negate the effects of most queuing. There is still the issue that the IBM drives with Legacy Queuing enabled were faster in quite a few applicatons at their time, on the other hand, there were enough benchmarks to show no benefits ---- even though these benchmarks might have missed the point. SATA uses first party DMA, that is, the drive itself sets up the DMA engine and further it can perform out of order data transfers and even if most benchmarks may not show an immediate benefit, it appears as if there may be some long-time reliability benefits at least
  14. unregistered

    What is NCQ ?

    in a nutshell, the drive itself can make a decision which commands or requests to service first. The best analogy is an elevator, if it services the floors in the order the buttons were pushed, then it does not queue any commands. If it alalyses the necessary halts and schedules them according to what is the most economic and fastest way to sevice the different floors, that's command queuing.
  15. Somehow, I had my hands in that one They are beautiful, though! Full story on the inquirer