androidi

Member
  • Content Count

    111
  • Joined

  • Last visited

Community Reputation

0 Neutral

About androidi

  • Rank
    Member
  1. czr Which motherboard/chipset and hdd controller were you using in these tests of 7200.11 if you remember?
  2. androidi

    Something Wrong With This Equation!

    You're lucky I noticed this because you can't expect serious responses with subject like that. "NCQ performance" would have been more appropriate subject. There's earlier threads on the same issue: http://forums.storagereview.net/index.php?showtopic=26006 See the first post there and my reply on 2nd page. I have nearly identical configuration except I tried NCQ on ICH9R with Samsung F1 1 TB, WD3200KS, Maxtor 6V300F0 and the results seem to tell that either the drives or the Intel driver is not implementing the NCQ. If you think the theory of how NCQ works on paper it's really ideal for multi-tasking in single user environment, however all the benchmarks from the years show it degrades performance. And the benchmarks are right. In my test enabling AHCI clearly made performance go down even in multi-tasking tests compared to no AHCI/NCQ. Given the thread I linked above says there is a consumer price drive that finally implements NCQ (outside of just mentioning 'support' for it in feature overview), I hope we see some respectable sites pick up this aspect and run some real world multi tasking tests that easily show the benefits so at some day the rest of the manufacturers actually implement it too.
  3. Samsung Spinpoint F1 HD103UJ (1TB) 12/2007 Rom Rev 1AA01107 with ICH9R AHCI in Vista x64 Q6600 4GB Gigabyte DS2R. 1 instance 100-110 MB/s 2 instances 7-10 MB/s per instance. (Total < 20 MB/s) http://www.samsung.com/global/business/hdd...&ppmi=1155# Features • SATA Native Command Queuing Feature I guess this translates to "we promise no data will be corrupted if you attempt to use NCQ but the support was hacked in there by a summer intern in 2 days." /sarcasm. I'll try to do test in 32bit 2003 server if I have time to see if there's any difference.
  4. My Samsung F1 Experience, to be continued as I'm going to return the current drive given the more severe errors on it! I ask you to look at the Size and Rom Rev and date on the top of the drive as they seem to differ.. HD103UJ (1TB) on ICH9R ide mode Drive #1 12/2007 Rom Rev 1AA01106, Size 953869MB LBA 1953525168 Drive #2 12/2007 Rom Rev 1AA01107, Size 953868MB LBA 1953523055 (Options/Drive info NATIVE SIZE: Size 953869MB LBA 1953525168) #1,2 Check M.C. Error : ECC error #1 ~30 Ecc errors in Full Read surface scan only, no other errors. #2 First test run, different results from later runs: Check M.C. Error : Undefined error Simple Surface Scan Error : Undefined error #2 Later tests, no more Undefined's, now it's the Ecc error for above. #2 Full Read surface scan: Sector errors at around 350 MB into the drive. Sector ID not found.. #2 seeks are slighty more noisy than #1. #1 vibrated more. Ecc error is less annoying than Sector error in one way atleast: When the drive hits the Sector error sector it really slows down during the scan so it takes many seconds to get to next sector. Given Maxtors have had SMART "hardware Ecc errors" as long as I remember maybe they are not so serious but I would def. return any drives with Sector error or the drive size less than Native size (indicates there's ton of bad sectors already if you ask me).
  5. androidi

    WMI Guru needed

    Experts-exchange might be a better place if that is all you got to go with. Or MS product support.
  6. Well too late now, but next time you get a 'new drive', first dump output from smartmontools to a file and then before going to disk management open WinHex and quickly browse through the physical volume to see it is full of 00. If all looks good then go to disk management and init the drive.
  7. androidi

    Recovering lost partition

    If all else fails, fixing a partition table is simple with WinHex (atleast after the first success).. If you know the drive was filled by single partition, it will avoid the need to scan through the disk. The values to enter to the partition table entry are pretty much the same as in any large ide hdd with the exception of the "sectors in partition". I don't remember right now but I think it was (total sectors in drive - empty sectors at end - 63 or 64).
  8. androidi

    Expresscard and SATA II?

    With the difference that I have a CardBus, not ExpressCard, in my laptop; I've been running like this with a CardBus SATA adapter for a while now. Like said above, the power is the issue. However in my case I had earlier bought another adapter (PATA-USB) than came with its own external power supply that has a molex plug. So by plugging a molex-sata power adapter to that I can run one 3.5" SATA drive on the laptop now. I doubt the power adapter is designed for powering two at once so I am not going to attempt that. The thing works fine, but a note about these molex-satapower converters: I've tried a few and they seem to have a problem where a tiny movement of the cable can cause the drive to power cycle! I've seen this with other internal molex connectors as well, touching the cables is not a good idea. As for booting from the external drive. It might work with ExpressCard based adapters, but I believe it would still require some level of support from BIOS. As for my CardBus/PCMCIA - it is very rare that a computer would support booting from these.
  9. androidi

    Removable ESATA drives - how do they work?

    And there are no "eSATA" DRIVES that I know about at the moment. More like ordinary drive inside an enclosure with the eSATA type of port. It seems to means mostly that at some end of the cable there is a new connector that has some grounding thing and is tested for more use. And that the cables with this new connector cost 5-10 times more. If your enclosure says eSATA then it likely requires this type of new cable but unless you also have controller with external eSATA port on it, you need a cable than is plain SATA in the other end and eSATA in the other.
  10. androidi

    Removable ESATA drives - how do they work?

    1) They do, but it is best to make sure nothing has been written to the drive in past few seconds before powering it down. I think this is even if the drives are "optimize for quick removal", but not sure about this. For peace of mind you could go through the Device Manager and Uninstall the drive you plan to unplug/power down, however whether there is actual benefit to this, I am not certain. 2) When you first insert a zeroed drive (factory formatted) and run disk management gui, it will create a GUID for the drive. You can then use the same diskmgmt.msc to assign/modify the drive letters or mount the drive under another drives empty folder (re-parse point). Upon later inserts, the GUID is matched and the drive always gets the same letter/mount point in that computer regardless of the controller it is attached to. 3) No and in my experience the drives never even show up there, atleast on the controllers I use. If you modify the drive to "optimize for performance" then you better do the procedures... 4) If the backup program support backing up to particular drive letter/location then I don't see why it could not work. I would also take a look of the procedure of restoring from backups and how it actually works out with this multidrive config. A note about multibay enclosures: If you plan to have more than 1 drive running in there constantly for hours then it is going to be quite noisy probably. The drives resonate the case likely and some of these have so loud fans that you can hear the 50 m away. Also the Port Multiplier (1 SATA cable = 5 drives externally) based solutions require that your controller supports that and AFAIK there are only a handful that do since this is still quite new thing.
  11. androidi

    Test on Hitachi 7K80

    a) are you actually planning to use 4x raid 0 for a purpose that will have measurable benefit from it? in most applications (non synthetic) you'd probably get equal or close to actual performance with 2x 300 GB DM10 running mirrored raid 1 and you could actually use the 300 GB to store some useful data instead of worrying when one of the drives in your raid 0 set fail. if single drive transfer/burst rate seems slower than 4x raid 0 (per 4) then it is obviously a driver issue when not using raid. http://gadgets.qj.net/Wall-O-Monitors-proj...pg/49/aid/21281 Of course if you, like the guy in that link, have money to spend on quite silly things, then why, go ahead! PS. I have data from 20 years ago that has gone through many HDDs, none raid though. In same time I've seen countless datalosses of many terabytes where that data was on raid set. I've also measured app perf RAID 0 vs no raid and while the 200+ MB/s looks nice, the real performance comes by using profiling to determine where most time is spent and unless your app is frequently doing sequential non buffered reads of gigs then the overall performance of the app is very unlikely to measurably increase from STR of 200+ vs the top performing single drive.
  12. androidi

    Vibration of current 250 GB drives

    Brands: In a similar pursuit of silent drive, I bought recently 250 GB Samsung, 250 GB Hitachi, 300 GB Maxtor DM10, 320 GB WD3200KS. I returned the February 2006 Samsung since while it didn't vibrate too much, it had other annoying noise issue. The ranking for vibration (least to worst): WD/Samsung, Hitachi, Maxtor. Maxtor has nice performance figures compared to the others in sequential access. (1 or 2 simult.) WD is my primary drive right now. Obviously most of the 2.5" vibrate considerably less.
  13. androidi

    Vibration of current 250 GB drives

    Your situation is all too familiar. I made a similar move couple years ago with Serial ATA drives supposed to sit in a 4x3.5" hotplug backplane. Well anything more than 1 drive in it running easily makes considerable noise from the case-resonance inducing vibration. I do not have a special "silent case" and since I swap the drives from the desktop to laptop (SATA CardBus adapter) on occasion, the silent solution for me was to get few of these: http://www.akasa.co.uk/akasa_english/spec_...ata_ex_bluv.htm And have only the least vibrating or infrequently powered drives in the drive cage and rest of the drives are outside the computer. When drives sit atop of few VHS casettes for example, the vibration is dampened enough to make Maxtor quite inaudible even though it is a 3 platter heavy weight vibrator when put to the aluminum cage. So I can have like 4-5 drives running and still have a silent computer with this external solution. The drive temperatures stick at around 44-46 celsius with no cooling when held above the "dampening" vhs casettes. I do not stack them though, so this takes some space and they gather dust too and are always visible.. But I can deal with the dust and the fact they are visible since they are still more silent with 30 cm away than what they would be when put in the case/cage and being few meters away.
  14. Device Model: HDT722525DLAT80 Firmware Version: V44OA96A User Capacity: 250,059,350,016 bytes Just bought and plugged in first time: (Interestingly it shows that someone had started it up once earlier, shouldn't the start stop be 0??) SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x000b 100 100 016 Pre-fail Always - 0 2 Throughput_Performance 0x0005 100 100 050 Pre-fail Offline - 0 3 Spin_Up_Time 0x0007 100 100 024 Pre-fail Always - 0 4 Start_Stop_Count 0x0012 100 100 000 Old_age Always - 2 5 Reallocated_Sector_Ct 0x0033 100 100 005 Pre-fail Always - 0 7 Seek_Error_Rate 0x000b 100 100 067 Pre-fail Always - 0 8 Seek_Time_Performance 0x0005 100 100 020 Pre-fail Offline - 0 9 Power_On_Hours 0x0012 100 100 000 Old_age Always - 0 10 Spin_Retry_Count 0x0013 100 100 060 Pre-fail Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 2 192 Power-Off_Retract_Count 0x0032 100 100 050 Old_age Always - 2 193 Load_Cycle_Count 0x0012 100 100 050 Old_age Always - 2 194 Temperature_Celsius 0x0002 196 196 000 Old_age Always - 28 (Lifetime Min/Max 24/28) 196 Reallocated_Event_Count 0x0032 100 100 000 Old_age Always - 0 197 Current_Pending_Sector 0x0022 100 100 000 Old_age Always - 0 198 Offline_Uncorrectable 0x0008 100 100 000 Old_age Offline - 0 Before going sleep I ran hdtach long test but did not check smart values. After I woke up I powered it again and wrote every sector full with WinHex... Then ran smartctl again: SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x000b 099 099 016 Pre-fail Always - 65537 2 Throughput_Performance 0x0005 100 100 050 Pre-fail Offline - 0 3 Spin_Up_Time 0x0007 100 100 024 Pre-fail Always - 0 4 Start_Stop_Count 0x0012 100 100 000 Old_age Always - 3 5 Reallocated_Sector_Ct 0x0033 100 100 005 Pre-fail Always - 0 7 Seek_Error_Rate 0x000b 100 100 067 Pre-fail Always - 0 8 Seek_Time_Performance 0x0005 100 100 020 Pre-fail Offline - 0 9 Power_On_Hours 0x0012 100 100 000 Old_age Always - 2 10 Spin_Retry_Count 0x0013 100 100 060 Pre-fail Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 3 192 Power-Off_Retract_Count 0x0032 100 100 050 Old_age Always - 3 193 Load_Cycle_Count 0x0012 100 100 050 Old_age Always - 3 194 Temperature_Celsius 0x0002 125 125 000 Old_age Always - 44 (Lifetime Min/Max 22/46) 196 Reallocated_Event_Count 0x0032 100 100 000 Old_age Always - 0 197 Current_Pending_Sector 0x0022 100 100 000 Old_age Always - 0 198 Offline_Uncorrectable 0x0008 100 100 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x000a 200 200 000 Old_age Always - 0 1 Raw_Read_Error_Rate 0x000b 099 099 016 Pre-fail Always - 65537 What??? After this I ran HDtach again for fun and what do you know: 1 Raw_Read_Error_Rate 0x000b 100 100 016 Pre-fail Always - 0 !?!?!? I thought the Pre-fail stuff stays put once it decreases, but apparently not? I read that these values are somehow normalized or something, is that what is happening here? Can someone explain thoroughly please?
  15. Your seem different from the unit I had though. My unit did not have the vibration-causing issue, but more of a steady whine that had 2 states: no whine and whine. Also my recording of the noise confirms that it wasn't a sound that shifted in/out from hearable range - this would have been evident in the recording. However with the sample size of couple units I'd not go round making too much conclusions.