Page 1 of 2 12 LastLast
Results 1 to 15 of 18
  1. #1
    Joined
    May 2002
    Posts
    178

    IBM deskstar 73.5 Gbyte LOW PERFORMANCE

    i tested my ibm disk with sandra 2001 professional and it has a low score compared to the other udma 100 disks.
    14800
    15300 with windows cache

    general ata100 7200 rpm disks have 24000!!!!

    i can verify low performance during normal windows operations like loading games, mooving files, defragmenting and during installations!

    My athlon 2000 runs with an elephant on its shoulders!!

    is there something to do??

    thanks in advance
    8RGA
    Barton 2500+
    Verax P14Cu 20 Db cooler
    2 x 256 corsair 3200 xms cl2
    2 x WD80JB 80 Gbytes
    sony DW U14a
    liteon 52x32x52 cdrw
    Audigy 2 platinum ex
    win xp professional sp2

  2. #2
    Joined
    May 2002
    Posts
    178
    yes it is a new system and before making the test i defragmented the disk....
    anyway: my second disk is a simple quantum fireball 20GB ata100. Comparing the 2 disks i noticed speed difference... quantum fireball is more " brilliant" can you understand what i mean?
    exploring folders, playing games, decompressing, file opening.... in all of these operations there is a slight difference between the 2 disks...
    instead the transfer rate is more or less equal, perhaps 35/36 MBytes/sec

    the point is that i was told that ibm works better and faster...
    maybe it is not true but i though the score gap in sandra tests was too high

    anyway i will try the program you told me
    thaks
    leo
    8RGA
    Barton 2500+
    Verax P14Cu 20 Db cooler
    2 x 256 corsair 3200 xms cl2
    2 x WD80JB 80 Gbytes
    sony DW U14a
    liteon 52x32x52 cdrw
    Audigy 2 platinum ex
    win xp professional sp2

  3. #3
    Joined
    May 2002
    Location
    New jersey
    Posts
    271
    Do you have any other drives (CD etc ) on the same IDE port?

    Could you give a little more information how drives are set up and connected to MOBO?

    If you have two devices on same IDE port then IDE controller will default all settings for that port to slowest device.
    Mr. Spock, how do I set my laser printer to stun?
    -------------------------------------------------------------------
    Box 1 Full Tower (2 80mm & 2 120 mm fans)--MSI P6N Plat--Intel E6700--Zalman 9500--G.Skill 6400PHU2-2GB--OCZ 700W--XFX-GF7600GT--WD Rapture 150 GB--2 Samsung 500 GB--optical Pioneer & LG

    Box2 Lian Li 70 series (6 120mm fans)--Abit IP35 Pro--Intel E8400 @3.3G--Zalman 9700--8 G memory G Skill F2-6400L4--GeForce 8600 GT--PC Power & Cooling 750 Quad--WD Raptor 150--2) WD 1 TB--1.5 TB Samsung--Asus & Pioneer Opticals

    Box3 Lian Li 70 serics (6 120mm fans)--ASRock X58 Extreeme---Intel i7 920 @3.6G--Cooler Master V8 (mod with 120mm silverstone fan)--12 G memery G Skill--G Force GTX 260--PC Power & Cooling 750 QB--WD 2.5 Raptor 300--2) Samsung 1.5 TB--2) WD 1 TB--Plextor & LiteOn optical

  4. #4
    Joined
    May 2002
    Posts
    178
    i tried to unplug all the drives exept this disk and its performances have remained the same...
    now i have a yamaha 24x10x40 and a standard 54x cdrom ( don't even know the brand )... i tried all possible configurations and nothing changed.... i can at least say my mobo is stable!

    leo
    8RGA
    Barton 2500+
    Verax P14Cu 20 Db cooler
    2 x 256 corsair 3200 xms cl2
    2 x WD80JB 80 Gbytes
    sony DW U14a
    liteon 52x32x52 cdrw
    Audigy 2 platinum ex
    win xp professional sp2

  5. #5
    Joined
    May 2002
    Location
    New jersey
    Posts
    271
    My understanding is that the Sandra scores are is about 15000 for IBM and 24000 for quantum?


    If drives bench mark in the 35 MB /sec this soundes about normal.
    Subtained transfer rates depend on platter rotation speeed more than anything else. The IBM has faster access time specs (the time to get the head to were it needs to be so it can read or write). Therefore open small files (folders, OS swap file, etc) should be faster on IBM due to faster access tiem.



    Run utility to completely check drive. (One that non-destruct checks clusters integrety)



    Have seen posts about some IBM's having bad clustor problems.

    If drive still have problems may have to clear CMOS?
    Mr. Spock, how do I set my laser printer to stun?
    -------------------------------------------------------------------
    Box 1 Full Tower (2 80mm & 2 120 mm fans)--MSI P6N Plat--Intel E6700--Zalman 9500--G.Skill 6400PHU2-2GB--OCZ 700W--XFX-GF7600GT--WD Rapture 150 GB--2 Samsung 500 GB--optical Pioneer & LG

    Box2 Lian Li 70 series (6 120mm fans)--Abit IP35 Pro--Intel E8400 @3.3G--Zalman 9700--8 G memory G Skill F2-6400L4--GeForce 8600 GT--PC Power & Cooling 750 Quad--WD Raptor 150--2) WD 1 TB--1.5 TB Samsung--Asus & Pioneer Opticals

    Box3 Lian Li 70 serics (6 120mm fans)--ASRock X58 Extreeme---Intel i7 920 @3.6G--Cooler Master V8 (mod with 120mm silverstone fan)--12 G memery G Skill--G Force GTX 260--PC Power & Cooling 750 QB--WD 2.5 Raptor 300--2) Samsung 1.5 TB--2) WD 1 TB--Plextor & LiteOn optical

  6. #6
    Joined
    May 2002
    Posts
    178
    i'm sorry i didn't explain my self correcty..

    in sandra test there is a referment score for ata100 7200rpm drives of 24000 on a 30 GB hard disk.

    I repeated the test many times with my IBM 73.5 GB and the score goes from 11.500 to 18.000.... last time i said different values because they where the ones i had at that moment...

    average access time goes from 8 to 10 msec without windows caching and 6 ms with windows caching
    instead my quantum 20GB disk gives always more then 18000 and has an average access time of 8 ms, and 4 ms with windows disk caching...

    i'm not sure about sandra's reliability on this tests but i'm sure i notice my quantum is a little bit faster and should be slower!
    Last edited by sensehi; 06-06-2002 at 04:51 PM.
    8RGA
    Barton 2500+
    Verax P14Cu 20 Db cooler
    2 x 256 corsair 3200 xms cl2
    2 x WD80JB 80 Gbytes
    sony DW U14a
    liteon 52x32x52 cdrw
    Audigy 2 platinum ex
    win xp professional sp2

  7. #7
    Joined
    May 2002
    Location
    New jersey
    Posts
    271
    Starting to sound like that IBM has problem. You may want to RMA drive and get tested or replace.
    Mr. Spock, how do I set my laser printer to stun?
    -------------------------------------------------------------------
    Box 1 Full Tower (2 80mm & 2 120 mm fans)--MSI P6N Plat--Intel E6700--Zalman 9500--G.Skill 6400PHU2-2GB--OCZ 700W--XFX-GF7600GT--WD Rapture 150 GB--2 Samsung 500 GB--optical Pioneer & LG

    Box2 Lian Li 70 series (6 120mm fans)--Abit IP35 Pro--Intel E8400 @3.3G--Zalman 9700--8 G memory G Skill F2-6400L4--GeForce 8600 GT--PC Power & Cooling 750 Quad--WD Raptor 150--2) WD 1 TB--1.5 TB Samsung--Asus & Pioneer Opticals

    Box3 Lian Li 70 serics (6 120mm fans)--ASRock X58 Extreeme---Intel i7 920 @3.6G--Cooler Master V8 (mod with 120mm silverstone fan)--12 G memery G Skill--G Force GTX 260--PC Power & Cooling 750 QB--WD 2.5 Raptor 300--2) Samsung 1.5 TB--2) WD 1 TB--Plextor & LiteOn optical

  8. #8
    Joined
    Jul 2001
    Location
    Bedford, England
    Posts
    458
    The IBM drives have an 'Acoustic Management' feature, whereby you can adjust them according to whether you want maximum performance or minimum drive noise - you can download it here
    There is a User Guide here
    and a list of supported drives here
    The IBM Feature Tool allows you to control some of the features in the newer IBM high performance ATA hard disk drives. The IBM Feature Tool allows you to:
    • Enable or disable the read-ahead or write cache
    • Change the drive Automatic Acoustic Management settings to the:
    • Lowest acoustic emanation setting (Quiet Seek Mode), or
    • Maximum performance level (Normal Seek Mode).


    This is worth trying, just to be sure that all is set as it should be.

    MSI Neo2 Platinum nForce3 Ultra S939
    Athlon 64 3500+ 90nm @2607 (11x237)
    Thermalright XP-90 + Papst 92mm 55cfm
    4x1GB Crucial PC3200
    Raptor 74GB SATA system drive + WD800JB data + Highpoint RocketRAID 100 (2x 80GB WD800JB RAID-1) + Onboard SATA (2x 120GB Maxtor DM9+ SATA RAID-0) + Maxtor Maxline Plus II 250GB backup drive
    Radeon X800 Pro (16 pipes) 500/500MHz
    Win XP SP2

    Part-time SETI crunchers: Athlon XP 2400+; Athlon XP 2100+; Athlon XP Barton 2500+@2.0GHz; Pentium-M 745

  9. #9
    Joined
    May 2002
    Posts
    178
    thanks but the default setting is set for maximum performance and maximum noise, i already had those utilities....

    leo
    8RGA
    Barton 2500+
    Verax P14Cu 20 Db cooler
    2 x 256 corsair 3200 xms cl2
    2 x WD80JB 80 Gbytes
    sony DW U14a
    liteon 52x32x52 cdrw
    Audigy 2 platinum ex
    win xp professional sp2

  10. #10
    Joined
    Aug 2001
    Location
    Stroud UK
    Age
    75
    Posts
    565
    Sensehi, Sandra 2001 or 2002 will never give a proper benchmark for any disks. The program is "upset" by fragmentation/amount of data on disk/OS etc. Download HDTach 2.61 (free) and you will get a totally reliable disk benchmark.

  11. #11
    Joined
    May 2002
    Posts
    178
    thanks a lot

    my test result with hd tach 2.61 are:
    IBM 73.5 GB 7200 RPM ata100
    access average time: 12.5
    read speed: 38.8 MB max, 9 MB min, 28 average

    quantum fireball 20 GB 7200 RPM ata100
    access average time: 12.6
    read speed: 37.2 MB max, minimum 19, average 30

    i though ibm disks where faster then my " cheap" quantum fireball!

    how can it be possible??

    sensehi
    Last edited by sensehi; 06-08-2002 at 07:37 AM.
    8RGA
    Barton 2500+
    Verax P14Cu 20 Db cooler
    2 x 256 corsair 3200 xms cl2
    2 x WD80JB 80 Gbytes
    sony DW U14a
    liteon 52x32x52 cdrw
    Audigy 2 platinum ex
    win xp professional sp2

  12. #12
    Joined
    Aug 2001
    Location
    Stroud UK
    Age
    75
    Posts
    565
    Sensehi, I don't think that the IBM disk access times look too good. I was expecting to see about 9m/secs average. Would you like to post the model type of the disk.

  13. #13
    Sensehi,

    WHich ibm you got there, Im an ibm fanatic, Ive got 120gxps, 60 gxps and 75's All trouble free by the way.

    those #s are low.
    have your checked the what dma there running in XP, if xp has read errors it will kick the dma down a notch, eventually leaving you at superslow

    Jungleman I think hdtach does count the average seek time differently, than say sandra does.

  14. #14
    Joined
    May 2002
    Posts
    178
    it is an IBM DTLA 307075 ( deskstar gxp 75)
    it should be a 8.5 ms seek time disk but hdtach gives always 12.5 or even more!
    the only thing that surprises me is that my quantum fireball 20GB is a little bit faster then it but should be slower and costs much less then ibm!

    the udma mode is correctly set to 5...
    thanks

    senshei
    8RGA
    Barton 2500+
    Verax P14Cu 20 Db cooler
    2 x 256 corsair 3200 xms cl2
    2 x WD80JB 80 Gbytes
    sony DW U14a
    liteon 52x32x52 cdrw
    Audigy 2 platinum ex
    win xp professional sp2

  15. #15
    Joined
    Aug 2001
    Location
    Stroud UK
    Age
    75
    Posts
    565
    Sensehi, my GXP75 30Gb disk gets an access time of 8.9m/secs with HDTach 2.61 and the GXP120 gets 8.3, and that's about what the specs say I should expect.

    I would backup that disk and run the IBM Drive Fitness Test and re-write the surfaces. Then select the advanced surface test. If all is well then re-try HDTach 2.61 before doing anything else. If it's still longer than 9m/secs RMA it.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •