Different devices on an IDE channel.

Discussion in 'NZ Computing' started by ~misfit~, Feb 3, 2004.

  1. ~misfit~

    ~misfit~ Guest

    I know nobody here is *that* stupid but I hang in alt.comp.hardware and
    there's a couple of guys in there who keep telling people not to keep their
    old HDD when upgrading. Some people post saying "Can I use my old drive for
    mp3s" or suchlike. I've had a long-running argument with these guys who
    claim that an IDE channel only runs as fast as the slowest device attached
    to it.

    I did an experiment this afternoon and thought I'd copy the post here. Just
    in case anyone is interested. The "Why? (Loaded question)" was my first
    foray into the thread. I've repeated over and over to these guys (mainly a
    guy called Stacey, who insists he's right) that they're talking shit.

    ~misfit~ wrote:
    > kony wrote:
    >> On Mon, 2 Feb 2004 23:19:27 +1300, "~misfit~" <~misfit~@his_desk.com>
    >> wrote:
    >>
    >>>>>> If it's not a UDMA drive, (i.e. it's an old PIO drive) don't
    >>>>>> slave it to the fast drive.
    >>>>>
    >>>>> Why? (loaded question).
    >>>>>
    >>>>
    >>>> Try it and see. Hook a 400MB vintage PIO mode drive up with an
    >>>> ata33> UDMA drive and post the benchmarks.
    >>>
    >>> And can you reccomend a small program to use to benchmark? I have a
    >>> Seagate ATA100 in my PC at the moment, on it's own on an IDE
    >>> channel, running ultra DMA mode 5, and a 212MB drive in the junk
    >>> drawer.
    >>>
    >>> Lets do this. Gotta be a small proggy, I'm on dial-up with a crap
    >>> connection.

    >>
    >> Atto is about as small as it gets, but doesn't show CPU utilization.
    >> Perhaps that doesn't matter on XP, you still have a built-in monitor.
    >>
    >> http://69.36.189.159/usr_1034/26k.zip

    >
    > Thanks, I'll have a play a bit later and post results. I have MBM5
    > set to show CPU utilisation in my tray anyway, although it's always
    > showing 100% as I run SETI CLI. Might turn it off for this.


    Ok, this will take a while, I ran some tests and wrote out the results by
    hand, I don't know how it will come out format-wise, maybe use fixed text
    setting. I fitted a Quantum LPS170A 170MB HDD (couldn't find the 212MB I
    mentioned earlier) as a slave to my 80GB Seagate and ran ATTO Disk
    Benchmark, at default settings. Transfer size was 0.5Kb through to 1024Kb,
    total length (whatever that means) was 4Mb. The old Quantum would run at
    Multi-Word DMA mode 1 but I limited it to PIO as that is what we were
    discussing. I have heard people say that the IDE bus is limited to the speed
    of the slowest drive and have always argued against this where modern mobos
    are concerned. Here are the results, I don't know how it will look though.
    There are seven columns, the first is transfer size, the second and third
    are write and read for my 80GB alone on the channel. the fourth and fifth
    are for the same drive with the 170MB as slave and the sixth and seventh are
    for the slave drive:

    0.5 87 108 75 78 79 58
    1.0 175 159 156 147 113 127
    2.0 324 327 270 302 211 210
    4.0 678 601 830 582 421 512
    8.0 1221 1168 1114 913 984 855
    16 3076 2318 1815 2308 1657 1677
    32 3994 4862 3423 4122 1946 1992
    64 7932 9274 9845 8966 1959 1946
    128 18459 17306 16543 15570 1959 1992
    256 29523 28714 26277 28100 1866 1994
    512 41020 39160 40590 36384 1984 2000
    1024 41323 42757 38130 44422 1996 1994

    So there you have it. It seems that, with a PIO drive as slave, the ATA100
    drive is a little slower in general but nowhere near as slow as the slowest
    device on the channel. It is my contention that any other drive on the IDE
    channel will slow it down slightly. Unfortunately I don't have a spare
    ATA100 drive to throw in as slave to test it without pulling the missus' PC
    apart, something I'm not prepared to do at this stage.

    I thought about doing several runs on each and averaging out the results but
    this has already taken a chunk of my day. In a few tests, as you can see
    above, the ATA100 drive was actually faster with a PIO slave. You can see
    above that the PIO dive tops out at below 2MB/sec while the ATA100 gets
    close to 40MB/sec in some tests. Even with a PIO slave.

    The ATA100 drive is formatted NTFS, the 170MB drive was already formatted
    FAT32 so I left it as it was. The motherboard is a Soltek SL 75FRN2-L.
    nForce2 Ultra 400 with ATA100 controllers. OS is WinXP Pro.

    What say you Stacey?
    --
    ~misfit~
     
    ~misfit~, Feb 3, 2004
    #1
    1. Advertising

  2. ~misfit~

    Steven H Guest

    On Tue, 3 Feb 2004 17:35:33 +1300, ~misfit~ wrote:

    > What say you Stacey?


    hehehehe - good one


    --
    -----------------------------------------------------------
    Steven H - Dunedin, New Zealand
    ..net Geek
     
    Steven H, Feb 3, 2004
    #2
    1. Advertising

  3. ~misfit~

    Gavin Tunney Guest

    On Tue, 3 Feb 2004 17:35:33 +1300, "~misfit~" <~misfit~@his_desk.com>
    wrote:

    >I know nobody here is *that* stupid but I hang in alt.comp.hardware and
    >there's a couple of guys in there who keep telling people not to keep their
    >old HDD when upgrading. Some people post saying "Can I use my old drive for
    >mp3s" or suchlike. I've had a long-running argument with these guys who
    >claim that an IDE channel only runs as fast as the slowest device attached
    >to it.
    >
    >I did an experiment this afternoon and thought I'd copy the post here. Just
    >in case anyone is interested. The "Why? (Loaded question)" was my first
    >foray into the thread. I've repeated over and over to these guys (mainly a
    >guy called Stacey, who insists he's right) that they're talking shit.
    >


    It used to be that way, few years back now, so they're not completely
    wrong..... they're probably old-timers or have read it somewhere.
    They've had separate timing on IDE channels for quite some time now,
    can see in the BIOS that it autodetects the PIO or DMA modes for each
    device whereas in the (distant) past it would run at the slowest
    detected PIO mode.

    Even in the old days it wasn't so much it would run at the 'speed' of
    the slowest drive, it ran at the I/O mode of the slowest which isn't
    quite the same. A newer drive would still be heaps faster at say PIO
    mode 4 than an old drive at PIO mode 4 on the same channel.

    But I doubt your antagonists will be interested, sounds like they're
    in a fixed mindset.

    GT
     
    Gavin Tunney, Feb 3, 2004
    #3
  4. ~misfit~

    ~misfit~ Guest

    Gavin Tunney wrote:
    > On Tue, 3 Feb 2004 17:35:33 +1300, "~misfit~" <~misfit~@his_desk.com>
    > wrote:
    >
    >> I know nobody here is *that* stupid but I hang in alt.comp.hardware
    >> and there's a couple of guys in there who keep telling people not to
    >> keep their old HDD when upgrading. Some people post saying "Can I
    >> use my old drive for mp3s" or suchlike. I've had a long-running
    >> argument with these guys who claim that an IDE channel only runs as
    >> fast as the slowest device attached to it.
    >>
    >> I did an experiment this afternoon and thought I'd copy the post
    >> here. Just in case anyone is interested. The "Why? (Loaded
    >> question)" was my first foray into the thread. I've repeated over
    >> and over to these guys (mainly a guy called Stacey, who insists he's
    >> right) that they're talking shit.
    >>

    >
    > It used to be that way, few years back now, so they're not completely
    > wrong..... they're probably old-timers or have read it somewhere.
    > They've had separate timing on IDE channels for quite some time now,
    > can see in the BIOS that it autodetects the PIO or DMA modes for each
    > device whereas in the (distant) past it would run at the slowest
    > detected PIO mode.


    Hi Gav. Yeah, I know, I keep telling them that too, the guys they are
    advising not to use the old drive have new builds, with new mobos.

    > Even in the old days it wasn't so much it would run at the 'speed' of
    > the slowest drive, it ran at the I/O mode of the slowest which isn't
    > quite the same. A newer drive would still be heaps faster at say PIO
    > mode 4 than an old drive at PIO mode 4 on the same channel.
    >
    > But I doubt your antagonists will be interested, sounds like they're
    > in a fixed mindset.


    You're right there, it was quite an active thread until I posted that
    yesterday. Not one reply since, heaps of other posts but nothing in that
    thread.

    Cheers,
    --
    ~misfit~
     
    ~misfit~, Feb 3, 2004
    #4
  5. ~misfit~

    RecyclerMan Guest

    Gavin Tunney wrote:
    > On Tue, 3 Feb 2004 17:35:33 +1300, "~misfit~" <~misfit~@his_desk.com>
    > wrote:
    >
    >
    >>I know nobody here is *that* stupid but I hang in alt.comp.hardware and
    >>there's a couple of guys in there who keep telling people not to keep their
    >>old HDD when upgrading. Some people post saying "Can I use my old drive for
    >>mp3s" or suchlike. I've had a long-running argument with these guys who
    >>claim that an IDE channel only runs as fast as the slowest device attached
    >>to it.
    >>
    >>I did an experiment this afternoon and thought I'd copy the post here. Just
    >>in case anyone is interested. The "Why? (Loaded question)" was my first
    >>foray into the thread. I've repeated over and over to these guys (mainly a
    >>guy called Stacey, who insists he's right) that they're talking shit.
    >>

    >
    >
    > It used to be that way, few years back now, so they're not completely
    > wrong..... they're probably old-timers or have read it somewhere.
    > They've had separate timing on IDE channels for quite some time now,
    > can see in the BIOS that it autodetects the PIO or DMA modes for each
    > device whereas in the (distant) past it would run at the slowest
    > detected PIO mode.
    >
    > Even in the old days it wasn't so much it would run at the 'speed' of
    > the slowest drive, it ran at the I/O mode of the slowest which isn't
    > quite the same. A newer drive would still be heaps faster at say PIO
    > mode 4 than an old drive at PIO mode 4 on the same channel.
    >
    > But I doubt your antagonists will be interested, sounds like they're
    > in a fixed mindset.
    >
    > GT

    Year I got one of those old buggers best thing I did for speed was to
    compress the drive :) old amiga floppy trick my mane machine does not
    care at all.



    --
    / __/ / / / / /__ / / ___/ / __/ / / / |/ / /__ /
    / / / /_/ / / /_/ / _-' / __/ / / / /_/ / / /| / _-'
    ___\ ____/ ____/ /___/ /____/ /_/ ___\ ____/ /_/ /_/ |_/ /___/
     
    RecyclerMan, Feb 4, 2004
    #5
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Unk
    Replies:
    0
    Views:
    1,214
  2. Replies:
    1
    Views:
    537
    kaydigi
    Oct 10, 2005
  3. VanguardLH

    Re: Dual channel or triple channel?

    VanguardLH, Jun 6, 2010, in forum: Computer Support
    Replies:
    1
    Views:
    1,253
    VanguardLH
    Jun 6, 2010
  4. Tony

    Re: Dual channel or triple channel?

    Tony, Jun 6, 2010, in forum: Computer Support
    Replies:
    0
    Views:
    642
  5. Zeke

    Re: Dual channel or triple channel?

    Zeke, Jun 7, 2010, in forum: Computer Support
    Replies:
    0
    Views:
    538
Loading...

Share This Page