Conference hydra::amiga_v1

    
    Several weeks ago  I also went throught the series-II upgrade om my GVP
    Impact 2000-HC.  Same comments as above.  Nice, simple, reliable
    process (and the process doesn't damage any of the data that was on
    your drive).  I also "feel" that there's been a performance improvement
    although nither diskspeed or diskperfa show any significant
    differences.    
    
    I also tried the advice of of another note in here somewhere and 
    experimented with my buffer sizes to get it to cache. I gave it lots of
    buffer and it actually began caching nicely on repeated accesses (no
    additional disk activity).  There's nothing magical about these numbers
    but I think I left them set at 200/100 for DH0/DH1.  
    
    What did surprise me was the following:  I'm used to the rate at witch
    icons fill the screen when I pop open my workbench windows. I would'nt
    call it slow but I wouldnt call it fast either.  It is very
    characteristic though. I fully expected when I got the disk caching
    that the icons were gonna blast right up there.  They didn't, in fact
    it made no difference at all whether they came from disk or cach.  I
    also did some other measurements (crude, stop watch type) by simply
    checking load times of software with and without cache.  I really
    couldn't perceive a difference.  I thought that this was good because it
    suggsted to me that my system is not spending it's time waiting for
    data to come from the drive.  It also convinced me that setting aside
    memory for cache and seeing it work, was neet,  but it didn't buy me
    anything in terms of performance (and it costs memory).
    
    My only criticism is: "I'm glad the installation procedure works as
    well as it does because I've never found GVP documentation adequate"
    
    Frank
    
    
    there are times when it 
    
    I did get one s

    Defrag the disk with QB Tools V1.2a and you'll see your icons blaaast
    up onto the screen.

    U42 was included in the upgrade for the benefit of owners of early
    models of the controller. My first try at the upgrade failed badly.
    After the installation I could not run my JX100 scanner or TAD. 
    I tokk it all back,and redid everything. It is alright now. I believe
    my problem was due to the seating of the board. 
    	I ended up rebuilding my whole hardrive in the end. I also noticed
    my 40 meg flushed out at 39meg running the auto software,and it was
    40.4 under the older version. While you are supposed to be able to use
    all the exsisting software after the installation I question this.
    The fact that the phisical size maps out differently is one of the
    reasons I feel this. Also I think you benefit by doing everything from
    scratch as far as performance is concerned. I see several factors that
    change with the upgrade,and wonder how they manage the things they do
    if you elect to keep ypur data on installation.(just reprep the drives
    and go). As a matter of fact I was warned that my old settings exceded
    the capacity of the drive. GVP told me that was because the software
    claims space to do itas remap of bad blocks when needed. How will this
    happen if you keep the old partitions. I wanted to redo it anyway
    because I needed a 2meg partition for Pro Video Post. One thing I did
    notice is that on startup it takes three times longer to get it
    running.
    							bill

    A nice featuire is the ability to get rid of those DH10: and DH11:
    two-digit drive designations. I changed then to DH1: and DH2: and
    selected WRITE on the FaaastPrep manual mode screen to change it.

    
    I learned a couple of things from GVP ... this driver and controller
    do not require any mountlist entries for attached drivers, nor do they
    require any BINDDRIVERS or MOUNT commands in the startup-sequence.

    RE: .3
    
    If you set a check on the LAST DRIVE gadget for the last drive on the
    chain, it will speed things up a bit at boot time. Otherwise, the boot
    code will search for the 7 allowable devices on the scsi bus.
    
    Also, if you have more than 1 harddrive, set the slowest drive to scsi
    device 0. That has the effect of forcing the boot through the slowest
    device. This mught actually help speed things up a bit.

    re .7 - check the last drive for a faster boot
    
    ahhhh thankyou thankyou thankyou thankyou........
    
    kevin

    Here's some additional info on getting past the slow spin-up time of
    Seagate drives that are daisy-chained onto the faster Quantums used by
    the GVP controllers.
    
    1.  Change the SCSI ID of the Quantum on the hardcard from 0 to 1. This
        requires placing a jumper on the SCSI ID block at the pair of pins 
        marked A0. You'll need to unscrew the drive from the card to do
        this.
    
    2.  Change the SCSI ID of the second drive from 1 to 0. This requires
        removing the jumper on the SCSI ID jumper block.
    
    NOTE:  Do not change the terminating resistors. They are meant to
           identify the drive furthest from the controller. By changing
           the SCSI ID, you are not changing the location of the drive.
    
    3.  Go into FaastPrep and select Manual mode. Change the BootPri of the
        second "slow" drive to -10. This will force the Quantum to act as
        the boot drive even though it has a SCSI ID of 1.
    
    4.  Then, check of LUN for the second "slow" drive.
    
    5.  Write this new information to the second drive and you are all set.
    
    

    *I hope my usage of "RDB" is correct, if not pls straighten me out
    
    
    I completed my upgrade from an A500+Supra500RX to an A2000+GVP Series
    II last night.  The GVP conversion was excellent.  I just plugged in
    the drive that was formatted on my A500/Supra combination, set the last
    lun/drive bit, wrote it to the *RDB, and booted.  That RDB standard is
    great!!!  I do plan to reformat the drive with the GVP software after
    I restructure my partitions but the option to "plug 'n play" the hard
    drive really sped up my transition.  The GVP gave me about a 100K
    speedup on reads according to diskspeed v3.1.  The max read with 32K
    buffer on the Supra500RX was about 524K/s and is about 662K/s using
    the GVP series II.  I bought the Series II with 4meg of memory.  Those
    SIMMs are inexpensive.  4meg of memory costs $190.00!  The SIMMs sell
    mail order for around $55. each.
    
    The documentation that GVP supplied is a little skimpy if you want
    to use the "manual" mode of installation, but their automatic
    mode works ok and is quite intuitive.  One thing I didn't like
    is that it increased the size of the partition label by appending
    the partition size at the end of the label; I label a partition "xyz"
    and when it amigados formatted the partition the label becomes
    "xyz{120mb}".  I had to manually relabel my partitions.  GVP's 
    software gives you more information on your drive type and
    format than Supra's.  To configure a drive on the Supra you had
    to access "drive parameters" enbedded in the software to configure
    the drive.  If it wasn't supported by the particular version of
    SupraFormat you were using then you had to find the information else-
    where.  GVP reads the RDB on the drive then displays the information.  
    After using them both I prefer the GVP over Supra although
    both allow you to read, write, and modify the drive's RDB parameters.
    GVP mounts the partitions *much* faster than Supra once you disable
    GVP's "scsi bus probe" on boot.  The Supra mount command allows
    you to mount individual partitions and/or drives.  I haven't discovered
    that capability with the GVP hardware, yet.  The title of this reply
    gives you my impressions.
    
    As an aside, I did notice that the colors displayed from the A2000 are
    noticably "lighter" than when displayed from the A500.  It is as if
    I turned up the "saturation" slider on my workbench screen.  Hmmmmm?
    I use an 8-color workbench and had to adjust most of the "G" settings.
    Any comments?
    
    mark