WB 3.1 and HDToolbox
Tony Leneis
finally got around to installing it. I also decided to add a new hard drive
at the same time. This means I got to play with the new HDToolbox. Problem
is, all the shiny new WB 3.1 books refer me to the _Amiga_Hard_Drive_User's_
Guide_, which, of course, was not deemed important enough to include with the
rest of the documentation. (Are there any A3000's out there which don't have
a hard disk?) Anyway, I dug out my WB 2.04 docs, but there are a few changes.
For example, when specifying the filesystem, I now can choose between Standard
File System, Custom File System, and UNI\01. What exactly is this UNI\01? Is
it an actual UNIX filesystem accessible from the Amiga side, or is it just a
new name for the 'Reserved Partition' option from WB 2.04? Also, there's this
really cool option that lets me change the file system block size. I assume
that the default value for this is 512 bytes? Seems like a nice way to speed
up handling of large files. On the partitioning screen, there's a new field
that I can enter the HostID in. Is this the SCSI ID for the Amiga itself?
i.e. if I changed it to 0, I could put a hard disk in addressed as 7? Anyway,
those are the unimportant questions... on to the more pressing problem...
When I defined my new hard drive for HDToolbox, I tried to get it to
read the configuration from the drive. It got the name right - a Quantum
LPS540S revision 5900, but it came up with a size of -62K... hardly close to
540 megs. The HDToolbox under WB2.04 had the same problem. Anyway, I looked
at the hard drive, and found some chicken scratches that seemed to indicate
number of cylinders, heads, and sectors (I took sectors to be the same as
blocks), and entered the numbers. The new size is much better, but it's still
a few megs shy of 540 like the drive claims. Is this normal? It kinda bugs
me that the number that pops up in the size field isn't the same as what is
stamped on the hard drive. Finally, since the read configuration got all the
other numbers wrong, I'm not sure I trust what it chose for the cylinder to
park the head at. It chose 0, but the WB 2.04 documentation says to use the
last cylinder unless otherwise specified by the manufacturer. Problem is, I
didn't get any documentation with the hard disk. Anyone know where Quantum
says the head should be parked?
Tony Leneis
len...@cse.ogi.edu
p.s. Other than the lack of documentation, the new HDToolbox is a nice
improvement over the one in version 2.04. If nothing else, it finally
has the same general cosmetic look as all the other WorkBench programs.
No more funky colors or buttons.
--
Tony Leneis __
--------------------------------------------------///-From the developer of:-
len...@cse.ogi.edu Hillary, __ /// Monster II, ODETOOLKIT,
ale...@newton.math.hmc.edu Gennifer, Only\\\/// Vogon Constructor Fleet
Oliver B. Warzecha
> 540 megs. The HDToolbox under WB2.04 had the same problem. Anyway, I looked
> at the hard drive, and found some chicken scratches that seemed to indicate
> number of cylinders, heads, and sectors (I took sectors to be the same as
> blocks), and entered the numbers. The new size is much better, but it's still
> a few megs shy of 540 like the drive claims. Is this normal? It kinda bugs
> me that the number that pops up in the size field isn't the same as what is
> stamped on the hard drive. Finally, since the read configuration got all the
This is quite normal. The reason for this is that the HD Manufacturers don't
count the size in MByte, instead they count in byte and divide by 1.000.000.
That would made 540.000.000:(1024*1024)
(tick...tick)
around 514.98 MB
Am I right?
Anyway, I was bugged by this first ("defect cylinders?") until I applied
some math to it. I still can live with a LPS240 with 233 MB space.
--
Ceterum censeo ed#(it)ieren esse delendam!
"Man sollte eben keine Frauen an Computer lassen..."
Angela Schmidt (!) in <31jaf0$9...@nz12.rz.uni-karlsruhe.de>
OBW
Ian Dean
> The new size is much better, but it's still
>a few megs shy of 540 like the drive claims. Is this normal? It kinda bugs
>me that the number that pops up in the size field isn't the same as what is
>stamped on the hard drive. Finally, since the read configuration got all the
>other numbers wrong, I'm not sure I trust what it chose for the cylinder to
>park the head at. It chose 0, but the WB 2.04 documentation says to use the
>last cylinder unless otherwise specified by the manufacturer. Problem is, I
>didn't get any documentation with the hard disk. Anyone know where Quantum
>says the head should be parked?
It all depends how you calculate MegaBytes, some people use 1000000 bytes (eg.
Quantum), some people use 1000 Kb and some people use 1024 Kb (HDToolBox)...
The Quantum LPS270, LPS340 & LPS540 use something called AirLock which
according to the specs is "automatic shipping lock and dedicated landing
zone", so (apart from wasting a cylinder) the setting in HDToolBox does
nothing...
Ian
Litrik De Roy
I can't find the original posting anymore so i'll jump in here...
I have almost the same problem. I have a new LPS270S but hdtoolbox says its size
is only -62 K :(
So I wonder what the correct settings for Cylinders,Heads,Blocks/Track and
Blocks/Cylinder are. My /old/ LPS105 has 2097,1,98,98 and it looks like
5396,1,98,98 are pretty good values for the 270 but I'm not sure. The "Check
data on drive" gives an awful lot of bad blocks, so it's possible the drive is
damaged.
Anybody any idea ?
OBW> Anyway, I was bugged by this first ("defect cylinders?") until I
OBW> applied some math to it. I still can live with a LPS240 with 233 MB
OBW> space.
And what are your settings in hdtoolbox ?
|| Litrik De Roy -- FidoNet : 2:292/603.61 -- InterNet : lit...@augfl.be ||
|| MultiUser Beta Tester -- Author of MuMu -- First Class Loser ||
|| root@arwen "To boldly go where no A500 has gone before!" ||
... won't somebody help me, is it hard, to let me find my way
won't somebody love me (for a start) i'm laughing as i pray
WILLIAM THOMAS CENTER
doesn't read the information from the drive. I originally came up with the
following settings:
Cylinders - 4130
Heads - 1
Blocks per Track - 135
Blocks per Cylinder - 135
which gave me 272 megs of disk space and it would format correctly. I called
Quantum's tech line and the amiga tech guy told me to use the following
settings:
Cylinders - 944
Heads - 14
Blocks Per Track - 40
Blocks per Cylinder - 560
which gives you 257 megs. When I asked if there were really 14 heads he said
"no" but it is only a mathmatical abstraction. I can't remember exactly but
it is the number of cylinders times the number of heads times 512 (or
something of that nature) and whatever gets you closest to 272 meg will work.
Bill Center
Bruce Elrick
#Heads x #blocks/track = blocks/cylinder
(makes sense as a track is a slice of the cylinder)
#cylinders x #blocks/cylinder = total number of blocks = #kB x 2
The abstraction is that with SCSI, the HD hides the above and only
requires the total be correct. The computer just uses the above
as a throwback.
Cheers...
--
Bruce Elrick EMail: Elr...@Physics.UToronto.CA
Department of Physics Office: McLennan Labs 1022
University of Toronto Work Phone: (416) 978-5207
Toronto Ontario M5S 1A7 Home Phone: (416) 928-0736
Tony Leneis
>Tony Leneis (len...@blue.cse.ogi.edu) wrote:
>:The new size is much better, but it's still
>:a few megs shy of 540 like the drive claims. Is this normal? It kinda bugs
>:me that the number that pops up in the size field isn't the same as what is
>
>number somewhere around 527M?
Nope - it was 515M. As another reader pointed out, Quantum says that one meg
equals 1,000 kilobytes equals 1,000,000 bytes. This actually is much easier
to work with than units of 1024, but somehow it feels like a marketing ploy.
Anyway, 540*1000*1000/(1024*1024) = 515.
David Oakes
: When I defined my new hard drive for HDToolbox, I tried to get it to
:read the configuration from the drive. It got the name right - a Quantum
:LPS540S revision 5900, but it came up with a size of -62K... hardly close to
:540 megs. The HDToolbox under WB2.04 had the same problem. Anyway, I looked
:at the hard drive, and found some chicken scratches that seemed to indicate
:number of cylinders, heads, and sectors (I took sectors to be the same as
:blocks), and entered the numbers.The new size is much better, but it's still
:a few megs shy of 540 like the drive claims. Is this normal? It kinda bugs
:me that the number that pops up in the size field isn't the same as what is
The numbers that manufacturers claim are not always correct. Was your actual
number somewhere around 527M?
*************************************************************
**** **** Replies to: ****
**** David Oakes **** aa...@cfn.cs.dal.ca ****
**** ****or: d.oa...@genie.geis.com ****
*************************************************************
Eric Krieger
> In article <121...@cup.portal.com> d...@cup.portal.com (Douglas Gary Bullard) writes:
> >Tony, if you got a capacity of 515 Mb, you're on target. Quantum told me
> >that the Amiga BIOS can't figure out the logical stuff on the drive.
Sometimes HDToolBox (or RDPrep) reports 'strange' values for number
of heads, tracks, blocks/track, sectors, etc. (like 1 head, and
25000 tracks, etc), but if the number of total blocks are correct,
there will be no problem.
> >Correct numbers are: 1120 cylinders, 16 heads, 59 blocks/track,
> >944 blocks/cylinder, park head at 1120.
> You'll find this info on the sticker on the top of the drive, most modern
> drives have this info there too except for blocks/cylinder which is
> blocks/track * heads...
I've several HD's from Quantum (my latest, 3 months old, and biggest
is a Empire 1080S = 1 GB), but none of them has a sticker on top,
bottom, front, back,.... with these kind of information.
--
| Eric Krieger : ekri...@quasar.xs4all.nl | Fido-Net: 2:285/1.10 |
| Q U A S A R : Last Point of Endless Time | AmigaNet: 39:102/1.21 |
Ian Dean
>Subject: Re: WB 3.1 and HDToolbox
>Date: Tue, 6 Sep 94 21:32:55 PDT
>Stuff about the Quantum 540...
>>Tony Leneis
>>len...@cse.ogi.edu
>Tony, if you got a capacity of 515 Mb, you're on target. Quantum told me
>that the Amiga BIOS can't figure out the logical stuff on the drive.
>Correct numbers are: 1120 cylinders, 16 heads, 59 blocks/track,
Douglas Gary Bullard
>Tony Leneis
>len...@cse.ogi.edu
Tony, if you got a capacity of 515 Mb, you're on target. Quantum told me
that the Amiga BIOS can't figure out the logical stuff on the drive.
Correct numbers are: 1120 cylinders, 16 heads, 59 blocks/track,
944 blocks/cylinder, park head at 1120.
Hope this helps!
Greg Bastow
wrote:
L> In article <CvqE9...@cs.dal.ca> aa...@cfn.cs.dal.ca (David Oakes)
L> writes:
L> >Tony Leneis (len...@blue.cse.ogi.edu) wrote:
L> >:The new size is much better, but it's still
L> >:a few megs shy of 540 like the drive claims. Is this normal? It
L> kinda bugs
L> >:me that the number that pops up in the size field isn't the same as
L> what is
L> >
L> >The numbers that manufacturers claim are not always correct. Was
L> your actual
L> >number somewhere around 527M?
L> Nope - it was 515M. As another reader pointed out, Quantum says that
L> one meg
L> equals 1,000 kilobytes equals 1,000,000 bytes. This actually is much
L> easier
L> to work with than units of 1024, but somehow it feels like a
L> marketing ploy.
L> Anyway, 540*1000*1000/(1024*1024) = 515.
It's not Quantum, it's ALL drive manufacturers. That is called a
DISK-Megabyte. I've requested the tech-spec books for every drive I've ever
purchased. Seagate, Quantum and Maxtor all refer to this math for coming up
with drive size.
Greg
-- Via DLG Pro v1.0
___________ _ _ *-----------------------------------------*
//////////// \\ // | 6.2 GIG Online, Gaming, Usenet, Fido! |
// \\ // | *> 30,000 Files Online and Growing ! <* |
// \\ // *-----* All Fish Disks Online [DLG BBS] |
// \\// | - 4 Line BBS - |
//_unnel \/_ision_BBS *-----------------------------------*
\
\__Greg_...@tvbbs.wimsey.com OR 1:153/910.0 (FidoNet)
Brett Watson
>Nope - it was 515M. As another reader pointed out, Quantum says that one meg
>equals 1,000 kilobytes equals 1,000,000 bytes. This actually is much easier
>to work with than units of 1024, but somehow it feels like a marketing ploy.
>Anyway, 540*1000*1000/(1024*1024) = 515.
It is unlikely that any manufacturer would deem a Megabyte to be 10^6
bytes (incorrect) rather than 2^20 bytes (correct). For casual
descriptions, this is okay - there is an error factor of only about 5%
between the two measures - but where the data is already neatly divided
up into even powers of two, power of ten measurements are a no-no.
Discrepancy between the claimed capacity of a hard disk and the amount
of data it can actually hold in files is caused by several things.
Firstly, manufacturers typically list the *unformatted* storage
capacity of their disk. When a disk is formatted, storage space is
lost in overheads such as sector headers and spare sectors (reserved
for remapping errors). This overhead can be substantial -
approximately 17% of the capacity of the disk in the case of the
Seagate ST11950N and ST12550N drives.
Once a disk has had a low-level format such as this applied, there
are a given number of blocks of a given size (usually 512 bytes)
available. Before any files can be written, though, there are a few
more overheads. In the case of the Amiga, the RDBs (Rigid Disk Blocks)
have to be written. This is a reserved part at the start of the disk
where partition information is written, as well as other optional
information such as bad block lists and filesystems.
Further overhead occurs when we get to the next stage - the
high-level format. At this stage sectors are used for file and
directory nodes, reserved blocks and the free block bitmap. On an
Amiga, two blocks are reserved (a mere 1k) and there is a one block
overhead for each file and directory, including the root directory.
Large files have extra blocks in proportion to their size - I believe
the figure is roughly one extra block per hundred data blocks.
The most substantial loss during high-level formatting, though, is
the free block bitmap. A 1Gb drive might have 2097152 blocks, which
require 512 blocks (256k) of bitmap to keep track of free space.
Another 5 blocks (about one per 100 blocks of bitmap) go on top of
that to keep track of where the bitmap blocks are.
The rest is available for files, but bear in mind that each file
introduces further overhead. Even a one byte long file takes two disk
blocks, and large files have approximately 1% overhead. If there is
one megabyte free on your hard disk, this doesn't mean you have room
for a one megabyte file.
--
/ The Famous Brett Watson <br...@abc.gov.au> x=sin(t), y=cos(3t) \
/ Working on digital audio at the Australian Broadcasting Corporation. \
\ "How many gigabytes worth of hard disks have you formatted today?" /
\ Disclaimer: Opinions herein might not be my employer's - or mine /
Ron Charlton
<len...@blue.cse.ogi.edu (Tony Leneis) writes:
<
<>Nope - it was 515M. As another reader pointed out, Quantum says that one meg
<>equals 1,000 kilobytes equals 1,000,000 bytes. This actually is much easier
<>to work with than units of 1024, but somehow it feels like a marketing ploy.
<>Anyway, 540*1000*1000/(1024*1024) = 515.
<
< It is unlikely that any manufacturer would deem a Megabyte to be 10^6
<bytes (incorrect) rather than 2^20 bytes (correct).
Unlikely? Nope. All drive manufacturers specify drive size
in 10^6 = a megabyte. It makes the unsuspecting buyer thinks (s)he is getting
more for the money. Caveat emptor.
See Expert C Programming - Deep C secrets by van der Linden.
Ron
--
A just machine to make big decisions | Ron Charlton
Programmed by fellows with compassion and vision | char...@cs.utk.edu
We'll be clean when their work is done ----------------------
We'll be eternally free yes and eternally young -- Donald Fagen, I.G.Y.