Read <MY DISCLAIMER> First
First--My Hard-Drive Info-Page Below the Index:
Index of All my other pages are listed below in NO frills text:
Up-grade your Windows-95 To 95a and Up-Date Tool (at Billies
Place)
http://support.microsoft.com/support/downloads/LNP195.asp?PR=ALL&FR=0&M=F&
Some info you'll need to Understand
Your Hard-Drive:
This Page with get you started in understanding of the Hard-Drive. Even though we'll be
addressing the IDE-at system, all drives Mfg.'s have to deal with Windows-95 FAT. If you
need help with FAT FDISK or FORMAT or Partition Magic. More
Information, use the e-mail Page. Thanks.
A User's View:
Let's take a look at what happens when you retrieve data from a hard disk drive. When you
issue a command to open an existing file, the application program you're running prompts
you to enter the name of the file to open. It then passes the file name to the operating
system, which determines where the file is located on the disk drive - the head number,
cylinder, and sector identification. The operating system transfers this information to
the disk controller, which drives an actuator motor connected to the actuator arm to
position the heads over the right track. As the disk rotates, the appropriate head reads
the address of each sector on the track. When the desired sector appears under the
read/write head, the entire contents of the sector containing the necessary data are read
into a special, ultra-fast memory, called cache, on the drive's PCB. Then, the disk drive
interface chip sends the necessary information to the computer's main memory. (in
short-more cache buffer, better performance). Storing data on a hard drive is a similar
process to retrieving data, only reversed. The host computer operating system is
responsible for remembering the addresses for each file on the disk and which sectors are
available for new data. If the file you want to store is large - for example, (NOTE: I put
this in for my friend SLIM), a 10 MB CAD/CAM drawing - the operating system instructs the
controller where to begin writing information to the disk. The controller moves the
read/write heads to the appropriate track and writing begins. When the first track is
full, the heads write to the same track on successive platter surfaces. If still more
track capacity is required to store all the data, the head moves to the next available
track with sufficient contiguous space and writes the data there. Although an
extraordinary amount of care and effort goes into making the platters for hard disk
drives, it is not economically feasible to manufacture 100 percent defect-free media.
Therefore, all modern hard drives have a defect management strategy built into the disk
controller to provide defect-free operation in the field. Defect
management involves setting aside some spare sectors on each disk surface to replace a
limited number of defective sectors. At the end of the manufacturing process, the entire
disk surface is scanned for defects and the disk controller stores a map of their
locations. When the operating system requests that information be written to one of the
bad sectors, the disk controller transparently maps it to one of the spares. The disk
controller continuously updates the defect map.
NOTE: a word about (IDE);
EDIE is a marketing program started by Western Digital to promote certain ATA-2 features
including ATAPI. WD has encouraged other product vendors to mark their products as
"EIDE compatible" or "EIDE capable".
Unlike humans, who use a 10-digit, decimal system for everyday computation, digital
computers and most other electronic equipment rely on a 2-digit, binary system. Using the
binary system, all data - letters, numbers, and other objects - are represented by a
series of binary digits called "bits." Consisting only of 0s and 1s representing
on or off switch positions, bits are recorded on a data storage medium, such as the
magnetic coating on the platters of a computer's hard drive. By combining individual data
bits into larger, 8-bit groupings called "bytes," computers encode data for
computing. For example, the letter "B" is encoded as "01000010" in the
most widely used method for representing alphanumeric characters for computer storage,
display, and printing.
Back Top
To fully appreciate the vital role mass data storage devices play in storing and
retrieving information, you need to understand the basics of computer systems. A computer
consists of both hardware and software components working together to help you accomplish
your tasks. At the most basic level, computers perform computations. They rapidly add,
subtract, divide, and multiply numbers that represent encoded data (letters, numbers,
charts, images, colors, etc.). We work with this data everyday when we use application
software such as word processors, spreadsheets, and graphics packages.
Let's take a look at what happens when you retrieve data from a hard disk drive. When you
issue a command to open an existing file, the application program you're running prompts
you to enter the name of the file to open. It then passes the file name to the operating
system, which determines where the file is located on the disk drive - the head number,
cylinder, and sector identification. The operating system transfers this information to
the disk controller, which drives an actuator motor connected to the actuator arm to
position the heads over the right track. As the disk rotates, the appropriate head reads
the address of each sector on the track. When the desired sector appears under the
read/write head, the entire contents of the sector containing the necessary data are read
into a special, ultra-fast memory, called cache, on the drive's PCB. Then, the disk drive
interface chip sends the necessary information to the computer's main memory.
Are there supposed to be bad sectors on the drive?
No. but all modern drives support error management, which completely hides any bad sectors
that may be on the disk off factory. Even a single bad sector is sufficient grounds to
return the drive under warranty. If you want to continue using it, the drive should be
viewed with the utmost suspicion.
Western Digital has a utility wdat_ide.exe that can hide grown bad sectors on many Caviar
disks. There is one exception. Under rare circumstances, use of bad (too fast) timing by
the disk adapter can cause bad sectors on a disk. This type of error can be fixed simply
by writing fresh data to these sectors, as there is no actual media
defect.
How Data is Organized on a Hard Disk Drive:
The surface of the drive platter is organized with coordinates, much like a map. Data is
stored in concentric tracks on the surfaces of each platter. (A platter has two sides, and
thus, two data recording surfaces.) A typical disk drive can have more than 2,000 tracks
per inch (TPI) on its recording surface. A cylinder describes the group of all tracks
located at a given head position across all platters. To allow for easier access to data,
each track is divided into individually addressable sectors. The process of organizing the
disk surface into tracks and sectors is called formatting, and almost all hard disk drives
today come preformatted by the manufacturer. The process of formatting a hard drive
applies addressing data to the platter's surface. In almost all systems, including PCs and
Macintoshes, sectors typically contain 512 bytes of user data plus addressing information
used by the drive electronics (although some proprietary systems use other sector
lengths). The disk drive controller, which resides on the drive's PCB, uses the formatting
information and addresses - much like a tourist uses a city map - to guide data into and
out of a specific location on the hard drive. Without formatting instructions, neither the
controller nor the operating system would know where to store data or how to retrieve it.
In earlier hard drive designs, the number of sectors per track was fixed and, because the
outer tracks on a platter have a larger circumference than the inner tracks, space on the
outer tracks was wasted. The number of sectors that would fit on the innermost track
constrained the number of sectors per track for the entire platter. However, many of
today's advanced drives use a formatting technique called Multiple Zone Recording to pack
more data onto the surface of the disk. Multiple Zone Recording allows the number of
sectors per track to be adjusted so more sectors are stored on the larger, outer tracks.
By dividing the outer tracks into more sectors, data can be packed uniformly throughout
the surface of a platter, disk surface is used more efficiently, and higher capacities can
be achieved with fewer platters. The number of sectors per track on a typical 3.5-inch
disk ranges from 60 to 120 under a Multiple Zone Recording scheme. Not only is effective
storage capacity increased by as much as 25 percent with Multiple Zone Recording, but the
disk-to-buffer transfer rate also is boosted. With more bytes per track, data in the outer
zones is read at a faster rate. Multiple Zone Recording on 2.5-inch disk drive products.
Read/Write Heads:Skimming the Surface Read/write heads are the single most costly
component of a hard disk drive, and their characteristics have a great impact on drive
design and performance. Despite their expense, the head's basic design and objective are
relatively simple: a head is a piece of magnetic material, formed almost in the shape of a
"C" with a small opening or gap. A coil of wire is wound around this core to
construct an electromagnet. In writing to the disk, current flowing through the coil
creates a magnetic field across the gap that magnetizes the disk coating layer under the
head. In reading from the disk, the read/write head senses an electronic current pulse
through the coil when the gap passes over a flux reversal on the disk.
Back Top
As technology increases, bits are packed more densely, and the space required to store a
bit shrinks. At the same time, the tiny size of the stored data bit causes the signal
produced by the head when reading the bit to become weaker and harder to read. As a
result, the fundamental challenge in packing bits closer together is finding a way to fly
the heads closer to the media to increase the amplitude of the signal. The hard disk drive
industry has made great strides on this front. In 1973, flying heights averaged 17
microinches. Today's heads fly at just three microinches, with 2- to 2.5-microinch flying
heights expected soon. And, in the not too distant future, read/write heads might even
make contact with the media, enabling data to be packed even more densely on the platter
surface but offering the additional challenge of eliminating added wear on the disk media
and read/write heads.
Typically, a drive head must settle lateral movement of the actuator, it must stop before
a read and write operation can begin and it must settle before a write can occur. Drive
Mfg.'s firmware with read-on arrival lets the drive start a normal read operation before
settling is complete and if an error occurs it is corrected via advanced ECC algorithms.
In Auto transfer ASIC technology, an interrupt occurs in an IDE system at each sector (512
bytes) of data transferred and this takes away from primary work. A drive Mfg.'s firmware
can significantly reduce the number of interrupts during processing of the I/O request
allowing the transfer of multiple sectors of data per interrupt and is similar to DMA or
fast multiword use.
Regardless of the physical capacity of the disk drive itself, with DOS prior to 1989,
there was a 32 MB partition limitation due to FAT fixed 512 byte cluster size. In 1985,
the 16-bit FAT can in to support larger partitions with the use of larger clusters. Then
there was the issue of the FAT and VFAT systems 2.1 GB barrier where of the maximum of
65,536 clusters. Fat clusters vary in size according to the size of the partition. This
means for DOS, Windows and Windows 95 users currently cannot address a disk logical
partition larger than 2.1 GB because larger drives require multiple partitions to provide
storage.
(NOTE) Windows NT and OS/2 users are not affected.
The increasing demand for drives with larger capacities means and requires more powerful
error correction schemes called (on-the-fly) error correction to save millisecond burst
errors of which takes 1 full disc revolution or (app. 13ns) and without it, the rates
decrease output. This means that the sequencer continues running unless more than one
error occurs in the same sector. In this case, a more rigorous correction algorithm
enables correction of double-burst errors of up to 3 bytes and a unrecoverable error rate
of 1 error in 10-14 bits read, and the head must settle it.
After almost 3 months of trial and error, I have finally benchmarked a 10 to 26 % increase
in speeding up W-95/W-95b/W-98 operating system and here is what I ended up with as the
procedure.
Swap-file partition for (speed):
(1) One way or the other (FDISK or PM 3), make a (80mb) partition. After you have created the partition, go to Control Panel, click on System, click the Performance tab, click on Virtual Memory. You'll see a drop box arrow listing the available partitions. Just choose the 80mb. You can also choose the size of the file. I choose 20- & 30+.
(2) It's where you want Windows to operate: Again--make a (120mb) partition.
Re-install your Windows in this partition any way you want--Partition Magic 3 makes it
easy.
NOTE: To get the full effect--you have to do both BUT either one will help.