Docster said:
What is the largest internal hard drive WindowsXP SP3 32 bit NTFS will
recognize and handle without problems?
I have heard 2TB and and smaller.
Using XP Pro with SP3 on a modern mother board and want to install a 1TB
internal HDD, can I do it without conflict?
This article, discusses a method of booting from and handling
disks, over 2.2TB in size. 512 byte sector * 2^32 is what the
number is derived from. But a number of elements have to come
together, to boot from a GPT prepared disk. I would need a UEFI
BIOS, a GPT capable OS, and a GPT prepared disk.
http://en.wikipedia.org/wiki/GPT_Disk
"For those disks with 512-byte sectors, the MBR partition
table entries allow up to a maximum of 2.20 TB (2.20 × 10^12 bytes)."
For data-only disks, the problems aren't nearly as bad.
There are workarounds to deal with the capacity issue.
So if your OS isn't living on there, is should be OK.
You can get information about large disk purchases, on the
disk manufacturer site. For example, if I was shopping for
a 4TB disk drive, I would visit the manufacturer site, and read any
"white paper" or FAQ entries, on what to do. For example, one
interim solution, was a special driver, that converts a large
disk, into a bunch of smaller virtual disks, each with its
own partition table. That's a workaround to shoehorn a big
disk, into a small disk OS.
NTFS isn't the problem here, as the Wikipedia article on
NTFS will tell you that there aren't issues yet with
capacities there. It's more a boot issue, if you're
wanting to boot from a drive like that.
The OSes also vary, on their support for sector size.
Legacy sectors are 512 bytes. Advanced format drives
switched to 4KB sectors (for greed reasons). Some of the
OSes, simply don't support 4KB sectors directly, but they
might have been patched to do so. Currently, 4KB disks
use "512e emulation", as a means to bridge both worlds.
And emulation, means the drive has "physical" sector
size (4KB) and "virtual" sector size (512e). It is up
to the drive controller board, to do read-modify-write cycles
to take a commanded 512 byte operation, and properly
fit it into an already occupied 4KB sector. This has
also meant, that the controller on the disk drive,
actually supports caching in a meaningful way (the
command rate on drives is a lot higher now, as a result).
Commands terminating in cache on the controller, you can
do a lot more of those per second. On older drives,
it wasn't apparent (to me at least), that there was
any difference between a 2MB and 8MB cache IDE drive.
The cache didn't seem to do anything.
If you're lucky, a disk manufacturer FAQ, may do a better
job of writing all this stuff up, than I can. Initially,
the disk manufacturer web pages, were woefully inadequate.
But there is some info now.
To me, the disadvantage of large disks, are time for
maintenance operations to complete, the liabilities with
that much content stored in one place. You should
really budget for two identical drives, so one drive
functions as your (offline) backup. That way, if
there is a disaster, you're covered. I generally
buy drives in pairs, like Noah's Ark
4000GB at 0.13GB/sec, takes 8.5 hours. So just to
read-verify the drive, takes that long. If you fill
drives like that up with content, doing maintenance on
them, takes all day. Running a backup, takes all night.
And so on.
The last drive I bought, was 2TB, and I specifically
selected that capacity, to avoid any special handling
issues (2.0TB is less than 2.2TB). Drive works with
WinXP SP3, Win2K SP4, and Windows 8, no problem.
The drive has no OS on it, and is a backup drive.
If I had to, I could move an OS onto it.
Paul
