Cy
Tue Sep 25 23:00:31 PDT 2007
That would actually only apply on a sequentially defragmented drive. And
would certainly no longer apply in the day of NTFS since there IS no FAT. I
do however recall that in that you are correct that xcopy would read as many
files as would fit in memory before writing out what was in memory in
cycles. I would guess it could be a little faster IF you didn't have any
cache running as the primary purpose for those is/was to eliminate exactly
that problem. Along with the fact that many caches did lazy writes and had
access to much more memory than xcopy did (most ran in extended memory which
xcopy couldn't do) pretty much did away with any advantage for xcopy.
The cache I use (hyperdisk IIRC) was very good at that. Far better than
MS's cache, and it made more reads and writes seem faster than the drive.
That was also in the day of MUCH slower drives than we have today. My slow
laptop drive (5400 RPM) is many many times faster than the fastest drives
from back then.
--
Cy Welch
Senior Programmer/Analyst
MetSYS Inc.
http://www.metsysinc.com
"Man-wai Chang ToDie" <toylet.toylet@gmail.com> wrote in message
news:eBmw1N$$HHA.536@TK2MSFTNGP06.phx.gbl...
> Man-wai Chang ToDie wrote:
>>>> XXCOPY and XCOPY should be faster.
>>> ...and you say this because... ??
>>
>> XCOPY uses a more efficient way of copying files. I forgot how it did it,
>> something to do with the FAT access I barely remembered from my DOS
>> days... Oh... maybe NTFS wouldn't allow that kind of access now. In the
>> latter case, I would apologize.
>>
> I remembered, it's the sequential read on the FAT for all the files needed
> to be copied, thus reducing disk head movement...
>
> --
> @~@ Might, Courage, Vision, SINCERITY.
> / v \ Simplicity is Beauty! May the Force and Farce be with you!
> /( _ )\ (Xubuntu 7.04) Linux 2.6.22.8
> ^ ^ 12:11:01 up 15:01 0 users load average: 0.00 0.03 0.00
> news://news.3home.net news://news.hkpcug.org news://news.newsgroup.com.hk