OT: Windows 2000 Pro to XP Pro upgrade without having to reinstall applications?

On 14/06/2013 09:04, The Natural Philosopher wrote:

Oink flap.
If I create a thousand files, then delete every other one how is it not fragmented? (feel free to send a link with explanations).
Of course with SSDs this is ceasing to be much of an issue.
Andy
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

In the short term it may well be. But at least OS X will tidy up for you.
--
Tim

"That excessive bail ought not to be required, nor excessive fines imposed,
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 16/06/2013 19:09, Tim Streater wrote:

Only if it becomes a "hotfile" it will stay fragmented if it doesn't, unless you run a defragger.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

No, OS X just moves files around for you to keep things unfragmented. Never any need for a defragger, or even to think about the matter.
--
Tim

"That excessive bail ought not to be required, nor excessive fines imposed,
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 16/06/2013 22:42, Tim Streater wrote:

It doesn't move every file, it only moves "hotfiles". The OS uses some usage pattern to decide what a "hotfile" is.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 17/06/2013 11:09, dennis@home wrote:

I was under the impression, possibly entirely mis-guided, that there are two features.
Hot File Clustering currently available only on boot volumes
and
On-the-fly Defragmentation
Further, the 1000 file idea would be causing fragmentation of free-space - a related but different problem to fragmentation of files.
--
Rod

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Have a look here:
http://osxbook.com/software/hfsdebug/fragmentation.html
--
Rod

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 17/06/2013 11:24, polygonum wrote:

Good read. So Apple's primary tools are delayed allocation and an on-the-fly defragmenter - and that seems to be enough.
Andy
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 16/06/2013 18:52, Vir Campestris wrote:

ITYM and then filled the free space with a file, how isn't it fragmented.
Of course the old way to avoid this is to reserve a huge space so there is always going to be a free space to fit the file.
However this is only true if you don't use the machine much. Its how unix used to do it but there are better ways now, the same ways that windows does it.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

The same ways that Windows dows what? Avoid fragmentation? NTFS is renowned for being poor at avoiding fragmentation.
--
Today is Pungenday, the 22nd day of Confusion in the YOLD 3179
RIP Iain M Banks, 16 February 1954 – 9 June 2013
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 17/06/13 12:26, Huge wrote:

Precisely. Its no use saying 'windows could' when windows patently doesn't. Its no use saying there is nothing inherently impossible about making windows use a better file system, when windows doesn't actually use a better file system.
Its no use saying 'but windows delivers a better user experience' when we are not talking about a user experience, we are talking about 'designed to sell, not designed to be stable and reliable'
Chrome and tailfins on a chassis with cart springs, basically, and its always been that way.
That mind set of sales volume and profit against quality is evinced in every single statement Microsoft makes about itself. Look at this thread, the fanbois are there saying how many units are sold, how many people use it.,, how much money it makes. AS if that was some kind of measure of its quality.
--
Ineptocracy

(in-ep-toc’-ra-cy) – a system of government where the least capable to lead
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 17/06/2013 12:26, Huge wrote:

Which version? The ten year old version in XP? How well do ten year old linux and OSX systems work?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Unable to distinguish between and O/S and a file system, eh, dennis? It's either NTFS or it isn't.
--
Today is Pungenday, the 22nd day of Confusion in the YOLD 3179
RIP Iain M Banks, 16 February 1954 – 9 June 2013
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 17/06/2013 20:35, Huge wrote:

Its you that claimed NTFS fragments and I that asked which version as to me it obvious that the OS can write to the same file system in many ways and it will still be NTFS.
There have also been versions of NTFS BTW.
So the question stands.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 17/06/13 20:09, dennis@home wrote:

MUCH better than XP. EXT2 etc have been around for donkeys years. Unix had far far netter filesystems than FAT years ago. It had to. It was the OS for large multi-user machines featuring seriously heavy access, not a toy operating system, for single users. Linux simply picked the best around to do the job.
But of course no one has such a thing as ten year old LInix , since the upgrades are free, they get upgraded to the latest mostly.
--
Ineptocracy

(in-ep-toc’-ra-cy) – a system of government where the least capable to lead
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

That reference was talking about OSX Panther (10.3), which must be 10 years old now. I'm running 10.8.
--
Tim

"That excessive bail ought not to be required, nor excessive fines imposed,
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Extremely well. And fragmentation is a problem with all Wundows versions.
--
<•DarWin><|
_/ _/
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 16/06/2013 18:52, Vir Campestris wrote:

That logic only follows if the 1000 files are placed in consecutive locations (or some such approach).
If a system uses something like "use the smallest unfragmented area large enough for the file" or "alternate between locations A and B" as placement algorithm, then it _might_ work out very differently.
In my experience the worst culprits for fragmentation on Windows boxes are the various files such as logs which grow, and grow, and grow. At each bit of growth, albeit only a few bytes may be added, another fragment appears. The simple process of copy file/rename old/rename new/delete old manages to get rid of vast swathes of fragmentation. This has been known for ages and yet Windows never had any tools for addressing the issues, or even for accessing the huge log files in a finite universe (that is, no command like tail).
--
Rod

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 16/06/13 18:52, Vir Campestris wrote:

Because te algo goes broadly like this (or one of the does)
Create first file at outside of disk Create second file in middle of disk.
To create any new file, stick it in the middle of the largest available space.
This ensures that by and ;large files are always contiguous.
Then of course there is read ahead caching.
If getting part of a file, get a bit more as well, and stuff it in memory in case its wanted'
And the block write daemon. When writing a file, dont update the disk, shove it in a buffer. When you are near the part of the disk that a bit of file to be written exists, write it. Otherwise leave it there until you have a spare moment.
The genius of later versions of Linux is that it trends to say 'any spare RAM is diskk buffers' so provided that you are not opening a brand new file,. it tends to be the case that you never write directly to or read directly from the disk anyway. You do it from, memory. even temporary file policy. all log files and other temporary files are rotated daily and archived, or if truly temporary, are erased on reboot. Or under a timer if you set it up.

er no. SSDS have limited write lifetimes. You need even more careful strategies to avoid erasing them as much as possible.
--
Ineptocracy

(in-ep-toc’-ra-cy) – a system of government where the least capable to lead
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 16/06/2013 22:02, The Natural Philosopher wrote:

This algorithm minimises the cost of an individual file being fragmented at the expense of ensuring that any collection of files written at the same time are scattered widely all over the disc. I bet we can find benchmarks favouring both approaches.

All of the above could equally well be written about Windows, with minor variations.

Er yes. The SSD itself deals with all this rubbish, it is totally invisible to the OS.
Andy
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.