Windows ME compared to Windows 98SE

"cquirke (MVP Windows shell/user)" in
:

On Sun, 24 Sep 2006 06:49:31 -0600, Dan wrote:

I use and like 98SE and was wondering what Windows ME gives a user that
98SE does not have.

Native support for USB storage devices, e.g. card readers and flash
drives. That's the big plus, in 2006. For an older but more in-depth
assessment, see http://cquirke.mvps.org/9x/WinME.htm

The system restore feature in ME is nice but can be overcome through a
third party solution like GoBack.

SR's a mixed bag, and the WinME implimentation SUCKS compared to the
completely different way it's done in XP.

it's much easier to exclude drives from sr in winxp.

See http://cquirke.mvps.org/9x/sr-sfp.htm

Windows 98(SE) has its own resource kit and I think that Windows
ME does not have one.

True - this is one reason why so much sware was so poorly adapted to
WinME, e.g. antivirus from Norton and McAfee, etc.

"sware" rhymes with "swear"

Windows ME may have better generic drivers but that limitation has been
overcome by using a Windows ME driver in 98SE as in the case of the Ati
Radeon 9800 XT which uses the Windows ME driver in 98SE. I like the
fact of easy shut down to MS-DOS that is given in 98SE but not ME.

See http://cquirke.mvps.org/9x/ME-DOS.htm

Was Microsoft's intention to do away with the 9x source code and focus
on the NT (New Technology) source code

Yes; it had been since the Win98 days, but the consumer market didn't
consider NT 4 or Windows 2000 to be adequate replacements, let alone
worth the extra cost.


probably didn't know. i didn't know that 2k had cost more than me. i think i've seen boxes
of 2k pro, so i've assumed 2k's unpro vs pro pricing paralleled xp's home vs pro pricing

Only when XP Home offered the same cost and
better cutting-edge media and DirectX support, did we swallow NT.

and that was why Windows ME was rushed out the door?

WinME was two things:
- the last chance to get Win9x right
- a test-bed for new technologies in preparation for XP
- a way of emulating NT's weaknesses in preparation for XP

-merchandising

So it's a strange mix of polished, mature code and brand-new
semi-assed new features still dripping amniotic fluid.

i like your prose

With a bit of work, it can be better than Win98SE, but without curbing
some of the illbegotten features and artificial limitations, many
users preferred Win98SE. In fact, our trade suppliers carried stock
of Win98SE beyond the newer WinME, right into early XP days.

soon after i bought my computer i noticed ads continued to offer either.

I am not trying to diss Windows ME just because of all the bad press.

WinME can be made very nice with a bit of work, and will stay nice if
you avoid certain apps that are known not to work well with it. Many
WinME-era systems may be able to run XP if they have enough RAM, so
some folks made that switch, but if price and availability of old RAM
keeps you to under 128M, you'd do better to stay on WinME.

i think cpu matters, too. with low end xmas 2000 (800mhz) 512 ram, basic moves get a bit
slow on winxp. maybe also due to scrimped L2 cache?

On Sun, 01 Oct 2006 15:15:09 -0700, "
"cquirke (MVP Windows shell/user)"
On Sun, 24 Sep 2006 06:49:31 -0600, Dan wrote:

SR's a mixed bag, and the WinME implimentation SUCKS compared to the
completely different way it's done in XP.

it's much easier to exclude drives from sr in winxp.

Yep - and that's just one advantage; there are two more...

1) No more "full-thickness copying"

WinME stored all SR backup data in C:\_RESTORE, irrespective of which
volume it came from. Combine that with an inability to turn OFF SR
for particular volumes, and join the dots... what is normally good
performance practice (rationalize head movement) via partitioning,
becomes mired in the extra overhead of copying the full contents of
EVERY deleted or altered file all the way over to C:

Here's how the logic would have to go...

File addition
- XP and ME: Track timedate and filespec (cheap)
File deletion
- ME on C: Copy pointer to file to SR data
- ME off C: Copy full contents of entire file to C:\_RESTORE
- XP, any volume: Copy pointer to file to SVI on same volume
File change (FATxx)
- ME on C: Copy pointer to old file to SR data
- ME off C: Copy full contents of entire file to C:\_RESTORE
- XP, any volume: Copy pointer to old file to SVI on same volume

See the massive extra traffic and head travel when WinME's SR is
combined with multiple partitions esp. on same physical HD?

2) Installation-unique SR data subtree

WinME stores all SR data in C:\_RESTORE, so there should be no risks
to the SR information on other HDs dropped into the system, right?

Nope; in a stunning bit of bad judgement, SR creates \_Restore
subtrees on all other HD volumes to contain a single file that tracks
what the volume's drive letter ought to be. \_Restore may give visual
feedback that it's different to C:\_RESTORE, but the names are
identical at the file system level - so dropping in a HD from another
WinME system automatically destroys that HD's SR data, when the
newly-created \_Restore overwrites its own C:\RESTORE.

WinME will also mix up SR data if there are multiple OS installations
in the same PC, or on same HD, adding to contention over "C:\Program
Files" and installation-unique files in C:\ (e.g. MSDOS.SYS).

In contrast, XP uses a unique identifier within the SVI path, so that
each installation will avoid tripping over the SR of others. Only if
you were to clone a HD, and thus the installation, would you run the
risk of tangling things up.

3) Poos its nappies less often

In (2), we saw how dropping a HD into a WinME PC destroyed that HD's
SR data. But WinME is worse than that; it auto-destroys all of its
own SR data whenever there's a change in HD visibility. Whenever a
new HD volume is found that lacks a recorded drive letter, or that
recorded drive letter is at variance with what is assigned, all SR
data its purged. It's as if no-one had ever thought through drop-in
HD management or multiple partitions before.

Having said that, both WinME and XP share some residual SR bad design
features; chiefly, that SR is always automatically enabled whenever a
new HD volume is found, and that SR settings (which volumes are
disabled for SR and how much space to use) are lost whenever SR is
disabled and then re-enabled.

Also, XP's UI to change SR settings uses a pathetic little display
window that can show only 4 volumes without scrolling - and yep, it's
no re-sizable. For each volume, you have to click multiple times
(select, Settings, disable, OK, YES I WANT TO DISABLE) to turn it off,
and any capacity limit setting will live only as long as the SVI on
that volume lives. The SR team hasn't "got" the notion of
controllable settings for things that don't yet exist, so you can't
pre-set SR to Off for any new HD volumes discovered.

Windows 98(SE) has its own resource kit and ME does not have one.

True - this is one reason why so much sware was so poorly adapted to
WinME, e.g. antivirus from Norton and McAfee, etc.

"sware" rhymes with "swear"

Yep :-)

Yes; it had been since the Win98 days, but the consumer market didn't
consider NT 4 or Windows 2000 to be adequate replacements, let alone
worth the extra cost.

probably didn't know. i didn't know that 2k had cost more than me. i think
i've seen boxes of 2k pro, so i've assumed 2k's unpro vs pro pricing
paralleled xp's home vs pro pricing

There was no "unpro" version of any NT until XP Home, which broke the
mold on NT pricing. Until XP, NT always cost more than Win9x; when XP
Home matched Win9x in price, it powered the move off Win9x to NT.

With a bit of work, it can be better than Win98SE, but without curbing
some of the illbegotten features and artificial limitations, many
users preferred Win98SE. In fact, our trade suppliers carried stock
of Win98SE beyond the newer WinME, right into early XP days.

soon after i bought my computer i noticed ads continued to offer either.

I saw plenty of Win98SE passing over the despatch counter in the early
days of XP - and that's not just warez-bunnies ducking Product
Activation, as these were new OS sales.

WinME-era systems may be able to run XP if they have enough RAM, so
some folks made that switch, but if price and availability of old RAM
keeps you to under 128M, you'd do better to stay on WinME.

i think cpu matters, too. with low end xmas 2000 (800mhz) 512 ram,
basic moves get a bit slow on winxp. maybe also due to scrimped L2 cache?

These days I find CPU speed to be almost irrelevant, especially things
like base RAM MHz and L2 cache size. But each Windows version is
compiled to be optimized for a particular CPU generation, and that may
affect matters - AFAIK, XP is compiled for the P4 generation.

Having said that, I doubt if we'd see the same dramatic difference as
we did when Pentium Pro vs. Pentium was compared in NT, Win9x and
Win3.yuk - a situation caused by PPro's total disregard for 16-bit
performance. One of the things that changed when Pentium II came out,
was more attention paid to 16-bit code performance.

Win9x was sensitive to that because it uses quite a bit of stable,
highly-optimized GUI code that just happened to be 16-bit. It was one
of the ways Win95 could rock in 4M RAM, as NT never could.



------------ ----- --- -- - - - -
Drugs are usually safe. Inject? (Y/n)
------------ ----- --- -- - - - -

Sorry cquirke, (remember where we've been before) simply potential
discussion:

"cquirke (MVP Windows shell/user)" wrote in
message ...
| On Sun, 01 Oct 2006 15:15:09 -0700, "
| "cquirke (MVP Windows shell/user)"
| On Sun, 24 Sep 2006 06:49:31 -0600, Dan wrote:
|
| SR's a mixed bag, and the WinME implimentation SUCKS compared to the
| completely different way it's done in XP.
|
| it's much easier to exclude drives from sr in winxp.
|
| Yep - and that's just one advantage; there are two more...
|
| 1) No more "full-thickness copying"
|
| WinME stored all SR backup data in C:\_RESTORE, irrespective of which
| volume it came from. Combine that with an inability to turn OFF SR
| for particular volumes, and join the dots... what is normally good
| performance practice (rationalize head movement) via partitioning,
| becomes mired in the extra overhead of copying the full contents of
| EVERY deleted or altered file all the way over to C:

Wait a minute, if your going to addess the aspects of ME, you should also
addresss the aspects regarding XP PRO at least.
What about ALL the activity XP PRO does pursuant the restorative aspects of
the OS?

|
| Here's how the logic would have to go...
|
| File addition
| - XP and ME: Track timedate and filespec (cheap)
| File deletion
| - ME on C: Copy pointer to file to SR data
| - ME off C: Copy full contents of entire file to C:\_RESTORE
| - XP, any volume: Copy pointer to file to SVI on same volume
| File change (FATxx)
| - ME on C: Copy pointer to old file to SR data
| - ME off C: Copy full contents of entire file to C:\_RESTORE
| - XP, any volume: Copy pointer to old file to SVI on same volume
|
| See the massive extra traffic and head travel when WinME's SR is
| combined with multiple partitions esp. on same physical HD?
|
| 2) Installation-unique SR data subtree
|
| WinME stores all SR data in C:\_RESTORE, so there should be no risks
| to the SR information on other HDs dropped into the system, right?
|
| Nope; in a stunning bit of bad judgement, SR creates \_Restore
| subtrees on all other HD volumes to contain a single file that tracks
| what the volume's drive letter ought to be. \_Restore may give visual
| feedback that it's different to C:\_RESTORE, but the names are
| identical at the file system level - so dropping in a HD from another
| WinME system automatically destroys that HD's SR data, when the
| newly-created \_Restore overwrites its own C:\RESTORE.

Are you saying XP PRO does it better, or is somehow more end user friendly?
Or that ME failed to address issues as it should?

|
| WinME will also mix up SR data if there are multiple OS installations
| in the same PC, or on same HD, adding to contention over "C:\Program
| Files" and installation-unique files in C:\ (e.g. MSDOS.SYS).
|
| In contrast, XP uses a unique identifier within the SVI path, so that
| each installation will avoid tripping over the SR of others. Only if
| you were to clone a HD, and thus the installation, would you run the
| risk of tangling things up.
|
| 3) Poos its nappies less often

So that's supposedly better?

|
| In (2), we saw how dropping a HD into a WinME PC destroyed that HD's
| SR data. But WinME is worse than that; it auto-destroys all of its
| own SR data whenever there's a change in HD visibility. Whenever a
| new HD volume is found that lacks a recorded drive letter, or that
| recorded drive letter is at variance with what is assigned, all SR
| data its purged. It's as if no-one had ever thought through drop-in
| HD management or multiple partitions before.
|
| Having said that, both WinME and XP share some residual SR bad design
| features; chiefly, that SR is always automatically enabled whenever a
| new HD volume is found, and that SR settings (which volumes are
| disabled for SR and how much space to use) are lost whenever SR is
| disabled and then re-enabled.
|
| Also, XP's UI to change SR settings uses a pathetic little display
| window that can show only 4 volumes without scrolling - and yep, it's
| no re-sizable. For each volume, you have to click multiple times
| (select, Settings, disable, OK, YES I WANT TO DISABLE) to turn it off,
| and any capacity limit setting will live only as long as the SVI on
| that volume lives. The SR team hasn't "got" the notion of
| controllable settings for things that don't yet exist, so you can't
| pre-set SR to Off for any new HD volumes discovered.

Gosh, are you indicating XP dumbs down the user, or perhaps the control?

|
| Windows 98(SE) has its own resource kit and ME does not have one.
|
| True - this is one reason why so much sware was so poorly adapted to
| WinME, e.g. antivirus from Norton and McAfee, etc.
|
| "sware" rhymes with "swear"
|
| Yep :-)
|
| Yes; it had been since the Win98 days, but the consumer market didn't
| consider NT 4 or Windows 2000 to be adequate replacements, let alone
| worth the extra cost.
|
| probably didn't know. i didn't know that 2k had cost more than me. i
think
| i've seen boxes of 2k pro, so i've assumed 2k's unpro vs pro pricing
| paralleled xp's home vs pro pricing
|
| There was no "unpro" version of any NT until XP Home, which broke the
| mold on NT pricing. Until XP, NT always cost more than Win9x; when XP
| Home matched Win9x in price, it powered the move off Win9x to NT.

Uh, correct in part, but then the "betas" would be unpro, wouldn't they?
And, wasn't there several version "overseas" which would or could be labeled
"NON-PRO"?
As for the "move", there where other compelling reasons, though the "sales"
would be a primary consideration.
|
| With a bit of work, it can be better than Win98SE, but without curbing
| some of the illbegotten features and artificial limitations, many
| users preferred Win98SE. In fact, our trade suppliers carried stock
| of Win98SE beyond the newer WinME, right into early XP days.
|
| soon after i bought my computer i noticed ads continued to offer either.
|
| I saw plenty of Win98SE passing over the despatch counter in the early
| days of XP - and that's not just warez-bunnies ducking Product
| Activation, as these were new OS sales.
|
| WinME-era systems may be able to run XP if they have enough RAM, so
| some folks made that switch, but if price and availability of old RAM
| keeps you to under 128M, you'd do better to stay on WinME.
|
| i think cpu matters, too. with low end xmas 2000 (800mhz) 512 ram,
| basic moves get a bit slow on winxp. maybe also due to scrimped L2 cache?
|
| These days I find CPU speed to be almost irrelevant, especially things
| like base RAM MHz and L2 cache size. But each Windows version is
| compiled to be optimized for a particular CPU generation, and that may
| affect matters - AFAIK, XP is compiled for the P4 generation.

Single core for the most part, until the advent of the 64 bit (Yes there
was dual processor support, don't go there, unless you think something needs
addressed.).
So do you think, as I do, that XP, as a whole, was (is) merely a test bed
for Microsoft?
One need go no further than the special versions created: HOME;
PROFESSIONAL; MEDIA; etc. (and their error reports), to see the indication
,or do you see something different?

|
| Having said that, I doubt if we'd see the same dramatic difference as
| we did when Pentium Pro vs. Pentium was compared in NT, Win9x and
| Win3.yuk - a situation caused by PPro's total disregard for 16-bit
| performance. One of the things that changed when Pentium II came out,
| was more attention paid to 16-bit code performance.
|
| Win9x was sensitive to that because it uses quite a bit of stable,
| highly-optimized GUI code that just happened to be 16-bit. It was one
| of the ways Win95 could rock in 4M RAM, as NT never could.
|
|
|
| ------------ ----- --- -- - - - -
| Drugs are usually safe. Inject? (Y/n)
| ------------ ----- --- -- - - - -

--
MEB
http://peoplescounsel.orgfree.com/
BLOG http://peoplescounsel.spaces.live.com/ Public Notice or the "real
world"

"Most people, sometime in their lives, stumble across truth.
Most jump up, brush themselves off, and hurry on about their business as if
nothing had happen." Winston Churchill
Or to put it another way:
Morpheus can offer you the two pills;
but only you can choose whether you take the red pill or the blue one.
_______________

cquirke (MVP Windows shell/user) wrote:
On Sun, 01 Oct 2006 15:15:09 -0700, "
"cquirke (MVP Windows shell/user)"
On Sun, 24 Sep 2006 06:49:31 -0600, Dan wrote:

SR's a mixed bag, and the WinME implimentation SUCKS compared to the
completely different way it's done in XP.

it's much easier to exclude drives from sr in winxp.

Yep - and that's just one advantage; there are two more...

1) No more "full-thickness copying"

WinME stored all SR backup data in C:\_RESTORE, irrespective of which
volume it came from. Combine that with an inability to turn OFF SR
for particular volumes, and join the dots... what is normally good
performance practice (rationalize head movement) via partitioning,
becomes mired in the extra overhead of copying the full contents of
EVERY deleted or altered file all the way over to C:

Here's how the logic would have to go...

File addition
- XP and ME: Track timedate and filespec (cheap)
File deletion
- ME on C: Copy pointer to file to SR data
- ME off C: Copy full contents of entire file to C:\_RESTORE
- XP, any volume: Copy pointer to file to SVI on same volume
File change (FATxx)
- ME on C: Copy pointer to old file to SR data
- ME off C: Copy full contents of entire file to C:\_RESTORE
- XP, any volume: Copy pointer to old file to SVI on same volume

See the massive extra traffic and head travel when WinME's SR is
combined with multiple partitions esp. on same physical HD?

2) Installation-unique SR data subtree

WinME stores all SR data in C:\_RESTORE, so there should be no risks
to the SR information on other HDs dropped into the system, right?

Nope; in a stunning bit of bad judgement, SR creates \_Restore
subtrees on all other HD volumes to contain a single file that tracks
what the volume's drive letter ought to be. \_Restore may give visual
feedback that it's different to C:\_RESTORE, but the names are
identical at the file system level - so dropping in a HD from another
WinME system automatically destroys that HD's SR data, when the
newly-created \_Restore overwrites its own C:\RESTORE.

WinME will also mix up SR data if there are multiple OS installations
in the same PC, or on same HD, adding to contention over "C:\Program
Files" and installation-unique files in C:\ (e.g. MSDOS.SYS).

In contrast, XP uses a unique identifier within the SVI path, so that
each installation will avoid tripping over the SR of others. Only if
you were to clone a HD, and thus the installation, would you run the
risk of tangling things up.

3) Poos its nappies less often

In (2), we saw how dropping a HD into a WinME PC destroyed that HD's
SR data. But WinME is worse than that; it auto-destroys all of its
own SR data whenever there's a change in HD visibility. Whenever a
new HD volume is found that lacks a recorded drive letter, or that
recorded drive letter is at variance with what is assigned, all SR
data its purged. It's as if no-one had ever thought through drop-in
HD management or multiple partitions before.

Having said that, both WinME and XP share some residual SR bad design
features; chiefly, that SR is always automatically enabled whenever a
new HD volume is found, and that SR settings (which volumes are
disabled for SR and how much space to use) are lost whenever SR is
disabled and then re-enabled.

Also, XP's UI to change SR settings uses a pathetic little display
window that can show only 4 volumes without scrolling - and yep, it's
no re-sizable. For each volume, you have to click multiple times
(select, Settings, disable, OK, YES I WANT TO DISABLE) to turn it off,
and any capacity limit setting will live only as long as the SVI on
that volume lives. The SR team hasn't "got" the notion of
controllable settings for things that don't yet exist, so you can't
pre-set SR to Off for any new HD volumes discovered.

Windows 98(SE) has its own resource kit and ME does not have one.

True - this is one reason why so much sware was so poorly adapted to
WinME, e.g. antivirus from Norton and McAfee, etc.

"sware" rhymes with "swear"

Yep :-)

Yes; it had been since the Win98 days, but the consumer market didn't
consider NT 4 or Windows 2000 to be adequate replacements, let alone
worth the extra cost.

probably didn't know. i didn't know that 2k had cost more than me. i think
i've seen boxes of 2k pro, so i've assumed 2k's unpro vs pro pricing
paralleled xp's home vs pro pricing

There was no "unpro" version of any NT until XP Home, which broke the
mold on NT pricing. Until XP, NT always cost more than Win9x; when XP
Home matched Win9x in price, it powered the move off Win9x to NT.

With a bit of work, it can be better than Win98SE, but without curbing
some of the illbegotten features and artificial limitations, many
users preferred Win98SE. In fact, our trade suppliers carried stock
of Win98SE beyond the newer WinME, right into early XP days.

soon after i bought my computer i noticed ads continued to offer either.

I saw plenty of Win98SE passing over the despatch counter in the early
days of XP - and that's not just warez-bunnies ducking Product
Activation, as these were new OS sales.

WinME-era systems may be able to run XP if they have enough RAM, so
some folks made that switch, but if price and availability of old RAM
keeps you to under 128M, you'd do better to stay on WinME.

i think cpu matters, too. with low end xmas 2000 (800mhz) 512 ram,
basic moves get a bit slow on winxp. maybe also due to scrimped L2 cache?

These days I find CPU speed to be almost irrelevant, especially things
like base RAM MHz and L2 cache size. But each Windows version is
compiled to be optimized for a particular CPU generation, and that may
affect matters - AFAIK, XP is compiled for the P4 generation.

Having said that, I doubt if we'd see the same dramatic difference as
we did when Pentium Pro vs. Pentium was compared in NT, Win9x and
Win3.yuk - a situation caused by PPro's total disregard for 16-bit
performance. One of the things that changed when Pentium II came out,
was more attention paid to 16-bit code performance.

Win9x was sensitive to that because it uses quite a bit of stable,
highly-optimized GUI code that just happened to be 16-bit. It was one
of the ways Win95 could rock in 4M RAM, as NT never could.



------------ ----- --- -- - - - -
Drugs are usually safe. Inject? (Y/n)
------------ ----- --- -- - - - -

I like your last comment and I sincerely agree with it. On my box, I
have worked hard to make sure everything works and it is over three
generations. ----- 98 Second Edition (my favorite -- due to ability to
get into roots of operating system and none of the annoying hide this
and that because XP Professional just amuses people are dumb and does
not give any credibility to the user unless needed. BTW, Chris Quirk,
MVP -- I have now been given remote assistance rights to service the XP
Professional computers at work. The thing is that the 98 Second Edition
computers exist much more so as individual units and this is so nice.
The kids at our school love the 98 Second Edition computers much better
and gravitate to them first before the XP Professional computers if
there is a choice. That is yet another reason why Microsoft must make
the right choice and release my version of Windows Classic Edition!

--
Dan W.

Computer User

MEB wrote:
Sorry cquirke, (remember where we've been before) simply potential
discussion:

"cquirke (MVP Windows shell/user)" wrote in
message ...
| On Sun, 01 Oct 2006 15:15:09 -0700, "
| "cquirke (MVP Windows shell/user)"
| On Sun, 24 Sep 2006 06:49:31 -0600, Dan wrote:
|
| SR's a mixed bag, and the WinME implimentation SUCKS compared to the
| completely different way it's done in XP.
|
| it's much easier to exclude drives from sr in winxp.
|
| Yep - and that's just one advantage; there are two more...
|
| 1) No more "full-thickness copying"
|
| WinME stored all SR backup data in C:\_RESTORE, irrespective of which
| volume it came from. Combine that with an inability to turn OFF SR
| for particular volumes, and join the dots... what is normally good
| performance practice (rationalize head movement) via partitioning,
| becomes mired in the extra overhead of copying the full contents of
| EVERY deleted or altered file all the way over to C:

Wait a minute, if your going to addess the aspects of ME, you should also
addresss the aspects regarding XP PRO at least.
What about ALL the activity XP PRO does pursuant the restorative aspects of
the OS?

|
| Here's how the logic would have to go...
|
| File addition
| - XP and ME: Track timedate and filespec (cheap)
| File deletion
| - ME on C: Copy pointer to file to SR data
| - ME off C: Copy full contents of entire file to C:\_RESTORE
| - XP, any volume: Copy pointer to file to SVI on same volume
| File change (FATxx)
| - ME on C: Copy pointer to old file to SR data
| - ME off C: Copy full contents of entire file to C:\_RESTORE
| - XP, any volume: Copy pointer to old file to SVI on same volume
|
| See the massive extra traffic and head travel when WinME's SR is
| combined with multiple partitions esp. on same physical HD?
|
| 2) Installation-unique SR data subtree
|
| WinME stores all SR data in C:\_RESTORE, so there should be no risks
| to the SR information on other HDs dropped into the system, right?
|
| Nope; in a stunning bit of bad judgement, SR creates \_Restore
| subtrees on all other HD volumes to contain a single file that tracks
| what the volume's drive letter ought to be. \_Restore may give visual
| feedback that it's different to C:\_RESTORE, but the names are
| identical at the file system level - so dropping in a HD from another
| WinME system automatically destroys that HD's SR data, when the
| newly-created \_Restore overwrites its own C:\RESTORE.

Are you saying XP PRO does it better, or is somehow more end user friendly?
Or that ME failed to address issues as it should?

|
| WinME will also mix up SR data if there are multiple OS installations
| in the same PC, or on same HD, adding to contention over "C:\Program
| Files" and installation-unique files in C:\ (e.g. MSDOS.SYS).
|
| In contrast, XP uses a unique identifier within the SVI path, so that
| each installation will avoid tripping over the SR of others. Only if
| you were to clone a HD, and thus the installation, would you run the
| risk of tangling things up.
|
| 3) Poos its nappies less often

So that's supposedly better?

|
| In (2), we saw how dropping a HD into a WinME PC destroyed that HD's
| SR data. But WinME is worse than that; it auto-destroys all of its
| own SR data whenever there's a change in HD visibility. Whenever a
| new HD volume is found that lacks a recorded drive letter, or that
| recorded drive letter is at variance with what is assigned, all SR
| data its purged. It's as if no-one had ever thought through drop-in
| HD management or multiple partitions before.
|
| Having said that, both WinME and XP share some residual SR bad design
| features; chiefly, that SR is always automatically enabled whenever a
| new HD volume is found, and that SR settings (which volumes are
| disabled for SR and how much space to use) are lost whenever SR is
| disabled and then re-enabled.
|
| Also, XP's UI to change SR settings uses a pathetic little display
| window that can show only 4 volumes without scrolling - and yep, it's
| no re-sizable. For each volume, you have to click multiple times
| (select, Settings, disable, OK, YES I WANT TO DISABLE) to turn it off,
| and any capacity limit setting will live only as long as the SVI on
| that volume lives. The SR team hasn't "got" the notion of
| controllable settings for things that don't yet exist, so you can't
| pre-set SR to Off for any new HD volumes discovered.

Gosh, are you indicating XP dumbs down the user, or perhaps the control?

|
| Windows 98(SE) has its own resource kit and ME does not have one.
|
| True - this is one reason why so much sware was so poorly adapted to
| WinME, e.g. antivirus from Norton and McAfee, etc.
|
| "sware" rhymes with "swear"
|
| Yep :-)
|
| Yes; it had been since the Win98 days, but the consumer market didn't
| consider NT 4 or Windows 2000 to be adequate replacements, let alone
| worth the extra cost.
|
| probably didn't know. i didn't know that 2k had cost more than me. i
think
| i've seen boxes of 2k pro, so i've assumed 2k's unpro vs pro pricing
| paralleled xp's home vs pro pricing
|
| There was no "unpro" version of any NT until XP Home, which broke the
| mold on NT pricing. Until XP, NT always cost more than Win9x; when XP
| Home matched Win9x in price, it powered the move off Win9x to NT.

Uh, correct in part, but then the "betas" would be unpro, wouldn't they?
And, wasn't there several version "overseas" which would or could be labeled
"NON-PRO"?
As for the "move", there where other compelling reasons, though the "sales"
would be a primary consideration.
|
| With a bit of work, it can be better than Win98SE, but without curbing
| some of the illbegotten features and artificial limitations, many
| users preferred Win98SE. In fact, our trade suppliers carried stock
| of Win98SE beyond the newer WinME, right into early XP days.
|
| soon after i bought my computer i noticed ads continued to offer either.
|
| I saw plenty of Win98SE passing over the despatch counter in the early
| days of XP - and that's not just warez-bunnies ducking Product
| Activation, as these were new OS sales.
|
| WinME-era systems may be able to run XP if they have enough RAM, so
| some folks made that switch, but if price and availability of old RAM
| keeps you to under 128M, you'd do better to stay on WinME.
|
| i think cpu matters, too. with low end xmas 2000 (800mhz) 512 ram,
| basic moves get a bit slow on winxp. maybe also due to scrimped L2 cache?
|
| These days I find CPU speed to be almost irrelevant, especially things
| like base RAM MHz and L2 cache size. But each Windows version is
| compiled to be optimized for a particular CPU generation, and that may
| affect matters - AFAIK, XP is compiled for the P4 generation.

Single core for the most part, until the advent of the 64 bit (Yes there
was dual processor support, don't go there, unless you think something needs
addressed.).
So do you think, as I do, that XP, as a whole, was (is) merely a test bed
for Microsoft?
One need go no further than the special versions created: HOME;
PROFESSIONAL; MEDIA; etc. (and their error reports), to see the indication
,or do you see something different?

|
| Having said that, I doubt if we'd see the same dramatic difference as
| we did when Pentium Pro vs. Pentium was compared in NT, Win9x and
| Win3.yuk - a situation caused by PPro's total disregard for 16-bit
| performance. One of the things that changed when Pentium II came out,
| was more attention paid to 16-bit code performance.
|
| Win9x was sensitive to that because it uses quite a bit of stable,
| highly-optimized GUI code that just happened to be 16-bit. It was one
| of the ways Win95 could rock in 4M RAM, as NT never could.
|
|
|
| ------------ ----- --- -- - - - -
| Drugs are usually safe. Inject? (Y/n)
| ------------ ----- --- -- - - - -


Windows XP as a test version --- hmm -- interesting concept and I will
see what Microsoft says about this theory.

XP does have the 100% CPU flaw that pops up every now and then
especially when the system is trying to read data and cannot read it.
XP Professional seems to be instructed by Microsoft --- okay we cannot
read x in current state so then let us jump x current state to maximum y
state and perhaps this will let us read x current state --- although it
sadly hardly ever seems to help and this feature should just be removed
since it is worth than useless in my opinion and may cause faster wear
and tear on the Central Processing Unit by bringing it up to 100% for an
extended time if XP Professional cannot read what it is given. 98
Second Edition will just eventually balk with an error and not stupidly
try to go to 100% to somehow achieve a state of readability that hardly
ever happens to help the system.

Fortunately, I have yet to notice this aka "feature" in Windows Vista
and hopefully was thankfully removed from Windows Vista. One can only
hope that the feature will be removed soon from XP Professional and the
sooner the better but definitely by Service Pack 3 for XP.

--
Dan W.

Computer User

On Wed, 4 Oct 2006 20:46:52 -0400, "MEB" meb@not
"cquirke (MVP Windows shell/user)"
| On Sun, 01 Oct 2006 15:15:09 -0700, "
| "cquirke (MVP Windows shell/user)"
| On Sun, 24 Sep 2006 06:49:31 -0600, Dan wrote:

| it's much easier to exclude drives from sr in winxp.

| Yep - and that's just one advantage; there are two more...

| 1) No more "full-thickness copying"

| WinME stored all SR backup data in C:\_RESTORE, irrespective of which
| volume it came from. Combine that with an inability to turn OFF SR
| for particular volumes, and join the dots...

Wait a minute, if your going to addess the aspects of ME, you should also
addresss the aspects regarding XP PRO at least.

OK...

What about ALL the activity XP PRO does pursuant the restorative aspects of
the OS?

Sorry, I don't understand what you're asking?

| 2) Installation-unique SR data subtree

| WinME stores all SR data in C:\_RESTORE, ... SR creates \_Restore
| subtrees on all other HD volumes to contain a single file that tracks
| what the volume's drive letter ought to be. ...the names are
| identical at the file system level - so dropping in a HD from another
| WinME system automatically destroys that HD's SR data

Are you saying XP PRO does it better,

XP (Home or Pro) does it MUCH better, oh yes ;-)

Each XP installation uses a unique identifier within the path to the
SR backups and associated data, so that one installation's SR never
overwrites that of another.

A common caveat applies, i.e. that this safety is lost if the same
installation is cloned, and the clones see each other.

Or that ME failed to address issues as it should?

Yep, the WinME team should have seen this issue coming at the
pen-and-napkin stage of the design. OTOH, you could say XP learned
from the "SR 1.0" mistakes made in WinME.

| 3) Poos its nappies less often

So that's supposedly better?

Well yes; if you're going to suffer the overhead of creating and
maintaining SR backup material and restore points, then you want to
get the most benefit from this as possible - especially if you are
going to encourage users to rely on SR, and/or rely on it yourself.

For example, drivers and other sware often make a restore point before
installing, on the assumption that you can undo the installation by
doing a System Restore - and therefore, you can skip providing a
formal uninstaller, right?

For another example, XP lacks Win98's robust automatic registry
backup. Whereas Win98 will keep the last 5 full file copies of all
registry files, XP backs up a small part of one hive within the same
file as the hive it's backing up, 1 copy deep. The assumption is that
because SR includes registry backup with each restore point, there's
no need to maintain registry backups anymore.

One could use SR's process to log registry and file system changes,
i.e. the potential to act as a "change reporter" is built into it -
95% of the work is done, all they'd have to do is generate difference
lists and present these through some sort of UI. Not done.

OTOH, WinME irreversably discards all SR data at the drop of a hat (or
more literally, the change of a drive letter). How can you rely on
something that's likely to have failed when you need it?

| Also, XP's UI to change SR settings uses a pathetic little display
| window that can show only 4 volumes without scrolling - and yep, it's
| no re-sizable. The SR team hasn't "got" the notion of controllable
| settings for things that don't yet exist, so you can't pre-set SR to
| Off for any new HD volumes discovered.

Gosh, are you indicating XP dumbs down the user, or perhaps the control?

There are three common and recurrent weaknesses he

1) Use the Win3.x feature set, Luke

Ever since Win3.yuk, we've had the ability to resize windows. So why
are so many dialog boxes non-resizable, even in Vista? You'd think
after over a decade, programmers would start to use the GUI feature
set. IMO, ANY dialog box that contains an unbounded number of items
should be resizeable, and should ideally remember the last size used.

We know we might have up to 24 drive letters used within the system
for HD volumes subject to SR control. Yet there's a gratuitous waste
of UI space within the SR dialog that could and should be used to
double the number of letters shown, thus halve the amount of
scrolling, even before you consider making the dialog resizeable.

Check it out... there's a solid slab of useless do-nothing dialog-box
grey under a 4-item-deep keyhole. Un believable!

2) Treat user settings with respect

MS is quite prone to losing user settings, especially when updates are
applied, Windows is "just" re-installed, or during more subtle context
changes. I see this as a reflection of coding quality in general,
i.e. the visible part of the same iceberg that opens up various
cross-context exploits and escalations.

For example, XP appears to store the SR capacity quota within the
volume's SVI subtree, rather than in the registry. There are good and
bad aspects to that; the good is there's an unambiguous link between
the volume and its associated setting, but the bad is that if you
disable SR and then re-enable it, and you clear the SVI while SR is
disabled, the setting will be lost.

This reflects a particular settings retention problem with MS; that
settings are lost when the applicable state changes, and are
re-defaulted when that state is restored.

3) Manage items that don't yet exist

This is a well-known requirement in database programming and record
locking. To create a new record, you have to lock that record (even
though it does not yet exist) so that no other entity can create the
same record while you do.

Newly-created items - think user accounts, newly-discovered HD
volumes, view options for new folders - inherit their settings from
some sort of prototype. All too often, that prototype is a set of
hardwired MS duhfaults, or there is poor or no user access to
controlling that prototype.

That would matter less if the defaults were maximally safe, but
invariably they are not - think maximum SR space allocation and
activity, hiding of crucial risk indicators, that sort of thing.

| probably didn't know. i didn't know that 2k had cost more than me. i
think
| i've seen boxes of 2k pro, so i've assumed 2k's unpro vs pro pricing
| paralleled xp's home vs pro pricing
|
| There was no "unpro" version of any NT until XP Home, which broke the
| mold on NT pricing. Until XP, NT always cost more than Win9x; when XP
| Home matched Win9x in price, it powered the move off Win9x to NT.

Uh, correct in part, but then the "betas" would be unpro, wouldn't they?

Eh?? Cummer gain? I don't understand what you're asking here.

Betas are usually "ultimate" edition, i.e. the one with everything in
it. Understandable; means the most functionality is exposed for
testing and the rich feature set attracts testers to the programme.

But the weakness is that the cut-lines defining sub-"ultimate"
editions may be under-tested. You may well find there's some hidden
functionality from a "higher" edition that's been left in the "lower"
edition, with no UI to control it. Hello, exploits.

And, wasn't there several version "overseas" which would or could be labeled
"NON-PRO"?

Overseas editions may be limited for various reasons:
- new features haven't been internationalized yet
- local law or litigation prohibits the provision of a feature
- a market need for a far cheaper edition (e.g. to combat piracy)

So we see Windows without Media Player to comply with EU strictures,
and a "starter" edition of Windows as an entry-level product in poorer
markets where piracy is rife, etc.

All NT was "pro" until XP, when XP Home was spun off to fit the Win9x
price structure. Some features stripped in XP Home are
understandable, e.g. support for domain login, while others are
arguably welcome risk management, e.g. EFS, network access to hidden
"admin" shares. However, some are gratuitous, such as limiting users
to 5 rather than 10 incoming network connections....
- 10: Win95, Win98, NT Workstation, Win2000 Pro, XP Pro
- 5: WinME, XP Home

That's one reason for some users to choose Win98SE over WinME.

| These days I find CPU speed to be almost irrelevant, especially things
| like base RAM MHz and L2 cache size. But each Windows version is
| compiled to be optimized for a particular CPU generation, and that may
| affect matters - AFAIK, XP is compiled for the P4 generation.

Single core for the most part, until the advent of the 64 bit (Yes there
was dual processor support, don't go there, unless you think something
needs addressed.).

Like most processor faster-at-same-GHz enhancements, dual core is only
faster if sware uses the enhancement, in this case the spare core.

So do you think, as I do, that XP, as a whole, was (is) merely a test bed
for Microsoft?

No, I don't - that test-bed was WinME :-)

XP was a point revision of Win2000 (i.e. NT 5.1 to Win2000's 5.0),
which was itself a mature evolution of NT. NT had always enjoyed a
slower pace of development - which is why it took so long to catch up:
- FAT32; not there until Win2000
- Defrag; not there until Win2000
- Scandisk; still waiting...
- maintenance OS; still waiting...
- Plug-n-Play; evolved through Win2000 to XP
- DirectX; always a version or few behind, until XP

The slower pace means more reliable code, as a rule.

What changed with XP was that ongoing development of rapidly-evolving
subsystems - media, DirectX, Plug-n-Play, new hardware standards - was
now done first on the NT code base, and then back-ported to Win9x, if
at all. So XP broke the mold; it came out with the very latest in
DirextX, media and hardware support. Combine that with Win9x pricing
and a dwindling interest in DOS/Win16 compat, and it paid off.

OTOH, XP was the first attempt to really expose NT to consumer useage.
I'd like to tell you that MS has learned from this, but to be honest,
I don't think they have - if you talk "home users", invariably MS will
talk about pretty UI, media flash, and ease-of-use. There's no
awareness that the fundamental design of pro-IT network client is
dangerously inappropriate for consumer use on the Internet.



------------ ----- --- -- - - - -
Drugs are usually safe. Inject? (Y/n)
------------ ----- --- -- - - - -

"cquirke (MVP Windows shell/user)" wrote in
message ...
| On Wed, 4 Oct 2006 20:46:52 -0400, "MEB" meb@not
| "cquirke (MVP Windows shell/user)"
| | On Sun, 01 Oct 2006 15:15:09 -0700, "
| | "cquirke (MVP Windows shell/user)"
| | On Sun, 24 Sep 2006 06:49:31 -0600, Dan wrote:
|
| | it's much easier to exclude drives from sr in winxp.
|
| | Yep - and that's just one advantage; there are two more...
|
| | 1) No more "full-thickness copying"
|
| | WinME stored all SR backup data in C:\_RESTORE, irrespective of which
| | volume it came from. Combine that with an inability to turn OFF SR
| | for particular volumes, and join the dots...
|
| Wait a minute, if your going to addess the aspects of ME, you should
also
| addresss the aspects regarding XP PRO at least.
|
| OK...
|
| What about ALL the activity XP PRO does pursuant the restorative aspects
of
| the OS?
|
| Sorry, I don't understand what you're asking?

Well it was originally posted pursuant to creation of the restoration
points in XP and the other activities which you have now described, but
having been sweetly reminded by mae of the "purpose" of the threads, and
reminded of lengthy discussion elsewhere, perhaps, a separate discussion
would have been in order. This was ME verses 98SE, hence my move to XP was
questionable, whereas yours merely responded to outside input.

The topic is particularly ripe, especially per your response. I'll wait for
an OKAY here for my response, or the new thread.

BTW, nice to see you back.

--
MEB
http://peoplescounsel.orgfree.com/
BLOG http://peoplescounsel.spaces.live.com/ Public Notice or the "real
world"

"Most people, sometime in their lives, stumble across truth.
Most jump up, brush themselves off, and hurry on about their business as if
nothing had happen." Winston Churchill
Or to put it another way:
Morpheus can offer you the two pills;
but only you can choose whether you take the red pill or the blue one.
_______________

On Sat, 7 Oct 2006 16:52:04 -0400, "MEB"

Well it was originally posted pursuant to creation of the restoration
points in XP and the other activities which you have now described, but
having been sweetly reminded by mae of the "purpose" of the threads, and
reminded of lengthy discussion elsewhere, perhaps, a separate discussion
would have been in order. This was ME verses 98SE, hence my move to XP was
questionable, whereas yours merely responded to outside input.

Oh, OK... most SR topics come up in WinME, and it seems inevitable to
compare how SR was done then and how it's been done since.

BTW, nice to see you back.

Thanks! I'm alone again (my partner's on a distant land mass) so
there's a bit more of me to go round :-)



------------ ----- --- -- - - - -
Drugs are usually safe. Inject? (Y/n)
------------ ----- --- -- - - - -

I'm glad to see you back, too, cquirke.


--
Thanks or Good Luck,
There may be humor in this post, and,
Naturally, you will not sue,
should things get worse after this,
PCR

"cquirke (MVP Windows shell/user)" wrote
in message ...
| On Sat, 7 Oct 2006 16:52:04 -0400, "MEB"
|
| Well it was originally posted pursuant to creation of the
restoration
| points in XP and the other activities which you have now described,
but
| having been sweetly reminded by mae of the "purpose" of the threads,
and
| reminded of lengthy discussion elsewhere, perhaps, a separate
discussion
| would have been in order. This was ME verses 98SE, hence my move to
XP was
| questionable, whereas yours merely responded to outside input.
|
| Oh, OK... most SR topics come up in WinME, and it seems inevitable to
| compare how SR was done then and how it's been done since.
|
| BTW, nice to see you back.
|
| Thanks! I'm alone again (my partner's on a distant land mass) so
| there's a bit more of me to go round :-)
|
|
|
| ------------ ----- --- -- - - - -
| Drugs are usually safe. Inject? (Y/n)
| ------------ ----- --- -- - - - -