Avoid 10 !

Ilya Chernykh wrote:
wrote in message
...

Starting with the introduction of 10, I posted ( repeatedly ) for
anyone to give me one good reason to "upgrade" from 7 to 10. So far
nobody has been able to do that.

I have a few reasons to upgrade from Win7 to Win8.1. They apply to Win10 as
well.
On the other hand I see no reason to upgrade from Win8.1 to Win10.

The reasons to upgrade from Win7 to Win8.1 include:

* On Win8.1 one can use Classic theme with DWM enabled, on Win7 one cannot.

* A few bugs/glitches in the theme were fixed in Win8, such as a black pixel
on the
leftmost tab in tabbed dialogs and garbage on the window corner after
resizing.

* Quick start. Also the fact it keeps AMD Overdrive settings unlike full
reboot.

* Boot menu on the external monitor (I use laptop in the role of desktop,
with external monitor).

* Better hardware support, more universal printer drivers, newer driver
model,
USB 3.0 support out of the box.

* Seems working faster for me.

------------------------

Win10 on the other hand, has tons of disadvantages compared to Win8.1.

When I tested 7 versus 8.1 versus 10, the 7 OS won by a small amount.
The difference between the three was not important. If I "needed more horsepower",
I would not waste the time rebooting to use another OS. That's how
tiny the difference is.

Now that Meltdown/Spectre patches are issued, that will change a bit,
and the results will vary more with application type. Now, it will
be harder to devise a test plan for a benchmark that is fair.

Paul

On Fri, 12 Jan 2018 05:58:51 -0500, wrote:

Anyone with half a brain, would have stayed with 7, which was and is
optimal. I did. I occasionally peruse this group to see what I'm
missing. Whew ! ...

Another troll that never tried 10. It can be tamed and works great.


KenW

"GS" wrote

| I don't know all the details. I just made sure
| none of my software would work until I'd got
| around to updating and testing.
|
| Keep in mind that since Vista is v6.0, that Win7 (v6.1) is really just a
Vista
| Service Pack. (IMO)
|
I'd agree with that. That's why I often refer
to Vista/7. I think of them as pretty much the
same. I think Vista got a bad rap because the
hardware wasn't ready for the bloat, and it was
rushed out after Longhorn failed.

What I meant was that I had written a number of
ActiveX components, mostly for scripting, on XP.
With Vista I didn't want them to run until I'd tested
them, but it turned out Windows was lying about
what version they were running on with Vista!

I got one email from an angry man who seemed
to think I'd cheated him by not yet supporting Vista.
He spitefully pointed out that he'd got around the
limitation. I realized someone like that would
probably also blame me if he was managing to
run my components on Vista and something
didn't work. Eventually I rewrote/tested many
of the components, but in the meantime I added
code to each one to bypass Windows lying and
find the real version, then block functionality if
it was later than XP.

I suspect the systematic Windows lying and
"virtualization" boondoggle are connected with the
mixed results people have using compat mode.
Microsoft figured they'd just trick older software
into thinking it could write to the Registry and
save files on restricted paths, using "virtualization",
and that would magically create compatibility.
But it was really more like an unofficial, half-baked
sandboxing, in which Microsoft were really
overstepping their responsibilities in providing
a software platform, in order to make the new
product look good.

On Sat, 13 Jan 2018 03:28:37 -0000 (UTC), Roger Blake
wrote:



Windows 10 in its default, out-of-the-box state is obnoxious as hell.



You're saying it much more strongly than I would, but I generally
agree. But as far as I'm concerned, much the same is true of most
software. Lots of major configuration improvements can be made to
their out-of-the-box states.

If you were to look at my computer, running Windows 10, you might have
a hard time even recognizing it as Windows 10, since it's very far
from its out-of-the-box state.




That stuff should be opt-in rather
than opt-out.



Yes. Again, I feel much the same way about most software.

And what's most important to me is that if I can configure a piece of
software to be the way I like it, I really don't care what its
out-of-the-box state was. If I like it when it's in the state I
configure it to, then it's a product I like.

Here's an example of what I mean by the paragraph above: I greatly
dislike FireFox, Chrome, and Edge in their out-of-the-box states. But
I can configure FireFox (and have configured it) to be in a state I
like very much. Since I can't configure Chrome or Edge the same way, I
like FireFox and I hate Chrome and Edge.

"GS" wrote

I don't know all the details. I just made sure
none of my software would work until I'd got
around to updating and testing.

Keep in mind that since Vista is v6.0, that Win7 (v6.1) is really just a
Vista Service Pack. (IMO)

I'd agree with that. That's why I often refer
to Vista/7. I think of them as pretty much the
same. I think Vista got a bad rap because the
hardware wasn't ready for the bloat, and it was
rushed out after Longhorn failed.

What I meant was that I had written a number of
ActiveX components, mostly for scripting, on XP.
With Vista I didn't want them to run until I'd tested
them, but it turned out Windows was lying about
what version they were running on with Vista!

I got one email from an angry man who seemed
to think I'd cheated him by not yet supporting Vista.
He spitefully pointed out that he'd got around the
limitation. I realized someone like that would
probably also blame me if he was managing to
run my components on Vista and something
didn't work. Eventually I rewrote/tested many
of the components, but in the meantime I added
code to each one to bypass Windows lying and
find the real version, then block functionality if
it was later than XP.

I suspect the systematic Windows lying and
"virtualization" boondoggle are connected with the
mixed results people have using compat mode.
Microsoft figured they'd just trick older software
into thinking it could write to the Registry and
save files on restricted paths, using "virtualization",
and that would magically create compatibility.
But it was really more like an unofficial, half-baked
sandboxing, in which Microsoft were really
overstepping their responsibilities in providing
a software platform, in order to make the new
product look good.

I don't follow this stuff as a rule, but your brand of thoroughness in your
posts makes for very interesting reading. I really appreciate you sharing you
knowledge!

--
Garry

Free usenet access at http://www.eternal-september.org
Classic VB Users Regroup!
comp.lang.basic.visual.misc
microsoft.public.vb.general.discussion

On Fri, 12 Jan 2018 13:17:00 -0800, Freely wrote:

Oh please.

Most Win 10 users do not know what they are doing and use the very
minimal capability of this overly complicated OS.

With the exception of maybe corporate users that have IT folks down the
hall.

MS keeps changing the user interface as an example of poor ergonomics.
You have to relearn the OS all over again each time a now major release
comes out.

Although some deep components might not change much, they way you get to
them are totally different.

We should all go back to CP/M !!!

80 or 86?

On 1/13/2018 3:05 AM, Paul wrote:

When I tested 7 versus 8.1 versus 10, the 7 OS won by a small amount.
The difference between the three was not important. If I "needed more
horsepower",
I would not waste the time rebooting to use another OS. That's how
tiny the difference is.

Now that Meltdown/Spectre patches are issued, that will change a bit,
and the results will vary more with application type. Now, it will
be harder to devise a test plan for a benchmark that is fair.

Â*Â* Paul
As I see it, the problem with this kind of "testing" is that you have to
restrict the apps to those which run on Win7. That is awfully
presumptuous about what users are doing these days.

--
best regards,

Neil

On 13/01/2018 13:21, KenW wrote:
Another troll that never tried 10. It can be tamed and works great.




Not another troll!! It is the same SteveGG and Windows10Hater. The
name keeps changing but the spots remains the same. It's very difficult
to change the spots. Black people can try all the bleaches in the world
but the skin reverts to blackness when the bleach effect has died out.




--
With over 600 million devices now running Windows 10, customer
satisfaction is higher than any previous version of windows.

"GS" wrote

| I don't follow this stuff as a rule, but your brand of thoroughness in
your
| posts makes for very interesting reading. I really appreciate you sharing
you
| knowledge!
|

Thank you. It's nice to know I'm not
straining *everyone's* patience.

Neil wrote:
On 1/13/2018 3:05 AM, Paul wrote:

When I tested 7 versus 8.1 versus 10, the 7 OS won by a small amount.
The difference between the three was not important. If I "needed more
horsepower",
I would not waste the time rebooting to use another OS. That's how
tiny the difference is.

Now that Meltdown/Spectre patches are issued, that will change a bit,
and the results will vary more with application type. Now, it will
be harder to devise a test plan for a benchmark that is fair.

Paul

As I see it, the problem with this kind of "testing" is that you have to
restrict the apps to those which run on Win7. That is awfully
presumptuous about what users are doing these days.

Are you talking about Store Apps (Metro flavor) ?

Of course they don't run on Win7.

But, on the other hand, they're not worth owning, either :-)
Who would need to benchmark a piece of glitzy graphical trash
written in HTML/Javascript ? I don't. Can we benchmark
"Groove" or "Skype" ? What would that even mean ?

A Win32 executable is "close to the iron" and shows
what the main engine (kernel and process/thread scheduler)
can do. I can run 7ZIP on the three platforms, and
get to see scheduling, page table handling for large
memory arrays, and what percentage of CPU the OS is
willing to give up, all in one run.

I'm suffering no loss here, by not visiting the App Store.

If I want Angry Birds, my back yard is full of them.

Paul

On 1/13/2018 4:49 PM, Paul wrote:
Neil wrote:
On 1/13/2018 3:05 AM, Paul wrote:

When I tested 7 versus 8.1 versus 10, the 7 OS won by a small amount.
The difference between the three was not important. If I "needed more
horsepower",
I would not waste the time rebooting to use another OS. That's how
tiny the difference is.

Now that Meltdown/Spectre patches are issued, that will change a bit,
and the results will vary more with application type. Now, it will
be harder to devise a test plan for a benchmark that is fair.

Â*Â*Â* Paul

As I see it, the problem with this kind of "testing" is that you have
to restrict the apps to those which run on Win7. That is awfully
presumptuous about what users are doing these days.

Are you talking about Store Apps (Metro flavor) ?

Of course they don't run on Win7.

No, not necessarily, but those also won't run on Win7.

But, on the other hand, they're not worth owning, either :-)
Who would need to benchmark a piece of glitzy graphical trash
written in HTML/Javascript ? I don't. Can we benchmark
"Groove" or "Skype" ? What would that even mean ?

I see. What people in the world are doing only matters if it matches
your opinion of what is worth doing. That's a ****-poor test design.

--
best regards,

Neil

On 01/12/2018 10:21 PM, GS wrote:

[snip]

Keep in mind that since Vista is v6.0, that Win7 (v6.1) is really just a
Vista Service Pack. (IMO)

Some of the early (prerelease) versions of 10 identified it as v6.4,
which has since changed to 10.

--
Mark Lloyd
http://notstupid.us/

"The beating of humanity's heart cannot be felt by placing the finger on
the church's pulse." [Lemuel K. Washburn, _Is The Bible Worth Reading
And Other Essays_]

Neil wrote:


I see. What people in the world are doing only matters if it matches
your opinion of what is worth doing. That's a ****-poor test design.

Bud, if you add too many layers to a software
stack, what are you testing exactly ? Which layer
is broken ? Which layer sucks ?

Let's take an absurd example. I drive to your
company at work Dec.2016 and ask how many widgets you
made that year. I drive up Dec.2017 and ask
how many widget you made that year. Now, I compare.
Can I draw any conclusions ? Have I controlled
for the economic conditions ? How many variables
could account for more or fewer widgets. Fewer staff ?
Union trouble ? Bad government tax scheme ? High
interest rates ? How do we isolate for just
productivity issues on your plant floor ?

The deeper a stack of stuff you're looking at, the
harder it is to draw any conclusions.

The "speed of an OS" is pretty simple. On a compute
bound process, how many cycles did it get ? How many
cycles did the OS waste ? It's that simple. The DE that
rests on top of it is irrelevant. Does it "compute"
or doesn't it ?

Now, if you wanted to say something like "the Win10 DE
is 79% faster than the Win7 DE", then that would
be a proper "statement-in-isolation". But, when you
make that statement, you're going to have to
demonstrate to people, how you isolated for
only that one factor. Which would be difficult to
do. If I'd thought of a simple way to do that
by now, I would have impressed you with facts
and figures about Win7 DE versus Win10 DE. As long as
"DE isolation" remains illusive, I can only report the
stuff that is easy to test, and also actually aligns with
the operating system proper. Which is the underneath part.

If I did say "the Win10 DE is 79% faster than the Win7 DE",
people would proceed to tell me I was full of ****,
and pick apart the experimental design. That would be
my reward for such an effort. "Prove it" would be
the response I'd get.

In Linux, I'd have more flexibility. I could move
a DE from the Linux equivalent of Win7 versus Win10.
*But*, the subsystems the OS provide, change with
time. DBUS appears and disappears. The DE I want to
test, might be making DBUS calls. Then I'm screwed.
So Linux comes closer to making such testing
possible, but without any guarantees. You might
not get very far trying that. And even without
doing such experiments, I can tell you that running
a Compiz based DE on top of ancient hardware,
is *dreadfully slow*. For some things, I don't
even need to present a test case, as the user
will have no trouble spotting just how bad it is.

And now that I've thought of the Compiz example, let's
consider Windows 10 running on top of an FX5200 with the
Microsoft Basic Display Adapter. There's no video card
acceleration. If you watch the animations (I have), the
animations in Windows 10 are perceptibly slower than
for video cards where the OS has a Nvidia/AMD driver.
But I don't have nice numbers for you, and of course
you're free to reject hardware that isn't convenient
in a discussion.

Paul

wrote in message ...

Anyone with half a brain, would have stayed with 7, which was and is
optimal. I did. I occasionally peruse this group to see what I'm
missing. Whew ! ...

Ha ha ha ha. Please stay in the appro NG, or perhaps you are thinking of
upgrading to 10 for better security etc and, and ,and.
Go bitch to your other half.
--
Buffalo

On 01/13/2018 7:54 PM, Buffalo wrote:
wrote in message ...

Anyone with half a brain, would have stayed with 7, which was and is
optimal. I did. I occasionally peruse this group to see what I'm
missing. Whew ! ...

Ha ha ha ha. Please stay in the appro NG, or perhaps you are thinking of
upgrading to 10 for better security etc and, and ,and.
Go bitch to your other half.

Didn't the OP say he only had half a brain in his first post?

Rene