A Windows XP help forum. PCbanter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » PCbanter forum » Microsoft Windows 7 » Windows 7 Forum
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

Questions about the "end of Windows 7"



 
 
Thread Tools Rate Thread Display Modes
  #271  
Old March 4th 19, 11:27 PM posted to alt.windows7.general
Dominique
external usenet poster
 
Posts: 343
Default Questions about the "end of Windows 7"

Big Al écrivait :

alt.comp.os.windows-10


Which USENET server has this group?

It doesn't seem to be on "Eternal".

I'm using an old version of Xnews.
Ads
  #272  
Old March 4th 19, 11:31 PM posted to alt.windows7.general
Paul[_32_]
external usenet poster
 
Posts: 11,873
Default Questions about the "end of Windows 7"

Mayayana wrote:
"Ken Blake" wrote

In general, how many applications you have open hardly matters. Much
more important is what applications they are, and especially whether
they are doing something at the moment.

Not necessarily. Pale Moon is using 150 MB RAM here just to
sit there. Firefox is similar. And they're the older versions that
don't run each instance in a separate process. If you have
40 tabs open, doing nothing, in recent vintage Firefox, you'll
still have each one loading independent instances of at least
some parts of the program.

On the other hand, visual Studio 6 (VB) loaded with a
large project loaded is using 1/8th as much RAM. Even
with the additional load of the entire MSDN help system it's
using about 1/5 of what FF takes to sit there. And I
still have more than 2/3 of my 3+ GB free.

I suspect many of the people who complain about RAM
are running bloated browsers with loads of tabs open,
in which they're logged into various sites like Google
or Facebook, and allowing videos to load in pages they're
not even looking at. Most people also don't block auto-refresh.
So things like news sites are reloading the whole thing
every few minutes.
Tabs have arguably been a disaster in that sense, making
it easy for people to keep a multitude of webpages open at
once, for no reason, at a time when webpages have become
amazingly bloated. A few years ago 100 KB was too big to load.
Now a single page that loads 20 MB is not unusual. That's
bigger than most software programs.


For Firefox, check out the size of XUL.dll. At
least on the pre 52-ESR ones, that was a "player".
It's one of the larger DLLs you'll run into.

When you build Firefox, during the linking stage,
XUL.dll takes a bit more than 3GB of RAM to link.
One time I tried it, I had to change the kernel:userspace
split, for the build to finish. A later attempt to build
32-bit Firefox, needed Visual Studio in a 64-bit OS, so
there would be sufficient RAM to finish the linking step.
I didn't keep notes, but it's possible it zoomed up
to 10GB during linking.

(If you build Chrome, there's certain value in having
32GB of RAM for your build machine.)

But once XUL.dll is loaded at runtime, if you fork multiple processes,
wouldn't they used shared code segments ? All the processes
should be able to read the code, out of a common segment.
Because code segments are not allowed to "self modify" when
running. The mapper would be set to read-only, once the
segment is loaded (attempting to write it would give some
sort of access error).

I'm sure the RAM they are wasting, is being put to good
usage storing cat photos. It can't possibly be bloated code :-)

Paul
  #273  
Old March 4th 19, 11:48 PM posted to alt.windows7.general
Bill in Co[_3_]
external usenet poster
 
Posts: 303
Default Questions about the "end of Windows 7"

Paul wrote:
Mayayana wrote:
"Ken Blake" wrote

In general, how many applications you have open hardly matters. Much
more important is what applications they are, and especially whether
they are doing something at the moment.

Not necessarily. Pale Moon is using 150 MB RAM here just to
sit there. Firefox is similar. And they're the older versions that
don't run each instance in a separate process. If you have
40 tabs open, doing nothing, in recent vintage Firefox, you'll
still have each one loading independent instances of at least
some parts of the program.

On the other hand, visual Studio 6 (VB) loaded with a
large project loaded is using 1/8th as much RAM. Even
with the additional load of the entire MSDN help system it's
using about 1/5 of what FF takes to sit there. And I
still have more than 2/3 of my 3+ GB free.

I suspect many of the people who complain about RAM
are running bloated browsers with loads of tabs open,
in which they're logged into various sites like Google
or Facebook, and allowing videos to load in pages they're
not even looking at. Most people also don't block auto-refresh.
So things like news sites are reloading the whole thing
every few minutes.
Tabs have arguably been a disaster in that sense, making
it easy for people to keep a multitude of webpages open at
once, for no reason, at a time when webpages have become
amazingly bloated. A few years ago 100 KB was too big to load.
Now a single page that loads 20 MB is not unusual. That's
bigger than most software programs.


For Firefox, check out the size of XUL.dll. At
least on the pre 52-ESR ones, that was a "player".
It's one of the larger DLLs you'll run into.

When you build Firefox, during the linking stage,
XUL.dll takes a bit more than 3GB of RAM to link.
One time I tried it, I had to change the kernel:userspace
split, for the build to finish. A later attempt to build
32-bit Firefox, needed Visual Studio in a 64-bit OS, so
there would be sufficient RAM to finish the linking step.
I didn't keep notes, but it's possible it zoomed up
to 10GB during linking.

(If you build Chrome, there's certain value in having
32GB of RAM for your build machine.)

But once XUL.dll is loaded at runtime, if you fork multiple processes,
wouldn't they used shared code segments ? All the processes
should be able to read the code, out of a common segment.
Because code segments are not allowed to "self modify" when
running. The mapper would be set to read-only, once the
segment is loaded (attempting to write it would give some
sort of access error).

I'm sure the RAM they are wasting, is being put to good
usage storing cat photos. It can't possibly be bloated code :-)

Paul


XUL is 51 MB in size here (using FF 52.9.0 ESR). XUL Build ID is
52.9.0.6746


  #274  
Old March 4th 19, 11:55 PM posted to alt.windows7.general
J. P. Gilliver (John)[_4_]
external usenet poster
 
Posts: 2,679
Default Questions about the "end of Windows 7"

In message , Mayayana
writes:
"J. P. Gilliver (John)" wrote

[]
| I think it's supposed to be a cat (and ironic; he likes cats). (And the
| .exe includes several other icons if you want.)
|
It might be a cat. And yes, I found the other icons
ages ago. I use the sunset one. It's simple and recognizable.


It is a cat - see https://www.irfanview.com/faq.htm#PAGE15 . (He likes
cats - see https://www.irfanview.com/images/me2.jpg [though that's some
years old].)
[]
Personally I also organize shortcuts, putting them
into categories and dumping all the crap like shortcuts
to websites, readmes, etc. So I have a nice, compact
Start Menu, with menu items like Office, Internet, etc.


I do similar.
[]
--
J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

I don't have an agree that our language torture is a quality add
- soldiersailor on Gransnet, 2018-3-8
  #275  
Old March 4th 19, 11:57 PM posted to alt.windows7.general
Big Al[_5_]
external usenet poster
 
Posts: 1,588
Default Questions about the "end of Windows 7"

On 3/4/19 5:27 PM, Dominique wrote:
Big Al �crivait :

alt.comp.os.windows-10


Which USENET server has this group?

It doesn't seem to be on "Eternal".

I'm using an old version of Xnews.

And I'm using Eternal Sept too. It's there.
make sure it's -10 not .10


  #276  
Old March 4th 19, 11:57 PM posted to alt.windows7.general
Paul[_32_]
external usenet poster
 
Posts: 11,873
Default Questions about the "end of Windows 7"

Bill in Co wrote:


It would be interesting to know what you consider "boots very quickly"
means, but if it does, I'm assuming you're using a SSD. Otherwise I'm
gonna guess it's several minutes.

I wonder if anyone has a computer using Windows 7 or 10 that boots up in a
couple of minutes without a SSD. IOW, not 4 or 5 or 10 minutes, to the
finish screen.


I got 26 seconds here, on a relatively fresh W7SP1 install.

(Installed two days ago for testing.)

That's from the time the BIOS strings stop printing to screen
(equivalent to seeing an underline character in the upper left
corner) until the desktop appears. And that's a HDD install.

But my "real" Win7 install, is busy after the desktop appears,
and isn't really "ready" as such, so the measurement on that
one is overly generous. The video card "gamma" shifts after
maybe 45 seconds to a minute, implying some initialization
process was just finishing.

I've had some Windows installs, where .NET is a culprit. The
Windows Firewall won't load, because some part of the OS feels
the associated .NET assembly isn't compiled. And it won't run
until ngen.exe compiles the assembly again. Which can delay
network availability for 30 seconds to a minute. You can
actually run ngen manually to try and fix that.

That makes the boot metric rather useless on "sick" OSes,
because we give the false impression you could actually use it.
That 26 second number (on a HDD boot device) is usable boot.
For some reason, it's ready to work when I am. But not all
my installs here, are as happy.

You have the option of using ProcMon or WPA (xbootmgr, xperf
and friends) if you want to study the component parts. But due
to the usage of SVCHOSTs, you won't always get answers to your
questions. You might discover a SVCHOST with a certain PID
is holding things up, but you might not be able to figure
out which service inside it, is the culprit.

Paul
  #277  
Old March 5th 19, 12:01 AM posted to alt.windows7.general
Bill in Co[_3_]
external usenet poster
 
Posts: 303
Default Questions about the "end of Windows 7"

Paul wrote:
Bill in Co wrote:


It would be interesting to know what you consider "boots very quickly"
means, but if it does, I'm assuming you're using a SSD. Otherwise I'm
gonna guess it's several minutes.

I wonder if anyone has a computer using Windows 7 or 10 that boots up in
a couple of minutes without a SSD. IOW, not 4 or 5 or 10 minutes, to
the finish screen.


I got 26 seconds here, on a relatively fresh W7SP1 install.

(Installed two days ago for testing.)

That's from the time the BIOS strings stop printing to screen
(equivalent to seeing an underline character in the upper left
corner) until the desktop appears. And that's a HDD install.

But my "real" Win7 install, is busy after the desktop appears,
and isn't really "ready" as such, so the measurement on that
one is overly generous. The video card "gamma" shifts after
maybe 45 seconds to a minute, implying some initialization
process was just finishing.

I've had some Windows installs, where .NET is a culprit. The
Windows Firewall won't load, because some part of the OS feels
the associated .NET assembly isn't compiled. And it won't run
until ngen.exe compiles the assembly again. Which can delay
network availability for 30 seconds to a minute. You can
actually run ngen manually to try and fix that.

That makes the boot metric rather useless on "sick" OSes,
because we give the false impression you could actually use it.
That 26 second number (on a HDD boot device) is usable boot.
For some reason, it's ready to work when I am. But not all
my installs here, are as happy.

You have the option of using ProcMon or WPA (xbootmgr, xperf
and friends) if you want to study the component parts. But due
to the usage of SVCHOSTs, you won't always get answers to your
questions. You might discover a SVCHOST with a certain PID
is holding things up, but you might not be able to figure
out which service inside it, is the culprit.

Paul


I need to correct my post in that it "only" takes two minutes to boot up to
the desktop, not 4, as I had originally guessed. Again, it's about twice as
long as on the Win XP equivalent computer next to it.


  #278  
Old March 5th 19, 12:29 AM posted to alt.windows7.general
Paul[_32_]
external usenet poster
 
Posts: 11,873
Default Questions about the "end of Windows 7"

Dominique wrote:
Big Al écrivait :

alt.comp.os.windows-10


Which USENET server has this group?

It doesn't seem to be on "Eternal".

I'm using an old version of Xnews.


Make sure you update your newsgroup list.

Things could get added to the list, since the
last time you downloaded it.

In the old days, this was called the .newsrc file.

https://i.postimg.cc/9QcwyBWs/refres...group-list.gif

Each newsreader has to implement that function, somehow.

Paul
  #279  
Old March 5th 19, 01:02 AM posted to alt.windows7.general
Mayayana
external usenet poster
 
Posts: 6,438
Default Questions about the "end of Windows 7"

"Paul" wrote

| For Firefox, check out the size of XUL.dll. At
| least on the pre 52-ESR ones, that was a "player".
| It's one of the larger DLLs you'll run into.
|
I noticed that 37 MB according to ProcExplorer.
But ProcExplorer doesn't come close to accounting
for all the 150 MB being used.


| But once XUL.dll is loaded at runtime, if you fork multiple processes,
| wouldn't they used shared code segments ? All the processes
| should be able to read the code, out of a common segment.

I'm not sure. That's why I said each instance would
be loading at least part of the load. If you need something
like kernel32 then it won't be loaded because it' already
running and your process is just one more refrence keeping
it in memory. But I don't know what else FF is doing that
takes 150 MB. For instance, 10-20 MB of history. Does
every FF process get a copy or is one copy somehow
shared? I don't know. I guess the answer would be to run
ProcExplorer with several separate-process browser
windows running and see what it says. I don't know
when/if FF switched to process-per-window, but that's
the fad these days, so I'm assuming they did that
somewhat recently.

| I'm sure the RAM they are wasting, is being put to good
| usage storing cat photos. It can't possibly be bloated code :-)
|

No doubt. I'm just judgemental. I see enough RAM to
run 6 copies of Win98 and I jump to conclusions. Luckily
for me, I don't have any cat photos, so I get by just
fine with 3 GB RAM.


  #280  
Old March 5th 19, 01:58 AM posted to alt.windows7.general
Apd
external usenet poster
 
Posts: 132
Default Questions about the "end of Windows 7"

"Mark Lloyd" wrote:
On 3/3/19 1:36 PM, Java Jive wrote:
For me, the only version of Windows that had a GUI that even came close
to meriting being described as 'professional' was 2k - it wasn't so
out of the box, because like every other it defaulted in everything to
icon and not list view, etc, but by dint of some time well spent when
first installed, it could be made acceptably professional.


Considering the GUI, W2K was my preferred version.


It's also mine...

BTW, I used it for about 3 years after that so-called "end of life".


....and I'm still using W2k today on the machine I use most. Of course,
there are some web sites that no longer work but no matter; I have
an XP laptop (disguised as a POS terminal so still being updated) and
an older iMac (dual boot between OSX and XP) which is fast and has a
nice big screen. I also have a Win7 laptop with a knackered battery
which I hardly use but it's there if I need it. One thing Win7 is good
for is copying stuff to my smartphone without needing extra software.
XP can read but can't write to it and the crap Samsung software won't
work properly on XP.

So my usage is:
1) W2k
2) XP (a close second)
3) W7 (rarely)

I also have a 486 running DOS 6 and Win 3.11, an ancient laptop with
NT4, some 9x machines and a few more XP ones for spares or checking
out possible malicious software.


  #281  
Old March 5th 19, 02:58 AM posted to alt.windows7.general
pyotr filipivich
external usenet poster
 
Posts: 752
Default Questions about the "end of Windows 7"

Sam E on Mon, 4 Mar 2019 12:07:32
-0600 typed in alt.windows7.general the following:
On 3/3/19 12:49 PM, Char Jackson wrote:

[snip]

These days, you could target the folks who have touch screens by adding
a widget that you swipe right to left when you get to the end of line.
You could call it a "Carriage Return!" :-)

I once heard it called "down and to the left" action.


CR & LF!

We keep this up, we may bring back Morse code.

Binary Coded Roman Numerals.
--
pyotr filipivich
Next month's Panel: Graft - Boon or blessing?
  #282  
Old March 5th 19, 03:52 AM posted to alt.windows7.general
Bill in Co[_3_]
external usenet poster
 
Posts: 303
Default Questions about the "end of Windows 7"

pyotr filipivich wrote:
Sam E on Mon, 4 Mar 2019 12:07:32
-0600 typed in alt.windows7.general the following:
On 3/3/19 12:49 PM, Char Jackson wrote:

[snip]

These days, you could target the folks who have touch screens by adding
a widget that you swipe right to left when you get to the end of line.
You could call it a "Carriage Return!" :-)


I once heard it called "down and to the left" action.


CR & LF!

We keep this up, we may bring back Morse code.


0Dh, 0Ah
.. _ . _ . _



  #283  
Old March 5th 19, 04:42 AM posted to alt.windows7.general
Gene Wirchenko[_2_]
external usenet poster
 
Posts: 496
Default Questions about the "end of Windows 7"

On Mon, 4 Mar 2019 10:07:42 -0500, Wolf K
wrote:

[snip]

I don't. If you have to work your way through a 1,000 page manual,
neither the software's features nor the user interface are intuitive
enough. It also means that there's no de-facto or other standard (==
user expectations, the "muscle memory" mentioned by Pyotr.)


It depends what you mean by working through. If there is good
guidance so that one does not have to read that much of it, I do not
have a problem with it. If you have read a substantial amount of that
large manual to get started, then yuck.

[snip]

Sincerely,

Gene Wirchenko
  #284  
Old March 5th 19, 06:17 AM posted to alt.windows7.general
Mike
external usenet poster
 
Posts: 185
Default Questions about the "end of Windows 7"

On 3/4/2019 11:05 AM, Bill in Co wrote:
Mark Lloyd wrote:
On 3/3/19 9:52 PM, Roger Blake wrote:

[snip]

Although I no longer use a typewriter, I do have a pretty ancient
Panasonic KX-P1124 dot matrix printer for use with multipart forms.
I actually bought it for use with a Commodore 64 what seems like
a lifetime ago. I'm continually amazed that it still works.


I had an Epson MX-80 for use with my C64 until the late nineties when I
sold it (with as 80386sx PC).

BTW, I still have a C64. It could last a long time, since it's seldom
used.


I had a LOT of fun messing around with its predecessor, the VIC-20. And I
didn't even have any backup, so after writing some programs in BASIC, or
occasionally assembly using the HesMon adapter, I'd have to leave the
computer continually, if I ever wanted to resume playing with it. Then I
got to heaven by finally purchasing the cassette tape backup recorder, so I
could finally save any programs I was writing, and turn the thing off. :-)


Ah, the good old days when ToysRus was the best place to buy desktop
computers. Only $200 for a 360K external floppy drive. At a time when
my house payment was $154/month.
All this nostalgia is urging me to go up into the attic and bring down some
history...and send it to Goodwill...
  #285  
Old March 5th 19, 07:55 AM posted to alt.windows7.general
Mike
external usenet poster
 
Posts: 185
Default Questions about the "end of Windows 7"

On 3/4/2019 11:19 AM, Java Jive wrote:
On 04/03/2019 18:32, Bill in Co wrote:

I wonder if anyone has a computer using Windows 7 or 10 that boots up
in a
couple of minutes without a SSD.Â*Â* IOW, not 4 or 5 or 10 minutes, to the
finish screen.


Yes, this oneÂ* -Â* I haven't actually timed it, but, unless it's doing
updates, it boots within about a minute, in fact my guess would be that
if measured from the grub menu comfortably within a minute.

FTR, it's a second-hand/used Dell Precision M6300, Intel Core 2 Duo
(T7500) @ 2.2GHz, 4GB RAM, running W7 64-bit Ultimate on a Toshiba
MQ01ABD050V, a conventional HD.Â* It's not as fast as my other W7 laptop
-Â* a Dell Inspiron 15RSE 7520 Core i5-3210M @ 2.5GHz with 8GB RAM
running Windows 7 Home Premium on a Western Digital WDC
WD10JPVT-75A1YT0, again a conventional HD (which has just taken 21
seconds from switch on to logon screen for a resume after hibernation,
no grub menu on that one)Â* -Â* but then no-one in their right mind would
expect it to be.

It groans sometimes loading a memory hungry program, for example Firefox
(I use Pale Moon which has a much smaller memory footprint for more or
less the same functionality), but generally it's fine running my
everyday stuffÂ* -Â* Explorer, Pale Moon, two copies of Thunderbird (one
for mail, another for news), Digiguide UK TV guide, a console or two, a
backup program which I run in the evening but otherwise just sits there,
and usually Explorer within Control Panel in a separate Window from the
above.Â* These are nearly always running, other programs are run as
needed.Â* I've used some fairly resource-intensive programs such as
FFMPEG and Handbrake on it, and it copes fairly well.

By contrast W10 took ages to get to the login screen, at a guess at
least about 3 minutes, and was unusable thereafter.

I should perhaps point out that I've customised this build, so Task
Manager shows about 40-50 processes running when the machine is idle
with no programs loaded (and, as already pointed out, that's still
double what I've been able to whittle W2k & XP down to), whereas W10 was
running as supplied by the reseller, but I didn't and still don't think
it would be worth even considering customising W10 to try and run it on
this PCÂ* -Â* I'm pretty sure that it would still be unacceptably slow.


We have to be careful not to compare apples to oranges and proclaim a
conclusion.
My definition of "boot time" is the time from power button cold start
until you can actually do something useful with the computer.
Restore from hibernate is more about how fast the disk can load the ram.
Wake from sleep is almost instantaneous, no matter which OS you use.

So, Dell Optiplex 360 2.8GHZ dual core, 4GB RAM.
Start with cold boot push the power button.
Time in seconds
What XP Win7 Win 10
Welcome screen 27 30 46
Desktop 45 48 59
Comodo display shows n/a 113 59
CPU usage drops n/a 128 78
disk activity drops n/a 128 150
You can actually do something 45 128 150

It appears that M$ has tuned win10 to look like it boots fast, but
it does take longer before you can actually do anything productive.
There's the whole other thing about fastboot.
I have hibernation disabled, so have no experience with that.
IIRC, fastboot is a combination of return from hibernation and cold boot
that reduces boot time. I'd rather have the disk space that hibernation
would steal. Most of my reboots are system demanded and wouldn't be
helped by hibernation anyway. Would be interesting to see numbers
for fastboot vs nofastboot.

Win 10 runs a lot of stuff in the background after you think it's
booted. After the initial disk activity drop, if I do nothing,
mine jumps back up to 50% cpu and 100% disk activity for up to several
minutes.
If I move the mouse, it backs down temporarily so I can still do stuff.

To be fair, the win10 drive is a 10KRPM server drive.
My slower win10 drives have different configurations and fewer metrics
implemented, but it seems to add 20 seconds or so to boot.
XP isn't loading Comodo, so it looks better there.

The optiplex 780 3GHz dual core with 8GB RAM and SSD is ready to do stuff
from cold start in 61 seconds.

I have a laptop with 2GB of RAM that takes windows 10 over 5 minutes to get
to the welcome screen...about half the battery life to get up to the point
that you can do something. After that, it's so slow that I go looking
for my shotgun to put it out of it's misery.

Boot time varies a LOT depending on your system and configuration and
exactly
what you consider "booted".

My conclusion is that XP may boot in 1/3 the time of Win10.
Win10 and win7 are close enough that it doesn't matter.

I still think that people should use what works for them.
For me, win10 is inevitable and its time has come...for me.

For those too stubborn to get the free digital entitlement, YMMV.
 




Thread Tools
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off






All times are GMT +1. The time now is 12:24 PM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 PCbanter.
The comments are property of their posters.