Re: Replacing laptop cpu

2015-05-27 Thread jd1008



On 05/26/2015 10:44 PM, g wrote:


On 05/26/2015 10:15 PM, jd1008 wrote:
<<>>


Thanx.
I saved the correction.

welcome.

how about running it to see if you can notice any changing in the
temps.

over 5 hrs, i am seeing change on this system;

   cores  +/- 2 c
   board  +/- 2 c
   fans   +/- 5 rpm

nothing drastic.

a sample of readings shows i am well within safe range.

Tue May 26 23:34:27 CDT 2015
coretemp-isa-
Adapter: ISA adapter
Core 0:  +49.0°C  (high = +86.0°C, crit = +100.0°C)
Core 1:  +49.0°C  (high = +86.0°C, crit = +100.0°C)

smsc47b397-isa-0480
Adapter: ISA adapter
fan1:   1301 RPM
fan2:  0 RPM
fan3:  0 RPM
fan4:   1212 RPM
temp1:   +64.0°C
temp2:   +51.0°C
temp3:   +32.0°C
temp4:  -128.0°C

temp3 and temp4 may well be err as i have not bothered to config. ;-)



I have to wait for the silver paste thing. It is probably
in the mail.
One thing I have to be extremely careful with is
where the grease will touch. The AMD cpu has exposed
contacts (super tiny chips and their spider contacts)
arrayed around the core. I have to be extra careful that the
conductive silver grease will never, in any way, touch those.

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-27 Thread Tim
On Tue, 2015-05-26 at 11:33 -0700, Joe Zeff wrote:
> I thought that crystal oscillators were used for the same reason 
> musicians use tuning forks: it doesn't matter how hard you hit them,
> you always get the same frequency.

They hum (the ringing after the blow) on pitch, for a prolonged period.
But you wouldn't use the initial strike, or rely on them as it's on the
verge of petering out.

Stringed instruments go flat (slightly) after the initial strike.  And
they should be tuned for how you play them.  If you play hard, then you
need to tune them while plucking the string strongly.  If you play
gentle, you need to tune them that way, too.  In neither case, pluck the
string once and fiddle with tuning while it simply rings.  I'd expect
the same for any mechanically vibrating musical instrument, particularly
ones with large parts.

There's a tension change with a strong initial strike or pluck, that
changes the pitch.  You can google how to properly tune a guitar to read
more about this (e.g. from recording studio techs), and see demo videos.

I would expect a similar thing with a tuning fork.  However, as they're
designed as a tuning tool, the settling period is probably so much
faster that you're not going to strike it as a problem (pun intended).

-

Even crystal controlled things do go off pitch, for unknown reasons.
I've certainly heard my (over-computerised) organ falter in the middle
of playing.  But without having a multimeter handy at the time, I
couldn't tell if it was due to a mains brown out or an internal fault.

We now return you to your regular programming...

-- 
tim@localhost ~]All mail to my mailbox is automatically deleted, there is no 
point trying
to privately email me, I will only read messages posted to the public lists.




-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-27 Thread Tim
Tim wrote:
>> Thermal pad - heatsink compound, similar job.  I wouldn't expect
>> both to be present, and I'd expect trying both to be a problem
>> in itself.

g:
> if one is good, both are better. ;-)

Not necessarily.

> pad and paste tim;
> 
> https://en.wikipedia.org/wiki/Thermally_conductive_pad

You seem to come away from that page with different things than I
noticed:

"As an alternative to thermal grease (sometimes called thermal transfer
material), AMD and Intel have included thermal pads on the bottom of
heatsinks shipped with some of their processors, as they are cleaner and
generally easier to install. However, thermal pads conduct heat less
effectively than a minimal amount of thermal grease."

Grease works better than a thermal pad.

And

> plus, see the 2 links shown under "References". good reading.

Regarding link 1:

It recommends one or the other, depending on the particular CPU.

Regarding link 2:

"promising start. Just 3°С higher compared to results with thermal
compound only. However, I could not make a screenshot of Linpack test
pass. The temperature sharply increased and went over 100° in seconds
which triggered CPU overheating function followed by an emergency PC
shut down."

"Later I realized that my feeling were right. I launched the system,
entered BIOS setup and went strait to monitoring section and… saw CPU
temperature reaching 95 ° C, 5 seconds after it, the computer shut down.
I checked the pressing, turned the pads on the other side — now the pink
side was facing the processor. I started the system and got similar
results. Test failed."

"... thick pads the temperature increased by 7°С"

Above, it talked about combined thermal pads and grease, or just just
thermal pads, alone.  Increasing temperature means that the CPU is not
being cooled properly (they're measuring the CPU temp, not the heatsink
dissipation temp).



-- 
tim@localhost ~]$ uname -rsvp

Linux 3.19.5-100.fc20.i686 #1 SMP Mon Apr 20 20:28:39 UTC 2015 i686

All mail to my mailbox is automatically deleted, there is no point trying
to privately email me, I will only read messages posted to the public lists.

George Orwell's '1984' was supposed to be a warning against tyranny, not
a set of instructions for supposedly democratic governments.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-26 Thread g


On 05/26/2015 10:15 PM, jd1008 wrote:
<<>>

> Thanx.
> I saved the correction.

welcome.

how about running it to see if you can notice any changing in the
temps.

over 5 hrs, i am seeing change on this system;

  cores  +/- 2 c
  board  +/- 2 c
  fans   +/- 5 rpm

nothing drastic.

a sample of readings shows i am well within safe range.

Tue May 26 23:34:27 CDT 2015
coretemp-isa-
Adapter: ISA adapter
Core 0:  +49.0°C  (high = +86.0°C, crit = +100.0°C)
Core 1:  +49.0°C  (high = +86.0°C, crit = +100.0°C)

smsc47b397-isa-0480
Adapter: ISA adapter
fan1:   1301 RPM
fan2:  0 RPM
fan3:  0 RPM
fan4:   1212 RPM
temp1:   +64.0°C
temp2:   +51.0°C
temp3:   +32.0°C
temp4:  -128.0°C

temp3 and temp4 may well be err as i have not bothered to config. ;-)


-- 

peace out.

If Bill Gates got a dime for every time Windows crashes...
 ...oh, wait. He does. THAT explains it!


in a world with out fences, who needs gates.

CentOS GNU/Linux 6.6

tc,hago.

g
.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-26 Thread jd1008



On 05/26/2015 09:01 PM, g wrote:

*WOW*

my bad. :-\

hard to believe error slide thru. ((GBWG))

see notation in script.

On 05/25/2015 03:17 PM, g wrote:
<<>>


forget that script. i thought i would expand script with a time stamp
so you can better relate to when things change.

=+=+=+=

#!/bin/sh
#   fn = log-sensors  v. 2  r. 01
# date = 2015-0525-1330
#   author = g e leem - copyleftist g e leem, gnu gpl applies.
# path = /home/users/scripts/
##
#   usage:  build a log of sensor readings at 10 minute intervals
#   add a time stamp every hour
##
#   notes:  sleep 600 = 10 minutes delay
#
#   chk-sensors = 6 starts with value of 6 to
#   count down sleep periods of 10 minutes in
#   first loop
#
#   chk-sensors = 7 so that time stamp with have
#   correct value for next loop
#
#   start logging from a terminal with command
#   log-sensors&
#   to run in background and free terminal

chksensors=6

while true ; do

 sensors >> ~/00-check-sensors

 sleep 600 ## set sleep period to an interval of your choice.

 if $chksensors = 0 ; then

   date >> ~/00-check-sensors

   chksensors=7

 fi

 chksensors=$chksensors - 1

*note*
above line is incorrect. should be;

  chksensors=$chksensors-1


done

=+=+=+=

now, if you are aware of time you are doing whatever, you will be
able to relate to when.

i have been running script to see how it is tracking and i see
changing in cpu and board temps and fans speeds. nothing drastic,
but interesting to see. :-)

when i returned inside from yard work, it was a while before i
got around to see what script had built in file.

after reading, i got busy with a few other things.

a few minutes ago when i finally got around to bring up window with
terminal i started script in, i got hit with;

   ]$ /home/geo/scripts/log-sensors: line 39: -: command not found

did not take me long to realize spaces are ok when used for testing
condition with " = ", but not with " - " and " + " changing value.

my apologies. [i blame it on chemo brain]

and i admit, i would have felt worse if someone else had informed
me of my err. ;-)



Thanx.
I saved the correction.

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-26 Thread g

*WOW*

my bad. :-\

hard to believe error slide thru. ((GBWG))

see notation in script.

On 05/25/2015 03:17 PM, g wrote:
<<>>

> forget that script. i thought i would expand script with a time stamp
> so you can better relate to when things change.
> 
> =+=+=+=
> 
> #!/bin/sh
> # fn = log-sensors  v. 2  r. 01
> #   date = 2015-0525-1330
> # author = g e leem - copyleftist g e leem, gnu gpl applies.
> #   path = /home/users/scripts/
> ##
> # usage:  build a log of sensor readings at 10 minute intervals
> # add a time stamp every hour
> ##
> # notes:  sleep 600 = 10 minutes delay
> #
> # chk-sensors = 6 starts with value of 6 to
> # count down sleep periods of 10 minutes in
> # first loop
> #
> # chk-sensors = 7 so that time stamp with have
> # correct value for next loop
> #
> # start logging from a terminal with command
> # log-sensors&
> # to run in background and free terminal
> 
> chksensors=6
> 
> while true ; do
> 
> sensors >> ~/00-check-sensors
> 
> sleep 600 ## set sleep period to an interval of your choice.
> 
> if $chksensors = 0 ; then
> 
>   date >> ~/00-check-sensors
> 
>   chksensors=7
> 
> fi
> 
> chksensors=$chksensors - 1

*note*
above line is incorrect. should be;

 chksensors=$chksensors-1

> done
> 
> =+=+=+=
> 
> now, if you are aware of time you are doing whatever, you will be
> able to relate to when.
> 
> i have been running script to see how it is tracking and i see
> changing in cpu and board temps and fans speeds. nothing drastic,
> but interesting to see. :-)

when i returned inside from yard work, it was a while before i
got around to see what script had built in file.

after reading, i got busy with a few other things.

a few minutes ago when i finally got around to bring up window with
terminal i started script in, i got hit with;

  ]$ /home/geo/scripts/log-sensors: line 39: -: command not found

did not take me long to realize spaces are ok when used for testing
condition with " = ", but not with " - " and " + " changing value.

my apologies. [i blame it on chemo brain]

and i admit, i would have felt worse if someone else had informed
me of my err. ;-)


-- 

peace out.

If Bill Gates got a dime for every time Windows crashes...
 ...oh, wait. He does. THAT explains it!


in a world with out fences, who needs gates.

CentOS GNU/Linux 6.6

tc,hago.

g
.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-26 Thread Joe Zeff

On 05/26/2015 03:16 AM, Tim wrote:

I'm not.  The other guy posed a theoretical question, and I gave a
supposed answer.  Smack it harder, and it'll ring differently.  I didn't
say it was going to be a huge difference.


I thought that crystal oscillators were used for the same reason 
musicians use tuning forks: it doesn't matter how hard you hit them, you 
always get the same frequency.

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-26 Thread g


On 05/26/2015 05:17 AM, Tim wrote:
> On Mon, 2015-05-25 at 16:21 -0500, g wrote:
>> also, is there a thermal pad between cpu and heat sink? there
>> should be one because they help thermal transfer.
> 
> Thermal pad - heatsink compound, similar job.  I wouldn't expect
> both to be present, and I'd expect trying both to be a problem
> in itself.

if one is good, both are better. ;-)


pad and paste tim;

https://en.wikipedia.org/wiki/Thermally_conductive_pad

plus, see the 2 links shown under "References". good reading.

after having read the AMD report, i do believe i will stay with
non conducting paste. comment about such is very reasonable.

[i like wikipedia. takes me back in thoughts of my early childhood
 days when i used to spend hours in library reading thru the many
 electronic books, only much easier.
 ]


-- 

peace out.

If Bill Gates got a dime for every time Windows crashes...
 ...oh, wait. He does. THAT explains it!


in a world with out fences, who needs gates.

CentOS GNU/Linux 6.6

tc,hago.

g
.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-26 Thread Tim
jdow:
> How far do you expect it to change, Grasshopper? Inquiring and bemused minds 
> want to know.

I wouldn't expect much, probably not even significantly.  After all, DC
biasing with varicap diodes across crystals are used in phase locked
loop circuitry (to do the phase locking).  And phase locking is really
only a slight pulling against the natural frequency of your electronic
tuning fork.

> I completely fail to see why you might be concerned about the oscillator 
> producing a frequency of say 100.00 MHz or 100.000100 MHz on a computer.

I'm not.  The other guy posed a theoretical question, and I gave a
supposed answer.  Smack it harder, and it'll ring differently.  I didn't
say it was going to be a huge difference.

Considering that quartz watches keep good time with diminishing battery
power over many years, being warmed by your wrist, and chilled by the
weather, I wouldn't think you'd get a significant change in BIOS clock
frequency depending on the BIOS battery, if it was designed to cope well
with battery changes, unless it was deliberately designed to go goofy
with significant battery charge drop.

And, yes I am well aware of using crystal ovens for greater clock
accuracy, in precision equipment.  But I'd only seen them make a few
Hertz difference in a 4.43 megahertz oscillator.

-- 
tim@localhost ~]$ uname -rsvp

Linux 3.19.5-100.fc20.i686 #1 SMP Mon Apr 20 20:28:39 UTC 2015 i686

All mail to my mailbox is automatically deleted, there is no point trying
to privately email me, I will only read messages posted to the public lists.

George Orwell's '1984' was supposed to be a warning against tyranny, not
a set of instructions for supposedly democratic governments.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-26 Thread Tim
On Mon, 2015-05-25 at 16:21 -0500, g wrote:
> also, is there a thermal pad between cpu and heat sink? there
> should be one because they help thermal transfer.

Thermal pad - heatsink compound, similar job.  I wouldn't expect both to
be present, and I'd expect trying both to be a problem in itself.


-- 
tim@localhost ~]$ uname -rsvp

Linux 3.19.5-100.fc20.i686 #1 SMP Mon Apr 20 20:28:39 UTC 2015 i686

All mail to my mailbox is automatically deleted, there is no point trying
to privately email me, I will only read messages posted to the public lists.

George Orwell's '1984' was supposed to be a warning against tyranny, not
a set of instructions for supposedly democratic governments.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread g


On 05/25/2015 04:28 PM, jd1008 wrote:
> On 05/25/2015 03:21 PM, g wrote:
<<>>

>> also, is there a thermal pad between cpu and heat sink? there
>> should be one because they help thermal transfer.
>
> I am not sure. Will have to get back to you on this.
> On another different laptop, there was no pad.

something to consider when it comes to cooling a cpu.


-- 

peace out.

If Bill Gates got a dime for every time Windows crashes...
 ...oh, wait. He does. THAT explains it!


in a world with out fences, who needs gates.

CentOS GNU/Linux 6.6

tc,hago.

g
.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread jd1008



On 05/25/2015 03:21 PM, g wrote:


On 05/25/2015 03:26 PM, jd1008 wrote:

On 05/25/2015 02:17 PM, g wrote:

<<>>


if you have not refreshed thermal paste yet, as a bit of curiosity,
run script before and after to also see if there is a change.



I ordered the thermal paste (Arctic silver) online.
Waiting for it's arrival.

.
ok.

i was not familiar with "Arctic silver", so i looked it up.

   http://www.arcticsilver.com/as5.htm

from specs, looks like a winner and something i just might order
some for myself. what i currently use does not have micro silver
and such seems like a very good addition.

thanks for passing along name.

also, is there a thermal pad between cpu and heat sink? there
should be one because they help thermal transfer.

I am not sure. Will have to get back to you on this.
On another different laptop, there was no pad.

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread g


On 05/25/2015 03:26 PM, jd1008 wrote:
> On 05/25/2015 02:17 PM, g wrote:
<<>>

>> if you have not refreshed thermal paste yet, as a bit of curiosity,
>> run script before and after to also see if there is a change.
>>
>>
> I ordered the thermal paste (Arctic silver) online.
> Waiting for it's arrival.
.
ok.

i was not familiar with "Arctic silver", so i looked it up.

  http://www.arcticsilver.com/as5.htm

from specs, looks like a winner and something i just might order
some for myself. what i currently use does not have micro silver
and such seems like a very good addition.

thanks for passing along name.

also, is there a thermal pad between cpu and heat sink? there
should be one because they help thermal transfer.


-- 

peace out.

If Bill Gates got a dime for every time Windows crashes...
 ...oh, wait. He does. THAT explains it!


in a world with out fences, who needs gates.

CentOS GNU/Linux 6.6

tc,hago.

g
.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread jdow

On 2015-05-25 06:18, Tim wrote:

On Mon, 2015-05-25 at 03:56 -0500, g wrote:

would you presume dropping of voltage would cause a great amount of
change in a crystal's oscillation?


Seems highly tangental to the prior conversation, but possibly yes.

Essentially, it's a mechanical vibration, even at a very high rate.

If you gently pluck a guitar string while tuning it, it doesn't play the
same note as if you very harshly pluck the string.  Or, to put in
another way, you pluck a string and let it ring, the note it plays goes
slightly flat before it peters out to not vibrating (the frequency
decreases).

So, I wouldn't be surprised if you kick a crystal with less voltage to
make it swing than you usually do, it mightn't do it so fast.



How far do you expect it to change, Grasshopper? Inquiring and bemused minds 
want to know.




You are right in that in general changing the voltage on a crystal oscillator 
will change the output frequency. Generally this is of concern to people who 
build precision oscillators but not mere mortals such as computer users who are 
bold enough to try to "fix" things inside their computers. We're talking single 
digit parts per million frequency changes until the voltage gets low enough the 
oscillator becomes unstable which may not happen before the oscillator simply stops.


The crystal oscillator frequency will change at PPM levels as voltages get 
higher until one of two things happens, the crystal blank fractures or a 
component in the oscillator fries.


Crystal oscillators bear no more resemblance to a guitar string than a hydra 
bears to a gorilla. In one case both vibrate. In the other case both are living 
creatures using DNA.


I completely fail to see why you might be concerned about the oscillator 
producing a frequency of say 100.00 MHz or 100.000100 MHz on a computer. The 
oscillator will vary as much as 100 ppm with modern floor sweeping crystals from 
oscillator to oscillator and over a wide temperature swing. The voltage induced 
changes will be small by comparison.


{^_^}   GPS Joanne

{^_^}
--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread jd1008



On 05/25/2015 02:17 PM, g wrote:


On 05/25/2015 12:13 PM, jd1008 wrote:

On 05/24/2015 10:22 PM, g wrote:

<<>>

.
before replacing or installing newer cpu, do what has already
been mentioned.

clean away old thermal paste, apply fresh _white_ *high temp*
thermal paste.

_do_not_ use a clear paste. clear does not give needed transfer.

as for installing newer type cpu, contact AMD support. who would
know better? ;-)

to find out if cpu is causing problem, use this script;

 #!/bin/sh
 while true ; do
   sensors >> ~/00-check-sensors
   sleep 600 ## set sleep period to an interval of your choice.
 done

when system locks up again, open "00-check-sensors" to see history
for temps and fans.

years back, i had a 586 system locking up. running above script
showed memory fan was slowing down. replaced fan, problem ended.



Good advice.

thank you.


Will try it out.

forget that script. i thought i would expand script with a time stamp
so you can better relate to when things change.

=+=+=+=

#!/bin/sh
#   fn = log-sensors  v. 2  r. 01
# date = 2015-0525-1330
#   author = g e leem - copyleftist g e leem, gnu gpl applies.
# path = /home/users/scripts/
##
#   usage:  build a log of sensor readings at 10 minute intervals
#   add a time stamp every hour
##
#   notes:  sleep 600 = 10 minutes delay
#
#   chk-sensors = 6 starts with value of 6 to
#   count down sleep periods of 10 minutes in
#   first loop
#
#   chk-sensors = 7 so that time stamp with have
#   correct value for next loop
#
#   start logging from a terminal with command
#   log-sensors&
#   to run in background and free terminal

chksensors=6

while true ; do

 sensors >> ~/00-check-sensors

 sleep 600 ## set sleep period to an interval of your choice.

 if $chksensors = 0 ; then

   date >> ~/00-check-sensors

   chksensors=7

 fi

 chksensors=$chksensors - 1

done

=+=+=+=

now, if you are aware of time you are doing whatever, you will be
able to relate to when.

i have been running script to see how it is tracking and i see
changing in cpu and board temps and fans speeds. nothing drastic,
but interesting to see. :-)


Thanx.

most welc. ;-)

if you have not refreshed thermal paste yet, as a bit of curiosity,
run script before and after to also see if there is a change.



I ordered the thermal paste (Arctic silver) online.
Waiting for it's arrival.
--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread g


On 05/25/2015 12:13 PM, jd1008 wrote:
> On 05/24/2015 10:22 PM, g wrote:
<<>>
>> .
>> before replacing or installing newer cpu, do what has already
>> been mentioned.
>>
>> clean away old thermal paste, apply fresh _white_ *high temp*
>> thermal paste.
>>
>> _do_not_ use a clear paste. clear does not give needed transfer.
>>
>> as for installing newer type cpu, contact AMD support. who would
>> know better? ;-)
>>
>> to find out if cpu is causing problem, use this script;
>>
>> #!/bin/sh
>> while true ; do
>>   sensors >> ~/00-check-sensors
>>   sleep 600 ## set sleep period to an interval of your choice.
>> done
>>
>> when system locks up again, open "00-check-sensors" to see history
>> for temps and fans.
>>
>> years back, i had a 586 system locking up. running above script
>> showed memory fan was slowing down. replaced fan, problem ended.
>>
>>
> Good advice.

thank you.

> Will try it out.

forget that script. i thought i would expand script with a time stamp
so you can better relate to when things change.

=+=+=+=

#!/bin/sh
#   fn = log-sensors  v. 2  r. 01
# date = 2015-0525-1330
#   author = g e leem - copyleftist g e leem, gnu gpl applies.
# path = /home/users/scripts/
##
#   usage:  build a log of sensor readings at 10 minute intervals
#   add a time stamp every hour
##
#   notes:  sleep 600 = 10 minutes delay
#
#   chk-sensors = 6 starts with value of 6 to
#   count down sleep periods of 10 minutes in
#   first loop
#
#   chk-sensors = 7 so that time stamp with have
#   correct value for next loop
#
#   start logging from a terminal with command
#   log-sensors&
#   to run in background and free terminal

chksensors=6

while true ; do

sensors >> ~/00-check-sensors

sleep 600 ## set sleep period to an interval of your choice.

if $chksensors = 0 ; then

  date >> ~/00-check-sensors

  chksensors=7

fi

chksensors=$chksensors - 1

done

=+=+=+=

now, if you are aware of time you are doing whatever, you will be
able to relate to when.

i have been running script to see how it is tracking and i see
changing in cpu and board temps and fans speeds. nothing drastic,
but interesting to see. :-)

> Thanx.

most welc. ;-)

if you have not refreshed thermal paste yet, as a bit of curiosity,
run script before and after to also see if there is a change.


-- 

peace out.

If Bill Gates got a dime for every time Windows crashes...
 ...oh, wait. He does. THAT explains it!


in a world with out fences, who needs gates.

CentOS GNU/Linux 6.6

tc,hago.

g
.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread Tim
Allegedly, on or about 25 May 2015, g sent:
> basically, crystals are cut to oscillate at designed frequency, with
> exceptions being those designed to vary with voltage.

Though, still dependent on temperature, and other tuning components
around them.

> both of which tend towards 'off topic', but i thought it might
> help others who may like to know more about the bios chip and
> it's oscillator. 

But DC supply rail voltages don't necessarily equate to oscillator drive
voltages.

I'm not hands-on familiar with playing with crystal oscillators, but
have played with LC oscillators, and those types of resonant circuits
tend to be somewhat less.

-- 
[tim@localhost ~]$ uname -rsvp
Linux 3.9.10-100.fc17.x86_64 #1 SMP Sun Jul 14 01:31:27 UTC 2013 x86_64

All mail to my mailbox is automatically deleted, there is no point
trying to privately email me, I will only read messages posted to the
public lists.

George Orwell's '1984' was supposed to be a warning against tyranny, not
a set of instructions for supposedly democratic governments.

ZNQR LBH YBBX



-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread jd1008



On 05/25/2015 02:56 AM, g wrote:


On 05/25/2015 12:40 AM, Tim wrote:

On Sun, 2015-05-24 at 23:22 -0500, g wrote:

before replacing or installing newer cpu, do what has already
been mentioned.
  
clean away old thermal paste, apply fresh _white_ *high temp*

thermal paste.

Even just mucking about with the fan (poking things about without
removing it) can disrupt how well a heatsink is pressing against a CPU
(with heatsink compound between them).  It dries out, over time.

A smooth, thin, consistent layer of compound should be spread over the
entire surface that mates with the heatsink, as per the instructions
that the compound should provide.

.
all very true.

especially following application instructions and using a transfer pad.
a 'box cutter' blade works well to spread paste on heat sink. after
applying and pressing heat sink on cpu, i wiggle heat sink to help
even out.

would you presume dropping of voltage would cause a great amount of
change in a crystal's oscillation?



I will try the compound and get back to to the list.

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread jd1008



On 05/24/2015 10:22 PM, g wrote:


On 05/24/2015 02:51 PM, jd1008 wrote:
<<>>


But this machine had been working for years without this recent
issue. So, I kind of dounbt it he the graphics chip issue. It's
history included the death of the cpu cooling fan, which I replaced.
The laptop belonged to a friend at the time. She sent it to me to
fix. I replaced the cooling fan with an identical one specified by
the same manufacturer and model number on the dead fan.

After fan replacement, it started having this intermittent problem.
So, I am led to strongly suspect that overheating may have caused
the problem before I even put in the new fan.

.
before replacing or installing newer cpu, do what has already
been mentioned.

clean away old thermal paste, apply fresh _white_ *high temp*
thermal paste.

_do_not_ use a clear paste. clear does not give needed transfer.

as for installing newer type cpu, contact AMD support. who would
know better? ;-)

to find out if cpu is causing problem, use this script;

#!/bin/sh
while true ; do
  sensors >> ~/00-check-sensors
  sleep 600 ## set sleep period to an interval of your choice.
done

when system locks up again, open "00-check-sensors" to see history
for temps and fans.

years back, i had a 586 system locking up. running above script
showed memory fan was slowing down. replaced fan, problem ended.



Good advice.
Will try it out.
Thanx.
--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread Patrick O'Callaghan
On Mon, 2015-05-25 at 22:48 +0930, Tim wrote:
> If you gently pluck a guitar string while tuning it, it doesn't play
> the same note as if you very harshly pluck the string.  Or, to put in
> another way, you pluck a string and let it ring, the note it plays
> goes slightly flat before it peters out to not vibrating (the
> frequency
> decreases).

The fundamental frequency of the string does not change, although the
energy in each harmonic does (higher harmonics decay faster, which
changes the audible character of the note). Not sure how that relates to
the behaviour of crystal oscillators.

poc

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread g


On 05/25/2015 08:18 AM, Tim wrote:
> On Mon, 2015-05-25 at 03:56 -0500, g wrote:
>> would you presume dropping of voltage would cause a great amount of
>> change in a crystal's oscillation?
>
> Seems highly tangental to the prior conversation, but possibly yes.
>
> Essentially, it's a mechanical vibration, even at a very high rate.  
>
> If you gently pluck a guitar string while tuning it, it doesn't play
> the same note as if you very harshly pluck the string.  Or, to put
> in another way, you pluck a string and let it ring, the note it plays
> goes slightly flat before it peters out to not vibrating (the
> frequency decreases).

a good analogy. but that applies slightly differently.

basically, crystals are cut to oscillate at designed frequency, with
exceptions being those designed to vary with voltage.

> So, I wouldn't be surprised if you kick a crystal with less voltage
> to make it swing than you usually do, it mightn't do it so fast.

true, but if frequency cut, they tend to reach designed resonance.

i did find a good lead in to crystals which refreshed my learning
from years ago at;

  https://en.wikipedia.org/wiki/Crystal_oscillator

more info relating to v/f swing near bottom;


https://en.wikipedia.org/wiki/Crystal_oscillator#Circuit_notations_and_abbreviations

[notable those with "VC" in name]

info on BIOS chip is at;

  https://en.wikipedia.org/wiki/BIOS

both of which tend towards 'off topic', but i thought it might
help others who may like to know more about the bios chip and
it's oscillator.


-- 

peace out.

If Bill Gates got a dime for every time Windows crashes...
 ...oh, wait. He does. THAT explains it!


in a world with out fences, who needs gates.

CentOS GNU/Linux 6.6

tc,hago.

g
.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread Tim
On Mon, 2015-05-25 at 03:56 -0500, g wrote:
> would you presume dropping of voltage would cause a great amount of
> change in a crystal's oscillation?

Seems highly tangental to the prior conversation, but possibly yes.

Essentially, it's a mechanical vibration, even at a very high rate.  

If you gently pluck a guitar string while tuning it, it doesn't play the
same note as if you very harshly pluck the string.  Or, to put in
another way, you pluck a string and let it ring, the note it plays goes
slightly flat before it peters out to not vibrating (the frequency
decreases).

So, I wouldn't be surprised if you kick a crystal with less voltage to
make it swing than you usually do, it mightn't do it so fast.


-- 
tim@localhost ~]$ uname -rsvp

Linux 3.19.5-100.fc20.i686 #1 SMP Mon Apr 20 20:28:39 UTC 2015 i686

All mail to my mailbox is automatically deleted, there is no point trying
to privately email me, I will only read messages posted to the public lists.

George Orwell's '1984' was supposed to be a warning against tyranny, not
a set of instructions for supposedly democratic governments.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread g


On 05/25/2015 12:40 AM, Tim wrote:
> On Sun, 2015-05-24 at 23:22 -0500, g wrote:
>> before replacing or installing newer cpu, do what has already
>> been mentioned.
>>  
>> clean away old thermal paste, apply fresh _white_ *high temp*
>> thermal paste.
>
> Even just mucking about with the fan (poking things about without
> removing it) can disrupt how well a heatsink is pressing against a CPU
> (with heatsink compound between them).  It dries out, over time.
>
> A smooth, thin, consistent layer of compound should be spread over the
> entire surface that mates with the heatsink, as per the instructions
> that the compound should provide.
.
all very true.

especially following application instructions and using a transfer pad.
a 'box cutter' blade works well to spread paste on heat sink. after
applying and pressing heat sink on cpu, i wiggle heat sink to help
even out.

would you presume dropping of voltage would cause a great amount of
change in a crystal's oscillation?


-- 

peace out.

If Bill Gates got a dime for every time Windows crashes...
 ...oh, wait. He does. THAT explains it!


in a world with out fences, who needs gates.

CentOS GNU/Linux 6.6

tc,hago.

g
.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-25 Thread Ed Greshko
On 05/25/15 05:39, Ronal B Morse wrote:
> Keep us posted. 

-1

I think there may be other lists better suited for PC hardware repair 
discussions.

-- 
If you can't laugh at yourself, others will gladly oblige.
-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread Joe Zeff

On 05/24/2015 10:46 PM, Tim wrote:

It's arcane knowledge that a clock might be off because of a battery on
a mains powered PC, not one that the general public would be expected to
know about.


Now more than ever, at least for the general public.  Still, I'd expect 
an honest computer tech to recognize the issue and not overcharge for 
the exchange.  Yes, you can grab some short-term profits by making 
things sound worse than they are, but you can get lots of 
recommendations to friends by telling the truth, and that means more 
customers and more long-term revenue.

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread Joe Zeff

On 05/24/2015 10:46 PM, Tim wrote:

I wonder if they've actually be designed to that, or people have just
observed that behaviour, and made an assumption.  I've certainly fixed
up a PC with a new battery, one that exhibited no clock problems.  The
battery was definitely low, resetting the BIOS parameters didn't help,
replacing the battery did.


My understanding is that it's intentional, to warn the user *before* the 
battery's dead.  This is what I learned back in the early 80's from my 
hardware geek the first time I had a CMOS clock issue.

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread Tim
On Sun, 2015-05-24 at 11:23 -0700, Joe Zeff wrote:
> Ever since mobos have had built-in clocks, they've been designed to 
> start running slow when the battery gets low so that you know it needs
> changing.

I wonder if they've actually be designed to that, or people have just
observed that behaviour, and made an assumption.  I've certainly fixed
up a PC with a new battery, one that exhibited no clock problems.  The
battery was definitely low, resetting the BIOS parameters didn't help,
replacing the battery did.

In this day and age you'd expect that there ought to be a positively
identifiable alert that the battery was going flat, such as a message
saying the CMOS/BIOS battery needs replacing.  One that the BIOS
displays while booting, and another that the OS displays while running.
It's arcane knowledge that a clock might be off because of a battery on
a mains powered PC, not one that the general public would be expected to
know about.  And I wouldn't put it past some people, with prompting from
a vendor, about a PC going wonky, to replace the whole machine instead
of the battery.

-- 
tim@localhost ~]$ uname -rsvp

Linux 3.19.5-100.fc20.i686 #1 SMP Mon Apr 20 20:28:39 UTC 2015 i686

All mail to my mailbox is automatically deleted, there is no point trying
to privately email me, I will only read messages posted to the public lists.

George Orwell's '1984' was supposed to be a warning against tyranny, not
a set of instructions for supposedly democratic governments.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread Tim
On Sun, 2015-05-24 at 23:22 -0500, g wrote:
> before replacing or installing newer cpu, do what has already
> been mentioned.
>  
> clean away old thermal paste, apply fresh _white_ *high temp*
> thermal paste.

Even just mucking about with the fan (poking things about without
removing it) can disrupt how well a heatsink is pressing against a CPU
(with heatsink compound between them).  It dries out, over time.

A smooth, thin, consistent layer of compound should be spread over the
entire surface that mates with the heatsink, as per the instructions
that the compound should provide.


-- 
tim@localhost ~]$ uname -rsvp

Linux 3.19.5-100.fc20.i686 #1 SMP Mon Apr 20 20:28:39 UTC 2015 i686

All mail to my mailbox is automatically deleted, there is no point trying
to privately email me, I will only read messages posted to the public lists.

George Orwell's '1984' was supposed to be a warning against tyranny, not
a set of instructions for supposedly democratic governments.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread g


On 05/24/2015 02:51 PM, jd1008 wrote:
<<>>

> But this machine had been working for years without this recent
> issue. So, I kind of dounbt it he the graphics chip issue. It's
> history included the death of the cpu cooling fan, which I replaced.
> The laptop belonged to a friend at the time. She sent it to me to
> fix. I replaced the cooling fan with an identical one specified by
> the same manufacturer and model number on the dead fan.
> 
> After fan replacement, it started having this intermittent problem. 
> So, I am led to strongly suspect that overheating may have caused
> the problem before I even put in the new fan.
.
before replacing or installing newer cpu, do what has already
been mentioned.

clean away old thermal paste, apply fresh _white_ *high temp*
thermal paste.

_do_not_ use a clear paste. clear does not give needed transfer.

as for installing newer type cpu, contact AMD support. who would
know better? ;-)

to find out if cpu is causing problem, use this script;

   #!/bin/sh
   while true ; do
 sensors >> ~/00-check-sensors
 sleep 600 ## set sleep period to an interval of your choice.
   done

when system locks up again, open "00-check-sensors" to see history
for temps and fans.

years back, i had a 586 system locking up. running above script
showed memory fan was slowing down. replaced fan, problem ended.


-- 

peace out.

If Bill Gates got a dime for every time Windows crashes...
 ...oh, wait. He does. THAT explains it!


in a world with out fences, who needs gates.

CentOS GNU/Linux 6.6

tc,hago.

g
.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread Ronal B Morse
I still like my guess better, but it's your machine and your money. Keep 
us posted.


RBM



On 05/24/2015 01:51 PM, jd1008 wrote:



On 05/24/2015 12:49 PM, Ronal B Morse wrote:

On 05/23/2015 11:03 AM, jd1008 wrote:

I have an HP laptop with
AMD Turion II X2 mobile processor RM-72 / 2.1 GHz CPU, Socket S1.
It is now causing blue screens in windows and freezes
linux (pclinuxos, knoppix, fedora live).

I have run the x86 mem test for more than a day, and
found no problems with the 4GB ram (2GB X 2).

I am wondering why the memtest does not freeze
could it be that only one core is causing the problem?

I think an earlier poster in this thread (Vacelaevus) might have 
identified a likely problem.


A few years ago there were a number of HP laptops sold with defective 
G8400/G8600 series nVidia graphics chips...out of spec solder or 
something as I recall. The symptoms were as you described...the 
machine would work normally for a period of time and then 
unexpectedly blue-screen, especially under heavy load. In some cases 
the problem occurred only intermittently. In some cases, the machine 
would appear to work normally after they had been allowed to cool down.


In your specific case, memtest doesn't put a very heavy load on the 
GPU, so it may not have warmed up sufficiently for the problem to 
manifest.


If your laptop has an nVIdia G8400/G8600/G9200/G9400 series GPU it is 
highly likely that is the cause of the problem and replacing the CPU 
will not help. The only recourse is to replace the entire motherboard 
as the GPU is not user replaceable.


The HP customer service website for your model laptop may contain 
additional information.


RBM

But this machine had been working for years without this recent issue.
So, I kind of dounbt it he the graphics chip issue.
It's history included the death of the cpu cooling fan, which I replaced.
The laptop belonged to a friend at the time. She sent it to me to fix.
I replaced the cooling fan with an identical one specified by the same
manufacturer and model number on the dead fan.

After fan replacement, it started having this intermittent problem.
So, I am led to strongly suspect that overheating may have caused
the problem before I even put in the new fan.



--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread jd1008



On 05/24/2015 12:49 PM, Ronal B Morse wrote:

On 05/23/2015 11:03 AM, jd1008 wrote:

I have an HP laptop with
AMD Turion II X2 mobile processor RM-72 / 2.1 GHz CPU, Socket S1.
It is now causing blue screens in windows and freezes
linux (pclinuxos, knoppix, fedora live).

I have run the x86 mem test for more than a day, and
found no problems with the 4GB ram (2GB X 2).

I am wondering why the memtest does not freeze
could it be that only one core is causing the problem?

I think an earlier poster in this thread (Vacelaevus) might have 
identified a likely problem.


A few years ago there were a number of HP laptops sold with defective 
G8400/G8600 series nVidia graphics chips...out of spec solder or 
something as I recall. The symptoms were as you described...the 
machine would work normally for a period of time and then unexpectedly 
blue-screen, especially under heavy load. In some cases the problem 
occurred only intermittently. In some cases, the machine would appear 
to work normally after they had been allowed to cool down.


In your specific case, memtest doesn't put a very heavy load on the 
GPU, so it may not have warmed up sufficiently for the problem to 
manifest.


If your laptop has an nVIdia G8400/G8600/G9200/G9400 series GPU it is 
highly likely that is the cause of the problem and replacing the CPU 
will not help. The only recourse is to replace the entire motherboard 
as the GPU is not user replaceable.


The HP customer service website for your model laptop may contain 
additional information.


RBM

But this machine had been working for years without this recent issue.
So, I kind of dounbt it he the graphics chip issue.
It's history included the death of the cpu cooling fan, which I replaced.
The laptop belonged to a friend at the time. She sent it to me to fix.
I replaced the cooling fan with an identical one specified by the same
manufacturer and model number on the dead fan.

After fan replacement, it started having this intermittent problem.
So, I am led to strongly suspect that overheating may have caused
the problem before I even put in the new fan.

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread Ronal B Morse

On 05/23/2015 11:03 AM, jd1008 wrote:

I have an HP laptop with
AMD Turion II X2 mobile processor RM-72 / 2.1 GHz CPU, Socket S1.
It is now causing blue screens in windows and freezes
linux (pclinuxos, knoppix, fedora live).

I have run the x86 mem test for more than a day, and
found no problems with the 4GB ram (2GB X 2).

I am wondering why the memtest does not freeze
could it be that only one core is causing the problem?

I think an earlier poster in this thread (Vacelaevus) might have 
identified a likely problem.


A few years ago there were a number of HP laptops sold with defective 
G8400/G8600 series nVidia graphics chips...out of spec solder or 
something as I recall. The symptoms were as you described...the machine 
would work normally for a period of time and then unexpectedly 
blue-screen, especially under heavy load. In some cases the problem 
occurred only intermittently. In some cases, the machine would appear to 
work normally after they had been allowed to cool down.


In your specific case, memtest doesn't put a very heavy load on the GPU, 
so it may not have warmed up sufficiently for the problem to manifest.


If your laptop has an nVIdia G8400/G8600/G9200/G9400 series GPU it is 
highly likely that is the cause of the problem and replacing the CPU 
will not help. The only recourse is to replace the entire motherboard as 
the GPU is not user replaceable.


The HP customer service website for your model laptop may contain 
additional information.


RBM
--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread Joe Zeff

On 05/24/2015 03:54 AM, Tim wrote:

Though you
usually have clock problems, at the same time, or as a precursor.


Ever since mobos have had built-in clocks, they've been designed to 
start running slow when the battery gets low so that you know it needs 
changing.

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread jd1008


On 05/24/2015 04:54 AM, Tim wrote:

On Sat, 2015-05-23 at 11:44 -0600, jd1008 wrote:

Cmos battery will not crash the OS.

Some hardware does odd things when the battery is going (*).  Though you
usually have clock problems, at the same time, or as a precursor.

You said your clock is fine, but is your hardware always connected to
the mains?  Also, are your observations about the clock not tainted by
network time servers resetting your clock?

Some BIOSs appear to be jointly powered by the mains (when available)
and the battery, others seem to be solely powered via their battery.


The bios clock stays accurate even when not connected to AC.
Main battery is in good condition.
I do not know if cmos battery is augmented by main battery

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread Tim
On Sat, 2015-05-23 at 11:44 -0600, jd1008 wrote:
> Cmos battery will not crash the OS.

Some hardware does odd things when the battery is going (*).  Though you
usually have clock problems, at the same time, or as a precursor.

You said your clock is fine, but is your hardware always connected to
the mains?  Also, are your observations about the clock not tainted by
network time servers resetting your clock?

Some BIOSs appear to be jointly powered by the mains (when available)
and the battery, others seem to be solely powered via their battery.

-- 
tim@localhost ~]$ uname -rsvp

Linux 3.19.5-100.fc20.i686 #1 SMP Mon Apr 20 20:28:39 UTC 2015 i686

All mail to my mailbox is automatically deleted, there is no point trying
to privately email me, I will only read messages posted to the public lists.

George Orwell's '1984' was supposed to be a warning against tyranny, not
a set of instructions for supposedly democratic governments.

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-24 Thread David Timms
On 24/05/15 03:03, jd1008 wrote:
> I have an HP laptop with
> AMD Turion II X2 mobile processor RM-72 / 2.1 GHz CPU, Socket S1.
> It is now causing blue screens in windows and freezes
> linux (pclinuxos, knoppix, fedora live).
Having recently replaced my notebook, I leave a short tale:
Machine would instantly power off during heavy tasks (one development
language) under windows, and trying any yum based activity in Fedora 20.
In Fedora, even setting nice at max niceness rarely got me through a yum
update (with the trashed rpm db that the (emergency) power off caused).

I opened all parts of the notebook, and used a compressed air spray to
remove any dust I could. This made no significant difference. I just
avoided those tasks (eg hit ctrl-z as the fan spun up, waited ten
seconds and let it go fg).

A few months later, I got access to an air compressor with nozzle. While
there are risks in doing this (damage components either physically or
via static/moisture), cleaning it in all directions worked well (jam any
fans with a cable tie to stop the air over-speeding the fan), and I was
able to use the laptop normally for the next few months. (I blame the
dog's fur ;-)

Other recommendations I saw included removing the cpu to replace the
heat transfer paste/material with fresh new material, could be worth a
shot...

> *Thermal Design Power* *35 Watt*
How does this compare to the existing CPU's dissipation ?

Another thing you can do with multi-core cpu (at least intel) is to
instruct the kernel to disable cores. However, I think this could only
help you if it is the second core that is a problem, because I think the
first core always runs the "boot process", ie it needs to stay on.

Dave.
-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-23 Thread Aradenatorix Veckhom Vacelaevus
Well, what you comment is difficult to diagnose. Years ago HP
struggled with computers with this processor but it was by overheating
GPU that is soldered to the motherboard. This is true for certain
models of GPUs from nVidia, but I don't know if this is your case, you
didn't specify the model of your laptop.

If the problem were the CPU as you suspicious, you can get another one
used (it is very difficult to find a new one in 2015) easily from
Amazon or eBay and installation is not difficult. The hard part is to
access it, because HP has an awfule design and practically you have to
gut the entire laptop to do it.

Good luck.
-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-23 Thread jd1008



On 05/23/2015 04:44 PM, Steven Rosenberg wrote:


On Sat, May 23, 2015 at 10:03 AM, jd1008 > wrote:


I have an HP laptop with
AMD Turion II X2 mobile processor RM-72 / 2.1 GHz CPU, Socket S1.
It is now causing blue screens in windows and freezes
linux (pclinuxos, knoppix, fedora live).

I have run the x86 mem test for more than a day, and
found no problems with the 4GB ram (2GB X 2).

I am wondering why the memtest does not freeze
could it be that only one core is causing the problem?

At any rate I wanted to replace it with
AMD Turion II Ultra M660 TMM660DBO23GQ 2.7GHz Dual-Core Mobile CPU
Processor, Socket S1


Will I be running into any problems?



It's difficult to tell what exactly is causing your problem. At some 
level, laptops are disposable computers. It's hard to really "fix" 
them. They are hard to disassemble and reassemble. At least on a 
build-it-yourself desktop you can replace the motherboard without too 
much pain and suffering and get an overall performance boost at the 
same time.


But laptops? They're getting harder to work on, not easier. I got one 
for my daughter that was really cheap, and one of the things 
contributing to the cheapness was the lack of a replaceable battery. 
You can't pop the battery out. There is no access to the RAM, hard 
drive or battery unless you take the thing apart completely.


It's not like there was some kind of laptop-repair nirvana of years 
past. Once I had to replace a dead hard drive in an Apple iBook G4, 
and with detailed instructions it took about three hours.


Other laptops allow swapping of a hard drive or RAM in minutes. But 
even one of my older Toshiba laptops (say 12 years old at this point) 
didn't offer easy access to the hard drive.


If you can swing it, I'd just get a new laptop.


Well,
the cpu is easily removable. It is under the heat sink
which is also easily removable. This laptop is indeed
very easy to open up and put back together.

I am waiting for a couple of CPU and laptop gurus (they create
unbranded laptops), to get back to me about this question.

Thanx.

JD
--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-23 Thread Steven Rosenberg
On Sat, May 23, 2015 at 10:03 AM, jd1008  wrote:

> I have an HP laptop with
> AMD Turion II X2 mobile processor RM-72 / 2.1 GHz CPU, Socket S1.
> It is now causing blue screens in windows and freezes
> linux (pclinuxos, knoppix, fedora live).
>
> I have run the x86 mem test for more than a day, and
> found no problems with the 4GB ram (2GB X 2).
>
> I am wondering why the memtest does not freeze
> could it be that only one core is causing the problem?
>
> At any rate I wanted to replace it with
> AMD Turion II Ultra M660 TMM660DBO23GQ 2.7GHz Dual-Core Mobile CPU
> Processor, Socket S1
>
>
> Will I be running into any problems?
>


It's difficult to tell what exactly is causing your problem. At some level,
laptops are disposable computers. It's hard to really "fix" them. They are
hard to disassemble and reassemble. At least on a build-it-yourself desktop
you can replace the motherboard without too much pain and suffering and get
an overall performance boost at the same time.

But laptops? They're getting harder to work on, not easier. I got one for
my daughter that was really cheap, and one of the things contributing to
the cheapness was the lack of a replaceable battery. You can't pop the
battery out. There is no access to the RAM, hard drive or battery unless
you take the thing apart completely.

It's not like there was some kind of laptop-repair nirvana of years past.
Once I had to replace a dead hard drive in an Apple iBook G4, and with
detailed instructions it took about three hours.

Other laptops allow swapping of a hard drive or RAM in minutes. But even
one of my older Toshiba laptops (say 12 years old at this point) didn't
offer easy access to the hard drive.

If you can swing it, I'd just get a new laptop.

--
Steven Rosenberg
http://stevenrosenberg.net/blog
http://blogs.dailynews.com/click
-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-23 Thread jd1008



On 05/23/2015 11:38 AM, Tod Merley wrote:

cmos battery

On Sat, May 23, 2015 at 10:21 AM, Dave Stevens > wrote:


On Sat, 23 May 2015 11:03:31 -0600
jd1008 mailto:jd1...@gmail.com>> wrote:

> I have an HP laptop with
> AMD Turion II X2 mobile processor RM-72 / 2.1 GHz CPU, Socket S1.
> It is now causing blue screens in windows and freezes
> linux (pclinuxos, knoppix, fedora live).
>
> I have run the x86 mem test for more than a day, and
> found no problems with the 4GB ram (2GB X 2).
>
> I am wondering why the memtest does not freeze
> could it be that only one core is causing the problem?
>
> At any rate I wanted to replace it with
> AMD Turion II Ultra M660 TMM660DBO23GQ 2.7GHz Dual-Core Mobile CPU
> Processor, Socket S1
>
>
> Will I be running into any problems?

have you tried reflashing the BIOS?

D

> Would the heat be an issue?
>
> These are the full technical data on it:
>
> General information
>
>
> Type  CPU / Microprocessor
>  Market segment   Mobile
> FamilyAMD Turion II Ultra Dual-Core Mobile
>


>
> Model number  *M660
>

*
>
> CPU part number
>
>   * *TMM660DBO23GQ* is an OEM/tray microprocessor
>
> Stepping code NAEIC AE
> 
> *Frequency*   *2700 MHz*
> *Bus speed*   *One 1800 MHz 16-bit HyperTransport link (3.6
> GT/s)* *Clock multiplier* *13.5*
> Package   638-pin lidless organic micro Pin Grid Array (UOL638)
> *Socket*  *Socket S1 (S1g3)*
> Architecture / Microarchitecture
>
>
> Microarchitecture K10
> Platform  Tigris
> Processor coreCaspian
> Core stepping DA-C2
> CPUID 100F62
> Manufacturing process 0.045 micron silicon-on-insulator
(SOI)
> technology *Data width*   *64 bit*
> *The number of cores* *2*
> *The number of threads*   *2*
> Floating Point Unit   Integrated, 128-bit wide
> Level 1 cache size2 x 64 KB 2-way set associative instruction
> caches 2 x 64 KB 2-way set associative data caches
> *Level 2 cache size*  *2 x 1 MB 16-way set associative caches*
> *Level 3 cache size*  *None*
> Multiprocessing   Uniprocessor
> Features
>
>   * MMX instructions
>   * Extensions to MMX
>   * 3DNow! technology
>   * Extensions to 3DNow!
>   * SSE / Streaming SIMD Extensions
>   * SSE2 / Streaming SIMD Extensions 2
>   * SSE3 / Streaming SIMD Extensions 3
>   * SSE4a
>   * AMD64 / AMD 64-bit technology
>   * EVP / Enhanced Virus Protection
>   * AMD-V / AMD Virtualization technology
>
> Low power featuresPowerNow! 2.0
> Integrated peripherals / components
>
>
> Integrated graphics   None
> Memory controller The number of controllers: 1
> Memory channels: 2
> Channel width (bits): 64
> Supported memory: DDR2-800
> Maximum memory bandwidth (GB/s): 12.8
> Other peripherals HyperTransport technology 3.0
> Electrical / Thermal parameters
>
>
> *Thermal Design Power**35 Watt*
>
>

--


Cmos battery???
I do not lose time. Time stays correct even if I do not
plug in the laptop for weeks.
Cmos battery will not crash the OS.

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-23 Thread Tod Merley
cmos battery

On Sat, May 23, 2015 at 10:21 AM, Dave Stevens  wrote:

> On Sat, 23 May 2015 11:03:31 -0600
> jd1008  wrote:
>
> > I have an HP laptop with
> > AMD Turion II X2 mobile processor RM-72 / 2.1 GHz CPU, Socket S1.
> > It is now causing blue screens in windows and freezes
> > linux (pclinuxos, knoppix, fedora live).
> >
> > I have run the x86 mem test for more than a day, and
> > found no problems with the 4GB ram (2GB X 2).
> >
> > I am wondering why the memtest does not freeze
> > could it be that only one core is causing the problem?
> >
> > At any rate I wanted to replace it with
> > AMD Turion II Ultra M660 TMM660DBO23GQ 2.7GHz Dual-Core Mobile CPU
> > Processor, Socket S1
> >
> >
> > Will I be running into any problems?
>
> have you tried reflashing the BIOS?
>
> D
>
> > Would the heat be an issue?
> >
> > These are the full technical data on it:
> >
> > General information
> >
> >
> > Type  CPU / Microprocessor
> >  Market segment   Mobile
> > FamilyAMD Turion II Ultra Dual-Core Mobile
> > <
> http://www.cpu-world.com/CPUs/K10/TYPE-Turion%20II%20Ultra%20Dual-Core%20Mobile.html
> >
> >
> > Model number  *M660
> > <
> http://www.cpu-world.com/CPUs/K10/AMD-Turion%20II%20Ultra%20Dual-Core%20Mobile%20M660%20-%20TMM660DBO23GQ.html
> >*
> >
> > CPU part number
> >
> >   * *TMM660DBO23GQ* is an OEM/tray microprocessor
> >
> > Stepping code NAEIC AE
> > 
> > *Frequency*   *2700 MHz*
> > *Bus speed*   *One 1800 MHz 16-bit HyperTransport link (3.6
> > GT/s)* *Clock multiplier* *13.5*
> > Package   638-pin lidless organic micro Pin Grid Array (UOL638)
> > *Socket*  *Socket S1 (S1g3)*
> > Architecture / Microarchitecture
> >
> >
> > Microarchitecture K10
> > Platform  Tigris
> > Processor coreCaspian
> > Core stepping DA-C2
> > CPUID 100F62
> > Manufacturing process 0.045 micron silicon-on-insulator (SOI)
> > technology *Data width*   *64 bit*
> > *The number of cores* *2*
> > *The number of threads*   *2*
> > Floating Point Unit   Integrated, 128-bit wide
> > Level 1 cache size2 x 64 KB 2-way set associative instruction
> > caches 2 x 64 KB 2-way set associative data caches
> > *Level 2 cache size*  *2 x 1 MB 16-way set associative caches*
> > *Level 3 cache size*  *None*
> > Multiprocessing   Uniprocessor
> > Features
> >
> >   * MMX instructions
> >   * Extensions to MMX
> >   * 3DNow! technology
> >   * Extensions to 3DNow!
> >   * SSE / Streaming SIMD Extensions
> >   * SSE2 / Streaming SIMD Extensions 2
> >   * SSE3 / Streaming SIMD Extensions 3
> >   * SSE4a
> >   * AMD64 / AMD 64-bit technology
> >   * EVP / Enhanced Virus Protection
> >   * AMD-V / AMD Virtualization technology
> >
> > Low power featuresPowerNow! 2.0
> > Integrated peripherals / components
> >
> >
> > Integrated graphics   None
> > Memory controller The number of controllers: 1
> > Memory channels: 2
> > Channel width (bits): 64
> > Supported memory: DDR2-800
> > Maximum memory bandwidth (GB/s): 12.8
> > Other peripherals HyperTransport technology 3.0
> > Electrical / Thermal parameters
> >
> >
> > *Thermal Design Power**35 Watt*
> >
> >
>
> --
> users mailing list
> users@lists.fedoraproject.org
> To unsubscribe or change subscription options:
> https://admin.fedoraproject.org/mailman/listinfo/users
> Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
> Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
> Have a question? Ask away: http://ask.fedoraproject.org
>
-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-23 Thread jd1008



On 05/23/2015 11:21 AM, Dave Stevens wrote:

On Sat, 23 May 2015 11:03:31 -0600
jd1008  wrote:


I have an HP laptop with
AMD Turion II X2 mobile processor RM-72 / 2.1 GHz CPU, Socket S1.
It is now causing blue screens in windows and freezes
linux (pclinuxos, knoppix, fedora live).

I have run the x86 mem test for more than a day, and
found no problems with the 4GB ram (2GB X 2).

I am wondering why the memtest does not freeze
could it be that only one core is causing the problem?

At any rate I wanted to replace it with
AMD Turion II Ultra M660 TMM660DBO23GQ 2.7GHz Dual-Core Mobile CPU
Processor, Socket S1


Will I be running into any problems?

have you tried reflashing the BIOS?

D


No, as I do not have the bios file, nor the flash util.

--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Re: Replacing laptop cpu

2015-05-23 Thread Dave Stevens
On Sat, 23 May 2015 11:03:31 -0600
jd1008  wrote:

> I have an HP laptop with
> AMD Turion II X2 mobile processor RM-72 / 2.1 GHz CPU, Socket S1.
> It is now causing blue screens in windows and freezes
> linux (pclinuxos, knoppix, fedora live).
> 
> I have run the x86 mem test for more than a day, and
> found no problems with the 4GB ram (2GB X 2).
> 
> I am wondering why the memtest does not freeze
> could it be that only one core is causing the problem?
> 
> At any rate I wanted to replace it with
> AMD Turion II Ultra M660 TMM660DBO23GQ 2.7GHz Dual-Core Mobile CPU 
> Processor, Socket S1
> 
> 
> Will I be running into any problems?

have you tried reflashing the BIOS?

D

> Would the heat be an issue?
> 
> These are the full technical data on it:
> 
> General information
> 
>   
> Type  CPU / Microprocessor
>  Market segment   Mobile
> FamilyAMD Turion II Ultra Dual-Core Mobile 
> 
>  
> 
> Model number  *M660 
> *
>  
> 
> CPU part number   
> 
>   * *TMM660DBO23GQ* is an OEM/tray microprocessor
> 
> Stepping code NAEIC AE 
> 
> *Frequency*   *2700 MHz*
> *Bus speed*   *One 1800 MHz 16-bit HyperTransport link (3.6
> GT/s)* *Clock multiplier* *13.5*
> Package   638-pin lidless organic micro Pin Grid Array (UOL638)
> *Socket*  *Socket S1 (S1g3)*
> Architecture / Microarchitecture
> 
>   
> Microarchitecture K10
> Platform  Tigris
> Processor coreCaspian
> Core stepping DA-C2
> CPUID 100F62
> Manufacturing process 0.045 micron silicon-on-insulator (SOI)
> technology *Data width*   *64 bit*
> *The number of cores* *2*
> *The number of threads*   *2*
> Floating Point Unit   Integrated, 128-bit wide
> Level 1 cache size2 x 64 KB 2-way set associative instruction
> caches 2 x 64 KB 2-way set associative data caches
> *Level 2 cache size*  *2 x 1 MB 16-way set associative caches*
> *Level 3 cache size*  *None*
> Multiprocessing   Uniprocessor
> Features  
> 
>   * MMX instructions
>   * Extensions to MMX
>   * 3DNow! technology
>   * Extensions to 3DNow!
>   * SSE / Streaming SIMD Extensions
>   * SSE2 / Streaming SIMD Extensions 2
>   * SSE3 / Streaming SIMD Extensions 3
>   * SSE4a
>   * AMD64 / AMD 64-bit technology
>   * EVP / Enhanced Virus Protection
>   * AMD-V / AMD Virtualization technology
> 
> Low power featuresPowerNow! 2.0
> Integrated peripherals / components
> 
>   
> Integrated graphics   None
> Memory controller The number of controllers: 1
> Memory channels: 2
> Channel width (bits): 64
> Supported memory: DDR2-800
> Maximum memory bandwidth (GB/s): 12.8
> Other peripherals HyperTransport technology 3.0
> Electrical / Thermal parameters
> 
>   
> *Thermal Design Power**35 Watt*
> 
> 

-- 
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org


Replacing laptop cpu

2015-05-23 Thread jd1008

I have an HP laptop with
AMD Turion II X2 mobile processor RM-72 / 2.1 GHz CPU, Socket S1.
It is now causing blue screens in windows and freezes
linux (pclinuxos, knoppix, fedora live).

I have run the x86 mem test for more than a day, and
found no problems with the 4GB ram (2GB X 2).

I am wondering why the memtest does not freeze
could it be that only one core is causing the problem?

At any rate I wanted to replace it with
AMD Turion II Ultra M660 TMM660DBO23GQ 2.7GHz Dual-Core Mobile CPU 
Processor, Socket S1



Will I be running into any problems?
Would the heat be an issue?

These are the full technical data on it:

General information


TypeCPU / Microprocessor 
Market segment  Mobile
Family 	AMD Turion II Ultra Dual-Core Mobile 
 

Model number 	*M660 
* 


CPU part number 

 * *TMM660DBO23GQ* is an OEM/tray microprocessor

Stepping code 	NAEIC AE 


*Frequency* *2700 MHz*
*Bus speed* *One 1800 MHz 16-bit HyperTransport link (3.6 GT/s)*
*Clock multiplier*  *13.5*
Package 638-pin lidless organic micro Pin Grid Array (UOL638)
*Socket**Socket S1 (S1g3)*
Architecture / Microarchitecture


Microarchitecture   K10
PlatformTigris
Processor core  Caspian
Core stepping   DA-C2
CPUID   100F62
Manufacturing process   0.045 micron silicon-on-insulator (SOI) technology
*Data width**64 bit*
*The number of cores*   *2*
*The number of threads* *2*
Floating Point Unit Integrated, 128-bit wide
Level 1 cache size  2 x 64 KB 2-way set associative instruction caches
2 x 64 KB 2-way set associative data caches
*Level 2 cache size**2 x 1 MB 16-way set associative caches*
*Level 3 cache size**None*
Multiprocessing Uniprocessor
Features

 * MMX instructions
 * Extensions to MMX
 * 3DNow! technology
 * Extensions to 3DNow!
 * SSE / Streaming SIMD Extensions
 * SSE2 / Streaming SIMD Extensions 2
 * SSE3 / Streaming SIMD Extensions 3
 * SSE4a
 * AMD64 / AMD 64-bit technology
 * EVP / Enhanced Virus Protection
 * AMD-V / AMD Virtualization technology

Low power features  PowerNow! 2.0
Integrated peripherals / components


Integrated graphics None
Memory controller   The number of controllers: 1
Memory channels: 2
Channel width (bits): 64
Supported memory: DDR2-800
Maximum memory bandwidth (GB/s): 12.8
Other peripherals   HyperTransport technology 3.0
Electrical / Thermal parameters


*Thermal Design Power*  *35 Watt*


--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org