Celeron frequency setting?

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

I just got a PC from a friend to sort out a bit. At bootup it showed Intel
Celeron 2.2G. I went into BIOS, set FSB to 133 (before it was just FSB
control disabled, nothing about current frequency). Now at start it shows
Celeron 2.93 G and below DRAM frequency 354 MHz. In BIOS RAM freq is set to
133/266 . MOBO is Gigaset GA-8PE800 Ultra.

After more than half hour of work(like defrag, working with disc, sorting
etc), temp is 41 degrees, so it's not overheating.
IS this setting correct ?
THX

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

The default FSB frequency for the Celeron should be 100MHz....therefore by
setting it to 133 you have made a 33% overclock to both the CPU and
Memory....as long as it works stable be happy!!!

--
*****Replace 'NOSPAM' with 'btinternet' in the reply address*****
"SleeperMan" <SleeperMan@too.sleepy> wrote in message
news:hpNrc.3507$37.427251@news.siol.net...
> I just got a PC from a friend to sort out a bit. At bootup it showed Intel
> Celeron 2.2G. I went into BIOS, set FSB to 133 (before it was just FSB
> control disabled, nothing about current frequency). Now at start it shows
> Celeron 2.93 G and below DRAM frequency 354 MHz. In BIOS RAM freq is set
to
> 133/266 . MOBO is Gigaset GA-8PE800 Ultra.
>
> After more than half hour of work(like defrag, working with disc, sorting
> etc), temp is 41 degrees, so it's not overheating.
> IS this setting correct ?
> THX
>
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

BigBadger typed:

> The default FSB frequency for the Celeron should be
> 100MHz....therefore by setting it to 133 you have made a 33%
> overclock to both the CPU and Memory....as long as it works stable be
> happy!!!



Aaaa...I didn't know that Celeron uses 100 MHZ and not 133. I thought that
all modern CPU's use 133...
So, in long term, it's still better to go back to 100...

Thanks!

>
>> I just got a PC from a friend to sort out a bit. At bootup it showed
>> Intel Celeron 2.2G. I went into BIOS, set FSB to 133 (before it was
>> just FSB control disabled, nothing about current frequency). Now at
>> start it shows Celeron 2.93 G and below DRAM frequency 354 MHz. In
>> BIOS RAM freq is set to 133/266 . MOBO is Gigaset GA-8PE800 Ultra.
>>
>> After more than half hour of work(like defrag, working with disc,
>> sorting etc), temp is 41 degrees, so it's not overheating.
>> IS this setting correct ?
>> THX

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

As long as it's stable and the temps are ok just leave it at 133....it's an
extra 730Mhz for free!...that's what overclocking is all about

--
*****Replace 'NOSPAM' with 'btinternet' in the reply address*****
"SleeperMan" <SleeperMan@too.sleepy> wrote in message
news:1QNrc.3508$37.427335@news.siol.net...
> BigBadger typed:
>
> > The default FSB frequency for the Celeron should be
> > 100MHz....therefore by setting it to 133 you have made a 33%
> > overclock to both the CPU and Memory....as long as it works stable be
> > happy!!!
>
>
>
> Aaaa...I didn't know that Celeron uses 100 MHZ and not 133. I thought that
> all modern CPU's use 133...
> So, in long term, it's still better to go back to 100...
>
> Thanks!
>
> >
> >> I just got a PC from a friend to sort out a bit. At bootup it showed
> >> Intel Celeron 2.2G. I went into BIOS, set FSB to 133 (before it was
> >> just FSB control disabled, nothing about current frequency). Now at
> >> start it shows Celeron 2.93 G and below DRAM frequency 354 MHz. In
> >> BIOS RAM freq is set to 133/266 . MOBO is Gigaset GA-8PE800 Ultra.
> >>
> >> After more than half hour of work(like defrag, working with disc,
> >> sorting etc), temp is 41 degrees, so it's not overheating.
> >> IS this setting correct ?
> >> THX
>
>
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Is this setting correct? Only if you want to overclock. You have set the
FSB speed to 33% above normal, giving a 33% greater CPU clock speed,
something that is fairly easy to do with almost any Intel Pentium CPU that
isn't already pushing the limits of the type (for example, a Pentium III 1
GHz is going to be difficult to get 33% higher because it is already the
top of that particular line.) It is not the standard setting, but most
people in this newsgroup would do it in a heartbeat (or clock cycle.0

The 41 degree C temperature (if that is the temperature reading from the CPU
on-die thermal diode) is well within the limits of the operating range of
the Celeron 2.2, but be aware that CPU usage is not particularly intense
when there is a lot of disk activity, because the limit is disk I/O speed
rather than CPU availablity. A better check for stability under CPU
stressful conditions is ripping an mp3 file from a music CD.

--
Phil Weldon, pweldonatmindjumpdotcom
For communication,
replace "at" with the 'at sign'
replace "mindjump" with "mindspring."
replace "dot" with "."

"SleeperMan" <SleeperMan@too.sleepy> wrote in message
news:hpNrc.3507$37.427251@news.siol.net...
> I just got a PC from a friend to sort out a bit. At bootup it showed Intel
> Celeron 2.2G. I went into BIOS, set FSB to 133 (before it was just FSB
> control disabled, nothing about current frequency). Now at start it shows
> Celeron 2.93 G and below DRAM frequency 354 MHz. In BIOS RAM freq is set
to
> 133/266 . MOBO is Gigaset GA-8PE800 Ultra.
>
> After more than half hour of work(like defrag, working with disc, sorting
> etc), temp is 41 degrees, so it's not overheating.
> IS this setting correct ?
> THX
>
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

BigBadger typed:

> As long as it's stable and the temps are ok just leave it at
> 133....it's an extra 730Mhz for free!...that's what overclocking is
> all about

Well, sure, if it would be mine...maybe ill ask him about it. PC is still
under warranty and i don't want to die and then i'll take the blaim...

>
>> BigBadger typed:
>>
>>> The default FSB frequency for the Celeron should be
>>> 100MHz....therefore by setting it to 133 you have made a 33%
>>> overclock to both the CPU and Memory....as long as it works stable
>>> be happy!!!
>>
>>
>>
>> Aaaa...I didn't know that Celeron uses 100 MHZ and not 133. I
>> thought that all modern CPU's use 133...
>> So, in long term, it's still better to go back to 100...
>>
>> Thanks!
>>
>>>
>>>> I just got a PC from a friend to sort out a bit. At bootup it
>>>> showed Intel Celeron 2.2G. I went into BIOS, set FSB to 133
>>>> (before it was just FSB control disabled, nothing about current
>>>> frequency). Now at start it shows Celeron 2.93 G and below DRAM
>>>> frequency 354 MHz. In BIOS RAM freq is set to 133/266 . MOBO is
>>>> Gigaset GA-8PE800 Ultra.
>>>>
>>>> After more than half hour of work(like defrag, working with disc,
>>>> sorting etc), temp is 41 degrees, so it's not overheating.
>>>> IS this setting correct ?
>>>> THX

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Phil Weldon typed:

> Is this setting correct? Only if you want to overclock. You have
> set the FSB speed to 33% above normal, giving a 33% greater CPU clock
> speed, something that is fairly easy to do with almost any Intel
> Pentium CPU that isn't already pushing the limits of the type (for
> example, a Pentium III 1 GHz is going to be difficult to get 33%
> higher because it is already the top of that particular line.) It is
> not the standard setting, but most people in this newsgroup would do
> it in a heartbeat (or clock cycle.0
>
> The 41 degree C temperature (if that is the temperature reading from
> the CPU on-die thermal diode) is well within the limits of the
> operating range of the Celeron 2.2, but be aware that CPU usage is
> not particularly intense when there is a lot of disk activity,
> because the limit is disk I/O speed rather than CPU availablity. A
> better check for stability under CPU stressful conditions is ripping
> an mp3 file from a music CD.
>

Temp is from BIOS PC Health screen after rebooting. I'll try some heavy CPU
usage for a while and see. MP3 sounds perfect.


>
> "SleeperMan" <SleeperMan@too.sleepy> wrote in message
> news:hpNrc.3507$37.427251@news.siol.net...
>> I just got a PC from a friend to sort out a bit. At bootup it showed
>> Intel Celeron 2.2G. I went into BIOS, set FSB to 133 (before it was
>> just FSB control disabled, nothing about current frequency). Now at
>> start it shows Celeron 2.93 G and below DRAM frequency 354 MHz. In
>> BIOS RAM freq is set to 133/266 . MOBO is Gigaset GA-8PE800 Ultra.
>>
>> After more than half hour of work(like defrag, working with disc,
>> sorting etc), temp is 41 degrees, so it's not overheating.
>> IS this setting correct ?
>> THX

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

CPU's cool off VERY quickly, after all, a heatsink is used. The better the
heatsink and the contact between the heatsink and the CPU, the faster
cool-off. Maybe you can use MotherBoard Monitor to get the readings from
the CPU on-chip thermal diode during operation; all Pentium and Celeron
CPU's have the on-chip thermal diodes, but the motherboard must have a
monitoring chip. If the temperature appears a BIOS screen, then there is a
monitoring chip; MotherBoard Monitor can work with almost any monitoring
chip.

--
Phil Weldon, pweldonatmindjumpdotcom
For communication,
replace "at" with the 'at sign'
replace "mindjump" with "mindspring."
replace "dot" with "."

"SleeperMan" <SleeperMan@too.sleepy> wrote in message
news:1pXrc.3520$37.431122@news.siol.net...
> Phil Weldon typed:
>
> > Is this setting correct? Only if you want to overclock. You have
> > set the FSB speed to 33% above normal, giving a 33% greater CPU clock
> > speed, something that is fairly easy to do with almost any Intel
> > Pentium CPU that isn't already pushing the limits of the type (for
> > example, a Pentium III 1 GHz is going to be difficult to get 33%
> > higher because it is already the top of that particular line.) It is
> > not the standard setting, but most people in this newsgroup would do
> > it in a heartbeat (or clock cycle.0
> >
> > The 41 degree C temperature (if that is the temperature reading from
> > the CPU on-die thermal diode) is well within the limits of the
> > operating range of the Celeron 2.2, but be aware that CPU usage is
> > not particularly intense when there is a lot of disk activity,
> > because the limit is disk I/O speed rather than CPU availablity. A
> > better check for stability under CPU stressful conditions is ripping
> > an mp3 file from a music CD.
> >
>
> Temp is from BIOS PC Health screen after rebooting. I'll try some heavy
CPU
> usage for a while and see. MP3 sounds perfect.
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Phil Weldon typed:

> CPU's cool off VERY quickly, after all, a heatsink is used. The
> better the heatsink and the contact between the heatsink and the CPU,
> the faster cool-off. Maybe you can use MotherBoard Monitor to get
> the readings from the CPU on-chip thermal diode during operation; all
> Pentium and Celeron CPU's have the on-chip thermal diodes, but the
> motherboard must have a monitoring chip. If the temperature appears
> a BIOS screen, then there is a monitoring chip; MotherBoard Monitor
> can work with almost any monitoring chip.


Good point. I'll do that. THX

>
>
> "SleeperMan" <SleeperMan@too.sleepy> wrote in message
> news:1pXrc.3520$37.431122@news.siol.net...
>> Phil Weldon typed:
>>
>>> Is this setting correct? Only if you want to overclock. You have
>>> set the FSB speed to 33% above normal, giving a 33% greater CPU
>>> clock speed, something that is fairly easy to do with almost any
>>> Intel Pentium CPU that isn't already pushing the limits of the type
>>> (for example, a Pentium III 1 GHz is going to be difficult to get
>>> 33% higher because it is already the top of that particular line.)
>>> It is not the standard setting, but most people in this newsgroup
>>> would do it in a heartbeat (or clock cycle.0
>>>
>>> The 41 degree C temperature (if that is the temperature reading from
>>> the CPU on-die thermal diode) is well within the limits of the
>>> operating range of the Celeron 2.2, but be aware that CPU usage is
>>> not particularly intense when there is a lot of disk activity,
>>> because the limit is disk I/O speed rather than CPU availablity. A
>>> better check for stability under CPU stressful conditions is ripping
>>> an mp3 file from a music CD.
>>>
>>
>> Temp is from BIOS PC Health screen after rebooting. I'll try some
>> heavy CPU usage for a while and see. MP3 sounds perfect.

 

[augustus]

Archived from groups: alt.comp.hardware.overclocking (More info?)

"SleeperMan" <SleeperMan@too.sleepy> wrote in message
news:hpNrc.3507$37.427251@news.siol.net...
> I just got a PC from a friend to sort out a bit. At bootup it showed Intel
> Celeron 2.2G. I went into BIOS, set FSB to 133 (before it was just FSB
> control disabled, nothing about current frequency). Now at start it shows
> Celeron 2.93 G and below DRAM frequency 354 MHz. In BIOS RAM freq is set
to
> 133/266 . MOBO is Gigaset GA-8PE800 Ultra.
>
> After more than half hour of work(like defrag, working with disc, sorting
> etc), temp is 41 degrees, so it's not overheating.
> IS this setting correct ?
> THX

Should be 100Mhz....not 133Mhz. If it was your own system , go ahead and
leave it like that. Might be OK indefinitely, might last a few months or a
year or two. I rarely ever overclock other's systems due to the simple fact
that you're responsible for any instability and shortened component life.
Overclocking a Celeron is like polishing a turd. Even when it's done, it's
still a turd. A 2.0Ghz P4 Prescott will outperform that Celeron running at
2.93Ghz.

 

[augustus]

Archived from groups: alt.comp.hardware.overclocking (More info?)

> Should be 100Mhz....not 133Mhz. If it was your own system , go ahead and
> leave it like that. Might be OK indefinitely, might last a few months or a
> year or two. I rarely ever overclock other's systems due to the simple
fact
> that you're responsible for any instability and shortened component life.
> Overclocking a Celeron is like polishing a turd. Even when it's done, it's
> still a turd. A 2.0Ghz P4 Prescott will outperform that Celeron running at
> 2.93Ghz.

Just so there's no confusion, the original .18 fab Socket 478 400Mhz FSB
256K L2 lineup was code named Prescott initilally. Now the controvertial new
version is also called Prescott. There was only a 1.9 and 2.0 Ghz Prescott
P4 produced before the Northwood .13 production started.

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

When you find reliable information about an Intel Pentium family CPU
destroyed by overclocking, please post it here because that would be a
first.

--
Phil Weldon, pweldonatmindjumpdotcom
For communication,
replace "at" with the 'at sign'
replace "mindjump" with "mindspring."
replace "dot" with "."

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Phil Weldon typed:

> When you find reliable information about an Intel Pentium family CPU
> destroyed by overclocking, please post it here because that would be a
> first.

That would make sense, since as i learned up to now, Intel just stops if max
temp has reached. That's very usefull, i think AMD stops, too. But not from
CPU stop, but from CPU burn-out...

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

At a certain point, AMD put a temperature limiting device on their CPU's but
it required circuitry on the motherboard to function, and that was not
present on some. There were some spectacular pictures posted of the
results. I'd guess that now all AMD CPU's now being produced have
self-contained protection against disaterous overheating. Intel CPU's since
introduction of the first Pentium have had that; a thermal-stop diode on
the chip that shuts the CPU down and another thermal diode on the chip that
is connected to two pins, allowing reading a internal temperature for the
CPU. Thermal stop is self-contained, the thermal diode requires motherboard
circuitry to function.

--
Phil Weldon, pweldonatmindjumpdotcom
For communication,
replace "at" with the 'at sign'
replace "mindjump" with "mindspring."
replace "dot" with "."


"SleeperMan" <SleeperMan@too.sleepy> wrote in message
news:KTqsc.3599$37.445547@news.siol.net...
> Phil Weldon typed:
>
> > When you find reliable information about an Intel Pentium family CPU
> > destroyed by overclocking, please post it here because that would be a
> > first.
>
> That would make sense, since as i learned up to now, Intel just stops if
max
> temp has reached. That's very usefull, i think AMD stops, too. But not
from
> CPU stop, but from CPU burn-out...
>
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Phil Weldon typed:

> At a certain point, AMD put a temperature limiting device on their
> CPU's but it required circuitry on the motherboard to function, and
> that was not present on some. There were some spectacular pictures
> posted of the results. I'd guess that now all AMD CPU's now being
> produced have self-contained protection against disaterous
> overheating. Intel CPU's since introduction of the first Pentium
> have had that; a thermal-stop diode on the chip that shuts the CPU
> down and another thermal diode on the chip that is connected to two
> pins, allowing reading a internal temperature for the CPU. Thermal
> stop is self-contained, the thermal diode requires motherboard
> circuitry to function.
>
>
> "SleeperMan" <SleeperMan@too.sleepy> wrote in message
> news:KTqsc.3599$37.445547@news.siol.net...
>> Phil Weldon typed:
>>
>>> When you find reliable information about an Intel Pentium family CPU
>>> destroyed by overclocking, please post it here because that would
>>> be a first.
>>
>> That would make sense, since as i learned up to now, Intel just
>> stops if max temp has reached. That's very usefull, i think AMD
>> stops, too. But not from CPU stop, but from CPU burn-out...

I can read CPU temp. I'm just not sure, if it's form CPU internal of from
MOBO sensor, which is supposed to be fitted just below the CPU. My CPU is
Athlon 1700+ and mobo Soyo dragon plus k7v...
it's just if Intel CPU's weren't so bloddy expensive...

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

You're reading the temperature from a sensor under the CPU socket, perhaps
with a fudge factor to make the reading approximate a corresponding
temperature that might be measured inside the CPU. The temperature under
the socket is not going to track the CPU internal temperature very well,
and motherboard temperature is going to be a factor. With an NT based
operating system, and Intel CPU cools off very rapidly at idle, and the
internal diode shows the drop to near the temperature of the air entering
the heatsink in less than a second. On the other hand, your AMD CPU clock
multiplier can be unlocked, the better to fry it! But to tell the truth, I
think AMD meltdowns are caused by faulty heatsink installation rather than
by overclocking. After all, if the overclocking is too extreme, the system
won't boot! And when it comes to raising the core voltage unreasonably
high chips from both manufactures all die.

--
Phil Weldon, pweldonatmindjumpdotcom
For communication,
replace "at" with the 'at sign'
replace "mindjump" with "mindspring."
replace "dot" with "."


"SleeperMan" <SleeperMan@too.sleepy> wrote in message
news:aersc.3602$37.445830@news.siol.net...
> Phil Weldon typed:
>
> > At a certain point, AMD put a temperature limiting device on their
> > CPU's but it required circuitry on the motherboard to function, and
> > that was not present on some. There were some spectacular pictures
> > posted of the results. I'd guess that now all AMD CPU's now being
> > produced have self-contained protection against disaterous
> > overheating. Intel CPU's since introduction of the first Pentium
> > have had that; a thermal-stop diode on the chip that shuts the CPU
> > down and another thermal diode on the chip that is connected to two
> > pins, allowing reading a internal temperature for the CPU. Thermal
> > stop is self-contained, the thermal diode requires motherboard
> > circuitry to function.
> >
> >
> > "SleeperMan" <SleeperMan@too.sleepy> wrote in message
> > news:KTqsc.3599$37.445547@news.siol.net...
> >> Phil Weldon typed:
> >>
> >>> When you find reliable information about an Intel Pentium family CPU
> >>> destroyed by overclocking, please post it here because that would
> >>> be a first.
> >>
> >> That would make sense, since as i learned up to now, Intel just
> >> stops if max temp has reached. That's very usefull, i think AMD
> >> stops, too. But not from CPU stop, but from CPU burn-out...
>
> I can read CPU temp. I'm just not sure, if it's form CPU internal of from
> MOBO sensor, which is supposed to be fitted just below the CPU. My CPU
is
> Athlon 1700+ and mobo Soyo dragon plus k7v...
> it's just if Intel CPU's weren't so bloddy expensive...
>
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Phil Weldon wrote:
> You're reading the temperature from a sensor under the CPU socket, perhaps
> with a fudge factor to make the reading approximate a corresponding
> temperature that might be measured inside the CPU. The temperature under
> the socket is not going to track the CPU internal temperature very well,
> and motherboard temperature is going to be a factor. With an NT based
> operating system, and Intel CPU cools off very rapidly at idle, and the
> internal diode shows the drop to near the temperature of the air entering
> the heatsink in less than a second.

Well, it drops to the temperature of the heatsink (if it tried to drop
lower then heat would flow from the heatsink TO the die rather than away
from it), which is not necessarily the temperature of the air entering it.
If the processor is not putting out much thermal flux then the heatsink
will cool down too, the time depending on it's thermal mass.

> On the other hand, your AMD CPU clock
> multiplier can be unlocked, the better to fry it! But to tell the truth, I
> think AMD meltdowns are caused by faulty heatsink installation rather than
> by overclocking. After all, if the overclocking is too extreme, the system
> won't boot! And when it comes to raising the core voltage unreasonably
> high chips from both manufactures all die.
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

You're right here. My friend has newer AMD (tbred type) and he told that CPU
temp drops very quickly after dropping CPU usage. Mine doesn't. SO, i guess
that tbred has built-in sensor, while Athlon doesn't.
Also you're right about overclocking - when i tried to run my 1700 + at 150
M, it just didn't boot...

Phil Weldon typed:

> You're reading the temperature from a sensor under the CPU socket,
> perhaps with a fudge factor to make the reading approximate a
> corresponding temperature that might be measured inside the CPU.
> The temperature under the socket is not going to track the CPU
> internal temperature very well, and motherboard temperature is going
> to be a factor. With an NT based operating system, and Intel CPU
> cools off very rapidly at idle, and the internal diode shows the
> drop to near the temperature of the air entering the heatsink in less
> than a second. On the other hand, your AMD CPU clock multiplier can
> be unlocked, the better to fry it! But to tell the truth, I think
> AMD meltdowns are caused by faulty heatsink installation rather than
> by overclocking. After all, if the overclocking is too extreme, the
> system won't boot! And when it comes to raising the core voltage
> unreasonably high chips from both manufactures all die.
>
>
> "SleeperMan" <SleeperMan@too.sleepy> wrote in message
> news:aersc.3602$37.445830@news.siol.net...
>> Phil Weldon typed:
>>
>>> At a certain point, AMD put a temperature limiting device on their
>>> CPU's but it required circuitry on the motherboard to function, and
>>> that was not present on some. There were some spectacular pictures
>>> posted of the results. I'd guess that now all AMD CPU's now being
>>> produced have self-contained protection against disaterous
>>> overheating. Intel CPU's since introduction of the first Pentium
>>> have had that; a thermal-stop diode on the chip that shuts the CPU
>>> down and another thermal diode on the chip that is connected to two
>>> pins, allowing reading a internal temperature for the CPU. Thermal
>>> stop is self-contained, the thermal diode requires motherboard
>>> circuitry to function.
>>>
>>>
>>> "SleeperMan" <SleeperMan@too.sleepy> wrote in message
>>> news:KTqsc.3599$37.445547@news.siol.net...
>>>> Phil Weldon typed:
>>>>
>>>>> When you find reliable information about an Intel Pentium family
>>>>> CPU destroyed by overclocking, please post it here because that
>>>>> would be a first.
>>>>
>>>> That would make sense, since as i learned up to now, Intel just
>>>> stops if max temp has reached. That's very usefull, i think AMD
>>>> stops, too. But not from CPU stop, but from CPU burn-out...
>>
>> I can read CPU temp. I'm just not sure, if it's form CPU internal of
>> from MOBO sensor, which is supposed to be fitted just below the
>> CPU. My CPU is Athlon 1700+ and mobo Soyo dragon plus k7v...
>> it's just if Intel CPU's weren't so bloddy expensive...

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Credit participation in this newsgroup and a little research on the side.

--
Phil Weldon, pweldonatmindjumpdotcom
For communication,
replace "at" with the 'at sign'
replace "mindjump" with "mindspring."
replace "dot" with "."


"SleeperMan" <SleeperMan@too.sleepy> wrote in message
news:mJrsc.3608$37.446231@news.siol.net...
> You're right here. My friend has newer AMD (tbred type) and he told that
CPU
> temp drops very quickly after dropping CPU usage. Mine doesn't. SO, i
guess
> that tbred has built-in sensor, while Athlon doesn't.
> Also you're right about overclocking - when i tried to run my 1700 + at
150
> M, it just didn't boot...
>
> Phil Weldon typed:
>
> > You're reading the temperature from a sensor under the CPU socket,
> > perhaps with a fudge factor to make the reading approximate a
> > corresponding temperature that might be measured inside the CPU.
> > The temperature under the socket is not going to track the CPU
> > internal temperature very well, and motherboard temperature is going
> > to be a factor. With an NT based operating system, and Intel CPU
> > cools off very rapidly at idle, and the internal diode shows the
> > drop to near the temperature of the air entering the heatsink in less
> > than a second. On the other hand, your AMD CPU clock multiplier can
> > be unlocked, the better to fry it! But to tell the truth, I think
> > AMD meltdowns are caused by faulty heatsink installation rather than
> > by overclocking. After all, if the overclocking is too extreme, the
> > system won't boot! And when it comes to raising the core voltage
> > unreasonably high chips from both manufactures all die.
> >
> >
> > "SleeperMan" <SleeperMan@too.sleepy> wrote in message
> > news:aersc.3602$37.445830@news.siol.net...
> >> Phil Weldon typed:
> >>
> >>> At a certain point, AMD put a temperature limiting device on their
> >>> CPU's but it required circuitry on the motherboard to function, and
> >>> that was not present on some. There were some spectacular pictures
> >>> posted of the results. I'd guess that now all AMD CPU's now being
> >>> produced have self-contained protection against disaterous
> >>> overheating. Intel CPU's since introduction of the first Pentium
> >>> have had that; a thermal-stop diode on the chip that shuts the CPU
> >>> down and another thermal diode on the chip that is connected to two
> >>> pins, allowing reading a internal temperature for the CPU. Thermal
> >>> stop is self-contained, the thermal diode requires motherboard
> >>> circuitry to function.
> >>>
> >>>
> >>> "SleeperMan" <SleeperMan@too.sleepy> wrote in message
> >>> news:KTqsc.3599$37.445547@news.siol.net...
> >>>> Phil Weldon typed:
> >>>>
> >>>>> When you find reliable information about an Intel Pentium family
> >>>>> CPU destroyed by overclocking, please post it here because that
> >>>>> would be a first.
> >>>>
> >>>> That would make sense, since as i learned up to now, Intel just
> >>>> stops if max temp has reached. That's very usefull, i think AMD
> >>>> stops, too. But not from CPU stop, but from CPU burn-out...
> >>
> >> I can read CPU temp. I'm just not sure, if it's form CPU internal of
> >> from MOBO sensor, which is supposed to be fitted just below the
> >> CPU. My CPU is Athlon 1700+ and mobo Soyo dragon plus k7v...
> >> it's just if Intel CPU's weren't so bloddy expensive...
>
>
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

The heatsink very quickly drops very, very, close to the temperature of the
air entering it once CPU is producing no heat, and functioning mostly as an
insulator. It is possible the temperature of a black anodized heatsink
could even reach a temperature slightly below that of air entering the
heatsink under certain circumstance (the heatsink "sees" surfaces cooler
than the air entering the heatsink... yes, I was suprised; radiative
transfer at ~ 20 degrees C is not as trivial as I thought... not
significant, but not trivial.)

--
Phil Weldon, pweldonatmindjumpdotcom
For communication,
replace "at" with the 'at sign'
replace "mindjump" with "mindspring."
replace "dot" with "."


"David Maynard" <dNOTmayn@ev1.net> wrote in message
news:10b53dngn0fe144@corp.supernews.com...
> Phil Weldon wrote:
> > You're reading the temperature from a sensor under the CPU socket,
perhaps
> > with a fudge factor to make the reading approximate a corresponding
> > temperature that might be measured inside the CPU. The temperature
under
> > the socket is not going to track the CPU internal temperature very
well,
> > and motherboard temperature is going to be a factor. With an NT based
> > operating system, and Intel CPU cools off very rapidly at idle, and
the
> > internal diode shows the drop to near the temperature of the air
entering
> > the heatsink in less than a second.
>
> Well, it drops to the temperature of the heatsink (if it tried to drop
> lower then heat would flow from the heatsink TO the die rather than away
> from it), which is not necessarily the temperature of the air entering it.
> If the processor is not putting out much thermal flux then the heatsink
> will cool down too, the time depending on it's thermal mass.
>
> > On the other hand, your AMD CPU clock
> > multiplier can be unlocked, the better to fry it! But to tell the
truth, I
> > think AMD meltdowns are caused by faulty heatsink installation rather
than
> > by overclocking. After all, if the overclocking is too extreme, the
system
> > won't boot! And when it comes to raising the core voltage unreasonably
> > high chips from both manufactures all die.
> >
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Phil Weldon wrote:

> The heatsink very quickly drops very, very, close to the temperature of the
> air entering it once CPU is producing no heat, and functioning mostly as an
> insulator. It is possible the temperature of a black anodized heatsink
> could even reach a temperature slightly below that of air entering the
> heatsink under certain circumstance (the heatsink "sees" surfaces cooler
> than the air entering the heatsink... yes, I was suprised; radiative
> transfer at ~ 20 degrees C is not as trivial as I thought... not
> significant, but not trivial.)
>

Well, as I said, it depends on the thermal mass of the heatsink and what
you consider 'quickly' to be.

At any rate, I'm just pointing out that it's observable.

The same happens in reverse, of course. The thermal diode reading will go
to heatsink temp+thermal insulation(plus heatspreader) resistance almost
instantly under load. Then you'll observe a slower rise as the heatsink
thermal mass heats up and then an even slower rise as the case air temp rises.

It's a matter of the thermal mass, which must absorb heat flux or else it
could not dissipate the heat as it must have a temp higher than the
incoming air in order for heat to flow.

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

You are right, of course. Long, long ago, in the days of the Celeron 300A,
I had a drilled out heatsink with thermistor (isolated from the heatsink,
but in contact with the CPU heat spreader) and a motherboard that could
measure temperatures via the thermal diode. Somewhere I have the numbers
from that, but I didn't track the temperatures with short time intervals,
and it was in the days before rock and roll, er... Windows 2000 with the CPU
idle function.

--
Phil Weldon, pweldonatmindjumpdotcom
For communication,
replace "at" with the 'at sign'
replace "mindjump" with "mindspring."
replace "dot" with "."


"David Maynard" <dNOTmayn@ev1.net> wrote in message
news:10b5c4pmdshp15c@corp.supernews.com...
> Phil Weldon wrote:
>
> > The heatsink very quickly drops very, very, close to the temperature of
the
> > air entering it once CPU is producing no heat, and functioning mostly as
an
> > insulator. It is possible the temperature of a black anodized heatsink
> > could even reach a temperature slightly below that of air entering the
> > heatsink under certain circumstance (the heatsink "sees" surfaces cooler
> > than the air entering the heatsink... yes, I was suprised; radiative
> > transfer at ~ 20 degrees C is not as trivial as I thought... not
> > significant, but not trivial.)
> >
>
> Well, as I said, it depends on the thermal mass of the heatsink and what
> you consider 'quickly' to be.
>
> At any rate, I'm just pointing out that it's observable.
>
> The same happens in reverse, of course. The thermal diode reading will go
> to heatsink temp+thermal insulation(plus heatspreader) resistance almost
> instantly under load. Then you'll observe a slower rise as the heatsink
> thermal mass heats up and then an even slower rise as the case air temp
rises.
>
> It's a matter of the thermal mass, which must absorb heat flux or else it
> could not dissipate the heat as it must have a temp higher than the
> incoming air in order for heat to flow.
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Phil Weldon wrote:

> You are right, of course. Long, long ago, in the days of the Celeron 300A,
> I had a drilled out heatsink with thermistor (isolated from the heatsink,
> but in contact with the CPU heat spreader) and a motherboard that could
> measure temperatures via the thermal diode. Somewhere I have the numbers
> from that, but I didn't track the temperatures with short time intervals,
> and it was in the days before rock and roll, er... Windows 2000 with the CPU
> idle function.
>

Hehe. Doesn't rock and roll predate the written word? I'm sure I saw Ringo
as a cave man somewhere.

I hadn't thought about it either till I was working on a system with
thermal diode monitoring, which is why it took us by surprise: we had just
never thought about it.