P7350 only has a thermal spec of 90°C? WTH!

All of the other Core 2s have a thermal spec of 100-105°C, why is the P7350's so low?

Doing a full system scan with Avira, my P7350 hit 76°C, that doesn't leave much buffer before it needs to shut down.

and it doesn't help the cooling on HP notebooks suck to begin with.

 

90C is still pretty high. It is dangerous for any electical component to be subjected to temperatures over 80C. At this point the components which make the processor, memory, gpu or whatnameyou start to decompose.

You should try undervolting your processor with rmclock or notebook hardware control. That will help decrease the heat generated from your processor and keep your system running cooler.

Check out The Undervolting Guide for more information

K-TRON

 

The P7350 usually doesn't go over 80 degrees C even under full load.
Temps have a tendency of being below 80 C really, so you don't have anything to worry about.
Plus, if you get a laptop cooler such as Zalman NC2000B, it will reduce the temps by about 10 degrees.

The only reason I can think of for now that P7350 has a thermal threshold of 90 deg C is because it's an entry Montevina cpu.

Also ... undervolting P7350 is not doable for now through RMclock, because it has half-multipliers (which the said program doesn't support).

I would love to undervolt if possible, but for now ... I can't think of a way to do it.
But ultimately there shouldn't really be any thermal issues to begin with as long as you clean out the laptop vents once per month (or once per 2 months).

 

What is the best way to clean out the vents?

air duster cans?

 

90*C is the Tcase temp, which doesn't play any role in shutdown or throttling. Try using a software such as CoreTemp or Everest Ultimate and hope it recognizes your CPU, so you can check the Tj(max).

Usually Tj(max) will be lower than the max Tcase, and when the core temps reach a few degrees below the max Tj, throttling will reduce frequency, multiplier and voltage to minimum. If you exceed Tj(max), then shut down is triggered. Intel doesn't really give any info on Tj, so you'll have to depend upon software to know the Tj(max).

To undervolt a P7350, you can use CrystalCPUID, and undervolt without losing any frequency, since it does recognize half-multipliers. The P7350 is a very cool CPU, and being a medium voltage CPU, it won't really cross Tj(max) with adequate cooling.

Checkout this guide.

 

It actually only operates at 20 degrees celcius. can i upgrade to a 105 degree celcius TDP proccesor?

 

Look into undervolting like Andy and K-Tron said.

 

My P7450 thermal spec is also 90.......TJMAX needs to be set at 90 otherwise it reads 10DegC too high.
Try Real Temp & see if that gives the same temp reading as what your using.
HwMonitor/Everest were all 10deg too high on mine until i adjusted the TJMax after it was pointed out to me here

 

Not true. Please stop making things up.

 

My T9600 in my Mac goes to 92c for 10mins or so before the fans kick in and it only drops to 82c while i am converting videos.

 

is this another case of someone obsessing over a 'number' they really don't understand?

 

There's a tweak to undervolt half-multipliers, actually. I have a T9600 which has 10.5x multipliers, and it's undervolted right now.

Here's how.

Check that guy romanov's post.

Reason is, RMClock just can't show the half-multipliers in the Profiles tab, so you have to modify the registry to show an 11x multiplier for a 10.5x multiplier (or 10x for a 9.5x and so forth) which will serve as its VID for the half multiplier.

 

High temperatures will increase wear on all components.

Maybe "decompose" wasn't the best word chosen by K-TRON but any constantly high temperature will increase wear onyour laptop - same goes for quick changes between high and low temps.

 

Again, please stop posting misinformation. High temperatures will NOT degrade the life span of you CPU/GPU in any realistic way. It was calculated through atomic physics that you can run a CPU/GPU NON-STOP for 20 years at its max temp and it will still work fine.

 

You should not forget that there is more than just a CPU in a laptop.

And heat isn't just localized in one single space.
You don't get a 90°C CPU and 1mm away acomponent at 20°C.

Yes, you can possibly run the CPU alone at its designed maximum temperature - but what about other components?
Don't forget them.

And Edit:
And no point did I refer to the CPU or GPU directly - I referred to a laptop in general.
Read the post properly please - especially when quoting.

 

Yes you did. You stated on ALL components including CPU, GPU, ect. and Ktron specifically targeted CPUs, GPUs, memory ect....this is simply atomically NOT true and which is why CPU, GPU core failure under spec conditions is nearly non-existent even for obsolete old CPUs.

 

You do realize the P-series CPU's have 25W TDP, right? Actual temps of 90C are unheard of for 25W TDP processors. I have a P8400 clocked to 2.7GHz and it's barely touched 70C.

Also according to the laws of thermodynamics, cooling is directly related to the difference in temperature between the ambient air and the item being cooled. Thus the hotter it gets the more apt it is to cool, so 20C of headroom is a lot on the upper end.

Finally, I have bets on someone changing from a 25W CPU with a 90C Tmax to a 35W CPU with a 105+C Tmax, and then freaking out when they find higher temperatures under load.

 

Your logic is 100% flawed. Is it also unheard of of an Atom CPU having only 2.5W TDP having 99c max temp? Nope....this is why your logic is flawed.

 

I meant that in reality you probably wouldn't hit 90C. Probably should rephrase the above.

Tmax is dependent on materials (ie:silicon), not the TDP. Thus a 2.5W Atom should have a similar Tmax to a 45W Q9100.

Upon looking at Intel's spec sheets, Atoms are rated at 90C and the Mobile Core 2 Quad at 100C. So there we go.

 

Amazing....I don't want to be rude but please dont post anything like this again and please dont even get me started....

 

? I thought I posted decent content. I'll continue to post what I consider useful stuff till I get banned. But in an effort to stay on topic, if you have any qualms let's do it via PM.

Anyways I hope the preceding has helped others.

 

You don't get banned for trying to help, lol

I think a 2.5W atom has 2.5W TDP, and 45W q9100 has a 45W TDP.

If, I'm correct (if not then correct me) then 2.5W is not simmilar to 45W.

 

You have to understand what you believe what is correct and what reality is can be two VERY different things such as in this case.

Its great to help people out but you have to understand that reality is not what one makes believe it is.

 

Bah here I go screwing things up again lol
I believe you're correct, but in reality I could just be perceiving things differently....

Now reads:
a 2.5W Atom should have a similar Tmax to a 45W Q9100.

Thanks again

 

This was interesting so i gave my P7350 a test with wprime 2.00 and realtemp 3.00:

P7350 idle temp: 41C 41C
P7350 load temp: 50C 50C (1024M wprime test)

There's probably cpu loading utilities far more demanding than wprime but at least i know that wprime can not heat the P7350 to more than 50C.

 

I don't know why the OP is being so panicky... the P series run way more cooler than the T series so a lower Tjmax is allright.. this thread has become into a flame war.. motion to close thread plz.

 

I gave Crysis a spin for about an hour and reached 58C, 58C (realtemp) on the P7350 and 65C on the HD4650 (gpu-z log to file). Distance to Tjmax is still 32C for the P7350.

 

I removed a couple of off topic messages.

Please try to stay on topic guys (and girls).

 

Oops srry about that
Anyway those temps would even be possible with such low power usage.

 

banned again he is.....

 

I agree with K-TRON's first post at the beginning of the thread: really, if any processor gets into the 80-90 degree C range, it is already in the danger range. If your computer's CPU gets above 70 degrees on load, you should probably take steps to try to lower the temperature: lift the laptop off the desk surface using some sort of stand (get creative - I use a Knex stand), use a notebook cooler, clean out dust from the heat exhaust, etc etc.

Not really - the P-series and T-series processors have such similar temperatures that it really comes down to the cooling system of each notebook and the conditions in which the computer is being used (ie: ambient temperature, whether or not it is elevated from the surface, etc).

Since yesterday evening...

 

K-TRON's post is completely wrong. Would Intel really say that 90C is the max temperature the processor should be allowed to get to if it caused the "materials to decompose?" High temperature makes the electronics consume more power and keeps it from clocking as high. It may make the computer unstable at stock speeds. It isn't going to cause damage, however.

He was correct though.

 

I have to agree with you.

I was reading the intel datasheets for my CPU and chipset. It said the TJmax for my CPU was 100C. But it also said damage was caused after hitting 130C.

 

I imagine some of the solder starts coming undone in excess of that 130C or so which is why some people can bake their GPU's. I don't know what form of solder they use and thus what the exact melting point is, but some solders usually melt at <200C.

 

wow this guy is a complete joke and needs to take a massive chill pill.

Anyway i agree with Ktron and others saying you should't let it go too high as parts will degrade. In a desktop CPU and environment its not so susceptical to wear and tear, but in a mobile system where all components are touching and very much share the heat + also crammed into the small dies then yes they do wear easier.

That IGrim guy doesn't remmeber all mechanical things dies off. Nothing last forever