Gaming on a X61 - CPU temperature.

I haven't played games for ages (at least 4 years) but decided to pickup a copy of Counter Strike 1.6 yesterday. I feel like a kid again

Although I know it is an old game and not really graphics hungry, I'm still quite impressed by my X61. After all it is an ultraportable with an integrated graphics chip. I constantly get maximum FPS (99) so the gameplay is very smooth.

I however have two questions:

1) Is there any reason why OpenGL doesn't render as many FPS as D3D does on the GMA X3100?

2) The CPU temperature gets up to 75-80 degrees celcius during gameplay. While idle temperatures gets all the way down to between 40-43 degrees celcius. I am a bit worried about the high CPU temp during gameplay. I however read somewhere that the max temp for a mobile Core2Duo CPU is 100 degrees celcius. Can anyone enlighten me. Should I be worried?

How about the rest of you? Has anyone tried playing older 3D games on their X61? If so, what is the CPU temperature during play?

 

1) Vista might have to do with it, or the drivers for the X3100 which are still in development.
2) 80C isn't bad at all. I've seen machines running 95-105C fine, though I wouldn't recommend that. But yes, 100C is definitely pushing it but anything like 80C is just toasty but fine.

 

1)x3100 OpenGL driver support is not nearly as good as it's D3D driver support.
2)Most of the Core processors are fine up to 95C or 100C. 80C is definitely nothing to worry about.

 

Thanks a million guys! This is why I love NBR.

Greg: I see you changed your Avatar as well

 

Since when is 95C or 100C fine?
Are you guys kidding?
I consider 80C pushing it.

 

According to Intel Specs, 100C is the maximum. (=water's boiling point).
I woudn't want my processor to run anywhere near that point, for any period of time.

http://processorfinder.intel.com/Details.aspx?sSpec=SLA45#

 

Finally getting some back up on this!!!

unhooked, I provided the spec directly from Intel last night that shows the max temp rated at 100C. What more do you need? Also, the T61P review shows CPU temp of about 83C under load.

That rule of "max temp of 60C underload" doesn't apply to all CPUs. CPUs are not all rated the same.

 

What more do I need?
100C is the max.
Running the processor anywhere near that temp is absolute insanity.
Again, 80C is pushing it. That would be my limit in a notebook.
All of my desktop processors, overclocked or not, never go over 65C.

 

I'm no computer genius but I know that your desktops also have a MUCH more efficient way of moving heat away, BIGGER heatsink + BIGGER fan + a lot more space for components + more airflow + extra fans = a LARGER amount of heat moved away much FASTER. How can you compare the temps of a desktop CPU to a mobile at all?

 

Did you read my post?

 

well... ok then what was the point of that last sentence? besides informing us on how hot your desktop cpus run.

 

-_- my X61s runs hella hot when I am gaming. (hot enough to create lag in the game) I am still waiting for lenovo send the desktop. Hopefully, my X61s wont overheat before that.

 

Okay, that's fine. But I've had many notebooks that run under load at a constant 90-95*C, and it has not yet caused any lasting damage, nor do I suspect it ever will. 100*C is the rated maximum temperature, meaning that's the temp that Intel considers the processor is safe up to. Given the safety margin they inevitably put in, that means that 100C is a realistic maximum temperature. Meaning that up to 95C or so should be perfectly safe.

 

which utility did you use to take the temp reading?

 

Depends on the laptop. I usually use NHC, sometimes i8kfangui for Dell laptops, sometimes Speedfan. The temps are pretty consistent between programs on most machines.

 

90-95C is way too hot.
sorry.

 

I understand that's partly personal opinion, but what are you basing this on? I've never seen or heard of any studies for notebook processors that imply that those temperatures are bad for the machine. Of course in a desktop it's way too hot, but as mentioned previously you can't at all compare desktop temps to notebook temps.

 

Notebook processors are not any different than the desktop ones (manufacturing process wise).
If you get it too hot, the processor will just throttle down, which will keep it from "melting".
THAT kept your notebooks from failing.
But it's still not good for the longevity.

 

Correct, the processor will cap out at the maximum temperature, which keeps the processor from ever entering ranges where damage could be done.

As far as I can tell, you have nothing to back this up.

 

Wanna invest a few grand into a "reliability study"?
I'll conduct it just for you.

besides, do you run your can engine at the max allowed coolant temp?
Will the engine fail right away? Probably not. Is it good for the engine?
You decide...

 

Unhooked I appreciate your concern, but until you can backup your statements with documentation, I will choose to listen to Greg and Odin243.

From what I read 100C is the maximum temperature. See this report:

http://users.erols.com/chare/elec.htm

Here you will also see that the maximum temperatures vary a lot depending on the type of CPU. Especially if it is a Mobile Core Solo/Duo CPU.

 

You can listen to whoever you want to.
It's not of my concern.

 

That's too funny. I just picked up a copy too a while back and just got to installing it on my notebook. I grew up on this stuff. And while it's no match for today's complex games, sure is nice to play a familiar game that's better than the average Solitaire.

Paolo

 

What I am saying is please back up your statements with solid documentation. Otherwise it will just be a subjective opinion. According to the report I linked to in the previous post, one can clearly see the difference between the different CPUs.

 

So, what is your point?
Your link doesn't prove anything one way or another...

 

** The second part of the spec refers to mobile processors without an integrated heat spreader, measured by internal Digital Thermal Sensors (DTS).
These are the dual Core temperatures measured within the hot spot of each Core.
Thermal Junction temperatures of 75c is hot, 70c is warm, and 65c is safe .
The dual Thermal Junction sensors are how Tjunction is measured, and are the dual Core temperatures displayed in TAT, and SpeedFan: Core 0 / Core 1.

** L2 and G0 Stepping have higher Tjunction max specs, however, it is not recommended to operate overclocked processors above 75c.

-BIOS-/-CoreTemp-

-70-/-85--85- Shutdown
-65-/-80--80- Throttle
-60-/-75--75- Hot
-55-/-70--70- Warm
-50-/-65--65- TAT
-45-/-60--60- Game
-40-/-55--55- Apps
-35-/-50--50- Web
-30-/-45--45- Idle


Read more here:

http://www.tomshardware.com/forum/221745-29-core-temperature-guide

 

From your link:

Those specs mean nothing for mobile C2D processors, which are entirely different in terms of temperatures.

 

Read through the thread!


** The second part of the spec refers to mobile processors without an integrated heat spreader, measured by internal Digital Thermal Sensors (DTS). Since Intel's Thermal Analysis Tool (TAT) is a Notebook tool, and desktop C2D's have an integrated heat spreader, TAT will typically indicate ~ 2c lower than SpeedFan. These are the dual Core temperatures measured within the hot spot of each Core. Thermal Junction temperatures of 75c is hot, 70c is warm, and 65c is safe . The dual Thermal Junction sensors are how Tjunction is measured, and are the dual Core temperatures displayed in TAT, and SpeedFan: Core 0 / Core 1.

** L2 and G0 Stepping have higher Tjunction max specs, however, it is not recommended to operate overclocked processors above 75c."

What's so hard to understand???

 

This is a recommendation, and as such I (mostly) agree with it. However it does not appear that they actually tested mobile processors, and they admit that the max temp is still 100*C, as claimed by Intel.

 

Every overclocking site, thread and discussion followes the same basic rules.

 

Not notebook ones, no, they don't. And we're not talking about overclocking here, we're talking about normal laod operating temperatures for notebook processors, which are much higher than safe temperatures for desktop processors.

 

my x61 is around 60-70degrees when i play WOW on low settings....

 

Which program do you use to measure the temperature? I use core temp 0.95. I know you will get different temp levels depending on which application you use.

I don't know how demanding WOW is compared to Counterstrike 1.6 - maybe someone could enlighten me? Otherwise 60-70 degrees doesn't tell me much.

 

What rendering mode are you in in CS 1.6? Also, do you have the newest beta drivers for the x3100?

 

Rendering mode: D3D (gives more FPS than OpenGL on my system)
Resolution: 1024x768, 32 Bit

I am not using the latest beta drivers for the X3100. I am using the latest version from Lenovo as I don't want to take any chances.

 

The drivers could be why then. In the Lenovo drivers I suspect much of the hardware capabilities of the x3100 are still disabled, causing more of the work to be offloaded to your CPU. Your temps are still nothing to be seriously worried about, however if you got newer drivers I suspect they'd go down some.

 

70C sustained under load is safe in notebooks and DESKTOPS.
Anything higher is pushing it.
What are you arguing about???

85C and 90C is way too hot!!!
No matter desktop or notebook.
Both are made on the same manufacturing process!

 

Can you back it up?

 

I have absolutely no idea where your getting that from. Your article briefly mentions Intel's specifications for mobile processors, but does not spend any time testing or reviewing them. Plus, it says at the very beginning that it's guide only applies to air-cooled desktop processors. Any notebook OEM, should you ask them, will say that a normal load temp for mobile C2D's is 60-80*C, and the max safe load temp is 100*C. Just read through some of Intel's specifications, and you'll see whereas the max safe temp for desktop Core 2 Duo's is around 72*C, the max safe temp for mobile ones is 100*C. The desktop and mobile versions are not designed with the same thermal specifications, and thus the temperatures cannot be compared.

http://processorfinder.intel.com/default.aspx

 

Please post some examples, not just a link to a database.

 

unhooked is basically saying that just because the manufacturer says the temperature limit is X degrees Celsius, does not mean it is a good idea to run it constantly at or near that temperature. Running constantly near the thermal threshold is closer to unsafe side of things. What if the ambient temperature increases and then your processor is now running over the temperature threshold? Maybe your processor can cope with the heat stress, maybe it cannot.

The hotter a processor is, the more Electron Transmigration occurs. That degrades semi-conductors. Better cooling lessens this degradation. Cooler components should result in longer product life. How long a person wants her computer to last depends upon what she wants from her computer.

unhooked is telling people to err on the side of caution and try not to let the computer even get to that point. Not every processor is the same and may not cope with heat stress the same way.

 

Which examples would you like? That database has the thermal specs for all of the current desktop and mobile Core 2 Duos. Pick any two of them and you'll see the thermal specs for mobile processors are higher than their desktop counterparts. Thus safe temperatures do depend on whether the processor is mobile or desktop, contrary to what you've been claiming.

I have no problem with erring on the side of caution, but people like unhooked who are used to desktop temperatures get laptop owner's needlessly worried about their temperatures. I would not recommend running near 100C all the time either, but someone who's load temps get to "75 to 80" degrees centigrade during gaming is not anything to be worried about on a laptop.

 

Needlesly?
75C maybe not outragesly warm during hot ambient temps...
but 80C, during gaming?
As I've always said 80C is clearly pushing it.
I think it was you who stated that you've seen laptops running 90-100C without any problems.
I'm not specifically familiar with X61s, but on my 14" T61 the highest I've ever reached was 74C, 100% full load, stressing each core with two instances of Prime95, running simultaneously, at 78F ambient.
No game will stress both cores in the same way, not even close.
There are two members on this board (stallen and lowspeed), who've run the same benchmark, with very similar results.
Check out this page:
http://forum.notebookreview.com/showthread.php?t=161040&page=6
If your modern Core2 Duo running 80C or higher under less stressful condition, then you've got problems.

 

I have. I wouldn't recommend that you run a processor at these temps constantly, but I have most definitely seen a modern C2D run at 95C under full load in a laptop, and have observed no ill effects.

I assume that you mean 78F ambient, or 26C. And a 14" is much different than a 12" notebook in terms of cooling and internal temperature. The ambient chassis temperature of a 12" notebook will be higher, causing the processor to cool less efficiently. And a processor doing full software rendering in a 3d game will be stressed, maybe not 100% but very close to it.

You've yet to provide a single source that supports this idea for laptop processors.

 

You're going in circles, my friend.
What machine do you have and what temps does it run?

 

Thank you for your help. Obviously, my English is not as good. (non-native speaker).

 

Circles?? I'm merely asking you for reliable source that "Anything higher [than 70C] is pushing it" for laptop processors. As far as I can tell, and from my own personal experience, that is blatantly false. I would personally consider anything higher than 85C to be something concerning, depending on conditions. It's possible that temps over 85*C are being caused by improperly seated heat sinks or an otherwise faulty cooling system. It's also possible that those temps are normal, however. I will agree that running a processor close to the recommended thermal maximum (in this case 100C) for long periods of time is risky.

Anyhow, my experience is (almost) solely with Dell laptops. The highest I've seen under load is while gaming with integrated graphics in the following laptops:
89*C in a D620 with C2D 1.86GHz
94*C in an M1210 with C2D 2.16GHz

Normally I see load temps between 70 and 85 degrees. This is in ambient temps from 25 to 30 C, or so.

 

Now we're getting to some kinda an agreement.
And obviously under load I'd expect an Ultraportable like an X61 to run a little warmer than a fully stressed T61, but nothing major.
I'm sure there are many systems with improperly designed cooling out there.
Here we're talking Lenovo, most of the time at least.

 

Wow guys! I've been away from my computer for a few days. This bantering makes for a nice read.:twitcy:

I don't think either of you are going to buckle because you are both so firm in your stance... backing down now would be a matter of pride.

It seems to me like a conclusion of this would be to determine the maximum temp and to define "pushing it". I will attempt to do that AND back-it up.

First, what is the maximum temperature. FACT: It is 100C. There is nothing here to argue about. I won't bother referencing it anymore. Intel has it right in the spec. The maximum temp for Notebook and Desktop processors are different in the specs because they are not the same processors. These are facts. Arguing them is just ridiculous.

The only argument you can have is defining "maximum temperature". Odin is saying that anything up to 100C is safe. Not a good idea for long term, but safe for short term. Maybe for longer term use less than 90C would be more ideal. unhooked (and I'm trying not to put words in your mouth, but to summarize) is saying that if the max temp is 100C then that means at or maybe even near 100C your laptop will fry.

I define "pushing it" as being the temp BEFORE any damage is done for short durations. I don't think it is a good idea to have sustained temps of 95-100C. I don't think that short term (10-15 minutes here and there) are going to harm the processor. Think about it. If someone fried their processor from running temps in the mid 90s Intel would have to replace a lot of processors. That is why they will rate the max temp as being the max temp BEFORE damage occurs.

Analogy: Just like "maximum strength" medications. That means maximum safe. Not to mean maximum to the point you might die from taking it.

Intel is not saying "100C is great, but if you go 101C it will explode." It's simple. They are saying "don't go over 100C because your processor could be damaged if you do." End of discussion really. If they thought damage would be caused at 85C, they would have said "maximum temp 85C" but they didn't they said "maximum temp 100C". And they are probably being slightly conservative with that temp just to lean toward the side of caution.

In a nutshell....

max temp: 100C (It's a Fact)
"Pushing it": 95C (my opinion)
"Totally safe": anything less than 85-90C
"You aren't working (or playing) hard enough": anything less than 60C

Of course I think we would all agree that it would be nice if our computers ran as cool as possible. But what are you going to do if you have a tiny little laptop like an X61 where sustained temps around 85C are the norm. Are you just suppose to be in a panic every time you use it? No of course not. It was built to handle it.

It makes perfect sense that temps would run higher in the x61s being that everything is so much more cramped in the smaller footprint. You basically have the same hardware only in a tighter space, probably smaller fans, and less room for airflow. I'm sure it gets hotter that two rats f'ing in a wool sock.

 

Completely agree. Thanks for the concise summary stallen