Overheating?


Bright Shadow

 

Posted

CoH presents a rare case for my laptop.

It runs every other game perfect. Champions Online, Halo 2, Fallout 3...

However, with CoH...my computer overheats. My comp literally shuts off in the middle of playing because of this.

I have been able to play because I use a cooling mat. However, the cooling mat is very annoying.

I understand it's probably because I'm using a laptop...but I do not understand how every other game runs fine.

Is there possibly a cheap Desktop PC that I can buy from best buy that will enable this game to run ok?


 

Posted

Best Buy has some very good ~$1000 desktops that will run this game just fine.

I would never game on a laptop. I did so for a few years, but it always begins to overheat eventually on everything.

If you want to continue you will do best by turning the graphics options down, especially things like FSAA, water effects, shadows and particle count/physics.


Edit:
Update all your laptop's drivers as well. Make sure everything is up to date and working properly and maybe the problem will be fixed if it seems so out of place.


 

Posted

make sure to use a can of air to blow the dust out of any of your vents and fans. depending on the laptop, you can access the cpu/gpu fan from a panel underneath. just make sure it is dust free as well.


 

Posted

Quote:
Originally Posted by Vitality View Post
I have been able to play because I use a cooling mat. However, the cooling mat is very annoying.
What is so annoying about the cooling mat? I was thinking of getting one, so would be interested to get your feedback.

Thanks


 

Posted

Quote:
Originally Posted by SteelMountain View Post
What is so annoying about the cooling mat? I was thinking of getting one, so would be interested to get your feedback.

Thanks
It's probably more or less just MY cooling mat that is annoying. It's very old and loud.
I'm sure you can get some that are quiet.
Although, I have tried different ones in the past...and the "quiet" ones never seem to cool it enough.


 

Posted

I kind of want to go back to your laptop overheating.

All processors, be they central processing units (CPU's), graphics accelerators (GA's), graphics processing unit's (GPU's), network processing units (NPU), sound / audio processing units (SPU / APU), and even Physics Processing Units (PPU's), all have defined maximum clock-speeds. Outside of direct user-intervention, such as increasing the voltage to the processor, or increasing elements of the processor's timing mechanisms, this means there is a hard-cap to the amount of work that any individual processor can accomplish.

The hard limit to how much work a processor can accomplish means that there is a hard limit on how much heat that processor can put off, a limit that is referred to as the TPD envelope, or Thermal Design Power: http://en.wikipedia.org/wiki/Thermal_design_power

This heat output is often measured in Watts.

Most processor vendors rate their processors by some variant of the Thermal Design Power. AMD, for example, sells their processors with explicit maximum TPD ratings: List of Phenom Processors :: Phenom Processors For Sale.

Other vendors, such as Intel, have caught flack for selling their processors using figures from Average Thermal Load, instead of the maximum thermal load.

The maximum TPD is a incredibly important figure for vendors actually building products with various processors. This CK8 cooler from Cooler-Master, for example, is only certified for dissipating up to 110watts. If you were to try and use it to cool off a processor with a 125watt rating, you'd probably wind up with a fried processor very, very quickly.

Presuming that the manufacturer of the processor was accurate about the thermal performance of their processor, outside of user or external intervention, there should not a case where a processor being used in a default setting should be cable of reaching a temperature where automatic shut-offs engage. Referencing the external intervention, build-ups of dust, dirt, or other such debris can impedes adequate and proper airflow to system processors, creating an overheating situation separate of any software variables.

Beyond these external factors, there are cases where specific computer systems, using parts from a specific vendor, can reach an overheated state. Beleaguered Graphics Processor vendor Nvidia is currently in excrement up to it's neck, currently facing multiple civilian level class action lawsuits over an ever-increasing series of exploding mobile platforms, as well as pressure from State and Federal level organizations, as well as vendor ire, over having not given Original Equipment Manufacturers and Original Design Manufactuers (OEM's and ODM's) accurate thermal ratings for several generations of Nvidia produced graphics chips.

The past couple of years has seen the Nvidia company scrambling to work with vendors, implementing bios lock-downs, aggressive cooling fan operational strategies, as well as extensive clock manipulation strategies based on real time temperature readings, to try and bring the issues with Nvidia's mobile platforms under control.

Many of these methods developed by Nvidia and system vendors to try and manage Nvidia's out-of-control thermal issues rely on applications using Microsoft Application Program Interfaces (API's) such as DirectX. Applications that do not use the Microsoft API's, such as those leveraging OpenGL, may or may not trigger the Nvidia drivers and system bios modifications required to reign in Vendor flaws reliant on a much lower antipated and advertised Thermal Design Power envelope.

As such, applications such as City of Heroes could force an Nvidia system into an over-heated state and after-market safety measures fail to engage.

Such API centric hang-ups are not associated with products from processor vendors such as Intel, AMD, Via/S3, SiS, ARM, Matrox, Apple (A4), IBM, DEC; or past vendors such as XGI, Xabre, Trident, SGI, and ULI; or from licensed processor architectures such as MIPS. It is possible that over-heated states of operation can be achieved by these and other vendors in cases where thermal controls have failed. To-date, there has been no consistent series of failures associated with any of these vendors due to failure to adequately inform system vendors of the thermal designs of manufactured components.

* * *

Yes. This was an over flowery way of saying your laptop probably has an Nvidia graphics chip, it's not detecting CoH because CoH uses OpenGL, and yes, this over-heating problem is pretty much specific to Nvidia LYING to system vendors about how hot Geforce parts actually are.

If you've GOT a system with AMD graphics though, and you are reaching the point of thermal shut-down under City of Heroes, my money's on your system fans being clogged up with dust.

Also, AMD would /really/ like you to file a Catalyst Crew Feedback over this matter: http://www.amd.com/us/CatalystCrewSurvey


 

Posted

My laptop had the ATI Mobility Radeon HD 2600.

Under the ATI Catalyst Control Center...I can change the settings for 3D graphics.
It allows me to set...
Anti-Aliasing
Adaptive Anti-Aliasing
Anisotropic Filtering
Catalyst AI
Mipmap Detail Level



Can I change these to lower setting to go along with my lower settings in the CoH Options?


 

Posted

CoH has a lot of rendering inefficiency.

I run top-end games on maximum graphics on my PC, and my GPU rarely ever crosses the 65C border for them. With CoH, my GPU can heat up to temperatures as high as 80C.

It's just something that's in CoH that I've had issues with.

Same scenario occurred on my nVidia 9800 GT, that eventually burnt out.

And now I have just slightly lower temperature readings on my nVidia 260 GTX. I don't get nearly as much heat playing other games.


 

Posted

Quote:
Originally Posted by Bright Shadow View Post
CoH has a lot of rendering inefficiency.

I run top-end games on maximum graphics on my PC, and my GPU rarely ever crosses the 65C border for them. With CoH, my GPU can heat up to temperatures as high as 80C.

It's just something that's in CoH that I've had issues with.

Same scenario occurred on my nVidia 9800 GT, that eventually burnt out.

And now I have just slightly lower temperature readings on my nVidia 260 GTX. I don't get nearly as much heat playing other games.
How do you get temp readings?


 

Posted

SpeedFan: Jack-of-all-trades temperature reader. But it's a little quirky in my experience.
GPU-Z: Minimal, portable GPU monitor.
CPU-Z: Minimal, portable CPU monitor.


 

Posted

Quote:
Originally Posted by Bright Shadow View Post
CoH has a lot of rendering inefficiency.

I run top-end games on maximum graphics on my PC, and my GPU rarely ever crosses the 65C border for them. With CoH, my GPU can heat up to temperatures as high as 80C.
On my system, the only way I can maximize the temp on the GPU (GeForce GTX 460) is to run FurMark, which makes it top out at around 72C.

Running CoH in Ultra Mode, it rarely cracks the 54C mark.


TargetOne

"If you two don't work this out RIGHT NOW, I'm turning this invasion around and going home!" - Emperor Cole

 

Posted

Quote:
Originally Posted by Bright Shadow View Post
SpeedFan: Jack-of-all-trades temperature reader. But it's a little quirky in my experience.
GPU-Z: Minimal, portable GPU monitor.
CPU-Z: Minimal, portable CPU monitor.
If you're running Windows Vista or Windows 7, there are desktop/sidebar gadgets to monitor your temps, as well as processor load and other statistics. I personally use All CPU Meter to monitor the CPU, and GPU Observer for the GPU.

Another good tool is HW Monitor, from the people who make CPU-Z (there's free and payware versions, the latter having a few more features).


TargetOne

"If you two don't work this out RIGHT NOW, I'm turning this invasion around and going home!" - Emperor Cole

 

Posted

Unlike most games, this one uses OpenGL for the graphics. Interestingly enough, FurMark that TargetOne mentions, is an OpenGL benchmark program.


Father Xmas - Level 50 Ice/Ice Tanker - Victory
$725 and $1350 parts lists --- My guide to computer components

Tempus unum hominem manet