The XPS M1530 Temperature Survey Idea

NB: This thread is for discussing my idea, once everyone has made suggestions for the survey, we can make a nice new thread to do the survey

It seems that everybody who owns an XPS M1530 has an extremely varied set of temperatures for their GPU and CPU. Now, I am a complete freak about my temps, constantly checking them and always worried they are too high. I scour the notebook review pages looking for a benchmark set of temps, but everyone either has overheating problems warranting a nice replacement (CPU 80+ GPU 90+), or they have chilled out systems that sound very cool (CPU< 69 GPU <72).

Now, there are many variables that cause these temperatures to seem so varied and I have come up with a list of what might cause differences in everyone's readings.

Please correct me if you have evidence to suggest any of these variables do not effect your temps. Also, if people could give me details on any variables I have listed that I can edit into this O.P.


1. Temperature Monitoring Software (HWMoniter, Speedfan, Rivatuner etc.)
Now, because everyone uses different programs, is it possible that the software interprets the temperatures differently? I am not 100% sure on this.

2. Faulty Sensors
Maybe a fault in the hardware's sensors will give people abnormal readings or temperature spikes (causing the MAX temp to display false data).

3. 6v 9v Batteries
The 6v Battery causes the XPS to lie flat, while the 9v battery props the notebook up at the back, allowing more airflow into the vents at the bottom. This would mean systems with a 9v battery will have cooler temperatures

4. 5600 RPM 7200 RPM Hard Drives
Apparently the 7200 RPM hard drive runs a lot hotter, this could radiate heat throughout the computer raising the temperatures.

5. CPU Model
The Old Merom Core 2 Duo processors (e.g. T7800, T7500 etc.) sound like they run hotter than the latest Penryn processors (T9500). Could this be why some people have super cool systems?

6. CPU Clock
So faster processors probably produce more heat than those with lower clocks. Is it possible that a 2.6 GHz processor will produce a hotter temp than a 2.0 GHz?

7. Undervolted CPU
Often people do not state whether they have undervolted their CPU when they post temps. This is an important variable as it can significantly decrease temps.

8. GPU Clock
A temp of 85 under load on an overclocked GPU indicates a different state of health than a temp of 85 on stock clocks.

9. Heatsink
I have heard on some of the more in-depth forums to do with M1530 temps, that people can have different types of Heat sink. If anyone could help me discover if there are any differences between heatsinks in 1530s that would be great.

10. Thermal Paste
People somtimes open their laptops and put something like arctic silver on the heatsink, this should make it run cooler, but sometimes posters forget to mention this when they post temps.

11. Notebook Cooler
This can affect temperatures hugely, and often people with dubiously low temperatures don't mention wether they are using a notebook cooler.

12. BIOS revision
Apparently different BIOS can cause different fan activity, effecting temps.

13. Power setting
Some people accidentally use the power-saver setting when they check their temps. Different settings cause different temperatures.

14. Driver Version
Video Drivers not designed specifically for the laptop can run hotter than Dell ones.

OK, so it seems that anyone who knows about these variables will have a much tougher time being reassured by someone else's temperature readings, because they could be different or the same for so many reasons.

I have therefore created this thread so everyone who owns this notebook can clearly state what their temps are and how their system is set up. This will act as a far more effective gauge.

So here is my proposed format for a general survey, I made some suggestions of what people SHOULD use to control the variables ->

TEMPERATURES
- CPU Min/Max Idle
- GPU Min/Max Idle

- CPU Min/Max Load (running ORTHOS, which can be downloaded here: http://www.overclock.net/attachment...7486-orthos-v20060420-orthos_exe_20060420.zip )
- GPU Min/Max Load ( running rthdribl, which can be downloaded here: http://www.daionet.gr.jp/~masa/archives/rthdribl_1_2.zip )

- Whether you were watching your temperatures in real time, or relied on a temperature report. (I suggest everyone does both and reports both results)

SOFTWARE
- The program you used to get these readings (I suggest everyone uses the latest version of HWMoniter)
- Power Setting in Vista (I suggest everyone uses High Performance)
- Bios Revision (I suggest people use the latest revision)
- Video Driver Version (Found in system information in the Nvidia control panel, state wether it is a modded, Dell M1530 or Dell non-M1530 driver)

SYSTEM:
- CPU Model and Speed (Txxxx, xGHz)
- Hard Drive Speed
- GPU Clock ("stock" otherwise write your clock speeds)
- 9v or 6v battery

Physical Ajustments
- Notebook Cooler (if yes, then write the model)
- Thermal Paste (if yes, what type?)
- Airflow and Surface (What type of surface is the laptop on? Have you propped it up to simulate a 9v battery? How high?)
- Ambient Room Temperature (If you can report these in degrees C, otherwise state your location and time of year with a general statement whether it is cold or hot)

So, What do you think?

even if the survey has a hundred seperate setups and results, at least people can search the thread for a similar setup and can compare their temps.

 

I agree with your suggestion and would like to add that ambient room temp can have a large effect, it's freezing here in the UK at the moment so obviously my max readings are quit a bit different to people in California for instance.

To be honest, I'm a bit depressed by the whole issue.
When abnormal temps first started being reported some months ago, it was mostly 1330s that were affected, with a smaller construction and the fact they were released months before the 1530.
Now there are a lot of threads appearing about the 1530 temps being abnormal, so alas I feel the Nvidia GPU issue is gonna get us all eventually.

 

I am also a complete freak on temperature and fan noise. I must admit that I am very happy with my temperatures but not so much with fan noise. The fan switches on every 5 min for about 30 seconds to take the temp down from ~40ºC to ~30ºC.

I have undervolted my CPU but this has only affected my MAX temp under load. It went down from 60 to 48.

TEMPERATURES
- CPU Min/Max Idle: 30/40ºC
- GPU Min/Max Idle: 50/60ºC

- CPU Min/Max Load (playing a system intensive game crysis, fallout 3 etc.): 48º
- GPU Min/Max Load ( '' '' ): 70

- Whether you were watching your temperatures in real time, or relied on a temperature report. (I suggest everyone does both and reports both results): Both

SOFTWARE
- The program you used to get these readings (I suggest everyone uses the latest version of HWMoniter): HWMonitor + RMClock
- Power Setting in Vista (I suggest everyone uses High Performance): Recommended by Dell
- Bios Revision (I suggest people use the latest revision): A12

SYSTEM:
- CPU Model and Speed: T9300 2.5Ghz
- Hard Drive Speed: 5400
- GPU Clock ("stock" otherwise write your clock speeds): Stock
- 9v or 6v battery: 9v

Physical Ajustments
- Notebook Cooler (if yes, then write the model): No
- Thermal Paste (if yes, what type?): No
- Airflow and Surface (What type of surface is the laptop on? Have you propped it up to simulate a 9v battery? How high?): Wood table.
- Ambient Room Temperature: 15-20ºC

 

I concur with op about have having this kind thread

and here is my input this thread.

TEMPERATURES

- CPU Min/Max Idle : 39ºC/45ºC
- GPU Min/Max Idle : 56ºC/62ºC

- CPU Min/Max Load (playing a system intensive game crysis, fallout 3 etc.) : 77ºC
- GPU Min/Max Load ( '' '' ) : 83ºC

- Whether you were watching your temperatures in real time, or relied on a temperature report. (I suggest everyone does both and reports both results) : Realtime

SOFTWARE

- The program you used to get these readings (I suggest everyone uses the latest version of HWMoniter) : HWMonitor
- Power Setting in Vista (I suggest everyone uses High Performance) : High Performance
- Bios Revision (I suggest people use the latest revision) : A12

SYSTEM:

- CPU Model and Speed (Txxxx, xGHz) : T7500 2.2Ghz
- Hard Drive Speed : 7200RPM
- GPU Clock ("stock" otherwise write your clock speeds) : Stock
- 9v or 6v battery : 9V

Physical Adjustments

- Notebook Cooler (if yes, then write the model) : No
- Thermal Paste (if yes, what type?) : No
- Airflow and Surface (What type of surface is the laptop on? Have you propped it up to simulate a 9v battery? How high?) : Wooden Table / Bed
- Ambient Room Temperature (If you can report these in degrees C, otherwise state your location and time of year with a general statement whether it is cold or hot) : 14-26ºC

 

TEMPERATURES
- CPU Min/Max Idle 38/55
- GPU Min/Max Idle 54/62

- CPU Min/Max Load (playing a system intensive game crysis, fallout 3 etc.) 65/75
- GPU Min/Max Load ( '' '' ) 71/78

- Whether you were watching your temperatures in real time, or relied on a temperature report. (I suggest everyone does both and reports both results)
Report from HWMonitor (the min and max) and sometimes in real time.

SOFTWARE
- The program you used to get these readings (I suggest everyone uses the latest version of HWMoniter) HWMonitor 1.1
- Power Setting in Vista (I suggest everyone uses High Performance) Ballanced
- Bios Revision (I suggest people use the latest revision) A08

SYSTEM:
- CPU Model and Speed (Txxxx, xGHz) T8300 2.4
- Hard Drive Speed 7200
- GPU Clock ("stock" otherwise write your clock speeds) stock
- 9v or 6v battery no idea

Physical Ajustments
- Notebook Cooler (if yes, then write the model) No
- Thermal Paste (if yes, what type?) stock
- Airflow and Surface (What type of surface is the laptop on? Have you propped it up to simulate a 9v battery? How high?) mainly wooden desk + propped a bit higher then 9V battery
- Ambient Room Temperature (If you can report these in degrees C, otherwise state your location and time of year with a general statement whether it is cold or hot) about 24 C

 

Thanks for the responses, but this is just a thread to discuss the idea of the survey, I want to perfect it and have a nice layout when i create the thread for it :]

 

You should use an identical loading scenario for each test. Some people use different games or have a different idea of what 'under load' means, or different length of time. This can have a BIG effect on temps. Some guys mention their 'load temps' and their laptop isn't even close to being at full load.

The easiest thing to do is for everyone to use a simple (free) stress tester (both GPU and CPU) for a specified length of time. Its also important to know whether the GPU and CPU are being stressed at the same time, since this has a big effect on temps. I suggest running Orthos (CPU stress-tester) and RTHDRibl (GPU stress-tester) simultaneously for 10 mins.

 

Should I add the ACPI temps? what is the ACPI?

 

bump? should I just make a nice and tidy survey thread? or is anyone else going to add some ideas?

 

I'd try and go for a slightly less congested post design. ie try and minimise the standard post format -

e.g instead of getting everyone to post

SOFTWARE
- The program you used to get these readings (I suggest everyone uses the latest version of HWMoniter) HWMonitor 1.1
- Power Setting in Vista (I suggest everyone uses High Performance) Ballanced
- Bios Revision (I suggest people use the latest revision) A08

just post

Software
- program hwmonitor 1.1
- power balanced
- bios A08

or even something more consolidated woule be better in my opinion

 

TEMPERATURES

- CPU Min/Max Idle : 47ºC/55ºC
- GPU Min/Max Idle : 58ºC/66ºC

- CPU Min/Max Load: 66ºC
- GPU Min/Max Load: 70ºC

- Whether you were watching your temperatures in real time, or relied on a temperature report. (I suggest everyone does both and reports both results) : Realtime

SOFTWARE

- The program you used to get these readings (I suggest everyone uses the latest version of HWMoniter) : SpeedFan & CPUID
- Power Setting in Vista (I suggest everyone uses High Performance) : "recommended setting"
- Bios Revision (I suggest people use the latest revision) : A12

SYSTEM:

- CPU Model and Speed (Txxxx, xGHz) : T7250 2.0Ghz
- Hard Drive Speed : 5400RPM
- GPU Clock ("stock" otherwise write your clock speeds) : Stock
- 9v or 6v battery : 6V

Physical Adjustments

- Notebook Cooler (if yes, then write the model) : No
- Thermal Paste (if yes, what type?) : No
- Airflow and Surface (What type of surface is the laptop on? Have you propped it up to simulate a 9v battery? How high?) : No, Regualr Table
- Ambient Room Temperature (If you can report these in degrees C, otherwise state your location and time of year with a general statement whether it is cold or hot) : 20-26ºC

 

thanks man, good idea. I'll make it less convoluted

 

Yes, simplification would help a lot.

Also, this may sound kind of out there, but I think it would be AWESOME if, after obtaining a said amount of data, a few "drop down lists" can be made. There will be multiple menus like this, and in each one is for specific things. For example, in one, you may pick your CPU model, then your GPU in another, so on and so forth. After selecting something for all the menus, then, if it is on record, a list of results from multiple users pops up. I don't even know if this is plausible, but it would be very useful for centralizing all the information in a very organized manner.

Here's what I mean:

 

I hate to complicate things even more, but wouldn't video driver version also be a big factor? This might be even more important than the bios version.

I am looking forward to this.

 

Cheers, i will edit that in now.

 

Do you think, when i post this survey, that the mods could make it a sticky? Might give people an easy place to find it and refer.

 

You're absolutely right. Different driver versions have different temperature fluctuations. I think this would be important to have in the database.

 

So do people have any other reservations about the survey? or should i go ahead and put it up?

 

If you want to have results that are comparable, just make sure to give people simple guidelines about what is considered 'load' and 'idle'.

ie
Idle - close all programs. Allow notebook to sit for 5 mins to dissipate any heat that has built up before testing. Start HWMonitor. Allow to run for a further 5 mins to capture max and min.

Load - Run Orthos (CPU) and RTHDRibl (GPU) simultaneously. Allow to run for a couple of minutes, start HWMonitor, allow to run for a further 5 mins to capture min/max load temps.

You could change the details or times as you like (5 mins is not a lot, but most people don't have time to run extensive tests), but anyway, the point is that you should definitely give a list of guidelines. There can be a VERY big difference between the GPU/CPU activity and temps you see running different games etc, so unless you keep the test the same for everyone, the results are pointless.

 

Are those programs safe for normal working GPU/CPU?

 

Well they are perfectly safe to run simultaneously assuming you normally get safe temps anyway, but god help you if you overheat normally because running those two applications on an overheating system is gonna cause a fire:laugh:

BTY rthrdribl uses the cpu pretty heavily on its own - usually around 70/80% so running orthos and rthdribl together is gonna cause a slideshow.

 

Yeah, for a normal working computer 100% load should always be safe. Those programs are safe, and quite common, but you can use any programs of your choosing. Preferably small (quick to download) and you should link to a safe source in the opening post.

The purpose for running them together is just to get the most thorough and max temp in the shortest time.

It is true that RTHDRibl stresses the CPU a lot also, but for example a T7250 might be at 80% load, whereas a newer T9500 might only be at 60-70% (or whatever).

It is best just to ensure all systems are 100% stressed so you can know the max temperature, and not just the 'nearly max' temp.

 

would you advise getting a more 100% loading program other than RTHDRibl?

 

Whatever you want. Once it is simple to use and free to download, then it doesnt really matter, whatever your preference.

The only important thing is making sure that everyone's results can be compared, which is why it would be no good for one guy to claim his 'max temps' after running 3DMark06, and another guy to give his temps after 2hrs of Crysis, and another guy to play an old game like HL2....etc

If everybody uses the same loading scenario for the same amount of time, at least that way you'll be able to look at other systems and know if yours runs hotter/cooler.

 

TEMPERATURES
- CPU Min/Max Idle 39c/50c
- GPU Min/Max Idle 58c/69c (at 69 fan kicks on)

- CPU Min/Max Load 60/67
- GPU Min/Max Load 69/74 (77 when CPU at 100% as well)

- Real-time

SOFTWARE
- Everest to get temperatures and stress CPU
- Win XP Full Power
- A12
- 179.28

SYSTEM:
- T8100 2.1 Ghz (stock speed)
- 7200 RPM 120 GB Seagate 7200.2 Momentus
- Stock cooling
- Stock GPU clock speed
- 9v battery

Physical Adjustments
- Notebook Cooler: none
- Thermal Paste: Stock
- Airflow and Surface: Fake wood desk
- Ambient Room Temperature: 70 f

Test software:
Everest CPU and memory stress test, concurrent with Real Time HDR IBL test.

How does that look for temps?