[OpenSource] Fan Control Software for Windows XP and Vista

This is an alternative version program with:

Version 0.4 - Nov 18, 2008
- Added resume support when resuming Windows from standby or hibernation
- Added ACPI patch and unpatch support (when reading temperature from the EC embedded controller, Windows logs an error to the system event viewer; simply execute the patch when requested by the program to avoid this)
- Added a two-degrees interval when starting/stopping the fan so actually it starts/stops less times

Version 0.3 - Oct 07, 2008
- Fixed the support for Slow Fan Mode (this mode doesn't exists, the Aspire One automatically set the fan to 1, 2, or 3 when needed)
- Enhanced the tray icon management, now it displays the temperature

Version 0.2 - Oct 01, 2008
- Support for BIOS 3114 and 3304
- Support for slow fan mode (automatic)
- Checks for Acer Aspire One PC
- Support for the Fahrenheit scale

Version 0.1 - Sep 2008
- Autostart with Windows
- Tray area control
- Temperature monitor
- Fan control (deactivable)
- Shows the fan state (active, inactive)
- CPU Temperature graph
- Docked window

Link: http://www.hexagora.com/en_dw_aspireone.asp



Alpha version ;-)

Bye, Davide

 

thanks for this, I will try it out for sure. Now all we need are more modders out there for overclocking tools and acer to release better bios's!

Is there a better way to have it minimize to tray when I want? or is it something that's always up?

 

Great program, a couple of comments.

1) I noted some obviously wrong temps being displayed for a moment. I noted 3C after the fan came on, and 126C just before it came on.

2) I'd like to see some kind of hysteresis built in so that the fan does not cycle on and off rapidly. Maybe "start at this temp and run until cooled to this temp" or "always run for at least so many seconds".

3) is there any way to close or dock the window?

4) I'm wondering if the fan speed is actually controlled by the hardware. Without any software running it starts slow and gets faster when needed. With software like this, by the time it turns on it is usually on high.

 

Congratulations on sharing your work!!

I had being working on early stages of this hack for GNU/Linux and we did share the information, but that other guy on that other no Open Osurce program for Windows, didn't share the code nor even at least put a link for the technical information on how to do the hack :-(

Hope you guys on Windows can have a good solution also, and let's share ALL the information so this hack can go ahead!

Thank you.

 

In order to put the program into the tray area simply click on the Tray Icon with the left-mouse button ;-)

 

Ah, duh.

I'm using this, I really like being able to switch between the BIOS controlled routine and this one with the slider switch. I still do see some obviously erroneous temps, like a sudden jump from 53C to 144C for one reading, that will cause the fan to come on for 5 sec and then will be OK. It seems to occur when I'm using the keyboard, not sure yet. I have the default XP installation.

 

Great! Keep up the good work, I love the opensource. That way we don't have to wait for one dude to make it better like the other control.

 

good work !

 

Is this version compatible with both XP and Vista? Or they are seperate downloads.

 

the other non-open fan control software has problems with the latest bios version 3440. has anyone tested this open version using the new bios?

 

Yeah! Thanks a lot for putting trough an open-source software for this, that's the only way to go for me The other fan control software really bothered to be closed, as I wanted to add my own modifications.

Thanks again to share it with the community!

 

well, from what i can tell, this software doesn't work after updating the bios to version v3440. it's kind of funny actually, the display says fan is off when it is running at full blast and fan is on when it is semi quiet. seems to have the opposite effect possibly. either way, the fan does not shut off.

 

so it's not working, is it? oO

 

Hi - I've had the same problem with a brand new AAO/1GB/160GB/XP machine that I got today.

It's got the bios# 3301

The utility does not work - when it's supposed to be off, the fan is actually ON. Also tried with the other closed source utility - had the exact same issue with the 3301 bios.

Any ideas on a fix for this?

 

Some parts of code need to be changed.
It's possible to controle the fan with the new bios of AA1

For exemple A1Ctrl can be used with bios 3114 or 3304.
http://www.blogeee.net/forum/viewtopic.php?p=84338#p84338

Thanks to the programmer Noda :!:

 

Thanks for your efforts!

Well - this software also has the well known "acpiec warning" problem - Windows will report warnings in the system event log, because it just reads the EC registers to get the temperature (as literally all those fan control tools for the AAOne). And it is - aehm - Visual Basic, which means, you may need runtimes, have higher memory consumption etc. :|

According the new values for the fan control, you may want to have a look here: http://www.aspireoneuser.com/forum/viewtopic.php?f=44&t=1888&st=0&sk=t&sd=a&start=210#p29126

I also work on an open source solution - but in C++ and by using the CPU MSR to get the temperature - so there will not be any warning messages in the Windows event log (a command line tool similar to acer_ec.pl to play around with the EC registers is already working - i will publish this soon).

Stay tuned...

 

I found the solution for the Latest Bios version problem;
not tried but it will probably work ;-)

In a few days the fix ;-)


@askaaron

Please let me know, if you want, the method used to read CPU temperature from the CPU MSR ;-)

 

1) Read the MSR, which tells the temperature, as specified in the documentation of Intel

2) Subtract this value from the specified maximum junction temperature (which is 90 for Atom N270)

Well - unfortunately, reading the CPU MSRs means, you have to execute some privileged CPU commands (in fact some assembler code or the __readmsr() instrinsic in Visual C++), which has to be done in kernel mode, as this is not allowed for applications, which always run in user mode. I think, i will need to extend the existing kernel driver, which i already use, to access the I/O ports to do this. But i have no idea, how this could be used in languages other than C/C++.

 

Still, reading the MSR for the CPU temp won't resolve the ACPI EC problem, as we also need to read the 0x55 register to know the fan status and set it accordingly. Currently in my tool (v0.1 beta) I'm not using this as I tried another method, and this causes problems like when waking up from sleep and some other cases.

As for the fan values correspponding to the bios versions, for those who don't want to look at my code:

-bios v0.3114:
0x00-> auto mode
0x20-> slow mode
0x1F-> stopped

-bios v0.3304:
0x00-> auto mode
0x10-> slow mode
0x20-> stopped

The slow mode is really interesting to have (as in my tool) as it pretty much cool the A1 the same as in auto mode, but with nearly no noise.

 

That's right. At 100% CPU the temperature of the processor in the slow mode is only a little big higher (+2 +3°C) than auto mode but the noise of A1 is really different. :!: