Discussion in 'Lenovo (IBM)' started by g2tl, Jun 13, 2008.
whoa whoa, calm down haha
I'll see what I can do when I get some time
I think I found one of the culprits:
I updated the LAN driver to the one from the Intel site (184.108.40.206 instead of 220.127.116.11) setting all "Wake on Lan", driver and Bios to off (did not work with the previous version). In my first measurement I get ~20 minutes longer lifetime on battery. Weird. I'll keep an eye on that...
Unfortunately this was not true and just because of the wrong runtime estimation of Vista
But I just installed Ubuntu 7.10. on my x61t and get way better battery life (see other Thread). So this magical battery drain of the x61t versus the x61s seems to be somehow Vista/Wacom related?
Put in sleep for 8 hours. The time battery drops from 90% to 68%. 4-cell battery for x61t. Nothing connected.
This is a known issue. Switch off "Wake on Lan" (Bios+Device Manager) and disable USB external device powering in Bios. Then the sleep drain should go down to around 0,5 Watt to just power the RAM and some related chips.
x60t, check out this site:
I have a hypothesis and would like to ask for your help to figure out more:
I suspect SATA Controller not to go into sleep mode because of some driver fault with Vista. This may only apply to those running "Compatibility Mode" and Vista, as I do. You can switch the SATA mode in Bios settings, but unfortunately this requires a full new Vista installation.
If you run "AHCI Mode" with the "Intel Matrix Storage driver" seemingly this does put the SATA controller to sleep and gets the HDD to spin down. This might require write caching enabled. Or maybe works only with Turbo Memory. If true, all here getting good battery runtimes probably have AHCI with Maxtrix driver and get their HDD to spin down completely and let the SATA controller sleep. Together this can sum up to approx. 2-3 Watt.
Anybody can back up or disprove this hypothesis?
Ok, I tested the Wattage (using NHC) with and without the TM. (Both with write caching on and off, which didn't seem to matter)
I got about .1W LESS without the TM. But then, that's the idle W, so in theory, if the TM allows the HDD to stay low for longer, then although the TM itself consume some W, in the end it does save some battery life.
What's really bothering me is that I still cannot get the W lower than 10W in XP, whereas in Vista, as soon as the HDD spins down, the W goes down fairly quickly to around 8W.
g2tl, thanks for this test.
I think it has something to do with AHCI power management. There I suppose the Intel Matrix Storage driver to have some advantage over the standard one. Some devices seem to be delivered with AHCI on, some of which come with the Matrix driver and some are set to "Compatibility Mode" (mine is) and the standard driver. Usually XP should come with the latter, which might not allow for the deeper sleep modes of AHCI. HDD together with the SATA controller both can account for 1-3 Watt, depending on power saving settings.
Now this is a bit difficult to test, as switching from Compatibility Mode to AHCI changes all the Disk access. As far as I know this requires a full new install.
I do have it set as AHCI, and XP runs fine.
It does not require a reinstall, as I have tried a few times trying to install XP.
I still don't understand why the power consumption in XP is so high.
If what you say is true, then the AHCI mode should allow for better power management, but I cannot imagine having even higher power consumption by switching to compatibility mode.
Maybe I'll give it a try.
Separate names with a comma.