entirely ignoring the grammatical flaws in the sentence, above, I'd like to address your use of logic...

why is it not possible that both postulates are true? perhaps he is a caffeine addict and has too much time on his hands...

while possible, the thought scares me silly...

I've pointed out elsewhere that Nokia has another terribly counter-intuitive power-waster with the N800 (don't know about the N810).

When you turn off (shut down) your N800 and plug it into the wall for recharging, the tablet boots and then the screen backlight stays on (dimly) for the entire time that the tablet is charging (showing the extremely helpful "charging" graphic). I'm not sure if it turns off when charging is complete, but I somehow doubt it. That's gotta burn some electricity that could be put to better use.

The irony is that if you leave your tablet on and charge it, the screen stays off (except, maybe, when that message comes up on the screen). Many people would probably think that they are "doing it right" when they turn off their tablet to charge it, but that's actually more wasteful!

Qwerty12, you need to add a custom message to the power-off charging screen, "You are wasting energy charging your tablet like this! Quick! Quick! Turn it back on!"

I'm a bit of a greenie tree-hugger type myself. And as someone else said, just using your tablet instead of a laptop / desktop computer whenever possible is a good way to save energy. I've also found that, despite what people say about how throwing out old electronics is terrible for the environment, replacing your old Pentium II with a new one and sending the old girl off to the recyclers can have a significant, noticeable effect on your power usage, especially if you leave the computer turned on all of the time.

.

Yes, I agree that charging the N800 while turned off, appears to be wasting energy due to the LCD image. My Samsung A900 (Blade) cellphone did this also.

I am also concerned about unnecessary wear on the backlight. My Blade probably had LED backlight so it wasn't a wear issue, if it had Electroluminescent (EL panel) backlighting it would be an issue.

What type of backlight does the N800 have? Is it cold cathode florescent (CCFL)? If so, then leaving the backlight lamp off during charging would seem the better way to do it because those CCFL lamps do get dimmer and eventually stop lighting altogether with age. The fewer wasted hours on that lamp the better.

The N800 also flashes the Blue LED when charging in the power off state. It is at a low duty cycle so it doesn't use much electrical energy.

When the N800 has completed charging the battery, the LCD remains illuminated. There is a graphic stating "Battery Full". Additionally, the Blue LED stays on 100%.

qole is correct. Charging the N800 in the on but standby state is preferable to charging it in the shut off state.

.

In any event all the heating around me is electrical, so it doesn't matter a bit to me what amount of electricity the Nokia or its charger converts to heat..

Anybody?

bun

I haven't measured battery draw under different conditions. I'll try to quantify what bluetooth draws, and the LCD backlight but beyond that I don't think my numbers will be trustworthy.

see http://www.mjmwired.net/kernel/Docum...vernors.txt#81 for explanation. As I said you can just play with the max and min values too (setting them to same value).

This one is impossible to answer. They don't have power consumption themselves. it depends on what you do. Also in typical situation even direct cpu frequency does not matter much unless CPU is running constantly since normally CPU is idle and sleeping (= low power mode with zero frequency, tunable with /sys/power/sleep_while_idle).

Default values set early on boot are

I have found that setting the scaling_governor to "performance" shows a noticeable improvement in, well, performance of the tablet, especially in CPU-intensive activities. "on_demand" doesn't seem to be responsive enough.

Do you think that turning off sleep_while_idle would have any positive effect on performance? Or does that not really have any impact except after a noticeable period of idleness?

Also, when I work my processor hard (e.g., last night, I was trying to run a very large Java applet), the back of the tablet actually gets quite warm.

iirc, the difference between the governor are how fast they step down the cpu clock.

performance will lock it at max clock, while conservative will more or less lock it at min clock. closer to performance, faster jump to max clock and a more gradual down clock, closer to conservative and its more directly to min clock and a gradual increase in clock on load.