Samsung Q1 Ultra Premium: battery drain during use

I'll be running a Battery Eater Pro test later today, but I wanted to provide some general battery drain observations on the Samsung Q1 Ultra Premium. I installed Notebook Hardware Control to get a feel for things and see some interesting data. During the observation, I had the brightness at 50%, WiFi on, Bluetooth off and used the "Normal" profile in the Samsung Battery Manager power settings.

By and large the 1.33 GHz Intel Core Solo runs at 800 MHz when it can. If it doesn't need to, it won't run at 1.33GHz, which keeps the power consumption of the CPU down. When "idling", meaning the machine is just sitting there doing pretty much nothing, the power draw averages around 6 Watts. For basic web surfing, I saw the drain hop up in to the 8 to 10 Watt drain. When stressed with a high CPU load, it looks like a 12 to 13 Watt draw.

Now there's many ways to "stress" the device out for these tests, but I'm trying to keep it simple. We can debate the approach but I wanted to do a quick-and-dirty set of observations for you. And for Steve, who's sending me note after note looking for this type of information. Hey, I can't blame him as he's publicly becoming enamored with this device. ;)

As I said, just sitting around, the device is pulling around 6 Watts. It varies as background processes kick in of course, but ranges between 5.5 and 7.5 Watts. Don't get hung up on the 58 Watt/hour rating of the 6-cell battery at this point. Meaning: don't conclude you can get nearly 10 hours of battery life with the 6 Watt draw. Theoretically, you could, but if your UMPC is idle for 10 hours, why have it powered on? (Reminds of the "if a tree falls in the forest bit...)

Firing up Internet Explorer and doing some web work resulted in the CPU staying in the 800 MHz zone for a majority of the time. No surprise there. The power draw was around 9 Watts or so during this time, so a little math shows us that you could probably surf on the Q1UP for 6 to 6.5 hours or so. That's assuming you don't do anything else, have a fully charged battery with very little wear, etc... call it a best case, out of the box situation.

Now what about a little stress? I opened up Microsoft Word 2007 and did a little inking. A wee bit more of a CPU spike, but not much. So I fired up a 720p WMVHD video in full screen. Now we're getting somewhere! The CPU load was averaging around 70% during playback and the device was using the full 1.33 GHz clock cycle more often than not. Average battery drain was around 12 Watts... a few spikes higher, but only up to 13.2 Watts. Using the 12 Watt figure, you're still near 5-hours of battery life if you just sat and watched video playback. If I was going to do that, I'd likely shut down all other apps and turn my WiFi radio off to save a little bit of juice.

This should give you a bit of an idea on the power consumption. Again, many ways to do these types of test, but I wanted to share some preliminary information. The Battery Pro torture test I run will tell us the minimum run-time you should expect and I'll have that info later today or tomorrow since it takes hours to run that.

I'm finding that this device offers a very acceptable compromise between performance and run-time. I've got excellent CPU power and responsiveness when needed but can still work for 5 or 6 hours. No doubt the 6-cell battery has much to do with that; with a 3-cell, I'd still have the horsepower, but only have it for 2.5 to 3 hours. Speaking of the battery, my second 6-cell should be arriving any day. I'll be rotating the batteries to keep the wear level even. When out and about, I should easily have 10 - 12 hours of computing time. If I need more than that, someone call an ambulance because I'm likely sleep and food deprived!

Comments

[nomo]

Kevin: Is it possible to measure the power draw of the wireless card and display? I recently used my tablet to ink notes at a conference. By turning WLAN off and toggling the screen on and off as required, battery consumption dropped dramatically.

[Steve 'Chippy' Paine]

Excelent. Results. Samsung seem to have done a great engineering job.

Steve

[iNsAnEcLoWn]

What about power consumption with wifi/bluetooth off and the USB WWAN modem plugged in and on? Pretty please?

[Kevin C. Tofel]

I'm running the Battery Eater test right now, so the device is tied up for the next several hours. I'll have to see if I can find a way to measure the draw per component, but it's always a good strategy to shut off the wireless card when not in use and/or setting the display to dim when not in use for X number of minutes.

As far as WWAN modem, I can check it, but I'm not sure I need to. The USB 727 EV-DO modem I use has the following power requirements: 5V, 500 mAh when transmitting, 115 mAh when idle. Using the Watts = Amps x Volts equation this should mean 2.5 Watts draw when transmitting and just over a half-watt when idle. Does that make sense? I'm relying someone smarter than me to correct the approach, so keep me honest.

[iNsAnEcLoWn]

So in lamens terms how many hours of surfing can I expect with the USB WWAN modem?
thanks for your help, I just ordered my Q1 ultra premium and plan on using it as a GPS Device (My wwan card has GPS built in), Netflix watch it now anywhere device, a work device (stock trading on the go) and finally a music device (transmitted through my car stereo of course).

[Steve 'Chippy' Paine]

Guideline power drain for Q1U components:

Screen + backlighting: Max 1.5W
Wifi: Max 1W, idles much lower.
Bluetooth: negligable

USB components are difficult to asses in terms of drain. Many of them prevent the CPU going into power-saving states. The effect can be quite dramatic. For example on my Q1 U I can use Wifi and browse with an average 8W drain. Turning off the Wifi and turning on the 3G modem results in a minimum 25% *increase* in drain when normaly, it should be a decrease.

Worst case average on the Q1 Ultra (A110) is about 11W drain (5 hours on the extended battery)
On the Q1 Ultra Premium I expect that to be about 12W based on Kevins result there. Not much practical difference which is really nice to see. If you can afford it, the Premium is obviously the one to buy.

Steve.

[Tarin]

your numbers are way off Chippy, all parts being equal theres a 2.5w difference alone in the CPU at different states. Kev's numbers have even reflected a 2w difference the majority of the time (plus he's only using XP).

calling things negligible & not practical just because thats your impression doesnt dispute mathematical importance & is pretty ridiculous especially considering the extremely low wattage usage by these devices percentage wise.

[Clay]

Hi, I have a Q1B and can't view Netflix watch instantly movies. I was told my netflix that this is due to the graphics driver. I have updated the driver but no luck. Is anyone else having this issue?