How much power does your desktop PC use?

The Zonbu team make a big deal out of the low power consumption of the hardware they sell.

Not only does low power consumption have financial and ecological benefits, but it conveniently lets them chase the Toyota Prius (and amazingly Lexus LS hybrid) “I want to be ecological because its fashionable” crowd. Now I’ve got nothing against being eco-friendly, but I’ll be honest, I’m taking a serious look at this because, like many of you out there, I am alarmed by rising energy costs.

Presumably, the idea of fashion conscious buyers wasn’t lost on them either as they separately sell various “skins” to make your Zonbu all pretty or at least to make it match your handbag and your lap dog.

Where were we? Ah yes, power consumption…

According to the Zonbu Website the unit consumes a paltry 15 watts of power during “average” use. They quote third party statistics as saying that a typical desktop computer consumers 175 watts of power while performing similarly average tasks.

A Challenge for the Zonbu Marketers

I challenge the Zonbu marketing staff to come forth and share with the community where they got their numbers for average desktop PC watt consumption and total average kilowatt hours per year consumed.

Not that I necessarily think they’re not accurate, it’s just good science to be upfront about something so central to your theme. At first I was lulled in to believing one of the many links around the usage table on their green page must have supplied the data, but I took a look and couldn’t find it. Where did that data come from?

I’m no scientist or eco-crusader, but I do happen to have a power usage meter at home. I decided to use my Kill-A-Watt meter to see what my home desktop was consuming. I figured it had to be more than the stat on the Zonbu website.

I expected the figure to be quite high because my home system has a 430 watt power supply, an AMD Athlon 3800+ 2.4Ghz processor, a very high end ATI graphics card (the kind with dual power connectors and lots of warnings about power requirements), as well as dual optical drives, and multiple hard drives (I think there are three in there). On top of that its powering dual on-board LAN ports, on-board audio and 2GB of RAM. All in all a pretty hefty home system. Certainly more than “average” in my experience. It’s also about 3 years old. So I must draw more than the “average” system that Zonbu references, right?

I plugged the meter in and let it run for a while before checking the load.

As I write this, it is reading 147 watts, or just under 10x the stated power draw of the Zonbu. I have to say I’m somewhat shocked that its that low, based on the equipment I have connected. I’m essentially idle writing this in FireFox. I’ll fire up Rhapsody and see if it makes any difference but I suspect not. Of course things like MSN Messenger and Skype were also in my tray, as well as the FreeAVG virus scanner, the Java console and perhaps one or two other little apps.

Interestingly, the disk access involved in loading Rhapsody caused the consumption to spike to 160 watts momentarily. I wonder if writing to compact flash has the same result or not? You’re not moving any parts so my guess is no, but thats why we have measurement tools.

Running Rhapsody has cost me considerably. I’m constant at 159 watts of draw. I forgot that Rhapsody is going to stream in audio that is going to keep the network card busy as well as the processor, decoding the music stream (presumably some Real Networks proprietary format). I wonder how much of the modest processing power the Zonbu offers will be consumed by running Rhapsody while I attempt to web-surf? That will be interesting to see.

So the good news is I’m not consuming as much power as I feared.

The bad news is I’m beginning to think the number on the Zonbu site might be a bit generous in their favour.

Well, they probably included the monitor in the PC measurement, right? Ok, well following that logic they would need to include it in the Zonbu measurement, at least to compare apples to apples. So let me see, I took arts math in high school, but if I add the same thing to both sides doesn’t one cancel the other out?

Or maybe its a more general measurement including using various office applications too? Although I can’t imagine that word/excel, once loaded, put any serious draw on the CPU until you ask them to do something complex. I guess we’ll wait and see what the Zonbu people say…

The real takeaway is the load on standby.

I think I’m like most people in that I don’t like to wait for my PC to boot up when I wander over to do something with it.

I utilize the “Standby” mode in Windows XP to put my computer in to a light sleep so I can awaken it on short notice and resume working. Occasionally it asks me to spoon, but we’re beyond that stage of our relationship…

What I discovered, after the Kill-A-Watt was in place, is that my PC is sucking back 6 watts of power just idling in standby mode. Wow. Thats only mildly less than the low end (8 watts) of what PC Magazine said their Zonbu unit was drawing while in use.

Put in perspective, it sounds like replacing my desktop PC with a Zonbu for the 80% of the time I’m using my PC for web browsing would allow me to cut my energy usage for that same period to 1/10th of my current load. Moreover, depending on what the standby load is for the Zonbu, I may be able to experience similar savings when in standby (most of the day while I’m at work and all through the evening).

Real Savings.

As I stated previously, my local rate for power is eye-popping at ~$0.40/kilowatt hour.

Bear in mind most American consumers are paying something on the order of $0.15/kilowatt hour or less, and it may vary with time of day, so do your own calculations. Also, bear in mind that we burn fossil fuels to generate power so rising oil costs aren’t just hitting you at the pumps – watch for a “fuel surcharge” on your utility bill.
I don’t currently have any easy way to break out what portion of my monthly energy bill goes to my desktop PC, but the relative savings based on 80% of my load being done at 1/10 of the draw, will rapidly translate in to serious reductions in my energy bill.

Once Zonbu steps up and tells us the source of their “average annual usage in kilowatt hours” we can explore the real dollar savings. I’m looking foward to it.

Regardless, no matter how you slice it, there are real savings to be had here.

-Mr. Zonbu


I live in a warm climate and we run A/C all year long. If we didn’t, I think I’d melt from the humidity. Therefore the actual heat my desktop PC displaces contributes directly to my overall energy consumption as I need to expel the hot air and replace it with cool air. While the Zonbu does generate some heat, it will add a lot less to my overall household load, and therefore lead to potentially lower A/C loading – the single biggest consumer of energy in our home. That “secondary” savings could actually eclipse the savings of the device itself.

About these ads

5 Responses to How much power does your desktop PC use?

  1. laika says:

    this is interesting. wish i had a Kill-A-Watt device to measure my old box in use.

    using Puppy Linux (the myPup variant), i boot from the CD to RAM (512Mb). the CD drive is then idle. i have no hard drive – settings are saved to a Compact Flash. i wonder how much power the 750Mhz CPU and the couple of tiny fans are using as i go about my business?

  2. Graphical says:

    Remember about tiered electrical pricing. I measured my “old PC” running Shorewall at 90W with the monitor off. It runs 24 hours/day. I thought that at $.13/KWH that it really didn’t make that much difference.

    Then I looked at the bill and saw the pricing tiers. We pay .119/KWH for the first baseline. Then .139/KWH to the next baseline. The third baseline is .229/KWH. So, incremental power use is cosing us almost twice what baseline usage costs. Ouch.

    Since we are barely into the .229 zone, replacing my firewall and my notebook with a Zonbu should pay for itself within 18 months or so. Now, if we could just get a PG&E energy savings rebate program going!

    PS: The notebook computer spends a fair amount of its on time (2 hours/day) just playing Yahoo Music Jukebox. The big savings is in the firewall.

    How is the audio quality of the Zonbu and how well does it work with USB speakers? I guess we’ll find out.

  3. Mary Jo Graca says:

    A Kill-A-Watt device, or another energy meter only costs about $25. My UPM Energy Meter can take a load of up to 1800W.

  4. Rigas says:

    The Zonbu is a nice device and I enjoy having one in our living room.

    I would not overestimate the amount of energy you can save with this box.
    Using some energy saving bulbs will have the same or even more impact.

    Taking the bike or public transportation to work instead of the car can make the difference …but not the use of a tiny little lifestyle computer … :-)

    See my review of the Zonbu here:


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

%d bloggers like this: