POPULAR RECENT STORIES

Techlogg.com RSS news feed

feed image
feed image

Login for our FREE daily newsletter






Lost Password?
No account yet? Register

Who's Online

We have 3 guests online
Visits today: 759
Visits yesterday: 815
Visits month: 5204
Visits total: 138953
Max.monthly visits: 24649
  occurred: 2008-9
Pages this month: 21158
Pages total: 753941
Data since: 2008-04-09
Black Google power saving figures are wrong PDF Print E-mail
Written by Darren Yates   
Thursday, 26 July 2007




Del.icio.us!

StumbleUpon!

All the tree huggers in the world won’t save the planet using Black Google. We have the facts to prove it.

We all want to do the right thing when it comes to the environment but it’s amazing just how much questionable information is out there when it comes to cost savings in electricity.

One furphy we came across recently was Blackle (www.blackle.com). Apparently, the idea is that if Google switched from a white to a black background, they could save the world 750-megawatt-hours of electricity each year. Sounds like a great figure – that is until you dig down and find out where this figure actually comes from.

Blackle was set up by Heap Media on the basis of a blogpost at ecoIron. Blackle even goes as far as maintain a precise (although only God knows how accurate) watt-hour counter of how much energy Blackle has actually saved.

Now ecoIron claims that displaying a black page on a CRT monitor will consume 59-watts while a white page will chew through 74-watts. Of course, there’s nothing as useful as what size or brand of CRT monitor they’re talking about but the basic gist of the argument is that 15-watts of energy can be saved from using black instead of white page backgrounds.

Click on the ecoIron headline for that story and you find that the basic information comes from the US Department of Energy’s EnergyStar information page where it has a table indicating white colours use 74-watts and black using 59-watts.

What is surprising is how completely useless and unverifiable the information on this DOE page is.

Firstly, there’s no mention at all of the type of monitor used. Was it an LCD or CRT? So far, everyone basing their stories of this EnergyStar webpage is assuming CRT. But then there’s nothing mentioned about the size of the monitor either. Was it 15-inch? 17-inch? 21-inch? Power consumption of all monitors varies considerably on a number of factors, including the monitor’s size. So on the face of it, these wattage ratings are completely unusable to base any rational conclusions on because there’s no information about how those results were obtained.

Still, ecoIron uses the 15-watt differential in going from white to black, plus the fact that 25% of the world’s monitors are CRT for the basis of its 750-mega-watt-hour (750MWh) energy-saving claim per year.

Now to get the 25% portion, ecoIron used data from DisplaySearch that showed that “worldwide, as of first quarter 2006, they estimate LCD penetration at 74.7 percent.”

There are also assumptions about Google getting 200million queries per day and each query is displayed for about 10 seconds.

The problem is that most of these assumptions are vague at best, all used to make a nice pithy head-line.

The truth about black and white

While there are a number of assumptions, we decided to find out for ourselves just what is the power consumption difference between Blackle and Google. Using our own commercial AC power meter, we tested three monitors – a 19-inch LG L192WS widescreen LCD that is about two months old; a 17-inch Mitsubishi DV1770 CRT monitor about three years old; and a 15-inch LG StudioWorks 570LE LCD monitor, about four years old.

We measured their power consumption on Blackle and Google to see what actually happens.

Monitor

Google – power consumption (Watts)

Blackle – power consumption (Watts

19-inch LG L192WS

24.5W

25.1W

17-inch Mitsubishi DV1770

72.5W

65.4W

15-inch LG StudioWorks 570LE

26.2W

26.5W

First up, the 17-inch CRT power reduction between Google and Blackle was nowhere near the 15-watts being suggested and was closer to 7-watts.

But the most surprising result was from the two LCD monitors where the power consumption actually rose by an average of 0.5W. The reason is that LCD monitors consume energy to block light when creating a black screen as the backlight is on continuously. CRT monitors work on a completely different process where three cathode ray guns fire electron beams onto a phosphor coating so that the electron beam creates the light.

So at the very least, supporters of a Black-screen Google are overstating their power savings by about double. At worst, as CRT monitors are continually phased out in favour of LCD screens, the power consumption of a Black-screen Google will actually be more than what we already have with the standard white Google.

UPDATE August 1, 2007: If you want a guaranteed way to reduce monitor power consumption, try dropping the refresh rate. This works for both CRT and LCD monitors. Read the story on our testing here.
 
UPDATE August 2, 2007: We've now expanded testing to include four CRT and four LCD monitors, Read the results here.  
Comments
Add New Search
Matt  - And what of the energy it takes to run i   |78.32.70.xxx |2007-09-23 21:34:31
Quite aside from the bad estimations, what about the power consumed in running
Blackle's server? Even if their figures are accurate, in six months, even with
being highlighted on link sites and news stories aplenty, they claim to have
saved 200 KW/h, barely enough to run a large domestic refrigerator for the same
amount of time. A small dedicated server would consume that amount of power in
under two weeks. If they were to completely supplant Google, even just serving
pages and CSS while Google's datacenters do the grunt work, they would still
need multiple servers that would consume more energy than they can ever hope to
save. Frankly, to me this looks like nothing more than a "green guilt"
con to scoop earnings from AdSense hits.
Jek Rozitis   |203.26.122.xxx |2007-08-30 00:55:51
750 MWh/year isn't that much globally. That's the equivalent of switching off an
average sized power plant for 30 mins each year.
Dell Mustang  - CEO aspirant   |195.177.212.xxx |2007-08-08 11:36:06
There are many other sites that use similar concept. One of them is hosted
on the google owned blogspot: Power Google
Darren Yates  - I didn't say it didn't   |59.167.236.xxx |2007-08-02 05:39:58
Alan, if you actually read what I wrote , I never said it didn't work on CRT
monitors. What I am saying is that it doesn't work to the levels suggested by
Blackle. As for your PDF file, Table 8 does suggest that with one monitor, there
is an unusually high change in power consumption of 88%, but again, what was the
monitor type, make, model?
The LCD results in that table (top of page 19 for
those interested) make no sense however because they suggest that it requires
less power to produce black than white, which is simply not how LCD monitors
work.
You should be asking Blackle to prove its claims are true for LCD
monitors, which they agree now cover 75% of the market, and growing. At the
moment, Blackle makes no distinction between LCD or CRT.
Alan Liefting  - It does give a saving for CRT monitors.   |138.75.1.xxx |2007-08-02 05:21:19
See http://enduse.lbl.gov/Info/LBNL-48581.pdf for an indepth discussion on
monitor power consumption. CRT monitor power, whilst on, incleases by 18-88%
between black and white. This is dependant on the type of monitor.
Justin  - Engineer   |61.88.183.xxx |2007-08-02 00:31:11
I have a wattmeter and tried this with my Compaq S700 monitor.
It normally
consumes 97 watts with normal Google showing. With Blackle it was 77 watts. So
it does work with CRT displays. LCD displays there was an increase of about a
watt. The company I work for has 50/50 LCD's to CRT's so there are about 3000
CRT monitors here. That is a saving of 60Kw or 25 fan heaters' worth. Yes I know
that you don't leave your monitor on Google all day but the point is made.
Changing applications & screensavers to darker colours does save power. Multiply
this globally and it is a heap of energy
jeff   |203.103.221.xxx |2007-08-02 00:29:20
By definition, aren\'t headlines necessarily pithy?

An exercise like
this must use a set of approximations and estimates - better to make
use of some ranges or minimums, since few, if any, can be determined absolutely. Eg, # of CRT
monitors in use, % of users who use Google, avg length of Google
view (10 secs would be much too short for me), monitor sizes etc etc.
Darren Yates  - Two more LCDs tested   |202.168.106.xxx |2007-08-01 11:28:33
Naomi, have just tested two more LCD monitors, a 19-inch LG L1917S and an AOC
LM721 17-inch model. I'll update the results as soon as I can but the results
are the same - both monitors consume more power on Blackle than they do on
Google. I'll be doing an update of this story shortly with full details on the
test results.
Darren Yates  - Yes, we will but...   |203.18.241.xxx |2007-08-01 06:34:33
LCD technology is fairly standard and because of its nature, it must consume
more power to create a black screen because it has to block the backlight from
coming through.
But yes, we'll look at testing some more monitors in the next
week, particularly LCD monitors. I'd also hope you'd ask of the guys behind
Blackle the same request.
Naomi  - Too small a sample   |203.108.78.xxx |2007-08-01 05:10:01
I'm no statistician but I dont think 2 LCDs and 1 CRT are a representative
sample of the monitors that are out there. Please do more testing to increase
your sample size so your results have more cred. I'd love to see the results.
Great effort though.
John  - Quite Right   |195.47.237.xxx |2007-07-31 07:58:24
It is ridiculous that Mark, the guy who originated this nonsense even admitted
in his own figures that 75% of machines are now using LCDs, for which Blackle
not only doesn't save power but actually uses slightly more. There's more about
it here:

http://www.charcoaldesign.co.uk/articles/ blackle
inwal.com  - White is absolute   |71.222.73.xxx |2007-07-28 22:09:13
Yes, white not only save power, but and usable. Black screen look likes funeral
service :grin
Write comment
Name:
Email:
 
Website:
Title:
UBBCode:
[b] [i] [u] [url] [quote] [code] [img] 
 
 
Please input the anti-spam code that you can read in the image.

3.20 Copyright (C) 2007 Alain Georgette / Copyright (C) 2006 Frantisek Hliva. All rights reserved."





Google!Facebook!Slashdot!Netscape!Technorati!Newsvine!Furl!Yahoo!Ma.gnolia!Add this social bookmarking functionality to your website! title=
Last Updated ( Thursday, 02 August 2007 )
 
< Prev   Next >
© 2008 techlogg.com
Joomla! is Free Software released under the GNU/GPL License.