Power Usage

From AwkwardTV
Revision as of 18:20, 11 September 2007 by ZgzA26 (talk | contribs)
Jump to: navigation, search

dividendo borsa www nike cm epsilon regione campania progetto aifa crema corpo calendula canon 24 85 costantino nissan micra super d elezioni amministrative cosenza 2004 bidone aspirapolvere liquidi pascani canon lahdc10 colonna sonora la mia vita a stelle e carla barzotti disco polo venus mp3 antivirus panda titanium 2005 yamaha tmax 2004 biglietti d augurio albergo tortoli amd athlon 64 4600 nike shox gmb if i were you nawal ayoub annuncio lavoro pordenone una fan enamorada video de michel vieth epistole 90 herlinda gomez kramp willy bocci trucchi tigra rio cartoon natwork nostradamus patch nero memoria ram 128 recensione videogioco pc videogioco usato linee aerea callithrix macchina fotografica nikon digitale bovino raffaele sevizie karama diamocideltu nec dvdrw slim linn 2250 navigazione laghi dance 2004 mp3 migliorista quotidiani maglia nazionale italiana calcio supergirl reamon stampante canon pixma ip4000 gioco euro 2004 pc barbie it mulinello elettrico e la mia vita albano carrisi finalmente demo per gtr cbr 1000 rr fireblade luise we srl www progamming smartcard case mobili mp3 fotocamera professoressa di lingue cucine ala elett basilea ristorante curriulum tais auto porsche cayenne adobe audition 1 5 win gioca cucinando genius mouse ps2 netscroll norton cambio astrea creative itrigue 3400 taj coral reef pc athlon 3800 villaggi turistici in calabria sexo anal saintex rsmmc dv 256 mb nokia plettenberg sample videoclip childrens hoffmeister adolf calendario per totti sony ericsson k750i calendari sporchi 2004 ron jeremy luny tunes la corriera del west microfono senza fili foto capelli nuda citta capo ristorante accessori golf tettone nude lynyrd skynyrd lyve combustibili vaio centrino free tv ringtones free monophonic ringtones hoodia azithromycin cheap accutane jazz ringtones buy metformin cheap propecia free christian ringtones free ringtones phendimetrazine online buy bontril azithromycin free monophonic ringtones free fly ringtones philips ringtones skelaxin online phendimetrazine online buy cephalexin cheap metronidazole morphine online cheap zyban caller ringtones cheap ritalin cheap ativan midi ringtones online effexor free jazz ringtones free voice ringtones free rave ringtones ultram online cheap hoodia buy tenuate cheap valium cheap soma xenical free polyphonic ringtones cheap cialis cheap celebrex cheap valium cheap vicodin audiovox ringtones payday loan cheap lorazepam cheap ephedra telus ringtones lotrel online cheap xanax pantech ringtones hgh buy zoloft free alcatel ringtones free nextel ringtones samsung ringtones free caller ringtones cheap skelaxin cheap lortab cheap fluoxetine cheap codeine cheap valium alprazolam online cheap oxycontin cheap valtrex telus ringtones cheap protonix free turkey ringtones ionamin online replica watches cheap ionamin free humour ringtones free vocal ringtones cheap meridia ritalin online cheap cialis funny ringtones free caller ringtones cheap cyclobenzaprine cheap meridia cheap ativan cheap tamiflu cheap amoxicillin jungle ringtones dianabol online free pop ringtones ephedrine online amoxicillin online glucophage didrex online cheap lortab car rental propecia online cheap accutane buy adipex free audiovox ringtones polyphonic ringtones rivotril online cheap tenuate albuterol online ortho online free rave ringtones == Measured power usage == Measured with Kill-A-Watt, 119V input, tested with 720p output to LCD (not plugged in to wattmeter)
Note: These figures have also been confirmed with a 240V unit.

  • Main menu: 14-15W
  • Playing a 640x480 XviD file: 17-18W
  • dnetc -benchmark2: 20W
  • "Sleeping" (white light off, no video output after holding down Play button): 14-15W
  • After a "shutdown -h now" command: 15W


  • Is there any way to further reduce power usage, e.g., disabling the graphics processor?
  • Perhaps swapping out the hard drive for a flash based card like is being done with iPod Minis nowadays?
  • What about a "Shutdown" frappliance that physically turned it off. (Realising that some sort of hard power switch would probably be needed to turn it back on).

Also as much as 30% of the power could be used by the AC-DC transformer, running straight from DC could lower the amps used.


  • Doing a pmset -g yields:
-bash-2.05b$ pmset -g
  System-wide power settings:
     SleepDisabled          1
  Active Profiles:
     AC Power                -1*
  Currently in use:
   disksleep      15
   hibernatemode  0
   displaysleep   1
   powerbutton    0
   sleep          30
   autorestart    1
   hibernatefile  /var/vm/sleepimage

So it looks like sleep has been "disabled" in the sense of traditional mac sleep. I am not super adept at the pmset command, and can't seem to find a way to reverse that setting. You are able to play with the settings currently in use (setting disksleep to 5, or 10 for example), but trying to set womp gives no results. A pmset -g cap gives you:

-bash-2.05b$ pmset -g cap
  Capabilities for AC Power:

Additional, you can directly edit the file: /Library/Preferences/SystemConfiguration/com.apple.PowerManagement.plist. At the bottom is the setting for SleepDisabled, which is set to true by default.

Note: I just changed the file so SleepDisabled is false, saved changes and rebooted. The settings were saved, and now I will see if the system goes to "sleep" on its own. I'll also report back if the AppleTV cools down at all, or if it appears the hard drive has spun down.

I have tried SleepDisabled false, but I doesn't help. The consumption is always 14-15W when idle or "switched off".

On the other hand if you "switch off" the Apple TV by holding play/pause key for 5 seconds and then you look at pmset -g there will be

   disksleep      1
   displaysleep   1

while in normal state there are zeros. The harddrive seems to sleep (I feel a delay issuing commands in commandline), but the consumption stays the same 14-15W :(

Measurement Requests

  • Power supply efficiency
    • Efficiency is measured as effective output power divided by input power (Po/Pi x 100 %). Pi can be measure as above using a Kill-A-Watt meter (you should take the Pi measurements at the same time as you measure the DC output current). To obtain Po (recall that Po = VI), first measure the DC output voltage (V) of the supply using a multi-meter in parallel with the output supply (this value should remain relatively constant regardless of what function the AppleTV is performing). Next, measure the DC current (I) using the multi-meter remembering that the meter must be placed in series with the DC output voltage from the power supply for this measurement (the output current will vary with what function the AppleTV is performing - current is proportional to processing load). Then, the output power (Po) is computed by multiplying the measured DC voltage by the measured DC current (Po = VI).
  • The digital wattmeter I have can measure P and U*I. If I remember well P=U*I*cos phi, where phi is the phase and cos phi is the efficiency. For AppleTV I have measured in idle P=15W and U*I=31VA, the efficiency seems to be comparable with my laptop power supply, where I have P=30W and U*I=60VA.