How much electricity does a Computer consume?

Here are some data for computers and components that I have measured. These were measured with a Conrad "Voltcraft Energy Check 3000" on the AC input of the device; according to c't this measurement device is relatively accurate above 7W or so.

Complete setups

190W
K6-2 300MHz system idle (see below), 21" Nokia 445Xi monitor displaying much white, active speakers, cable modem, HP DeskJet 600 (soft-off).
244W
Athlon 64 system idle (see below), 21" Nokia 445Xi monitor displaying much white, aktive speakers, cable modem (HP LaserJet 1200 hard-off).
25W
The Athlon 64 system above, with main box soft-off, monitor soft-off, speakers hard-off, printer hard-off, cable modem on (no power switch).
12.7W
iBook G4 1066MHz idle, screen off, running Linux-2.6.8; 12.1W with disk on standby (spun down) in laptop-mode; 18W idle with screen on, brightest setting; 14.9W idle with screen on, darkest setting before total off; 22W loaded (kernel compile), screen off.

Boxes

360-550W
Core 2 Quad Q6600 (2.4GHz), Asus Striker Extreme motherboard (Nvidia 680i chipset), 2 EVGA Nvidia 8800 Ultra Superclocked graphics cards (for SLI operation), 3 hard disks, 650W power supply, running Windows Vista 64. 360-380W idle, 470-550W when loaded with Crysis (a game benchmark).
283W-423W
Dual Xeon 5160 (Socket 771, 3GHz, dual core, 4MB L2 cache, 65nm), Supermicro X7DBE+ board with on-board graphics (ATI ES1000), 24GB DDR2-667 RAM (12*2GB FB-DIMMs), 2 400GB SATA hard disks spinning, 1 DVD-RW drive, Tagan 700W power supply, lots of fans, running Linux 2.6.17.
clock   idle  load 1  load 2 load 3 load 4
2000MHz 283W   290W    297W   305W   311W
2333MHz 284W   296W    309W   317W   326W
2666MHz 285W   305W    324W   335W   347W
3000MHz 286W   313W    340W   354W   368W
With other programs producing the load we could get up to 412W with pure CPU load, 419W with CPU and memory load, and 423W by also accessing the hard disks.
283W-451W
The Dual Xeon box above, upgraded with two Xeon 5450s (Socket 771, 3GHz, quad core, 2*6MB L2 cache per socket, 45nm).
   load  0    1    2    3    4    5    6    7    8   
2000MHz 283W 292W 302W 311W 321W 329W 336W 344W 353W
3000MHz 291W 310W 328W 345W 362W 377W 396W 413W 431W
The load was produced using yes >/dev/null (IIRC that load produced 412W at load 4 3000MHz before the CPU upgrade). By replacing one yes process with a process that copies 1GB of memory repeatedly, the power consumption at load 8 3000MHz went up to 451W. Replacing more yes processes with memory-intensive ones reduces power consumption.
292W-351W
HP Workstation with 2 900MHz Itanium2 CPUs, 4GB RAM, 2 SCSI hard disks with 10000rpm and some other goodies.
   load  0    1    2
 900MHz 292W 321W 351W
The load was produced using yes >/dev/null
180W-225W
Dual Opteron 246 (Socket 940, 2GHz, 1MB L2 cache), some Tyan Thunder K8 board with on-board graphics, 2GB PC2700 RAM with ECC, 2 IDE hard disks, 1 DVD-ROM drive. 180W idle, 225W with both CPUs under load.
186W
Compaq XP1000 (500MHz Alpha 21264 CPU, 128MB RAM, 1 SCSI disk, 1 CD-ROM drive, Matrox Millenium II PCI graphics card).
121W-212W
Dual Opteron 270 (Socket 940, Dual-core 2GHz, 2*1MB L2 cache), Tyan S2892 Thunder K8SE board with on-board graphics (ATI Rage XL), 8GB PC3200 ECC RAM, 2 300GB SATA hard disks spinning, 1 DVD-RW drive, Tagan TG-480-U22 power supply, running Linux-2.6.14.3.
clock   voltage  idle  load 1  load 2  load 3  load 4
1000MHz 1100mV   121W   124W    127W    131W    134W
1800MHz 1350mV   161W   171W    181W    191W    202W
2000MHz 1350mV   167W   178W    190W    200W    212W
   ondemand      121W   154W  165W/189W
Linux-2.6.14.3 seems to prefer to put the second process on the second chip, so we usually got the 189W consumption with two non-nice processes and the ondemand frequency governor; we got the 165W number by starting three processes, then killing the middle one, so that both processes run on the same chip, and the other chip idles.
97W-185W
Core 2 Duo E8400 (45nm, 3GHz, 6MB L2), MSI P35 Neo2-FR (Intel P35+ICH9R chipset), PowerColor Radeon X850XT card with 256MB RAM, 4GB DDR2-800 RAM, Creative Audigy soundcard, NE2000PCI clone Ethernet card, 2 spinning 750GB SATA hard disks (WDC WD7500AACS-00ZJB0, SAMSUNG HD753LJ), 1 DVD-RW drive, 1 floppy drive, Enermax-EG365AX-VE(G) ATX12V power supply, Linux. Idle 97W-100W. Pure CPU loads (yes >/dev/null):
clock    idle   load 1  load 2
2000MHz  100W   111W    121W
2333MHz  101W   115W    127W
2666MHz  102W   118W    133W
3000MHz  103W   123W    140W
179W when running glxgears in addition to two instances of nice yes >/dev/null at 3000MHz. 150W-155W when running UT2004 without active background jobs. 185W peak when running Titanquest on Windows XP.

Variations on graphics cards:

Connect3D ASUS     Sapphire Sapphire Gainward Sapphire  
Radeon    Nvidia   Ultimate Radeon   Radeon   Ultimate
X850XT    EN8600GT Radeon   3870     4650     Radeon  
          silent   X1650Pro Ultimate          4670      
98W       110W      94W     148W      95W     102W     Linux idle (free drivers: radeon, nv), 2 HDs spinning
98W       103W      94W     104W      83W      87W     Windows idle, usually 1 HD spinning
185W      152W     150W     202W     121W     142W     Windows TitanQuest peak, usually 1 HD spinning
All but the X850XT and the 4650 are fanless; with the 3870 the power consumption reached 230W in one of the 3DMark06 benchmarks.
103W-156W
Xeon 3070 (=Core2 Duo E6700, Socket 775, 2.66GHz, 4MB L2), Supermicro PDSME+ (Intel E7230 chipset), 8GB DDR2 RAM, 2 320GB SATA hard disks, 1 DVD-ROM, 1 floppy drive, Supermicro case with Ablecom SP645-PS 645W power supply. With yes >/dev/null loads (and idle drives), I see:
clock   idle   load 1  load 2
1600MHz 103W    112W    122W
2133MHz 104W    120W    136W
2666MHz 104W    133W    156W
With the load gforth -e ": foo begin again ; foo" used in most other results here, I see:
clock   idle   load 1  load 2
2666MHz 104W    123W    132W
The difference between the loads was not as big on other machines where I tested both.
94W-160W
Core 2 Duo E6600 (Socket 775, 2.4GHz, 4MB L2 cache), ASUS P5LD2 SE board (i945P), 2 GB DDR2 RAM, Palit Radeon X800GTO with 256MB RAM, Realtek 8169 ethernet card, 1 250GB SATA hard disk, 1 DVD-ROM drive, 1 DVD-RW drive, DTK 400W power supply. 94W idle under Linux, 160W gaming under Windows. With a Radeon X550 (instead of the X800GTO) under Linux, we see the following with a pure CPU load (the same used on other boxes; there are more energy-hungry loads):
clock   idle  load 1  load 2
1600MHz  85W    94W    100W
2400MHz  86W   105W    115W
Moreover, we tried a few different graphics cards on this machine, and measured the following (under Windows):
Card            idle   UT2004  Aquamark
Radeon X550      86W    122W
Radeon X800GTO   94W    163W
Radeon X850XT   104W    181W    190W
83W-180W
Athlon 64 3200+ (Socket 754, 2GHz, 1MB L2 cache, Clawhammer C0 stepping), Asus K8VSE Deluxe (VIA K8T800), Gforce4Ti4200 AGP with 64MB RAM, 512MB PC2700 RAM with ECC, Creative Audigy soundcard, NE2000PCI clone Ethernet card, 2 IDE hard disks, 1 LG CDRW drive, 1 Liteon DVD+RW drive, floppy drive, Enermax-EG365AX-VE(G) ATX12V power supply. 120W idle under Linux (without OS support for Cool&Quiet) with one disk spinning, 160W when running oggenc under Linux, 180W when playing a game under Windows.

~15W less with a Radeon 9600 (instead of the Gforce 4200). ~20W less when idle under Linux with cpufreq (Cool'n'Quiet support) @800MHz. With these changes, ~83W when idle, ~145W compiling, ~160W gaming.

83W-143W
Athlon 64 X2 4400+ (Socket 939, 2.2GHz, 2*1MB L2 cache, cpu family 15 model 35), Tyan S2865AG2NRF Tomcat K8E, on-board graphics (ATI Rage XL) in text mode, 4GB PC3200 DDR ECC SDRAM, 2 300GB hard disks spinning, 1 DVD-RW drive, Tagan TG480-U22 power supply.
                |------ power ------|
clock   voltage idle   load 1  load 2
1000MHz  1200mV  83W     93W    102W 
1800MHz  1250mV  86W    103W    121W 
2000MHz  1300mV  88W    109W    130W 
2200MHz  1350mV  92W    116W    143W 
A very similar machine, but with an Athlon 64 X2 4600+ (cpu family 15 model 43) consumes as follows:
                |------ power ------|
clock   voltage idle   load 1  load 2
1000MHz  1200mV  89W     98W    106W
1800MHz  1200mV  90W    105W    120W
2000MHz  1250mV  93W    112W    131W
2200MHz  1300mV  98W    120W    144W
2400MHz  1300mV  98W    122W    149W
BTW, load is a pure CPU load. We found significantly higher power consumption for a memory-bound load (~147W on the second system with load 1, and up to 160W with load 2; additional core-intensive work and maybe some I/O to a PCIe graphics card should increase the power some more).
65W-110W
Athlon 64 3200+ (Socket 939, 2GHz, 512KB L2, Winchester (90nm)), MSI K8T Neo2 (VIA K8T800), Radeon 9600XT AGP with 256MB RAM, 1GB PC3200 RAM (2*512MB), on-board sound and network, 1 hard disk, 1 CDRW drive, 1 DVDRW drive. ~65W idle under WXP (no Cool'n'Quiet, should be ~6W less), ~78W idle under Linux (no Cool'n'Quiet, does not work properly on this box), ~100W compiling, ~110W gaming. The main difference to the other Athlon 64 3200+ box probably the CPU (early Clawhammer vs. Winchester).
60W-100W
Pentium 4 2.26GHz, i845E based board, ATI Rage128, 1GB PC2100 RAM with ECC, 1 IDE disk, 1 CD-ROM drive. 60W idle, 100W under load.
60W
K6-2 300MHz (idle), Soyo SY-5EHM (VIA chipset), Voodoo 3 3000 AGP, 192MB PC100 RAM, Soundblaster Pro ISA sound card, NE2000 PCI ethernet card, 3 hard disks, 1 CD-ROM drive (playing an audio CD).
53W-155W
Xeon 3460 (2.8GHz Socket 1156), Intel S3420GPLC motherboard (Intel 3420 chipset) with onboard graphics, 1x2GB DDR3 ECC memory (interim solution until we get our 2x4GB), 2 SATA hard drives (spinning), 1 DVD-ROM drive (idle), Corsair CMPSU-400CX 80+ power supply.
   load  0    1    2    3    4    5    6    7    8   
2800MHz 53W  80W  96W 118W 138W 145W 149W 151W 155W
Performance tests at different load levels showed that Turbo mode did not work.
16W
Igel Premium 532: diskless, fanless PC with 800MHz VIA C3 CPU used as X-Terminal, with on board graphics, idle.

Monitor

88W-163W
30" Dell 3008WFP LDC Monitor; 88W at lowest brightness (still almost too bright when the monitor is new), 163W at the highest brightness (unbearably bright when the monitor is new).
70W-110W
21" Nokia 445Xi CRT monitor; 70W displaying black, 110W displaying all white at the highest brightness level.
40W-50W
21" Viewsonic VP 211b LCD monitor. Power varies with brightness level set (but not with content).

Speakers

6W-7W
Microspot CP-300 active stereo speakers. 6W silent, 7W with ordinary volume.
10W-30W
Creative Inspire P580 5.1 system. 10W silent, 30W at max. volume with some film running.

Links


Anton Ertl