Power consumption of HW (not) relevant to MorphOS
  • Moderator
    Kronos
    Posts: 2119 from 2003/2/24
    Due to the current crisis I've bought a cheap device to measure power usage at the outlet.
    Code:

    idle compiling off(standby)
    iMac G5 72.4W 97W 3.3W 2.5GB RAM default screen brightness
    Peg2 G4 54.5W 57.2W 1.5W 1GB, Radeon 9250, some old HDD, passive PSU
    22" LG 35W ---- 0.5W 1680x1050


    Plan to add more PPC and non PPC-HW later, but the shocking result atm is that it makes more sense to run the iMacG5 standalone than running a Pegasos_G4 and an ancient TFT with similar specs.
  • »08.12.22 - 11:37
    Profile
  • Order of the Butterfly
    Order of the Butterfly
    ernsteiswuerfel
    Posts: 492 from 2015/6/18
    From: Funeralopolis
    Good idea! Could do some measurement on my G5 11,2 and G4 3,6 too.
    Talos II. [Gentoo Linux] | PMac G5 11,2. PMac G4 3,6. PBook G4 5,8. [MorphOS 3.17 / Void Linux / Gentoo Linux] | Vampire V4 SA [ApolloOS / Amiga OS 3.2.1]
  • »08.12.22 - 12:51
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 11795 from 2003/5/22
    From: Germany
    Quote:

    Code:

    off(standby)
    iMac G5 3.3W



    Interesting. In light of the old discussion from some years back, do you have access to any PowerMac G5 and can measure standby there?
  • »08.12.22 - 13:38
    Profile
  • Moderator
    Kronos
    Posts: 2119 from 2003/2/24
    HW that I plan to add in the coming weeks:

    PowerMacG4 1.8GHz (with and without the 22" ADC + the ADC on it's own PSU)
    PowerMacG5 2x2.3 (AGP)
    EFIKA
    QuadCore "Atom" PC
    HexCore 2010 MacPro
    Base model MacStudio
    Maybe some 68k HW, 2010 iMac and MacMini

    Various (LCD) monitors ranging from 1998 18" to 2020 32".
  • »08.12.22 - 13:47
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    Amigaharry2
    Posts: 1082 from 2010/1/6
    From: EU-Austria (Wien)
    From my own mesurement interessting behaviour G5/2,3GHz:

    Powerconsumption (with monitor, audio-subsystem and tiny-WLAN-hub is in OpenFirmware near 290W, at MOS about 260W and in OSX 196W! (ideling)

    OSX has a good powermanagement, that's for sure, but OF does'nt seem to have anything (diff=100W!!!)......thats a surprise to me.
    Peg2, 3xPowerMac G5, 2xPowerbookG4, 2x MacMiniG4, Efika (again), A3000T and life is never boring.....
  • »08.12.22 - 17:47
    Profile
  • MorphOS Developer
    jacadcaps
    Posts: 2789 from 2003/3/5
    From: Canada
    @Amigaharry2

    Interesting. My PCIe 2.3 takes about 190W when idling on MorphOS. That's with R1950, an Nvidia and a SSD drive and Realtek eth. Optical drive removed.

    Measured with Emporia Energy Vue.

    [ Edited by jacadcaps 08.12.2022 - 19:07 ]
  • »08.12.22 - 19:01
    Profile Visit Website
  • Moderator
    Kronos
    Posts: 2119 from 2003/2/24
    Quick test with the 2x2.3GHz G5 running a 3.10 beta (so any improvements to power management in the past 5.5 years does not apply).

    Idle: 211W compile: 215W standbye: 2.3W

    Needless to say this won't return to daily usage anytime soon....
  • »08.12.22 - 20:09
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 11795 from 2003/5/22
    From: Germany
    >> In light of the old discussion from some years back, do you have
    >> access to any PowerMac G5 and can measure standby there?

    > 2x2.3GHz G5 [...] standbye: 2.3W

    Thanks. This is the exact middle of the 0.9...3.7 W range mentioned back then (as opposed to the 30+ W some claimed to have measured).
  • »08.12.22 - 20:35
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    Amigaharry2
    Posts: 1082 from 2010/1/6
    From: EU-Austria (Wien)
    @Jacad: I think I'm not far away to yours: keep in mind, that monitor and WLAN-hub consumes about 60-70W too. Will check PMac allone tomorrow.
    Surprising for me is, that OF seems to have no real or efficient powermanagement (compared to OSX or even MorphOS) - did'nt expect that!

    Could be, that I have to test another G5 with different OF, because this one goes (in OF console) after 3 minutes allway to full power blowing (thermal calibratioon is OK).

    Measuringsystem is calibrated UNITEC EIM-812 performance measuring device (CATII) - should be exact enough.
    Peg2, 3xPowerMac G5, 2xPowerbookG4, 2x MacMiniG4, Efika (again), A3000T and life is never boring.....
  • »08.12.22 - 23:20
    Profile
  • Priest of the Order of the Butterfly
    Priest of the Order of the Butterfly
    analogkid
    Posts: 630 from 2004/11/3
    From: near myself
    Powermac7,3 (G5 2x1,8 GHz) with ADC 17" Studio Display and MorphOS 3.17, with dnetc running: 177 W.

    No comparison to OSX, since this machine is MorphOS-only.
  • »09.12.22 - 06:48
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Zylesea
    Posts: 2039 from 2003/6/4
    Quote:

    Kronos schrieb:
    Due to the current crisis I've bought a cheap device to measure power usage at the outlet.
    Code:

    idle compiling off(standby)
    iMac G5 72.4W 97W 3.3W 2.5GB RAM default screen brightness
    Peg2 G4 54.5W 57.2W 1.5W 1GB, Radeon 9250, some old HDD, passive PSU
    22" LG 35W ---- 0.5W 1680x1050


    Plan to add more PPC and non PPC-HW later, but the shocking result atm is that it makes more sense to run the iMacG5 standalone than running a Pegasos_G4 and an ancient TFT with similar specs.


    Measured the Efika years ago: https://via.bckrs.de/MorphOS/bench_en.html (sd card should lower the uptake quite a bit)

    Also on that site, measurements for the mac mini and pegasos1.
    I can confirm your conclusions from the total energy uptake POV as written in this thread already https://morph.zone/modules/newbb_plus/viewtopic.php?forum=11&topic_id=12759&viewmode=flat&sortorder=0&start=25
    --
    http://www.via-altera.de

    Whenever you're sad just remember the world is 4.543 billion years old and you somehow managed to exist at the same time as David Bowie.
    ...and Matthias , my friend - RIP
  • »09.12.22 - 07:51
    Profile Visit Website
  • Moderator
    Kronos
    Posts: 2119 from 2003/2/24
    @Amigaharry2

    There is no powermanagment in OF so it is normal to go into jet-mode after a while.

    The cooling on these is good enough to have the CPU(s) at full blast without damage for forever in contrast with current (x86) CPU that would easily kill themselves running full tilt on their stock coolers.
  • »09.12.22 - 08:57
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 11795 from 2003/5/22
    From: Germany
    > I can confirm your conclusions from the total energy uptake POV

    What's notably off though is the standby consumption of the iMac G5, where you measured 11 W and Kronos only 3.3 W.
  • »09.12.22 - 09:59
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    Amigaharry2
    Posts: 1082 from 2010/1/6
    From: EU-Austria (Wien)
    Quote:

    Kronos schrieb:
    @Amigaharry2

    There is no powermanagment in OF so it is normal to go into jet-mode after a while.

    The cooling on these is good enough to have the CPU(s) at full blast without damage for forever in contrast with current (x86) CPU that would easily kill themselves running full tilt on their stock coolers.


    No, depends on Firmware.

    PM G5 (OF V4):

    PM 2,3 GHz : 5.1.87f allways silent
    PM 2,3 GHz : 5.2.4f1 blowing like a jet
    PM 1,8 GHz : 5.1.5f0 allways silent
    PM 2,0 GHz : 5.1.8f7 blowing like a jet

    Seems that OF Versions 5.18f7 and above dont manage that as good as older FW-revisions.
    It's also very interessting that this 2 blowing G5 have that backside blower-problem under MorphOS too, which I discribed in another thread. Northbridge becomes much more hot, than the other two G5. Astonishingly under OSX it does'nt occour.
    Peg2, 3xPowerMac G5, 2xPowerbookG4, 2x MacMiniG4, Efika (again), A3000T and life is never boring.....
  • »09.12.22 - 10:33
    Profile
  • Moderator
    Kronos
    Posts: 2119 from 2003/2/24
    Quote:

    Andreas_Wolf wrote:
    > I can confirm your conclusions from the total energy uptake POV

    What's notably off though is the standby consumption of the iMac G5, where you measured 11 W and Kronos only 3.3 W.


    The device I've bought is really cheap so I wouldn't trust it to 100% precise.

    Difference can be up to variations in the PSU (and they decay of the notorious caps) or maybe some "wake on LAN" feature that may exist and be activated.

    FYI just did my old 27" (2560x1440) Dell and it claims 0.0W for standby and 37.3W when used. Given that is almost 1.5 times the surface of the 22" with just 2W more I am starting to suspect that the 22" is just bad.
  • »09.12.22 - 10:43
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    Amigaharry2
    Posts: 1082 from 2010/1/6
    From: EU-Austria (Wien)
    Tested now a G2 2.0GHz/ Dual_Cpu, Radeon 9600P, HD 1x2TB, 1xDVDRW, 1x IDE2CF, 2GB RAM
    No other el. consumers

    Off: 20W
    OF: 250W (allways)
    MOS: 220W - 230W (Idle - Wayfarer loading/Playing Youtube)
    OSX: 150W - 230W (Idle - Firefox loading/playing Youtube)
    Peg2, 3xPowerMac G5, 2xPowerbookG4, 2x MacMiniG4, Efika (again), A3000T and life is never boring.....
  • »09.12.22 - 19:42
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 11795 from 2003/5/22
    From: Germany
    > G2 2.0GHz/ Dual_Cpu [...] Off: 20W

    That's a remarkable (and daunting) difference to Kronos' 2.3 W.
  • »09.12.22 - 22:23
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    Amigaharry2
    Posts: 1082 from 2010/1/6
    From: EU-Austria (Wien)
    Yeah! Can't explain it, tested twice: 20W!
    Peg2, 3xPowerMac G5, 2xPowerbookG4, 2x MacMiniG4, Efika (again), A3000T and life is never boring.....
  • »09.12.22 - 22:53
    Profile
  • Priest of the Order of the Butterfly
    Priest of the Order of the Butterfly
    KennyR
    Posts: 852 from 2003/3/4
    From: #AmigaZeux, Gu...
    Just take note that some power measuring devices get completely confused if there's a bad power factor. PSUs tend to have pretty bad PF.
  • »10.12.22 - 16:30
    Profile