• Pi power

    From Tim Streater@3:770/3 to All on Tue Jan 5 22:54:36 2021
    Having received a Pi 4B today with all the bits to make a system, I let it install and update the OS. A bit later I shut it down, and then turned off the Pi power supply.

    The Pi carried on being powered. After dismissing the possibility of sentient self-powered computers taking over, I discovered that the volts were coming
    via the HDMI port.

    Why is this? Is it normal? Is it safe? Anything special I need to do on shutting down?

    Plugging the HDMI cable back in and restoring normal power and all appears
    fine But I'd like to be sure to avoid damage.

    The HDMI cable is connected to one of these:

    <https://www.amazon.co.uk/gp/product/B07SR1S2JL/>

    which is quite handy to allow two screens on the Mac and then switch one to
    the Pi.

    --
    Tim

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Andy Burns@3:770/3 to Tim Streater on Tue Jan 5 23:16:54 2021
    Tim Streater wrote:

    The Pi carried on being powered.

    Unless a 4B is different to any other rPi, they never turn off, they
    just halt.

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Mayayana@3:770/3 to Andy Burns on Tue Jan 5 19:59:04 2021
    "Andy Burns" <usenet@andyburns.uk> wrote

    | Unless a 4B is different to any other rPi, they never turn off, they
    | just halt.
    |

    ? They have a power switch. I turn it on and it goes through
    boot. I shut down through the GUI and it goes through
    shutdown. I don't see it as different from a PC, except that
    I have to turn off the power switch by hand. I suppose
    if you made it from scratch you might not have a power
    line switch, but it's standard with the kits.

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From A. Dumas@3:770/3 to Mayayana on Wed Jan 6 07:31:56 2021
    Mayayana <mayayana@invalid.nospam> wrote:
    "Andy Burns" <usenet@andyburns.uk> wrote

    | Unless a 4B is different to any other rPi, they never turn off, they
    | just halt.
    |

    ? They have a power switch.

    They don't. Like Andy said: if you do not physically disconnect the power
    plug, the Pi will be halted, not powered off.

    if you made it from scratch you might not have a power
    line switch, but it's standard with the kits.

    Not with my kits.

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Mike@3:770/3 to usenet@andyburns.uk on Wed Jan 6 09:44:02 2021
    In article <i5kab6FrpbpU2@mid.individual.net>,
    Andy Burns <usenet@andyburns.uk> wrote:
    Tim Streater wrote:

    The Pi carried on being powered.

    Unless a 4B is different to any other rPi, they never turn off, they
    just halt.

    But, from the original post, which you snipped :-

    A bit later I shut it down, **and then turned off the Pi power supply.**
    The Pi carried on being powered.

    Are you saying the Pi 4B is a stealth free-energy device? ;)
    --
    --------------------------------------+------------------------------------ Mike Brown: mjb[-at-]signal11.org.uk | http://www.signal11.org.uk

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Tim Streater@3:770/3 to All on Wed Jan 6 10:44:48 2021
    On 06 Jan 2021 at 10:31:01 GMT, Martin Gregorie <martin@mydomain.invalid> wrote:

    On Wed, 06 Jan 2021 09:44:02 +0000, Mike wrote:

    In article <i5kab6FrpbpU2@mid.individual.net>,
    Andy Burns <usenet@andyburns.uk> wrote:
    Tim Streater wrote:

    The Pi carried on being powered.

    Unless a 4B is different to any other rPi, they never turn off, they
    just halt.

    But, from the original post, which you snipped :-

    A bit later I shut it down, **and then turned off the Pi power
    supply.**
    The Pi carried on being powered.

    Are you saying the Pi 4B is a stealth free-energy device? ;)

    Its not. This just shows that, if there are common positive and ground
    rails on the Pi's PCB that are connected to ALL sockets capable of
    accepting a connection from an externally powered peripheral device,
    which AFAIK is the case for all Pi models, then a Pi will remain active
    as long as external device with a power source providing enough current
    to keep the RPi alive is connected to it.

    It also means that you can destroy a Pi by connecting it to anything
    which requires more than 5v to operate and that has its positive and
    ground rails connected to the plug it uses to talk to the Pi.

    OK. So what device is providing the power to the HDMI cables, and why? In my setup I have a Dell display connected via its DVI (?) socket and an adapter to a small HDMI switcher that can route the display either to my Mac Mini or to the Pi.

    I did notice that, after disconnecting the HDMI from the Pi, and reconnecting it (all this with the Pi's power adapter unplugged from the mains), the Pi did not power up again. So whatever is on teh HDMI is not a strong power source.

    --
    Tim

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Andy Burns@3:770/3 to Mike on Wed Jan 6 10:55:56 2021
    Mike wrote:

    Andy Burns wrote:

    Tim Streater wrote:

    The Pi carried on being powered.

    Unless a 4B is different to any other rPi, they never turn off, they
    just halt.

    But, from the original post, which you snipped :-

    A bit later I shut it down, **and then turned off the Pi power supply.**
    The Pi carried on being powered.

    Are you saying the Pi 4B is a stealth free-energy device? ;)

    The Pi is the HDMI source, and the monitor is the HDMI sink, per the
    spec the 5V power should be supplied from the Pi to the monitor, not the reverse ...

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Martin Gregorie@3:770/3 to Mike on Wed Jan 6 10:31:01 2021
    On Wed, 06 Jan 2021 09:44:02 +0000, Mike wrote:

    In article <i5kab6FrpbpU2@mid.individual.net>,
    Andy Burns <usenet@andyburns.uk> wrote:
    Tim Streater wrote:

    The Pi carried on being powered.

    Unless a 4B is different to any other rPi, they never turn off, they
    just halt.

    But, from the original post, which you snipped :-

    A bit later I shut it down, **and then turned off the Pi power
    supply.**
    The Pi carried on being powered.

    Are you saying the Pi 4B is a stealth free-energy device? ;)

    Its not. This just shows that, if there are common positive and ground
    rails on the Pi's PCB that are connected to ALL sockets capable of
    accepting a connection from an externally powered peripheral device,
    which AFAIK is the case for all Pi models, then a Pi will remain active
    as long as external device with a power source providing enough current
    to keep the RPi alive is connected to it.

    It also means that you can destroy a Pi by connecting it to anything
    which requires more than 5v to operate and that has its positive and
    ground rails connected to the plug it uses to talk to the Pi.


    --
    --
    Martin | martin at
    Gregorie | gregorie dot org

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Joe@3:770/3 to Tim Streater on Wed Jan 6 11:03:23 2021
    On 6 Jan 2021 10:44:48 GMT
    Tim Streater <timstreater@greenbee.net> wrote:



    OK. So what device is providing the power to the HDMI cables, and
    why? In my setup I have a Dell display connected via its DVI (?)
    socket and an adapter to a small HDMI switcher that can route the
    display either to my Mac Mini or to the Pi.

    I did notice that, after disconnecting the HDMI from the Pi, and
    reconnecting it (all this with the Pi's power adapter unplugged from
    the mains), the Pi did not power up again. So whatever is on teh
    HDMI is not a strong power source.


    HDMI is Smart, God help us. You cannot generally power up an HDMI
    source, then connect it to an HDMI sink, because the source will not
    generally supply signals to an open cable. It wants to see something
    alive on the other end before it will power up its output drivers. It's
    all about energy saving.

    Sadly, some HDMI sinks do the same. I found that my laptop could not
    supply an HDMI signal to a particular Grundig monitor, because neither
    end was willing to accept that the other existed. I tried many attempts
    at powering them both up with varying delays between them, but could
    not find a window in which both could see the other at the same time.
    Both devices worked correctly with other HDMI devices.

    The same will be true of HDMI 5 Volts. Not all devices do it at all,
    and those which do will generally want to see something already
    connected to the cable before they turn it on. Some cables do not have
    the 5V line. There are several HDMI standards, and many cheap cables.
    Only the most recent, for example, have an Ethernet connection.

    --
    Joe

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Tim Streater@3:770/3 to All on Wed Jan 6 10:32:59 2021
    On 06 Jan 2021 at 09:44:02 GMT, Mike <Mike> wrote:

    In article <i5kab6FrpbpU2@mid.individual.net>,
    Andy Burns <usenet@andyburns.uk> wrote:
    Tim Streater wrote:

    The Pi carried on being powered.

    Unless a 4B is different to any other rPi, they never turn off, they
    just halt.

    But, from the original post, which you snipped :-

    A bit later I shut it down, **and then turned off the Pi power supply.**
    The Pi carried on being powered.

    Are you saying the Pi 4B is a stealth free-energy device? ;)

    Ah, someone referring to my OP, instead of rabbiting on about irrelevancies.

    --
    Tim

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Andy Burns@3:770/3 to Tim Streater on Wed Jan 6 10:57:47 2021
    Tim Streater wrote:

    Mike wrote:

    Are you saying the Pi 4B is a stealth free-energy device? ;)

    Ah, someone referring to my OP, instead of rabbiting on about irrelevancies.

    What's relevant is that your monitor shouldn't be reverse feeding 5V to
    the Pi

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Chris Green@3:770/3 to Andy Burns on Wed Jan 6 11:19:51 2021
    Andy Burns <usenet@andyburns.uk> wrote:
    Mike wrote:

    Andy Burns wrote:

    Tim Streater wrote:

    The Pi carried on being powered.

    Unless a 4B is different to any other rPi, they never turn off, they
    just halt.

    But, from the original post, which you snipped :-

    A bit later I shut it down, **and then turned off the Pi power supply.** >> The Pi carried on being powered.

    Are you saying the Pi 4B is a stealth free-energy device? ;)

    The Pi is the HDMI source, and the monitor is the HDMI sink, per the
    spec the 5V power should be supplied from the Pi to the monitor, not the reverse ...

    I've heard it said that the HDMI spec is a real can of worms.

    This thread is just another thing that increases my dislike of HDMI.

    It's a real pity that there isn't really a modern video/display cable
    that isn't beset by the software and 'anti pirating' stuff that's in
    HDMI.

    --
    Chris Green
    ·

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From TimS@3:770/3 to Andy Burns on Wed Jan 6 11:44:17 2021
    On 06 Jan 2021 at 10:57:47 GMT, Andy Burns <usenet@andyburns.uk> wrote:

    Tim Streater wrote:

    Mike wrote:

    Are you saying the Pi 4B is a stealth free-energy device? ;)

    Ah, someone referring to my OP, instead of rabbiting on about irrelevancies.

    What's relevant is that your monitor shouldn't be reverse feeding 5V to
    the Pi

    Agreed, if that's what its doing. Could be the other Mini, I suppose. I'll
    have to do some more testing.

    --
    Tim

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Axel Berger@3:770/3 to Chris Green on Wed Jan 6 13:13:42 2021
    Chris Green wrote:
    that isn't beset by the software and 'anti pirating' stuff that's in
    HDMI.

    Exactly. And there is another thing. Originally video had to be fast,
    for the eye needed at least 70 Hz not to see annnoying flicker. Those
    days are long gone. Current dispays are static or buffered. Not all
    computing is video or games (though admittedly those are the uses that
    bring the revenue in), so cables, buffers and a lot of critical stuff
    could be made much easier by reducing to 10 frames per socond or even
    less. One frame per second is plenty for many use cases, at least all
    those requiring long cables and long distances.


    --
    /¯\ No | Dipl.-Ing. F. Axel Berger Tel: +49/ 221/ 7771 8067
    \ / HTML | Roald-Amundsen-Straße 2a Fax: +49/ 221/ 7771 8069
     X in | D-50829 Köln-Ossendorf http://berger-odenthal.de
    / \ Mail | -- No unannounced, large, binary attachments, please! --

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Andy Burns@3:770/3 to TimS on Wed Jan 6 12:10:01 2021
    TimS wrote:

    Andy Burns wrote:

    What's relevant is that your monitor shouldn't be reverse feeding 5V to
    the Pi

    Agreed, if that's what its doing. Could be the other Mini, I suppose. I'll have to do some more testing.

    Wouldn't be the biggest surprise in the world if the a cheap switch box
    just commoned all the 5V pins together (since it can act as either a 2>1
    or 1>2 device, it'll be dumb).

    But a source HDMI device is expected to protect itself from an
    unexpected 5V arriving into its output, so no harm "should" occur.

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From A. Dumas@3:770/3 to Tim Streater on Wed Jan 6 12:02:15 2021
    Tim Streater <timstreater@greenbee.net> wrote:
    Ah, someone referring to my OP, instead of rabbiting on about irrelevancies.

    Wow, you sound delightful.

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Ahem A Rivet's Shot@3:770/3 to Axel Berger on Wed Jan 6 12:47:44 2021
    On Wed, 06 Jan 2021 13:13:42 +0100
    Axel Berger <Spam@Berger-Odenthal.De> wrote:

    Chris Green wrote:
    that isn't beset by the software and 'anti pirating' stuff that's in
    HDMI.

    Exactly. And there is another thing. Originally video had to be fast,
    for the eye needed at least 70 Hz not to see annnoying flicker. Those

    That number seems to have been creeping up all my life - 24FPS was
    once considered good enough for movie cameras.

    days are long gone. Current dispays are static or buffered. Not all
    computing is video or games (though admittedly those are the uses that

    It is rare that any system has no video or similar high update
    speed requirement at all, not counting headless machines that have no
    monitor.

    bring the revenue in), so cables, buffers and a lot of critical stuff
    could be made much easier by reducing to 10 frames per socond or even
    less. One frame per second is plenty for many use cases, at least all
    those requiring long cables and long distances.

    However if you need *any* video or similar then you need better
    frame rates unless you can build a cable that supports high burst rates but
    not high continuous rates which seems unlikely without a lot of circuitry
    and latency.

    --
    Steve O'Hara-Smith | Directable Mirror Arrays C:\>WIN | A better way to focus the sun
    The computer obeys and wins. | licences available see
    You lose and Bill collects. | http://www.sohara.org/

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Ahem A Rivet's Shot@3:770/3 to Chris Green on Wed Jan 6 12:39:36 2021
    On Wed, 6 Jan 2021 11:19:51 +0000
    Chris Green <cl@isbd.net> wrote:

    It's a real pity that there isn't really a modern video/display cable
    that isn't beset by the software and 'anti pirating' stuff that's in
    HDMI.

    The software and anti-pirating stuff is HDCP not HDMI. HDCP runs
    over HDMI, DVI and DisplayPort as well as USB-3 and will probably be
    required in anything new that comes up even though it is essentially
    useless. HDMI itself doesn't seem worse than any of the others, better than
    DVI in that it carries audio as well as video.

    --
    Steve O'Hara-Smith | Directable Mirror Arrays C:\>WIN | A better way to focus the sun
    The computer obeys and wins. | licences available see
    You lose and Bill collects. | http://www.sohara.org/

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Mayayana@3:770/3 to A. Dumas on Wed Jan 6 08:18:00 2021
    "A. Dumas" <alexandre@dumas.fr.invalid> wrote

    | > ? They have a power switch.
    |
    | They don't. Like Andy said: if you do not physically disconnect the power
    | plug, the Pi will be halted, not powered off.
    |
    | > if you made it from scratch you might not have a power
    | > line switch, but it's standard with the kits.
    |
    | Not with my kits.
    |
    Interesting. I bought a kit at Microcenter. It seemed
    cheaper and easier than buying separate pieces.
    Included was a short length of cord with an inline switch.
    So I treat it just like a computer. Since I only use it
    to stream Internet/movies to a TV, there's no reason
    to have it on normally. I turn it on when I want to use it.
    When I'm through I go to shutdown, let the OS shut down,
    then turn off the power. But I suppose if you're using it as
    part of some kind of 24/7 device you'd never want to
    turn it off or shut down the OS.

    I've never tried leaving the power on after shutdown,
    but I know it doesn't self-boot because that HDMI input
    to the TV no longer has input, even if I haven't yet
    turned off the power switch. So presumably I'd have to
    cycle the switch to get it to boot again.

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From TimS@3:770/3 to Andy Burns on Wed Jan 6 13:25:33 2021
    On 06 Jan 2021 at 12:10:01 GMT, Andy Burns <usenet@andyburns.uk> wrote:

    TimS wrote:

    Andy Burns wrote:

    What's relevant is that your monitor shouldn't be reverse feeding 5V to >>> the Pi

    Agreed, if that's what its doing. Could be the other Mini, I suppose. I'll >> have to do some more testing.

    Wouldn't be the biggest surprise in the world if the a cheap switch box
    just commoned all the 5V pins together (since it can act as either a 2>1
    or 1>2 device, it'll be dumb).

    Are these 5V signal or power? From what Joe said upthread, both ends need to see the other. A 5V signal should suffice for that, I would have thought. No need for the source to be offering to power the remote end.

    But a source HDMI device is expected to protect itself from an
    unexpected 5V arriving into its output, so no harm "should" occur.

    A diode would suffice.

    --
    Tim

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Oscar@3:770/3 to alexandre@dumas.fr.invalid on Wed Jan 6 14:02:58 2021
    In article <5ff5676c$0$304$e4fe514c@news.xs4all.nl>,
    A. Dumas <alexandre@dumas.fr.invalid> wrote:
    They don't. Like Andy said: if you do not physically disconnect the power >plug, the Pi will be halted, not powered off.

    I'm not sure about the 4, because I don't own one, but this is true for
    at least all Pi's before the 4. My shiny new Pi400 does however have the capability of a soft power off. It even has a special key-combo for it.

    This is however completely irrelevant to this thread, because the OP had
    his PI still being powered after physically powering the powersupply off.
    --
    [J|O|R] <- .signature.gz

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Andy Burns@3:770/3 to TimS on Wed Jan 6 14:41:25 2021
    TimS wrote:

    Andy Burns wrote:

    Wouldn't be the biggest surprise in the world if the a cheap switch box
    just commoned all the 5V pins together

    Are these 5V signal or power?

    Power, but only 55mA

    From what Joe said upthread, both ends need to
    see the other. A 5V signal should suffice for that, I would have thought. No need for the source to be offering to power the remote end.

    The "seeing" is over the DDC pin

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From TimS@3:770/3 to Andy Burns on Wed Jan 6 15:17:59 2021
    On 06 Jan 2021 at 14:41:25 GMT, Andy Burns <usenet@andyburns.uk> wrote:

    TimS wrote:

    Andy Burns wrote:

    Wouldn't be the biggest surprise in the world if the a cheap switch box >>> just commoned all the 5V pins together

    Are these 5V signal or power?

    Power, but only 55mA

    Not enough to run a Pi, then, I imagine, but enuough to flutter the Pi's red/green LEDs.

    --
    Tim

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Martin Gregorie@3:770/3 to TimS on Wed Jan 6 16:05:39 2021
    On Wed, 06 Jan 2021 15:17:59 +0000, TimS wrote:

    On 06 Jan 2021 at 14:41:25 GMT, Andy Burns <usenet@andyburns.uk> wrote:

    TimS wrote:

    Andy Burns wrote:

    Wouldn't be the biggest surprise in the world if the a cheap switch
    box just commoned all the 5V pins together

    Are these 5V signal or power?

    Power, but only 55mA

    Not enough to run a Pi, then, I imagine, but enough to flutter the Pi's red/green LEDs.

    The red LED just shows that power is on - nothing more.

    The yellow and green ones show whether the Pi is active: stop a running
    Pi with the 'sudo stop' command and all the LEDs go out apart from the
    red one, which stays on until you switch off or unplug the wall rat.

    That you don't know that shows that you usually stop your Pis by powering
    them off. Not a good idea because it can corrupt the SD card - and Murphy
    says that you WILL soon or later power it off while the SD card is in the middle of wear balancing - this will cause unrecoverable errors.


    --
    --
    Martin | martin at
    Gregorie | gregorie dot org

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From TimS@3:770/3 to All on Wed Jan 6 16:14:19 2021
    On 06 Jan 2021 at 16:05:39 GMT, Martin Gregorie <martin@mydomain.invalid> wrote:

    On Wed, 06 Jan 2021 15:17:59 +0000, TimS wrote:

    On 06 Jan 2021 at 14:41:25 GMT, Andy Burns <usenet@andyburns.uk> wrote:

    TimS wrote:

    Andy Burns wrote:

    Wouldn't be the biggest surprise in the world if the a cheap switch >>>>> box just commoned all the 5V pins together

    Are these 5V signal or power?

    Power, but only 55mA

    Not enough to run a Pi, then, I imagine, but enough to flutter the Pi's
    red/green LEDs.

    The red LED just shows that power is on - nothing more.

    The yellow and green ones show whether the Pi is active: stop a running
    Pi with the 'sudo stop' command and all the LEDs go out apart from the
    red one, which stays on until you switch off or unplug the wall rat.

    Thanks - that's useful to know.

    That you don't know that shows that you usually stop your Pis by powering them off.

    No "usually" about it. This is my first Pi and it was unboxed approx 24 hrs ago.

    Not a good idea because it can corrupt the SD card - and Murphy
    says that you WILL soon or later power it off while the SD card is in the middle of wear balancing - this will cause unrecoverable errors.

    Roger that, Houston.

    --
    Tim

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Theo@3:770/3 to Chris Green on Wed Jan 6 16:41:25 2021
    Chris Green <cl@isbd.net> wrote:
    Andy Burns <usenet@andyburns.uk> wrote:
    The Pi is the HDMI source, and the monitor is the HDMI sink, per the
    spec the 5V power should be supplied from the Pi to the monitor, not the reverse ...

    I've heard it said that the HDMI spec is a real can of worms.

    This thread is just another thing that increases my dislike of HDMI.

    It's exactly the same with VGA.

    The purpose of this is for the computer to read the EDID ROM from the
    monitor. It supplies 5V to allow detection of the connected monitor even if it's turned off.

    The monitor/KVM should not be emitting power on this rail.

    Although it wouldn't have taken much for the Pi to incorporate a diode to prevent back-powered. Maybe there are use cases where people want to power their Pis from HDMI, I don't know.

    Theo

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Martin Gregorie@3:770/3 to Theo on Wed Jan 6 17:04:21 2021
    On Wed, 06 Jan 2021 16:41:25 +0000, Theo wrote:


    Maybe there are use cases where people want to
    power their Pis from HDMI, I don't know.

    Doing that would make the installation tidier if I wanted to glue the Pi enclosure to the back of the screen - provided that the screen
    can provide the wattage needed. Is that likely, bearing in mind that, if
    I did that, I'd also want it to power a keyboard and mouse via the USB
    sockets on the Pi.


    --
    --
    Martin | martin at
    Gregorie | gregorie dot org

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Chris Green@3:770/3 to Ahem A Rivet's Shot on Wed Jan 6 17:04:01 2021
    Ahem A Rivet's Shot <steveo@eircom.net> wrote:
    On Wed, 6 Jan 2021 11:19:51 +0000
    Chris Green <cl@isbd.net> wrote:

    It's a real pity that there isn't really a modern video/display cable
    that isn't beset by the software and 'anti pirating' stuff that's in
    HDMI.

    The software and anti-pirating stuff is HDCP not HDMI. HDCP runs
    over HDMI, DVI and DisplayPort as well as USB-3 and will probably be
    required in anything new that comes up even though it is essentially
    useless. HDMI itself doesn't seem worse than any of the others, better than DVI in that it carries audio as well as video.

    Yes, but you can't get an HDMI interface that doesn't implement HDCP
    can you? That's the whole point, it prevents (supposedly) you viewing
    things you shouldn't with your TV (or whatever) that has an HDMI
    connector.

    --
    Chris Green
    ·

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Theo@3:770/3 to Chris Green on Wed Jan 6 18:02:02 2021
    Chris Green <cl@isbd.net> wrote:
    Yes, but you can't get an HDMI interface that doesn't implement HDCP
    can you? That's the whole point, it prevents (supposedly) you viewing
    things you shouldn't with your TV (or whatever) that has an HDMI
    connector.

    HDCP is optional - both sources and sinks can elect not to have it (the Pi
    has HDCP support in hardware but doesn't enable it, for example). Obviously
    if you're a monitor and don't implement it, you won't work with HDCP
    sources.

    Many cheap HDMI splitters strip off HDCP (there's no way to pass through
    HDCP communication with 3 concurrent sinks and it's cheaper just to strip
    it) so it can easily be bypassed.

    Theo

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From druck@3:770/3 to Martin Gregorie on Wed Jan 6 18:50:04 2021
    On 06/01/2021 17:04, Martin Gregorie wrote:
    On Wed, 06 Jan 2021 16:41:25 +0000, Theo wrote:


    Maybe there are use cases where people want to
    power their Pis from HDMI, I don't know.

    Doing that would make the installation tidier if I wanted to glue the Pi enclosure to the back of the screen - provided that the screen
    can provide the wattage needed. Is that likely, bearing in mind that, if
    I did that, I'd also want it to power a keyboard and mouse via the USB sockets on the Pi.

    HDMI will never power a Pi, way too thirsty.

    Similar devices such as an Amazon Firestick have to have an external
    PSU, as they require more power than even the TV's USB port can supply.
    Note that if this does seem to work, you are probably overstressing your
    TVs electronics, which will lead to premature failure.

    ---druck

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From TimS@3:770/3 to All on Wed Jan 6 18:58:38 2021
    On 06 Jan 2021 at 16:41:25 GMT, Theo <theom+news@chiark.greenend.org.uk>
    wrote:

    Chris Green <cl@isbd.net> wrote:
    Andy Burns <usenet@andyburns.uk> wrote:
    The Pi is the HDMI source, and the monitor is the HDMI sink, per the
    spec the 5V power should be supplied from the Pi to the monitor, not the >> > reverse ...

    I've heard it said that the HDMI spec is a real can of worms.

    This thread is just another thing that increases my dislike of HDMI.

    It's exactly the same with VGA.

    The purpose of this is for the computer to read the EDID ROM from the monitor. It supplies 5V to allow detection of the connected monitor even if it's turned off.

    The monitor/KVM should not be emitting power on this rail.

    I shutdown the Pi and when it had, turned off its power at the switch on the socket strip. This caused the green-LED fluttering I mentioned before. I then disconnected the HDMI cable from my Mac (the cable that goes to the other
    input of the KVM unit). The green and red LEDs on the Pi then went out.

    So it's my Mini that's doing this, but that is perhaps not surprising given what I've learnt in this thread.

    --
    Tim

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Joe@3:770/3 to Ahem A Rivet's Shot on Wed Jan 6 20:20:26 2021
    On Wed, 6 Jan 2021 12:47:44 +0000
    Ahem A Rivet's Shot <steveo@eircom.net> wrote:

    On Wed, 06 Jan 2021 13:13:42 +0100
    Axel Berger <Spam@Berger-Odenthal.De> wrote:


    Exactly. And there is another thing. Originally video had to be
    fast, for the eye needed at least 70 Hz not to see annnoying
    flicker. Those

    That number seems to have been creeping up all my life -
    24FPS was once considered good enough for movie cameras.


    For filming, but it was never displayed at that rate. In 50Hz countries, telecines and cinema projectors had a double shutter to display each
    frame twice (though no interpolation) and NTSC telecines repeated
    alternately 2/3 times to get it up to 60. Not sure if US cinemas did
    that, or put up with 48Hz. Modern telecines can interpolate for
    smoother motion, and play at any film speed.

    --
    Joe

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Martin Gregorie@3:770/3 to druck on Wed Jan 6 20:39:02 2021
    On Wed, 06 Jan 2021 18:50:04 +0000, druck wrote:

    On 06/01/2021 17:04, Martin Gregorie wrote:
    On Wed, 06 Jan 2021 16:41:25 +0000, Theo wrote:


    Maybe there are use cases where people want to power their Pis from
    HDMI, I don't know.

    Doing that would make the installation tidier if I wanted to glue the
    Pi enclosure to the back of the screen - provided that the screen can
    provide the wattage needed. Is that likely, bearing in mind that, if I
    did that, I'd also want it to power a keyboard and mouse via the USB
    sockets on the Pi.

    HDMI will never power a Pi, way too thirsty.

    Similar devices such as an Amazon Firestick have to have an external
    PSU, as they require more power than even the TV's USB port can supply.
    Note that if this does seem to work, you are probably overstressing your
    TVs electronics, which will lead to premature failure.

    Makes sense in light of the content of earlier posts: if the HDMI cable
    only carries power so it can wake up some HDMI chip in the, possibly powered-off, device at the far end which tells the initiator what it is
    talking to and what that can or cannot do, then it follows that doing
    this only needs a few milliamps.

    I thought that might be the case: thanks for confirming it.


    --
    --
    Martin | martin at
    Gregorie | gregorie dot org

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From druck@3:770/3 to Joe on Wed Jan 6 20:43:44 2021
    On 06/01/2021 20:20, Joe wrote:
    For filming, but it was never displayed at that rate. In 50Hz countries, telecines and cinema projectors had a double shutter to display each
    frame twice (though no interpolation)

    They actually played it 4% faster at 25fps and then doubled it to 50fps.

    and NTSC telecines repeated alternately 2/3 times to get it up to 60.

    Which gives the correct speed, but terrible tearing of movement when
    different film frames are used for each field of an interlaced TV frame. Luckily not something you see any more with progressive formats.

    ---druck

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Eli the Bearded@3:770/3 to steveo@eircom.net on Wed Jan 6 23:08:51 2021
    In comp.sys.raspberry-pi, Ahem A Rivet's Shot <steveo@eircom.net> wrote:
    Axel Berger <Spam@Berger-Odenthal.De> wrote:
    Exactly. And there is another thing. Originally video had to be fast,
    for the eye needed at least 70 Hz not to see annnoying flicker. Those
    That number seems to have been creeping up all my life - 24FPS was
    once considered good enough for movie cameras.

    Possibly, but probably not. Thomas Edison believed 48 FPS was minimum
    quality projection. Is that during your life? Film was recorded at 24
    FPS, yes, but projected so that each frame was shown two or three
    times, 48 to 72 FPS.

    At around 16 FPS the motion stops looking jerky, but there's still a
    visible flicker and some eyestrain. Faster playback rates (with
    duplicated frames) removes the flicker, but depending on the duplication
    method can introduce other artifacts. 24FPS source shown on a
    television, for example.

    Televisions used AC frequency as a reference, so preferred 50 FPS in
    50Hz parts of the world and 60 FPS in 60 Hz areas for projection and
    half that for recording. Stuff recorded at other frame rates get extra
    frame duplication to keep pace.

    25FPS source ABCDEFGHIJKLMNOPQRSTUVWXY
    yeilds by simple doubling:
    50FPS display AABBCCDDEEFFGGHHIIJJKKLLMMNNOOPPQQRRSSTTUUVVWWXXYY

    24FPS source ABCDEFGHIJKLMNOPQRSTUVWX
    yeilds by mostly doubling but also padding:
    50FPS display AAAABBCCDDEEFFGGHHIIJJKKLLMMNNOOPPQQRRSSTTUUVVWWXX

    Theoretically you could do something like:
    AAABBCCDDEEFFGGHHIIJJKKLLMMMNNOOPPQQRRSSTTUUVVWWXX
    but I think TV signal pipeline is designed around 25 FPS input.

    24FPS to 60FPS works similarly, but has more doubled doubles.

    As technology has improved, so have standards, though. Artifacts of
    a slower displays were previously tolerated, but now a effort is made
    to avoid those problems. Now you can more easily accomodate that AAABBCCDDEEFFGGHHIIJJKKLLMMMNNOOPPQQRRSSTTUUVVWWXX version, for example.

    Totally as an aside:

    One interesting trick I've seen done is mixing "frame" rates to make
    things look jerky. In the _Lego Movie_ the camera and backgrounds
    update at 24FPS, but the characters in the foreground update at 12FPS.
    It gives the film a much more hand animated look.

    I've got a slowed down sample from a since removed Youtube clip that I
    made in 2014:

    https://qaz.wtf/C/lego-sample.gif

    Note that the character spinning in the background on the left animates
    every frame, while the walking hero front and center animates every
    other frame.

    Elijah
    ------
    severely compressed the colors in the gif to get the file size down

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Axel Berger@3:770/3 to Ahem A Rivet's Shot on Thu Jan 7 09:08:17 2021
    Ahem A Rivet's Shot wrote:
    In more recent times 60Hz was said to be needed
    for flicker free and now it's 70Hz I'm seeing and high end TVs are currently advertising 120Hz.

    The way I remember it is, that one of the reason for white on black was
    the lower sensitivity of the eye for flicker. The first monitors with a
    white background were hailed as a huge improvement but they all -- both
    highly expesive ISA card monitor combos and the Atari -- needed 70 Hz to
    be acceptable. 60 Hz was considered borderline for colour when white backgrounds were avoided.


    --
    /¯\ No | Dipl.-Ing. F. Axel Berger Tel: +49/ 221/ 7771 8067
    \ / HTML | Roald-Amundsen-Straße 2a Fax: +49/ 221/ 7771 8069
     X in | D-50829 Köln-Ossendorf http://berger-odenthal.de
    / \ Mail | -- No unannounced, large, binary attachments, please! --

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Ahem A Rivet's Shot@3:770/3 to Eli the Bearded on Thu Jan 7 07:40:42 2021
    On Wed, 6 Jan 2021 23:08:51 +0000 (UTC)
    Eli the Bearded <*@eli.users.panix.com> wrote:

    In comp.sys.raspberry-pi, Ahem A Rivet's Shot <steveo@eircom.net> wrote:
    Axel Berger <Spam@Berger-Odenthal.De> wrote:
    Exactly. And there is another thing. Originally video had to be fast,
    for the eye needed at least 70 Hz not to see annnoying flicker. Those
    That number seems to have been creeping up all my life - 24FPS
    was once considered good enough for movie cameras.

    Possibly, but probably not. Thomas Edison believed 48 FPS was minimum
    quality projection. Is that during your life?

    No, but 50Hz interlaced monitors (stripped TVs really) are and
    removing the interlace was considered enough to make it flicker free and "professional" circa 1980 - I sometimes turned the interlace back on and
    pushed the resolution up. In more recent times 60Hz was said to be needed
    for flicker free and now it's 70Hz I'm seeing and high end TVs are currently advertising 120Hz.

    Film was recorded at 24
    FPS, yes, but projected so that each frame was shown two or three
    times, 48 to 72 FPS.

    Weren't domestic cine projectors 24 fps and considered good enough
    for the use ?

    --
    Steve O'Hara-Smith | Directable Mirror Arrays C:\>WIN | A better way to focus the sun
    The computer obeys and wins. | licences available see
    You lose and Bill collects. | http://www.sohara.org/

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From The Natural Philosopher@3:770/3 to Ahem A Rivet's Shot on Thu Jan 7 08:16:10 2021
    On 07/01/2021 07:40, Ahem A Rivet's Shot wrote:
    On Wed, 6 Jan 2021 23:08:51 +0000 (UTC)
    Eli the Bearded <*@eli.users.panix.com> wrote:

    In comp.sys.raspberry-pi, Ahem A Rivet's Shot <steveo@eircom.net> wrote: >>> Axel Berger <Spam@Berger-Odenthal.De> wrote:
    Exactly. And there is another thing. Originally video had to be fast,
    for the eye needed at least 70 Hz not to see annnoying flicker. Those
    That number seems to have been creeping up all my life - 24FPS
    was once considered good enough for movie cameras.

    Possibly, but probably not. Thomas Edison believed 48 FPS was minimum
    quality projection. Is that during your life?

    No, but 50Hz interlaced monitors (stripped TVs really) are and
    removing the interlace was considered enough to make it flicker free and "professional" circa 1980 - I sometimes turned the interlace back on and pushed the resolution up. In more recent times 60Hz was said to be needed
    for flicker free and now it's 70Hz I'm seeing and high end TVs are currently advertising 120Hz.

    Film was recorded at 24
    FPS, yes, but projected so that each frame was shown two or three
    times, 48 to 72 FPS.

    Weren't domestic cine projectors 24 fps and considered good enough
    for the use ?

    16fps IIRC?


    --
    Ideas are more powerful than guns. We would not let our enemies have
    guns, why should we let them have ideas?

    Josef Stalin

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From John Aldridge@3:770/3 to All on Thu Jan 7 09:20:59 2021
    In article <5FF6C171.23E2CA59@Berger-Odenthal.De>, Spam@Berger-
    Odenthal.De says...

    Ahem A Rivet's Shot wrote:
    In more recent times 60Hz was said to be needed
    for flicker free and now it's 70Hz I'm seeing and high end TVs are currently
    advertising 120Hz.

    The way I remember it is, that one of the reason for white on black was
    the lower sensitivity of the eye for flicker. The first monitors with a
    white background were hailed as a huge improvement but they all -- both highly expesive ISA card monitor combos and the Atari -- needed 70 Hz to
    be acceptable. 60 Hz was considered borderline for colour when white backgrounds were avoided.

    It was interestingly variable from person to person. Some of my
    colleagues worked happily on monitors set at 60 Hz, but anything less
    than 70, preferably 75 Hz, made me feel sick.

    John

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From druck@3:770/3 to Ahem A Rivet's Shot on Thu Jan 7 09:53:53 2021
    On 07/01/2021 07:40, Ahem A Rivet's Shot wrote:
    No, but 50Hz interlaced monitors (stripped TVs really) are and
    removing the interlace was considered enough to make it flicker free and "professional" circa 1980 - I sometimes turned the interlace back on and pushed the resolution up. In more recent times 60Hz was said to be needed
    for flicker free and now it's 70Hz I'm seeing and high end TVs are currently advertising 120Hz.

    Several different issues there.

    Old CRT TVs worked with an interlaced signal so needed long persistence phosphors so the each field of alternating scan lines would still be
    visible when the next field was displayed. UK PAL TVs were set up for
    50Hz fields so had a longer persistence than US NTSC at 60Hz.

    CRT monitors (generally) worked with non interlaced signals, so had low persistence phosphors as all the scan lines would be updated every
    frame. Monitors were invariably designed for the US market so had
    phosphor persistence for a minimum of 60Hz frame rate, and would look
    terrible if driven at 50Hz.

    As monitor resolutions and sizes increased the frame rates also
    increased to 70Hz or 75Hz, to reduce flicker and smearing of screen updates.

    Now with LCD, LCD/LED and OLED technologies the screen does not flicker
    at the update rate. Panels used to have a 60Hz update rate to match US
    content (but would be fine for 50Hz), but active 3D drove the adoption
    of 120Hz or greater panels, so alternate views could be shown at 60Hz.
    3D has largely been dropped now, but 120Hz or greater panels still
    exist, so they can either be driven at a higher frame rate by games
    consoles, or additional intermediate frames are generated for lower
    refresh rate content, in both cases to make animation smoother.

    ---druck

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Martin Gregorie@3:770/3 to druck on Thu Jan 7 12:40:55 2021
    On Thu, 07 Jan 2021 09:53:53 +0000, druck wrote:

    Now with LCD, LCD/LED and OLED technologies the screen does not flicker
    at the update rate. Panels used to have a 60Hz update rate to match US content (but would be fine for 50Hz), but active 3D drove the adoption
    of 120Hz or greater panels, so alternate views could be shown at 60Hz.
    3D has largely been dropped now, but 120Hz or greater panels still
    exist, so they can either be driven at a higher frame rate by games
    consoles, or additional intermediate frames are generated for lower
    refresh rate content, in both cases to make animation smoother.

    Modern phones seem to have refresh rates of anywhere between 60Hz and
    120Hz - apparently the gaming crowd want it.


    --
    --
    Martin | martin at
    Gregorie | gregorie dot org

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Charlie Gibbs@3:770/3 to Ahem A Rivet's Shot on Thu Jan 7 19:29:02 2021
    On 2021-01-07, Ahem A Rivet's Shot <steveo@eircom.net> wrote:

    Weren't domestic cine projectors 24 fps and considered good enough
    for the use ?

    Yes, but panning across a scene could be a bit jarring
    when viewed on the big screen.

    --
    /~\ Charlie Gibbs | "Some of you may die,
    \ / <cgibbs@kltpzyxm.invalid> | but it's a sacrifice
    X I'm really at ac.dekanfrus | I'm willing to make."
    / \ if you read it the right way. | -- Lord Farquaad (Shrek)

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From David Higton@3:770/3 to Ahem A Rivet's Shot on Thu Jan 7 21:05:21 2021
    In message <20210107074042.df7dcb8f3d2a551a78cf1a64@eircom.net>
    Ahem A Rivet's Shot <steveo@eircom.net> wrote:

    Weren't domestic cine projectors 24 fps and considered good enough
    for the use ?

    I remembered 18 fps, with a 3-bladed shutter in the projector, giving
    54 flashes per second.

    I see someone else said 16 fps, so I turned to good old Wikipedia, which suggests that Super 8 was 18 fps and Standard 8 was 16 fps. So we
    were both right!

    David

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Dennis Lee Bieber@3:770/3 to All on Fri Jan 8 22:33:39 2021
    On 5 Jan 2021 22:54:36 GMT, Tim Streater <timstreater@greenbee.net>
    declaimed the following:


    The Pi carried on being powered. After dismissing the possibility of sentient >self-powered computers taking over, I discovered that the volts were coming >via the HDMI port.

    https://en.wikipedia.org/wiki/HDMI look at the pin-out... Pin 18 is intended to provide power for "active cables"
    https://www.hdmi.org/spec/hdmi2_1 https://www.hdmi.org/spec21sub/cablepower


    --
    Wulfraed Dennis Lee Bieber AF6VN
    wlfraed@ix.netcom.com http://wlfraed.microdiversity.freeddns.org/

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Anssi Saari@3:770/3 to Dennis Lee Bieber on Sat Jan 9 12:03:51 2021
    Dennis Lee Bieber <wlfraed@ix.netcom.com> writes:

    https://en.wikipedia.org/wiki/HDMI look at the pin-out... Pin 18 is intended to provide power for "active cables" https://www.hdmi.org/spec/hdmi2_1 https://www.hdmi.org/spec21sub/cablepower

    Cable power yes, but it's also very convenient that an HDMI port can
    power an HDMI to VGA converter. There are so many old projectors in so
    many meeting rooms with VGA cables even today. The minimum current to
    provide through pin 18 is only 55 mA though. Not enough to run a Pi.

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)