back to article Are Samsung TVs doing a Volkswagen in energy tests? Koreans hit back

Samsung has responded quickly to claims that its televisions were designed to cheat in official power consumption tests. Independent European testing lab ComplianTV has claimed that some of Samsung's televisions degrade their performance and lower the brightness of the screen during standard IEC testing, and then return to …

  1. Your alien overlord - fear me

    Pretty obvious answer - don't have just one film to use as a test bed. Maybe get a Netflix account or something?

    1. SuccessCase

      Unsurprising. This *is* Samsung after all

      This after all is Samsung, the same company as has already gamed benchmark applications. In the past in their handsets, they have implemented the equivalent to "defeat software" which ensures their chips speed up and don't use any low power modes when any benchmarking apps are run by users. This is effectively cheating. There are, I think, even greater parallels between what Samsung did then and what the car manufacturers are doing now (though clearly with a lower potential cost implication for the customer).

      On a common sense view what they did is clearly, wrong, but they were able to, in PR terms "control the narrative" better than the car manufacturers have and put out a tenuous argument that it was justified on the basis that they were revealing the actual unbridled performance capability of their chips - even if you would never see it in the real world, and they didn't say upfront they were unlevelling the playing field in a way they hoped no-one would notice and they hoped would be permanent and for all time. It seems to me the car manufacturers might have been able to claim a similar argument as Samsung claimed (it would be very interesting to know if the emissions standards/tests are worded in such a way as to avoid a get out on this form of technicality). Where Samsung were "kind of" able to stop the narrative running away from them (e.g, get out a semi rational counter view before the story had spread). BMW clearly haven't been able to do the same. The story was already framed as cheating beyond any reprieve before they were able to address it.

      Additionally Samsung have also previously sold flagship devices advertised as 8 core CPU's, which though they contained 8 cores, 4 of the cores were permanently disabled (not just due to lack of software support but actually in terms of the hardware design).

      Samsung are unconscionable! Always have been.

      1. StooMonster

        Re: Unsurprising. This *is* Samsung after all

        to;dr = Samsung caught cheating benchmarks before, they've got form thus no surprise.

        I would bet that it's not just their televisions that act differently when being tested for power consumption.

        Imagine if Samsung made cars!

        1. Ambivalous Crowboard

          Re: Imagine if Samsung made cars

          Ugh, that'd be awful! They'd only let you listen to the radio stations which they installed in your car (and there'd be THOUSANDS of them but they're ALL crap), and then one day you'd come out to your car to find that it'd had an update overnight, and now car doesn't work on certain roads...

      2. Flocke Kroes Silver badge

        Re: Unsurprising. This is *monitors* after all

        Monitor specs from all manufacturers have been dishonest for decades. CRT monitors were credited with a higher resolution than the number phosphor dots. Repeating the contrast ratio measurements required a lightless cave painted dark black. High scan rates were possible, but the pixels became blurred because of lack of bandwidth. These days its 4K resolution - with a 15Hz frame rate (interleaved).

        As soon as a particular specification influences purchasing decisions, every manufacturer hunts for ways to make the number better - whether or not it reduces the quality of the product. The solution was to go to PC World to select a monitor, then buy it somewhere that did not charge £30 for the cable.

        Wait a couple of weeks, and you will see what naughty things the other manufactures have done to get their high scores.

        1. toughluck

          Re: Unsurprising. This is *monitors* after all

          @Flocke: That's what contrast ratio is about. Why should anybody care about a test in bright sunlight? Oh, this screen has 1.5:1 contrast, and that one has 2:1 contrast, so the other one is better (conveniently ignoring that the second set has a glossy screen and gets less diffuse reflection, hence can get better contrast, it will get specular reflections unless perfectly placed). What you need to be looking at is peak luminance (in cd/m² = nit), then you can work out the minimum luminance from the contrast, and decide if that's going to be enough for your usage or not.

          As for 4K/UHD, you're way off base. The UHDTV standard specifies progressive framerates at 24, 25, 50, 60 and 120, the latter three can be broadcast interlaced. So at the very worst, you can get 25 Hz frame rate, never 15 Hz (there's no broadcasting standard that allows 30 fields per second).

          If you're using a HDMI 1.4 compliant cable, then yes, you'll get a lower refresh rate, but it's going to be 2160f30 (not exactly progressive, but the signal will be de-interlaced). But it's only your fault for getting the wrong cable.

          If you don't know how to interpret the results, you should ask somebody to help you pick the set.

          1. Justin Pasher

            Re: Unsurprising. This is *monitors* after all

            @toughluck: Everything sounded great until you said "If you're using a HDMI 1.4 compliant cable". There's no such thing as an "HDMI 1.4 compliant cable". That's a marketing thing (just like contrast ratio). There are only four types of HDMI cables.

            Standard

            Standard with Ethernet

            High Speed

            High Speed with Ethernet

            HDMI 1.4 is a software specification, not a hardware specification. A cable knows nothing about software, because, well, it's hardware. It's like saying an ethernet cable is "IPv6 compliant"

          2. DougS Silver badge

            @toughluck

            Where are you getting your information?

            Everything I've read about 4K says that interlacing is GONE, and GOOD RIDDANCE to that. It also must support 30 fps, as that's used for a lot of US broadcast TV - many dramas here are produced in 30 fps, and either frame doubled when broadcast on a 720p (60 fps) network like ABC, or interlaced when shown on a 1080i network like CBS.

      3. h4rm0ny

        Re: Unsurprising. This *is* Samsung after all

        I was coming here to post something similar about Samsung. This is the company that has literally tried to hand judges a briefcase full of cash, movie-style. This is kind of a re-post but last time I shared this story it was over a year ago, so here's an anecdote about Samsung.

        In 2006 Samsung was sued by Pioneer for infringing their patents on Plasma TV technology. A memo from a Samsung engineer used as evidence showed that they knowingly infringed on the patents. Rather than agree a licencing fee however, Samsung counter-sued and buried Pioneer under suits and appeals. Pioneer was awarded $59million in damages, but got buried in punitive legal actions from Samsung and a few years later shut down the television division, in large part because of this. Ten-thousand people who worked in that division directly or indirectly, lost their jobs

        They're a fun company.

        1. Evil Graham

          Re: Unsurprising. This *is* Samsung after all

          @h4rmony: Rather than agree a licencing fee however, Samsung counter-sued and buried Pioneer under suits and appeals.

          Really I am struggling to think of a large corporation that doesn't behave like that.

      4. itzman

        Re: Unsurprising. This *is* Samsung after all

        I remember the days of 6 transistor, 7 transistor, and yea, even 8 transistor radios.

        One of them being a RF demodulation diode, and another simply soldered to the board, but left unconnected to the circuitry.

      5. Anonymous Coward
        Anonymous Coward

        Re: Unsurprising. This *is* Samsung after all

        If you think a Samsung TV doing what it's supposed to do (ie. reduce brightness under certain conditions) is the same as VW installing a defeat device to cheat the testing procedure, then you basically don't understand what you're talking about.

        We all want TVs which give a nice picture whilst reducing power consumption, but when Samsung manage to achieve this and then the system does what IT IS INTENDED TO DO in real life, somehow that's cheating. Very strange.

  2. John Tserkezis

    "but not the spirit of the law, alleged Rudolf Heinz"

    Boo fucking hoo. I hear this all the time. The "law" observers ONLY the law, cry baby losers bring up the "spirit" which is ENTIRELY non-definable, not-recognised, and very much open to personal interpretation.

    Heinz, if you don't like how the law works, CHANGE IT, and quit your whining.

    That said, I specifically disabled this Motion Lighting bullshit when I first plugged in the TV and set it up. I'm sensitive to brightness changes where there should not be. If you don't want your six-thousand inch TV to consume so much power, buy something smaller.

    See what you've done? I'm ranting again. And I've run out of icecream too, which would make me feel better.

    1. Graham Marsden
      Thumb Down

      > the "spirit" which is ENTIRELY non-definable, not-recognised, and very much open to personal interpretation.

      I'm glad I'll never play board games against a rules lawyer like you...

      1. Mark 85 Silver badge

        Actually he does have a point. Try suing someone for not following the "spirit of the law" and see where that gets you.

        1. dkjd

          If they describe their tv as more economical than their rivals, and it is not, then they can presumably be sued under trades description act. If they have been gaming the test it will get you about £100 per set I would guess

      2. John Tserkezis

        "I'm glad I'll never play board games against a rules lawyer like you..."

        You won't have to, I'm not invited to play board games anymore anyway. And yes, after what I did to Monopoly, along withing bringing the statistical printouts to every game - people get sick of that. But I did quite well while I was still playing.

        However, playing by the rules is what the law is all about, ever had to deal with insurance companies? They could replace their entire course and fine print with "You're screwed", only reason they don't, is their fine print isn't readable or understood by mere mortal humans, which lets their marketing people sweet talk you with the "spirit" of their organisation.

        If you think they're going to stick to their "spirit", good luck, is all I have to say. You're going to need it.

        1. h4rm0ny

          I don't think anyone is disputing that prosecutions follow the law as written, they're disputing the "Boo hoo!" part. It's little different to any other complex system such as my computer OS. If someone finds an exploit, well that needs patching, but I don't suddenly express contempt for the victim of a hack and exalt the hacker for finding a way to get access without valid credentials.

          Samsung may have found a loophole way around the intent of the law, that is worthy of criticism.

          1. nijam

            > ... around the intent of the law

            Also irrelevant - the law is as written, and is nothing to do with what anyone says the "spirit" or the "intent" is.

            1. John Crisp

              >Also irrelevant - the law is as >written, and is nothing to do >with what anyone says the >"spirit" or the "intent" is.

              Not quite right I think.

              The law is as written indeed, but often wording can be ambiguous and the judges (in the higher courts) are there to decide what the intent was as you may note when cases are appealed on a 'point of law'.

              Plenty of references to this online.

              Spirits can go whistle obviously !

    2. P. Lee Silver badge

      > I specifically disabled this Motion Lighting bullshit when I first plugged in the TV and set it up.

      And herein lies the problem. After all, when you plug in a TV its normally "off" be default. That's quite energy efficient. Just because nearly everyone changes the setting to "on" is surely is no reason to have to measure that particular option...

      Maybe just the worst-case should be tested, rather than defaults?

  3. Henry Wertz 1 Gold badge

    This really depends on what's happening

    This really depends on what's happening.

    If, as Samsung says, this dimming thing really does kick in under real-world usage and save power, then I think it's legitimate. Perhaps the testing should be done with options like this on then once with them off (making sure brightness and contrast are adjusted, so the vendor doesn't just use inappropriately dim defaults), so you get kind of a typical and (somewhat) worst case figure.

    If there are sets that actually detect the IEC clip and start power saving right away, that's cheating and I assume they'll get fines and possible lawsuits. I do wonder how many would sue though, I couldn't get that worked up over a dollar or two a year on my power bill. But (in the US) if the extra power use pushes some monitor that claimed to be energy star into not meeting energy star standards then that company could have problems with the EPA.

    1. wolfetone Silver badge

      Re: This really depends on what's happening

      I bought a Samsung 46" LED TV a year or two ago, and there is an option for Eco mode (something like that) and it displays a power meter. If you have Eco mode on then the screen does dim and adjusts it's brightness according to the light in the room. Often if I watch the TV without the lights on it's quite dim, but that's only noticeable when you whack the big light on and it takes a few seconds to brighten up a bit.

      Still, who's bothered about a TV using a bit more juice when we're driving cars that could be doing much worse?*

      * I absolutely refuse to believe VW are the only ones doing this

      1. toughluck

        Re: This really depends on what's happening

        CAFEE tested two Volkswagens (a Jetta and a Passat) alongside a diesel BMW X5, and only the Volkswagens exceeded their emissions (and by a very considerable margin).

        And it's about emissions, not about fuel consumption. I thought it was a proven point that cars consume more fuel than the catalog says, wasn't it?

        --

        As for the TV using too much power -- I doubt that Samsung TVs detect the exact video that's being played. If they want to prove that Samsung gamed the test by detecting it after all, they should simply alter the video slightly (e.g., swap red and blue channels), assuming it's a normal RGB LED. If the consumption is markedly different, they have their answer.

      2. King Jack

        Re: This really depends on what's happening

        @wolftone I was with you until you used the childlike argument that X is ok because Y is worse.

  4. Adam T

    Samsung, cheat??

    Naahhh say it ain't so! They're as honest as the day is long.

    http://www.anandtech.com/show/7187/looking-at-cpugpu-benchmark-optimizations-galaxy-s-4

  5. Anonymous Coward
    Anonymous Coward

    Big business cheating and misleading the customer, surely not ?

    They always put the profits customer first.

  6. Ken Hagan Gold badge

    Thumbs up for "doing a Volkswagen"

    Excellent verbal boffinry there. I look forward to seeing it on a regular basis when I go vulture spotting with my mobe.

  7. Anonymous Coward
    Anonymous Coward

    Phones

    A certain popular (in 2011) SMG smartphone had known overheating issues yet nothing was done.

    I had the misfortune to run into this and sure it would work fine under "normal" ie nothing else running usage conditions but would reliably crash when warm or hot running some games.

    I managed to duplicate this behavior by getting CPU usage up to 70% and presumably lacking a fan some race condition happened that caused it to freeze and require Vulcan Nerve Pinch.

    I still wonder to this day if they fixed it in the later firmware update simply by detecting > 70% usage and dumping low priority threads to the least reliable core, aka load shedding or some other such method, in fact a lot of newer chips do this by design when running near their thermal limit.

    Needless to say, the new product from the same company also has thermal issues but it does seem to self regulate so maybe they fixed it?

  8. MJI Silver badge

    Unrealistic in the real world

    As soon as you buy a TV you sit down and set the thing up, you crank down the eye melting brightness for a start.

    Since power consumption appears to be connected to brightness on panels, perhaps they should test after setting up rather than out of box.

    1. Dan 55 Silver badge

      Re: Unrealistic in the real world

      They're all set bright for store display, a tested TV won't have the brightness set like that. Hopefully.

  9. Anonymous Coward
    Anonymous Coward

    Samsung has responded quickly

    funny, how it's a "big" thing, when cheating on mob tests never surfaced in mainstream media. It must be something in the air. Next thing, they'll discover that "up to" and "from" are shockingly far from the figures that stand next to them...

    1. Anonymous Coward
      Anonymous Coward

      Re: Samsung has responded quickly

      Next thing, they'll discover that "up to" and "from" are shockingly far from the figures that stand next to them...

      And that will be the end of renewable energy...and climate change, the movie...

      ...I'll get my coat

  10. TRT Silver badge

    Oh yes, motion lighting.

    I have a Samsung TV. That's the setting that enforces you to RTFM and scramble through the menus to turn it off because it makes the picture look shit on about a quarter of all films.

  11. Anonymous Coward
    Anonymous Coward

    A big part of me hope these companies have the book thrown at them and we get some tighter emissions laws / energy standards etc etc but another part of me has a little bit of sympathy. The tests are in a lab type environment and it's well known what the test will be. Can you imagine this situation in an exam: now students, these are the questions you'll be asked but you aren't allowed to do any special preparation for them. The whole testing system is flawed top to bottom.

    With cars I think the best approach would be to fit 10 (randomly selected ideally) cars with monitoring equipment for a month. I'm sure you could find enough takers to have the box of tricks attached to the car for a small financial reward.

  12. Efros

    Samsung Plasma here

    Otherwise known as living room auxiliary heater.

  13. Tromos
    Joke

    My broadband provider has a Samsung TV

    He said it saves him up to 99% of his electricity bill.

  14. Rob Crawford

    I would suggest that you have to use media that covers a large number of scenarios if you wish to test the energy use of a TV or media player.

    The FPGA on board will be working much harder when there is a lot of movement and changes between frames, no doubt the amount of value switching on the pixels will also be reflected in power usage.

    So you really need a selection of media types to reflect average usage.

    As for car emissions testing, everybody has known for at least 30 years they bear no relationship to reality, how can a rolling road reflect real world usage with varying surfaces, surface water, humidity, number of people in the car and temperature, especially when all of those vary during a single journey.

    All it is possible to do is to have fairly standard tracks, test several times and average the results,otherwise it's an entirely synthetic test that reflects nothing whatsoever.

    But most people know that anyway, but choose to ignore it

  15. Yugguy

    Does anyone ever actually ever use all these options ever?

    I've got a Samsung 40inch smart tv.

    It has like 9 million diferent options for the display, and perhaps 4 million for sound.

    I've never changed ANY of them since I put it on what I like when I first got it 2 years ago.

    Actually I tell a lie, I sometimes use the Movie sound setting for, er, movies.

    Does anyone actually change their settings a lot for different programmes?

  16. Marcus Fil

    The Simpler Solution

    Instead of berating a test mode mandate one - a mode the turns all the dials up to 11 to turn the set into a black body radiator on scale close to a Red dwarf. The equivalent of driving your car, windows down, heating up and every electrical accessorty, A/C etc. on full blast. stuck and left it in 2nd gear. Introduce huge fines, manadatory life sentences, genocide etc. for the manufacturer if anyone ever discovers a mode/feature combination worse than test mode. The challenge becomes who can actually make the best set that performs well in the worst case - because by definition it will always perform better in the real world.

  17. TRT Silver badge

    Can I get

    a diesel powered TV?

    1. Oldfogey

      Re: Can I get

      Easy. Just hook it up to a diesel generator.

  18. Anonymous Coward
    Anonymous Coward

    In the olden dayz

    TVs used leaded glass, lead solder, arsenic, asbestos (power resistors, heatsinks, etc), beryllium, mercury and cadmium in the tube phosphors.

    The best example of killer sets was the one where the 4 legged capacitors installed as a "safety feature" went bad and resulted in a very bright very small picture and copious X-rays.

    These days they just use extra arsenic and cadmium in the RGB LEDs to get the brightness up, and toxic solvents that make the worker's skin fall off and rot their brains because the next best solvent costs 6* as much.

    That and the toxins present in the LCD manufacture which include fun compounds like silane (SiH2), HF (to etch the glass), tetrahydrofuran, stannous chloride etc.

    Some of the LCD chemicals have been linked to birth defects that would make benzene run off and hide in a corner.

  19. John Brown (no body) Silver badge

    especially when manufacturers run the examinations themselves

    ...well, no shit Sherlock (a dig at govt, not El Reg)

    If a government wants to mandate regulations then it's surely up to them to test for and enforce them.

    Self certification only works in the perfect world of Julia Cordray of new slander-app Peeple fame where unicorns are pink and there are double rainbows in the sky all the time.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2019