The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.

    brightness issues with replacement panel and some calibration questions

    Discussion in 'Hardware Components and Aftermarket Upgrades' started by equalizer2000, Nov 3, 2018.

  1. equalizer2000

    equalizer2000 Notebook Consultant

    Reputations:
    25
    Messages:
    282
    Likes Received:
    71
    Trophy Points:
    41
    Hello, hoping someone can give some advice to point me in the right direction. Here is a synopsis. I tried to make it brief, but no way that was happening with this saga!

    I have a new Alienware 17 R5. I ordered it with the QHD screen, which is about 73% Adobe RGB, 400 nits. I wasn't sure if I wanted QHD or UHD (UHD is 4k of course, better color, but only 60Hz vs 120Hz of the QHD - this thread is NOT about which to pick!), and since the UHD cost as much additional as it appeared it would cost for an aftermarket UHD screen, I figured, why not start QHD, and upgrade later.

    I then proceeded to order five different UHD screens from eBay. (not all at once - various issues; two sellers cancelled, not sure if they were really suddenly out of stock, or if my pointed questions that it better be original AUO or else, made them decide against sending a knockoff) Of the three screens I received:

    1. First one was a ChiMei Innolux. Came in a nice box, looked new. Had a lot of backlight bleed, and several dead pixels. Also, since it reported as an Innolux in its EDID, G-Sync won't work. (only the AUO B173ZAN01.0 is white-listed for G-Sync) Color accuracy was good though, 99% Adobe RGB, 100% sRGB. Importantly, it was BRIGHT - I was very surprised, I was sure that the QHD was supposed to be brighter, but on this panel, I had to turn it down 2-3 notches in a darkened room.

    While I used my SpyderColor5 Pro to calibrate it, I unfortunately didn't take note of the nits brightness the calibration gives when starting. I THINK it was low 300s though. (which I thought odd, since it's supposed to be 400 - but I don't remember)

    2. Second one was from China direct - smelled weird, but no dead pixels, G-Sync worked. however, screen said "No Warranty" where it should have said AU Optronics. HWiNFO showed different manufacture date from sticker (year earlier). According to YouTube video I found, AUO panels with stickers that don't say AUO, but have Limited or No Warranty, are fakes. However, it was 4k, and also 99% Adobe RGB! The real issue was, it was very, very dim.

    My Spyder confirmed this, max brightness registered only 170 nits.

    3. Third screen - was sent as a replacement for my Innolux from the first place. Unfortunately, it was same sticker as #2, "No Warranty"; date of manufacture off by a year. But, no dead pixels, 100% Adobe RGB. Yet, seemed a bit dim! Not unusable like #2, but barely usable.

    The Spyder said this one was 280 nits. So, I thought, maybe it's just because the colors are more accurate.

    NEXT - I made some odd discoveries. This whole time, I had thought my Lenovo Thinkpad Yoga 14 was bright enough (but always wished it had one more notch). From reviews at that time, the screen is mediocre, about 267 nits. Which made sense, since screen #2 was reporting to be 170 nits. That seemed about right. However, I finally use my Spyder on the Lenovo (only got the Spyder recently), and IT measured at max brightness of 220 nits! How could this be, I said... the new screen I bought seemed barely usable, yet it is reporting 280 nits vs 220! I then did a calibration on the Lenovo, which made it appear dimmer (as calibration always does). Now it appeared closer - but it STILL looked to me like the Lenovo was more legible, and, if not still a bit brighter, at least EQUAL brightness. (but how - it is 220 nits, 60 nits less than the UHD screen!)

    Then I did one more test - I tested my original QHD screen for brightness. That has a factory rating of 400 nits. Yet the Spyder reported it at 240 nits! What the heck! How can that screen, which looks brightest of all (except equaled by the Innolux in #1), report that it is DIMMER by a bit than this new, dim-looking screen?

    So that wasn't too short, lol. I had to relay all that to get to my questions -

    1. Is it possible that a screen with more nits in its colorimeter reading actually looks darker, if its color accuracy is higher? I.e., the colorimeter brightness reading isn't how the human eye judges brightness?

    2. Is it possible the Spyder is a POS? I've read reviews going both ways. Some seem to say only get the xRite1 Pro. I WAS going to get that, but I'm not a super serious user, and the SPyder Pro is almost half the price of the xRite. I have an xRite on order now though, and will return the Spyder. But in general, am I putting to much faith in the Spyder? (or in any colorimeter, period?) After all, how come even the known good QHD screen, which LOOKS BRIGHT, and all reviews say is 400 nits, only measured 240, right?

    3. I haven't had time to try, but will more advanced software, like DisplayCAL, let me calibrate, AND also manually pump up the gamma and/or brightness for general use? I ask this, because, I've spent $100 on shipping now on these three screens. I don't want to keep trying more, it's a waste of time at this point. This screen is barely usable, BUT I found that if I don't use my calibration, and manually punch up the gamma to 1.5 from 1.0 in the nVidia control panel (this laptop is 100% on discrete all the time, so no need to worry about the iGPU settings), THEN it does look bright! But I assume I am losing contrast, and it's obviously not going to be calibrated. (although it looks decent) Or should I just return the screen? The Spyder5Pro software just punches out a calibration, no real adjustment can be made. I am assuming DisplayCal or the iRite Pro's software DOES allow for such adjustments though? (I assume that is how the display cals for sale online which have multiple versions for blue blocker, daytime, gaming, etc., but are all color calibrated for a particular monitor, can be made?)

    That is my most important question, because that means I can make this panel work. It's just very weird, because why is it that, at stock settings, 100% brightness is nowhere near where it should be (which explains low backlight bleed), yet the video driver can force it to be so bright? Obviously the backlight is not deficient, right. Is it possible the color LUT in the panel is just messed up or something from the factory, hence the low brightness? I.e., it just needs an extreme calibration (one that will pump gamma and/or brightness, which my Spyder software won't do), and isn't faulty in and of itself?

    Again, I really wonder what my Innolux had reported as its brightness. I do think it was 320 or 330 though, because I remember thinking, hmm, that's weird, shouldn't it be 400... but I wasn't too worried, because it was so bright anyway.

    And just to make the point again - I am worried about this panel because my #2 was the same sticker, and was VERY dim. I never tried manually correcting via the nVidia control panel though. But man, that was so dim, it was like the 20 or 30% setting on a regular functioning screen. #3 feels like it is 70%, maybe 80% if I'm being generous, of a regular screen. (I'd really say more like 60-70% though.)

    Also, I know for color work, you're supposed to operate at 100-120 nits anyway. But that's just not enough for me to do general computing work. This #3 screen is dim enough where I have to lean in to make out the words to read some things. Just not good.

    It looks good with my manual fix, I just know it will take time before I ever get to dive into display cal. So I am ok keeping it if that means there is a way down the road to pump the brightness AND keep it calibrated - but I am not familiar enough with all the calibration tools available yet to do so, and I don't want to tell the seller I am ok even though I already complained. Thus I am hoping for some good advice on whether that is something I can expect to do when I have time to try it out. (and also any feedback on why the Spyder measurements of nits don't seem to match what looks brighter to my eyes)

    Thank you to anyone who made it this far!

    Edit: all laptops screens were measured in the same room, same lighting conditions, no reflected light, just in case that wasn't clear.
     
    Last edited: Nov 3, 2018
  2. Starlight5

    Starlight5 Yes, I'm a cat. What else is there to say, really?

    Reputations:
    826
    Messages:
    3,230
    Likes Received:
    1,643
    Trophy Points:
    231
    Last edited: Nov 4, 2018
  3. equalizer2000

    equalizer2000 Notebook Consultant

    Reputations:
    25
    Messages:
    282
    Likes Received:
    71
    Trophy Points:
    41
    How do you disable it? I didn't even know these machines had one! The brightness always remains fixed to what I set it to (it only dims when Tobii detects that I look away for 10 seconds or more - but I disable Tobii and display sleep when I do colorimeter tests, since they can take upwards of an hour). When I tested with the colorimeter, I know it was at 100%. Is this some secret setting for the panel I have to change, or a Windows thing?
     
  4. Starlight5

    Starlight5 Yes, I'm a cat. What else is there to say, really?

    Reputations:
    826
    Messages:
    3,230
    Likes Received:
    1,643
    Trophy Points:
    231
    @equalizer2000 it's a windows thing, in power options. It may explain brightness variation in repeated tests.
     
  5. equalizer2000

    equalizer2000 Notebook Consultant

    Reputations:
    25
    Messages:
    282
    Likes Received:
    71
    Trophy Points:
    41
    Ok, the dimmed brightness setting in Power options - no, I keep that set at 100% as well. Again, the panel is just very dim even at 100% brightness... I have seen this before, even worse, with the previous one I tried (that one must have maxed out at 100 nits). The panel is definitely defective - I don't know why Spyder5Pro software gave consistent yet inaccurate brightness measurements, but I just tested all my machines with the same puck but the DisplayCAL software, and I got readings that seemed in line finally - my known good QHD display is rated at 400 nits, and tested at 383 nits; my Lenovo known to be 257 nits tested at 265 nits; and this new defective panel, which is supposed to be 350 or 400 nits, tested at 190 nits. (so basically, it's only half the brightness - they are probably selling it as a reject hoping a buyer doesn't notice or care?) The 190 nits makes sense, because only if I use it in a blacked out room with no lights on does it even seem close to normal brightness.

    Anyway I am probably just going to return it, and try another, this time just go to the source on AliExpress. I thought eBay might be a higher standard, but after my experience (two faulty ones from a U.S. seller, one faulty from China, and two that cancelled their orders on me when I asked specifically if it was genuine AUO and not refurbished.

    I also see on AliExpress you can buy an HDR 4k portable monitor, only 13mm thick! 17.3"... Kind of expensive at $520 but cheaper than an OEM 4k panel from Dell (they wanted $650 - ridiculous), and then I could use it as a second monitor, or just prop it on the laptop in front of the built-in screen (leave the built-in as the QHD 120hz Dell screen)... and get the best of both worlds that way... an even BETTER 4K screen (HDR), and use the built-in QHD for gaming.... maybe. :)