November 29, 2017 at 11:06 pm #45606
I believe that is the case.January 27, 2018 at 4:38 am #46568
PCM2: Check the thread “Samsung C32HG70 first impressions” started by lukiev3. You have totally co-opted every one of the original 4 pages of that thread, all of which now display a news release about the 27″ and 32″ versions of the HG70 series monitors. The thread itself appears to be lost. I have been trying desperately to find the thread in which I asked you about the presence or absence of PWM in the Samsung C32H711 monitor. I finally got through to someone at Samsung tech support on a chat line and this person confirmed that it does NOT have PWM. I wondered because they advertise it as “flicker-free,” but they also say the same thing about the C32HG70, which you said showed evidence of PWM when you saw it at last year’s CES show. (I said I was astounded that they would use a non-PWM display on the 27″ model in the same series, and you replied something like “Well, that’s that’s what they did.”) I suspect that this exchange was in the now-defunct thread that has been co-opted by the announcement. I certainly can’t find it in any of the other threads that I would have been likely to have followed.
I also need to start a discussion about problems with calibrating monitors using the Spyder5ELITE (which you also use). I can’t get it to work consistently and their tech people have been giving me a run-around for over a week now. I’m getting truly bizarre results with it. Should I start a new thread for this discussion? (It is also the reason I sent you an email [blind drop, actually] asking how to place images in posts.)January 27, 2018 at 8:16 am #46570
The thread has been redirected because there was too much off-topic drivel and pointless questions being asked over and over again. Not from you, specifically, but various users. And it wasn’t providing sufficient useful information to newcomers to the thread, which from my logs were few and far between anyway. I would also like to point out that this is not a general/technical help forum. So no, you do not need to start a discussion about problems calibrating monitors. My time is precious and I have better things to do than troubleshoot non monitor related issues. I’m sorry if the guys at Datacolor aren’t helping with your issue, even though they are paid to do so (I am not).
I’ve also told you quite explicitly that the C32HG70 uses PWM. It does. Fact. End of. You prefer to trust some Samsung technical support worker who likely has no clue over a trained expert? And ignoring evidence from TFT Central’s oscilloscope testing? If so go ahead, but you will have outstayed your welcome here.January 27, 2018 at 8:23 pm #46578
You read my post incorrectly. I did NOT contradict you on the PWM for the C32HG70 monitor. I used that as an example of where Samsung was claiming “flicker-free” when you had explicitly stated that it did use PWM. I said that the tech guy said the C32H711 monitor did not use PWM. You said you didn’t know about that monitor. The C32H711 is an entirely different monitor than the C32HG70.
I”m sorry that you cannot offer me any assistance in working out my problems with the Spyder5ELITE unit and its software. It would be helpful if you could point me to a forum somewhere that would be appropriate. Right now, I’m ready to throw the Sypder5ELITE in the junk heap. It is a monitor-related issue, as I cannot get the C32H391 to work correctly. I’ve traded black crush for white crush. Their so-called “Tone curve” (you more correctly call it a “gamma tracking curve”) shows wild deviations from the gamma=2.2 curve, with 100% brightness being attained at about an input of 85% RGB rather than at input of 100% RGB as it should. Yet their data info window shows the calibrated gamma as 2.2. Both cannot be true. So if I can’t depend on the data supplied by the equipment and software, how can I get a decent calibration of my monitor? BTW, Lagom’s banding data show the “white crush” clearly, so it appears that the gamma tracking curve is correct and the calibrated data info is a flat-out lie.
I am not ignoring your expertise in these matters. In fact, I’m using the information you supply as the “last word” on monitor-related subjects. There is no one else whom I trust. IMHO the technical support from DataColor is useless garbage. But if calibrating a monitor is not a monitor-related issue, I certainly don’t know what would be. If a monitor cannot be successfully calibrated (and mine apparently can’t), then it’s a piece of junk. That’s what I’m going to do with the C32H391. If it can’t be calibrated, I’m going to throw it away. My problem is that I can’t be sure that it’s something I’m doing in the calibration process that is producing the weird gamma-tracking curve. And Datacolor won’t tell me if it is.
Datacolor support wrote me that: “From the information you have provided, this is OK here for your monitor in calibrated mode. Please keep in mind that the aim of the calibration is that the colours are correct according to the ICC standard.” Well it isn’t OK if the data are internally inconsistent, and that bit about the ICC standard is just verbiage. I’m thoroughly disgusted.January 27, 2018 at 8:43 pm #46583
I’ve moved this discussion here as it’s a good enough place for more general discussion to take place.
I did misread your post and for that I apologise. You’re correct, I was not able to confirm that PWM was or wasn’t used on the C32H711. I would take anything a Samsung technical support person says with a large dose of salt, although I think the C32HG70 is the exception rather than the rule regarding its strange usage of PWM. I wouldn’t necessarily expect the C32H711 to use it. I would repeat what I said on the no redirected thread in that respect:
You could always pick up a C32H711 yourself and test. I could guide you through what to look for if you’re unsure. That way you could also help support the website, not just with the purchase but by confirming whether PWM is used.
You can post images using the ‘img’ button in the edit box, above where you enter text for a post. And I think this would help so I can see exactly what you’re referring to. It isn’t really a monitor issue, though, it is related to the Spyder5ELITE, the software or the ICC profile implementation. The monitor just sits there and outputs the image – that isn’t the part of the chain that’s causing the issue here, I can assure you.January 29, 2018 at 12:17 am #46598
I was afraid you would say that about the Samsung techie. Getting info from the Samsung people is like pulling hens’ teeth.
Here is my basic problem. The Spyder5ELITE (which I’ve seen that you use in your reviews) reports the target and calibrated values of gamma as 2.20. But their “Tone Response” curve looks totally different from the gamma = 2.2 curve. Unfortunately, I can’t use the “img” button to send the image because it is not on the Internet and therefore has no URL. I can supply a JPG, PNG, PDF, or a few other forms of file with the image in it, but I have no (easy?) way of putting it on the Internet so it would have a URL.
I have confirmed that even the weird gamma curve that is displayed by their Tone Response function is not correct, either. It shows the output as rising far above the gamma 2.2 reference curve from input levels of about 20% or more, and hitting 100% (saturation) at about an input of roughly 87% (R=G=B=221). It does not do that. There is a sudden transition from smooth gray scale values from 94% to 95% input (i.e., the output for RGB of 240 looks normal in a progression of gray scale images I made, but changes suddenly to a much brighter pinkish color for an RGB of 242). This is why I said I’ve traded black crush for white crush (if there is such a beast). I was going to insert an image of this gray scale progression (which runs from 81% to 100% in increments of 1% (the best one can do in Photoshop), but the Internet/URL thing stymied me.
The black crush is effectively gone now. On the Lagom black level test, I can see all of the squares in a darkened room (the first one is very difficult to make out, but it is there). In a bright room, I can see from #6 on up easily, #4 & #5 with averted vision. (Before the last calibration I could see only the bottom line of squares in a brightly lit room, although I could see #15 with averted vision.) I can also see all of the squares in the Lagom White saturation test, but the one labeled 254 is very difficult to make out. Interestingly, all of the squares in that test (except for the first at 200) fall into the “white crush” region of my setup.
Another interesting sidelight: In a previous calibration I could see all of the squares in the Lagom Black test when I took a screen shot (with the snipping tool) but on the monitor itself I could see only from square #7 upwards. Since screenshots get their data from what the GPU is sending out (not what’s on the screen), this implies that something in my monitor was introducing some black crush of its own. Or am I out to lunch again?
I know you have used the Spyder5 ELITE quite often in your reviews, so I was hoping you might have some clue as to what is going on with my setup. I’m thinking perhaps I have a bad hardware unit, or at least, poorly calibrated. But that doesn’t explain the discrepancies in the data the software displays and what the hardware is reporting (correctly or not). The more I think about this, the curiouser and curiouser it gets! (As Alice would say.)January 29, 2018 at 8:00 am #46599
You can email the curves over, but I assume you’re looking at the ‘Tone Response’ Advanced Report after calibrating? Often the colorimeter disables the ICC profile and what you’re seeing is likely the uncalibrated state. Check that you have ‘Calibration On’ rather than ‘Calibration Off’ when you right click on the Spyder5 icon in the notifications area (system tray). This doesn’t always guarantee anything, though, sometimes I actually need to manually re-activate the ICC profile before measuring gamma using something like DisplayProfile. The tricky part is trying to activate the profile after you’ve started the gamma test, but getting back to the right window again before the first measurement is taken.January 30, 2018 at 1:20 am #46600
Heh, heh!–“tricky” is a massive understatement for me. On my system, what you describe is literally impossible to do. Once the Tone Response test is initiated, it takes over the entire screen real estate. Even the taskbar will not come up if I move the cursor to the bottom of the screen. All I can say is this: Display Profile shows the correct ICM file (it’s not actually an ICC profile on my system) both immediately before running the test and immediately afterwards (even before cancelling the save to an XML file). Not only that, but running the test today yielded totally different results than it did yesterday. The gamma curve on this run was on top of the gamma 1.8 curve near the origin (input value of 0), whereas yesterday it followed the gamma 2.2 curve in the same region. It also had the display brightness as more than 100% at the top end, making a loop from 100 at an input of 87, rising to perhaps 105 at input of 92, and then dropping back to 100 at 100 input. (You don’t have a gamma 1.8 curve on the monitors you have tested–or at least, any that I’ve read, which is quite a few. We may have different versions of the software, or do you know how to suppress that additional curve? On my Tone Response graphs, that curve is in green, not blue like the gamma 2.2 curve.) A monitor that can display brightness of 105% of its own brightest value? Egad!
This is from running “Display Analysis” from the shortcut menu on the main screen, checking only the Tone Response box, and after the test, clicking on “View Report.” Also amazing is that today on the Gray Ramp graph (just below the Tone Response graph) it showed color temperatures ranging from 7900 K down to 7100 K at the white point (input of 100). Yesterday, it showed CTs ranging from 4050 K to 4350 K, ending at 4300 K at the white point (input 100). And these were for the same calibration ICM file! Another oddity: on the SpyderProof window, clicking on the “Switch” button to show the uncalibrated display does not change the profile that is displayed by Display Profile. (And, yes, I did stop Display Profile and restart it to get the uncalibrated measurement; I don’t think it will change once it is run.) How does the software manage to show the uncalibrated display without changing the ICM profile for the GPU, if only for the duration that one views the uncalibrated display?
Meanwhile, I can’t see a bit of difference in the Lagom test results for Black level, White saturation, or Gradient (banding) between what they showed yesterday and what they showed today. I”m beginning to believe that I have defective Spyder5ELITE hardware (or maybe the software?). the results I’m getting are all over the place, with no rhyme or reason.
I’ve about decided to buy one of the C32H711 monitors and see if it will work for me (as you suggested). Meanwhile, I think I’ll leave this monitor alone, since it passes the Lagom tests quite nicely IMHO. Let the Spyder software spout all the nonsense it wants to. I’ll rely on the Lagom tests from here on out and ignore the reports from the Datacolor software.
I think I know why it’s so difficult to find a decent review of monitors like my C32H391 or the C32H711: The only places that do a really thorough reviews–like the ones you do–are interested mainly in monitors for gaming. And these monitors, which have only a 60 Hz refresh rate, are hardly attractive to gamers. But since I never play games on my computer, that aspect is of no interest to me. Sigh. . . .January 30, 2018 at 4:21 am #46602
My estimation of your expertise just went up a couple of notches. You hit the nail right on the head. That’s exactly what’s happening. The Tone Response curve is for the uncalibrated monitor (GPU, really). I noticed when I switched from the calibrated to the uncalibrated view in SpyderProof, the entire screen changed. Display Profile does NOT catch this action, and still reports the calibrated ICM profile. I went back to Lagom, however, and did a capture (Snipping Tool) of the Gradient test for the uncalibrated mode. I read it into Photoshop and evaluated the colors. The sudden switch to a saturated white band occurred at an input of exactly 86, just as the Tone Response curve showed. When I switched back to calibrated mode and did the same thing, the sudden change to saturated white occurred at an input level of 95, the same as I had measured previously. This 95 is an input of 242 on the RGB scale (0 to 255). But it looks like almost 100%. That it is not is shown by the Lagom White saturation test (calibrated mode), in which all of the checkered patterns are visible except the one at 254, which can only be seen with averted vision. (They all have a pinkish cast, however.) The Black level test, oddly enough, looks better in uncalibrated mode because the squares are closer to a neutral gray. The low-numbered squares are a bit easier to see in calibrated mode, but all of them below the one at 40 have an olive-greenish caste to them. In the White saturation test in uncalibrated mode, all of the checkered patterns disappear except the one at 200. I’ll stick with the calibration because the off-color of the darker grays doesn’t seem to affect anything I do much.
BTW, the Gray Ramp graph for the latest test (tonight) was–once again–different from any previous ones. Similar to the previous one but values shifted somewhat, and both of them vastly different from the first time I ran the test on this profile.
Also BTW, I changed “review” to plural in my previous post but forgot to take out the “a” before “really.” Argh!January 30, 2018 at 7:49 am #46603
I’m glad you sorted this. I was also going to ask if you had F.lux enabled or something else. DisplayProfile does not reflect this or when Spyder5 force deactivates things, that’s right. A few things:
– ICC and ICM are used interchangeably. There are technical differences and they stand for different things, but do exactly the same job.
– You can use the tick boxes in the reports to disable the ‘1.8’ curve from showing.
– You purchased A C32H711 but didn’t use our link. I’ve now dedicated a lot of time to helping you and answering your questions. Any reason for not supporting our work?January 30, 2018 at 3:53 pm #46608
I must have mislead you. I have not yet purchased a C32H711 monitor, and when I do, I will purchase it from Amazon through your link, as suggested for supporting this forum. I did that when I ordered my current monitor, the C32H391, last spring. At the time I mentioned that your link directs to Amazon.com (main site), but I’m using Amazon Smile to support the Wounded Warriors project. When I arrived at Amazon.com, I was immediately prompted to redirect to Amazon Smile, which I did. I mentioned at the time that I was not sure if this somehow negated the link from your website, but no comment was ever made about that (to the best of my limited knowledge–as I’m 86 years old, my memory is not as good as it once was). If this is a problem, please let me know. I do want to support your website and deeply appreciate the help you have given me. If it is a problem, I will simply make a direct donation to your website. I value your input very highly.
I knew the ICC and ICM profiles are used interchangeably, but I had (and have) no idea what the technical differences are, or if they would have any effect on the calibration process. The fact remains that Datacolor’s Spyder5(ELITE) disables the calibrated profile when making a Tone Response chart (aka–more accurately–as a gamma response curve). This is just plain stupid, and IMHO is a defect in their product. I intend to raise a stink about it with Datacolor. Disabling the calibrated profile while making a chart that purports to represent the calibrated gamma data (and is so labeled on the chart) is the height of absurdity. And that’s not just my opinion. I think anyone with any brains would come to the same conclusion.
Thank you for the tip about the gamma 1.8 curve on the Tone Response charts.
BTW, you are not located in Bangladesh, are you? I noticed that when I posted my previous remarks last night, at 9:21 EST on January 29th, that the post was labeled as posted on January 30 at 4:21 a.m.–a time that corresponds with the time zone in which Bangladesh is located. I always had the impression that you were located either “down under” or in the UK, since you consistently use British syntax (e.g., spaces before and after an m-dash) and spelling (e.g., colour rather than the American color). Or is it just that the website itself is located at a server in Bangladesh? I’m just curious, and if you do not wish to disclose any of this, I’ll understand. 🙂
Full disclosure on my part:
G. David Thayer
Rapidsoft Press ® LLC
Sarasota, FL USA
aka flboffin (fl = Florida, boffin is British slang for an egghead–which you probably know already)
If you’d rather respond to this by email, my address is “gdthayer [at] verizon.net” (the [at] is to defeat bots that look for email addresses)January 30, 2018 at 4:13 pm #46660
Haha, I do apologise. I jumped the gun completely and read “I’ve about decided to buy one of the C32H711 monitors” as “I have bought x”. That’s what I get for reading your posts early in the morning, body awake but brain still asleep. 🙁 I’m located in England, not Bangladesh – the forum time should have been set to GMT/UTC but I see it was offset by an hour from there. Now fixed, hopefully.
I’m not sure what happens if you use Amazon Smile, although I don’t want to discourage you from using it. I think it should if you have clicked on one of our links, are redirected to Smile and then make your purchase. Could you remind me of the exact date you ordered your current monitor? I’d be able to check back and see if it registered as a sale.
I totally agree with you that disabling the profile is ridiculous behaviour for the Spyder software. It makes it very annoying to take measurements with profiles active or validate the profile as you have to manually activate them.
Edit: Will fix the time zone later. I fixed it and the forum got confused, putting my reply before your earlier reply. Because I replied within an hour and the offset for the forum time was changed… Gah!January 31, 2018 at 1:33 am #46664
PCM2: I bought the C32F391 monitor from Amazon Smile (Prime) on May 1, 2017. Should be easy to look up.
I believe I made a typo in an earlier post where I gave this monitor an “H” series designation instead of the correct “F.” I also bought a Dataport cable in case the HDMI didn’t work (that cable came with the monitor). Which begs the question: If one has either connection available, which is the better one to use? Or is it a case of “six of one and half a dozen of the other?”
I’m still quite disappointed in the text-rendering by the C32F391 monitor. Who would ever have guessed that a five-year-old 32-inch Sharp HDTV would render text better than a 32-inch monitor–equipment that is expressly intended for use with a computer?
Of course, the Sharp HDTV had black crush like you wouldn’t believe. I guess the guys at Spotify thought I was nuts when I wrote them that their new screen design (which uses dark gray sliders against a black background) made it impossible for me to find the sliders. Now they are crystal clear.
I shall probably set the size of text, icons, etc. to 125% with the WQHD monitor to make up for the small size of things at that resolution. That is, unless it screws up apps too badly. Windows should leave them alone and just adjust the desktop, taskbar, etc. Apps generally size themselves to the resolution available automatically. Text within apps can be adjusted from within the app.January 31, 2018 at 6:39 am #46665
Yes, that did go through. So that’s good to know – and a retrospective thank you for your support. 😉 I realised you were referring to to the CF391 rather than CH391 (a model they never made) as I read it as C32F391 in my mind anyway.February 2, 2018 at 11:40 am #46670
PCM2, You deleted your last post here (after confirming that my purchase of the C32F391 was credited to your website), and I don’t recall if you answered my question about the graphics connections. Is there any advantage re calibration in using Data Port as opposed to HDMI? I read somewhere that Windows will treat an HDMI input as a TV input and make some unwanted adjustments to the video, but it will leave Data Port inputs alone. Anything to that?
You must be logged in to reply to this topic.