No announcement yet.

Screen Burn

  • Filter
  • Time
  • Show
Clear All
new posts

  • Screen Burn


    I have been using 4d Systems for several years now and this is the first time I've encountered this issue. My conclusion is that the screen is defective but I will list what I've done and let you decide. I want to make sure I'm not doing anything wrong before I send it back.

    I have attached a couple of screenshots showing part of the issue. There are two things to note.
    1. The gradient doesn't blend into the background despite one of the colors being the same. I have four other screens of the exact same model and they don't have this issue. Also the colors appear to be darker than the others, and it is not as bright.
    2. My camera was not able to pick it up as well but if you look closely you can see a couple of fancy buttons faintly showing up toward the top of the gradient. This was from a project I was working on. But showed up when I deployed a completely different project. It almost looks as though the previous image was burned into the screen.
    I would like to dispel some of the initial questions that might be asked. The screen was brand new and I received it about a week ago and just now pulled it out of the box for the first time. I deployed my program to the screen and immediately noticed issue #1 listed above. I had another screen of the same model running the same program for comparison. I tried adjusting a couple of parameters such as the "contrast" from the GTX tool, and the gradient colors. These did not fix the issue. Also the screen was never on for more than a few minutes as I would unplug the USB cable as I made my changes. So there is no way that "screen burn" should have occurred.

    Thinking it might be something with my uSD or program, I started a new project and used a uSD from one of the screens working correctly. I simply added a gradient object to Form0 and then compiled and built. The results were the screenshots attached. I deployed this program to one of the screens that is working correctly and it displayed correctly.

    So to summarize I have already verified it is not the program, uSD, programmer, or cable. I am using the current version of Workshop4 IDE and the latest PmmC. Like I said before, the likely answer is a defective screen, but I just wanted to get a second opinion.
    Attached Files

  • #2
    I forgot to mention this is a Gen4-uLCD-35DCT-CLB model.


    • #3
      Hmm, not quite sure what is going on here.

      Can you please run the attached, it should cycle through two gradients and a full screen of green, it should also print a number.

      Run it on some of your other displays as well, and note which number corresponds to which driver version (-A, -B,, -C, etc.)

      Also note any displays that don't display correctly. (presumably only the noted one)
      Attached Files


      • #4
        On the display in question:
        The gradients appear the same as before, they vary in color across the object, but they don't merge into the proper color. To elaborate, if the colors are green and black, and the background is black, the black part of the gradient should merge into the background. Instead just a darker green is the last shade and a visible line separating the gradient and the background is very obvious. The number on the display is 9488. I see driver -B and -D both correspond to that number but I'm not sure how to differentiate.

        At this time all of my other screens of that model are in service and I can't pull them off their respective projects. There will be an opportunity next week while the station is down.

        I'm not sure if it will be helpful but I ran the supplied program on a Gen4-uLCD-35DT and the gradients appear correctly and the number is still 9488. The PmmC is rev 2.4, and the SPE2 is rev 1.2. When I tried to update the PmmC, I received the error message attached.

        I was able to update the PmmC on the display in question prior to trying to update this one. Could it be a bad file? I'm not sure how to restore to factory settings. I never deployed a program to the display before updating the PmmC.
        Attached Files


        • #5
          Sorry for another message but I also got my hands on a Gen4-uLCD-50DT. It displayed the gradient correctly as well. However it's number was 0000. It had PmmC rev 2.4 as well, but unlike the 35DT I was able to update it successfully. PmmC ver R2.5 still displayed the gradient correctly.


          • #6
            The message 'merely' means there is no update driver available, at least in this case. If you put orginal driver (attached) in C:\Users\Public\Documents\4D Labs\4DUpdates\PmmCs then it will update the PmmC.

            So, yeah, the 9488 indicates a B or D driver is correct, the aren't interchangeable as the comments say. the 5" uses an SSD1961/3 driver, the 0000 is because the command used to display the 9488 is 'undefined' on that driver, similar to the HX8357's 0000

            I'm not sure how to interpret your description, can you post a photo of each of the three different 'screens'
            Attached Files


            • #7
              ​​To better explain the gradient issue I took some pictures. Unfortunately it isn't as pronounced in the pictures as I would like to be able to show, but I believe it is enough to help illustrate what I'm talking about. For all displays I just uploaded a program with a black background and a gradient (green and black). The gradient is full width, and half height for each respective screen.

              Click image for larger version

Name:	Gradient2.jpg
Views:	18
Size:	4.49 MB
ID:	76003
              In this picture you can clearly see where the gradient ends and the final row of color at the bottom of the gradient is still many shades away from black. There is a clear line that separates the background from the gradient. This is not what I expect from a gradient.

              Click image for larger version

Name:	Gradient1.jpg
Views:	15
Size:	4.73 MB
ID:	76004
              In this picture it is hard to tell but you can see that the bottom of the gradient sort of "fades" into the background creating a more seamless transition from the green to the black. There isn't nearly as clear a distinction from the gradient to the background. In fact, the gradient should cover half the screen since it is 240 height, but because it transitions well into the black it appears as though it does not cover half the screen but less than half. This is how it should be.

              Click image for larger version

Name:	Gradient3.jpg
Views:	16
Size:	4.51 MB
ID:	76005
              This image also displays the gradient correctly. Even though the gradient covers half of the screen, because of the smooth transition you can't really tell.

              Click image for larger version

Name:	Gredient4.jpg
Views:	16
Size:	5.28 MB
ID:	76006 Click image for larger version

Name:	Gredient5.jpg
Views:	15
Size:	3.15 MB
ID:	76007
              This was another attempt to show the issue. The screen on the left has lots of black toward the bottom as it fades into the second color of the gradient. The screen on the right barely has any black toward the bottom and mostly appears to be a darker green but never quite makes it to black. This is why the problem display gradient does not fade into the background when the background color and second gradient color are the same. This appears to be a defect with the screen. I have tested this with other colors but I only showed green for this post.

              I apologize on the quality of the images but hopefully this better demonstrates what I am tryin to describe. All of the other displays that I have used gradients on appear correct to me. The screen in question is the only one that appears differently. In addition to this issue all the colors appear to be slightly darker/duller than the other displays. It doesn't show up well in the images I posted but it is very clear in person. So with the gradient, color difference, and screen burn effect, it appears something is defective in the screen physically since the two screens I used for comparison use the same driver. I would also like to note that even though these displays differ in the sense that one is a 35DT and the other is a 35DCT-CLB, I have three other 35DCT-CLB that look just like the 35DT images posted here.

              P.S. I also noticed through trial and error that only the bottom half of the screen has the screen burn effect and it isn't always consistent. That's why it is not visible in these images.


              • #8
                Don't you just hate it when the camera lies?

                Still finding it hard to get to the bottom of this.

                The problem is that there are quite a few 'natural' variables.

                Each of the -B, -C and (-A or -D) variants will look slightly different, primarily because of the different driver Chips, but also because each driver chip has a different Gamma setting (as recommended by the 'glass' manufacturer). The -B to -D difference is mainly due to the -D being an IPS (In Plane Switching) Glass. IPS should improve the display quality, of course, in certain circumstances it could just be seen as 'different'. i.e. -A and -D should look the same

                Also, remember that these displays are 565 colours, this means there is 5 bits of colour depth for Red and Blue and 6 bits of colour depth for Green. So 32 shades of Red and Blue and 64 shades of Green, including 'full off'.

                I've attached another program that shows all of the possible colour levels for each of the three colours, maybe you can take a photo using that that will make it clear?

                Which are the Driver revisions you have? Do you have multiples of the faulty one that you can put side by side?

                Attached Files