SUMMARY De file ScrollTestPAL.img contains an image for a movie DVD. This DVD contains some test patterns for testing horizontal scroll on DVD-TV Combinations. The image is free software. You can frelly distribute and copy. There is no warranty of any kind. HISTORY Before buying our first large flat panel TV I did some research and read about the trouble some of these had with fast moving images. One of the advantages of plasma was supposed to be that it suffered less from this problem, so I settled in the end for a new Panasonic TH37PV70. I connected this TV to my Panasonic S325 DVD-player via Component cables. The S-325 is capable of 576p (576p25 I think). I also connected it to a cheap digital TV (DVB-C) receiver using a good SCART-cable. Pretty soon I noticed that the image was very blurry/fuzzy under horizontal movement. It did not matter what the source was. Especially horizontally moving text was awful. Unreadable sometimes where it had been reasonably readable with our previous TV (Medium-Sized Sony CRT). I tried everything I could with noise reduction settings and such, but whatever I did, the result remained bad. So, a while back, I asked a Panasonic engineer to have a look at it. But this guy came and said it was all within specs. I was pretty disappointed. How come hardly readable horizontal moving text is "within specs"? Wasn't this movement issue supposed to be a thing of the past? I also noticed the blurrying effect with action movies and they get less enjoyable as a result. The clearest demo is with text, but what we see with text is a good measure for how good fast action will be as well. Investigating the source of the blurry movements I noticed at a certain moment that a blurry horizontal moving object was displayed not once, but twice *simultaneously* on the screen. No wonder it looked terrible. If I followed certain moving texts with my eyes I could clearly see the multiple images where there was supposed to be only one. How good this could be seen depended on speed of the moving text. I was wondering, where does this effect come from? Is there any modern DVD-TV setup that does this right? THE DEMO I need to do a first step. Create a good reproducible example. So, I created this DVD. It contains 9 short movies. Each movie consists of two lines of text running horizontally over the screen. One is moving from left to right and the other from right to left. They end up above each other, For each movie there are three speeds: normal, faster, and fastest. The fastest take roughly 1.5sec to move the text lines to their position. The three texts are: The Quick Brown Fox Jumper Over The Lazy Dog a series of 'l' characters a series of 'o' characters Especially the 'l' character moving is a good way to look at the echoes. The fast 'l' and 'o' movies give the best demo results for me. THE INITIAL RESULTS On my own setup (Panasonic S325 @ 576p via component to Panasonic TH37PV70) I can now clearly demonstrate the echoes. The fastest 'l' movie clearly displays 3(!) images for every one there should be. Just try to follow the incoming start of the text line with your eyes and it is pretty easy to see why this moving stuff becomes so blurry. Note that the DVD-player is sending the frames out progressively and not interlaced and the TV handles that fine. Then I tested this with friends who just had bought a brand new Philips LCD flat screen. To my surprise it did better on the echoes, there were only two images where I had three (and where there was supposed to be one). But that TV showed another distortion. Where the lines were close to each other moving against each other the top side of the bottom text line and the bottom side of the top text line showed nasty distortions. No good way to describe these distortions other than that they seem to originate in the LCD-panel where closely lying LCD cells may influence each other in some way. If from algorithms in trouble or some physical effect, I do not know. And when I connected my S325 via SCART to a very cheap SEG analog TV it also displayed the multiple images (two where there should be only one). SOURCE OF THE PROBLEM At this point, I have no idea what causes this. I cannot rule out for myself that interlacing, refresh rates or something has anything to do with it. But whatever the source, I am convinced there is a problem here. Display of horizontal movement should be a lot better than it is now. Feb 2008 Gerben Wierda