Andromeda Stardust #1 Posted November 4, 2015 I made a post in the New Pacman Atari thread a few days ago: http://atariage.com/forums/topic/229152-new-pacman-for-atari-2600/?p=3355400 My inquiry was how the New Pacman homebrew added white dots to the strawberries in Pacman 8k. I first noticed the sprite after doing the resistor mod to enhance the saturation on my 4-switch woody. Turns out the white dots were actually non existent black voids, but sure as heck showed up as white on my mom's CRT. It's got great contrast and black level but the beam is too fuzzy to show scanlines in detail. My smaller bedroom CRT shows off scanlines and the large RF only Zenit CRT I have packed away in the garage is immaculate. I'll need to check some other sources including the LCD in the living room to see if they all do this. Compare this real life macro shot of my mom's bedroom CRT, to this Stella screenshot (composite filter) of the exact same ROM: The VCS should be clearly outputting black pixels but they show up as white little dots on the strawberry stage on my mom's CRT. There is no multicolor or multisprite usage going on as Pacman lives are displayed independently on the same scanlines and no midscan manipulation is taking place in this portion of the display kernel. Even looking only at the red phosphors in the macro photograph, there is absolutely no dimming of the red color at all, only brightening of the green and blue phosphors making them appear as white when zoomed out. Assuming the luminosity and saturation components drop during the strawberry sprite to account for the change from red to black, and the individual components were somehow shifted, shoudn't there be some regional dimming of the red phosphors? I can't see any. What is the exact mechanism that produces the effect of white dots on read instead of black? Curious about the nature of the effect. To do: Photograph the strawberry sprite effect using the composite output of my Best AV7800, and compare the Woody RF to the 7800 AV on the LCD and my other CRT set. Also any other recommended games to test the RF/Phosphor effects with? My Harmony is fully loaded... 1 Quote Share this post Link to post Share on other sites
Osgeld #2 Posted November 4, 2015 (edited) https://en.wikipedia.org/wiki/Composite_artifact_colors in a nutshell its bleeding on the falling edge of the pixel before it, you can kind of see it on your first screen shot, they are not white, they are orangish purpilish on the edges, its a natural effect pretty much only on NTSC RS170 or black n white video which is the main portion of the signal runs at X speed (ie 15khz per line) the color phases were kind of kludged in at a later date with a base frequency of 3.15mhz so the scan line cant really keep up and it blurs color, to get into how and why your going to have to look up color burst, phase offset and figure out what the heck this means but in this case it looks like the red is turned off in the color signal but the pixel signal is dragging along to the next square (inside the tv, not the game) Edited November 4, 2015 by Osgeld Quote Share this post Link to post Share on other sites
DrWho198 #3 Posted November 7, 2015 It does make this strawberry look extra nice though Thanks Osgeld, when I was looking at these pictures I looked over this but it's true, you can clearly see that all the right side of the red has bright spots. And if you coulnt the remaining white dots in the middle of the strawberry then these are the same amount as the black spots on the Stella screenshot. I never saw this kind of bleeding, but I live in PAL land, where bleeding looks a bit different. Quote Share this post Link to post Share on other sites
Andromeda Stardust #4 Posted November 7, 2015 It does make this strawberry look extra nice though Yes it does. Thanks Osgeld, when I was looking at these pictures I looked over this but it's true, you can clearly see that all the right side of the red has bright spots. And if you coulnt the remaining white dots in the middle of the strawberry then these are the same amount as the black spots on the Stella screenshot. I never saw this kind of bleeding, but I live in PAL land, where bleeding looks a bit different. I'm still in awe how the black dots somehow became white. I still need to examine this sprite on my 7800 and the other TVs. Quote Share this post Link to post Share on other sites
+save2600 #5 Posted November 7, 2015 Re: Keystone Kapers, if you look closely at Keystone Kelly's buttons in his coat, you'll see that they're white playing through RF. Seems as if they take on whatever color the background is actually, but against the green background, they appear white through RF. Quote Share this post Link to post Share on other sites
Andromeda Stardust #6 Posted November 21, 2015 The artifacting must be in the TV set rather than the signal itself. I added a macro shot of the LCD Sanyo. Side by side comparison. Same 4-Switch Atari (with 820ohm color fix). Same RF NTSC channel 2. Click to blow up. Mind blown... Quote Share this post Link to post Share on other sites
Osgeld #7 Posted November 21, 2015 Yep it is totally dependent on the signal processing of the TV which includes the decay of phosper Quote Share this post Link to post Share on other sites
Andromeda Stardust #8 Posted November 21, 2015 Yep it is totally dependent on the signal processing of the TV which includes the decay of phosper I totally need to put in Tower Toppler in my Best AV-modded 7800 and see how my 2006 Sanyo LCD responds... Quote Share this post Link to post Share on other sites