I was browsing some TV docs today and noticed that PAL stands for Phase Alternating Line. This is a reference to the colour burst signal being inverted on alternate lines. Then it clicked as to what the issue is with these systems.
The phase of the composite colour pin is advanced, I'm guessing either due to age of the vic, or age of the capacitors in the colour output stage.
It always appeared to me that the colour burst was coming out inverted on alternate lines, but I never realised that's what's meant to happen.
So, has anyone attempted replacement of the capacitors (maybe even resisters) in the colour output stage? Or, if the VIC is dodgy, maybe altering the values of the capacitors to delay the phase back to the correct value.
It may be preferable to some, rather than doing the s-video mod.
An example of what I'm talking about...
Bad PAL display outputs revisited.
Moderator: Moderators
-
- The Most Noble Order of Denial
- Posts: 343
- Joined: Fri May 01, 2009 4:44 pm
- Mike
- Herr VC
- Posts: 4841
- Joined: Wed Dec 01, 2004 1:57 pm
- Location: Munich, Germany
- Occupation: electrical engineer
With PAL, the red component of the colour difference signal is alternated in sign on consecutive lines. Should a phase shift occur in the TV signal, with PAL this only leads to desaturation of the colour, not a shift in the hue as with NTSC.
The alternate zig-zag pattern (I presume you mean the image of the cyan border) results from cross-talk between chroma & luma. It cannot be avoided unless the combination of both within the video circuitry is *not* done, but both signals are left separated, as done in the S-Video mod.
In case you meant the two faint lines in the left cyan border, this is the VIC chip fetching information to display the first character on each line.
Michael
The alternate zig-zag pattern (I presume you mean the image of the cyan border) results from cross-talk between chroma & luma. It cannot be avoided unless the combination of both within the video circuitry is *not* done, but both signals are left separated, as done in the S-Video mod.
In case you meant the two faint lines in the left cyan border, this is the VIC chip fetching information to display the first character on each line.
Michael
-
- The Most Noble Order of Denial
- Posts: 343
- Joined: Fri May 01, 2009 4:44 pm
I replaced the electrolytics and it had no effect.
But... I replace the 230 ohm resistor on the diagonal from FB7 to R10 with a 3k and there was major improvement. The unreadable red cards on Poker are now readable. There is a *slight* drawback though, in that you get a little bit of blue bleading into the blacks.
Maybe a little bit of some sort of RC filter there would have a better effect.
But... I replace the 230 ohm resistor on the diagonal from FB7 to R10 with a 3k and there was major improvement. The unreadable red cards on Poker are now readable. There is a *slight* drawback though, in that you get a little bit of blue bleading into the blacks.
Maybe a little bit of some sort of RC filter there would have a better effect.
-
- The Most Noble Order of Denial
- Posts: 343
- Joined: Fri May 01, 2009 4:44 pm
-
- The Most Noble Order of Denial
- Posts: 343
- Joined: Fri May 01, 2009 4:44 pm
As far as I'm concerned, the only viable way to correct the VIC's video problems is to do the modification to S-Video, which can be found in the Wiki, or elsewhere on Denial.
Composite PAL and NTSC are never really much good for displaying computer graphics, and very little can be done to improve it.
Composite PAL and NTSC are never really much good for displaying computer graphics, and very little can be done to improve it.