Re: vic-ii pal colors

From: Michiel Boland <michiel_at_boland.org>
Date: Wed, 8 Jul 2020 20:15:46 +0200
Message-ID: <6b66204a-3843-f892-1a31-dfbf5d104630_at_boland.org>
On 7/8/20 9:05 AM, laubzega wrote:
> Michiel, were you able to measure the difference in phase between "same"
> colors in odd and even lines?

Hi. These are the measurements from the bt848 capture card, converted to analog 
yuv. (a=amplitude, phi=phase angle)

blue
  odd  y= 0.25 u= 0.22 v= 0.01 a= 0.22 phi=   3
  even y= 0.25 u= 0.25 v=-0.06 a= 0.26 phi= -14
brown
  odd  y= 0.25 u=-0.21 v= 0.10 a= 0.23 phi= 155
  even y= 0.25 u=-0.18 v= 0.16 a= 0.24 phi= 139
red
  odd  y= 0.32 u=-0.06 v= 0.25 a= 0.25 phi= 104
  even y= 0.32 u= 0.02 v= 0.21 a= 0.21 phi=  85
purple
  odd  y= 0.40 u= 0.11 v= 0.21 a= 0.24 phi=  62
  even y= 0.40 u= 0.17 v= 0.18 a= 0.25 phi=  47
orange
  odd  y= 0.40 u=-0.17 v= 0.16 a= 0.24 phi= 137
  even y= 0.40 u=-0.11 v= 0.21 a= 0.24 phi= 118
light_blue
  odd  y= 0.50 u= 0.22 v= 0.02 a= 0.22 phi=   6
  even y= 0.50 u= 0.25 v=-0.05 a= 0.26 phi= -11
green
  odd  y= 0.54 u=-0.08 v=-0.21 a= 0.22 phi=-111
  even y= 0.54 u=-0.14 v=-0.20 a= 0.24 phi=-125
light_red
  odd  y= 0.54 u=-0.06 v= 0.25 a= 0.25 phi= 104
  even y= 0.54 u= 0.02 v= 0.22 a= 0.22 phi=  85
cyan
  odd  y= 0.69 u= 0.05 v=-0.25 a= 0.25 phi= -78
  even y= 0.69 u= 0.00 v=-0.21 a= 0.21 phi= -90
yellow
  odd  y= 0.85 u=-0.20 v=-0.01 a= 0.20 phi=-176
  even y= 0.85 u=-0.24 v= 0.05 a= 0.25 phi= 169
light_green
  odd  y= 0.85 u=-0.08 v=-0.21 a= 0.22 phi=-111
  even y= 0.85 u=-0.13 v=-0.20 a= 0.24 phi=-124

I also attempted to measure the phase angle from scope readings, using color 
burst as a reference: (I didn't bother to measure the 'light' variations)

odd

blue     1
purple  59
red    102
orange 134
brown  154
yellow 180
green -113
cyan   -80

even

blue   -13
purple  48
red     90
orange 122
brown  142
yellow 169
green -123
cyan   -89

not sure how accurate these are, but there definitely is consistent phase 
difference between odd and even lines.

It would be interesting to see some actual vectorscope pictures. The pepto 
article[1] seems to suggest that all 'all angles are perfectly dividable by 22.5 
degrees', I don't see that at all.

Cheers
Michiel

[1] https://www.pepto.de/projects/colorvic/2001/
Received on 2020-07-08 21:00:04

Archive generated by hypermail 2.3.0.