[time-nuts] Re: World's most accurate PC clock!

Magnus Danielson cfmd at bredband.net
Sun Jul 3 17:35:18 EDT 2005


From: "Tom Van Baak" <tvb at leapsecond.com>
Subject: [time-nuts] Re: World's most accurate PC clock!
Date: Sun, 3 Jul 2005 11:12:24 -0700
Message-ID: <001e01c57ffa$c4861400$260ff204 at computer>

> > Tom,
> > 
> > A maser-driven synthesizer is definitely the best approach, but the 
> > frequency isn't exactly 14.31818. It's 14.3 (18)repeated, as the 
> > frequency is derived from 5.00000MHz  * 63 / 22. That means you need 
> > a divide-by-eleven to get it spot on. This is all described in the 
> > Wikipedia entry for NTSC.
> 
> I know what you're getting at. At one level you're
> correct. But since we're time-nuts allow me to
> digress into precision and history here. You see,
> spot on isn't good enough in this case. For a PC
> most NTSC articles are close but not the whole
> story.
> 
> The frequency of a perfect "3.58 MHz" colorburst
> crystal is 3 579 545.454 545... Hz (as a decimal
> approximation), or 5 MHz * 63 / 88 (as an exact
> rational number).

To bring the historical perspective in a little more...

Traditional NTSC grayscale television was 525 lines per frame, 30 frames per
second. However, for the colour variant, the frame rate was shifted to be
1000/1001 of the grayscale variant, thus 30000/1001 frames per second.
In order to sneak in the colour signal with minimal influence on the greyscale/
lumina channel while providing a high enought frequency for the colour
differenc I/Q modulation to have sufficient of bandwidth (about 600 kHz) the
colour subcarrier was chosen to be 455/2 (13*7*5/2) multiple of the line
frequency. The resulting subcarrier frequency thus becomes:

                                       3    3
           1000         455           2  * 5       2     5*7*13
f   = 30 * ---- * 525 * --- = 2*3*5 * ------- * 3*5 *7 * ------
 sc        1001          2            7*11*13              2

thus becoming

       4  2  7  2       3  2  7
      2 *3 *5 *7 *13   2 *3 *5 *7
f   = -------------- = ----------
 sc     2*7*11*13          11

                           7  6
since our 5 MHz is really 5 *2  Hz we need see that all of the 5s come into use
and only three of the 2s, the rest we need to synthesize, thus the 63/88 ratio.

On the other hand, the subcarrier frequency is +/- 10 Hz in ITU-R BT.470, so
so much for precission work. The only other requirement is that it may not
change more than 100 mHz per second, thus a drift-rate requirement. This
drift-rate requirement is there in order for the subcarrier modulated colours
not to shift in frequency so fast that the I/Q is turned (and thus colours
shift) due to frequency drift from the colour burst in the beginning of the
line. The infamous colour shift of NTSC is due to the phase delay and group
delay difference that especially reflections cause on signals, where PAL
excells by having one of the colour difference signals inverted on every other
field, so that the phase-shift goes the other direction and is visually almost
canceled.

Ah well, DVB and ASTC is changing all this - almost!

Cheers,
Magnus




More information about the time-nuts mailing list