NTSC ancient and modern

[Panrock]

Having just read about Darryl's impressive achievement of producing a 405-line NTSC colour converter I'm now eagerly awaiting seeing it in action, working with an original receiver at some future BVWS event!

Wonderful though it was, I've heard NTSC (never-twice-the-same-colour) had something of an bad reputation in the early days for incorrect hue caused by subcarrier phase errors. By reversing the phase every other line, PAL got over this to some extent, but at the cost of hanover bars - or reduced saturation.

I understand that today's NTSC though, isn't at all bad. What's happened? Is everything just much more stable now, or have deeper changes been made to the signal and circuitry ?

Steve

[ppppenguin]

While not wishing to detract from Darryl's work let's not forget that David Robinson's converter provides a 405 line NTSC output. Alas it has never been tried with a suitable receiver. Some of us saw it demonstrated at the last NVCF and we could see the subcarrier on the waveform monitor and as patterning in saturated areas of colour on a monochrome monitor.

PAL certainly overcame the phase** problems that plagued early NTSC. In practice all receivers except the early Sony sets used delay line PAL decoding and so did not suffer Hanover bars unless there was a fault. The desaturation effects were generally imperceptible, even with quite severe phase errors. The Sony sets got round the PAL patents by effectively treating the PAL signal as NTSC. Thus they required a hue control.

NTSC with modern engineering is a perfectly good and stable system. PAL (and SECAM) are more tolerant of phase problems. You can get away with more and still have decent colour. When NTSC started in 1954 the valve technology of the day was only just about up to the job. Circuit stability and consistency were major issues. By the start of PAL transmissions in 1967 all (virtually all?) the studio equipment was transistorised and much more stable.

**I'm using phase problems as a collective for differential phase, differential gain and any other chroma phase errors.

[evingar]

By the start of PAL transmissions in 1967 all (virtually all?) the studio equipment was transistorised and much more stable.

Hmmm. When I first started out, I worked on Quantel's 5000 (NTSC) 1st generation "Effects Machine”. Not a valve in sight, but "stable" would not be a word I'd use to describe it . This was due to the chains of monostables in the digital timing / sync generation area and the "capacitors" around the MC1495's (4 quadrant multipliers). These consisted of bits of Kynar wire twisted together, then cut back to "trim" (literally !) the capacitance. Never mind, always always a good excuse for an extra cuppa, "Jus waitin' for the machine to drift in Guv'nor" - Happy days

[tubesrule]

Chris,
Not sure if you use the same term over there, but these capacitors are called gimmicks here and they were fairly common in early high frequency products. You'll still see amateur radio guys using them.

As for NTSC equipment, it has certainly come a long way from the early days. While my personal recollection of color television starts in the mid 60's, I worked on many 50's - 70's sets when I worked in repair. The sets of the 50's and 60's were actually quite stable once setup, but you did have to twiddle the tint (or hue) control sometimes when changing channels, but rarely when just watching a program. By the 80's the sets were very stable and almost never required adjustment. My daily set is a 27" Sony from 1986 and the only time I have ever adjusted any of the controls is when I'm using it for development of things like the 405NTSC converter. In cases like this I use a blue filter and color bars to evaluate how I'm doing.

Darryl

[CheapSlider]

VIR - Vertical Interval Reference on line 19 - proposed in the 1980s

Is it possible that VIRS was quietly incorporated into NTSC decoder chips with no fanfare?

I do remember that in the early 80s colour TV news pictures from the USA were often washed out and always tinged with a colour cast. (I could never work out if Ronald Reagan's face was really purple or green).

from danalee site:
"VIRS consists of a 70 unit IRE signal modulated with subcarrier of the same level and phase as colour burst, with 50 and 7.5 unit IRE level signals. It is normally used for either manual or automatic adjustment of chrominance gain and phase, luminance and setup parameters of a video signal. The modulated 70 unit signal is of average chrominance phase, at average Caucasian skin-tone luminance level. The 50 unit IRE signal represents an average picture luminance level, and the 7.5 unit IRE pedestal is used for picture black level setup. VIRS is normally found on line 19, both fields.

Some television sets can make use of this signal to align themselves automatically for the best picture quality. For this to work, each source is supposed to generate its own VIRS signal, and shouldn't be stripped out or re-generated later - the idea being that, as you go from source to source, the viewing or monitoring system adjusts for each one. This concept has never really caught on, however."

there is also a reference to VIRS at the bottom of the sencore page.


Ghost Cancelling Reference -this was proposed in the 90s and is also referred to in the danalee site:
"GCR - is a relatively new signal transmitted on line 19 of some television stations. It’s a sweep in frequency from 0 Hz to 4.2 MHz, occurring over one video line. It's only useable by new television receivers that have ghost cancelling ability. The GCR is transmitted with the regular TV picture and is compared to a clean version that resides in the television set. Any differences found are used to tune out ghosts in the received transmission"

I don't know if these 2 signals (VIRS and GCR) would be compatible or mutually exclusive with each other.