I've been attempting to recreate the effect of an old NTSC television broadcast lately with the use of DVDs. The DVD-Video spec allows for up to 720 x 480 pixels at a display rate of 29.97 frames per second, interlaced.

My DVD player is outputting composite video to an RF modulator which is then connected to my TV and the signal is demodulated by the receiver, effectively simulating the NTSC transmission model.

What I'm wondering is if RF modulators output NTSC System M spec video for everything you feed it, allowing for a maximum video bandwidth of 4.2 MHz or about 330 lines of horizontal resolution (440 x 486 from a digital perspective), or if it's converting the composite video signal to RF however it sees fit and the resulting image will look more like 540 line DVD?

Any help would be greatly appreciated.


3 Answers 3


That would depend on the specific modulator, but RF signal output is just basically the same composite video it gets but modulated to RF.

So the RF output will be NTSC-M if the baseband CVBS that is sent to modulator is also NTSC-M compliant, and that the modulator modulates it in NTSC-M compliant way.

Some sources can/may output non-broadcast compliant CVBS baseband signals, because the output is meant to be connected to a TV or other equipment, and not meant to be transmitted over RF in a compliant fashion. The modulator can and will most likely limit the bandwidth to fit the signal to the RF channel, but it will still modulate whatever compliant or non-compliant signal it gets in.

  • \$\begingroup\$ Nope nope nope. Not for any real DVD player sold on the market. \$\endgroup\$ May 7, 2020 at 21:18
  • \$\begingroup\$ I suggest to read the book Video Demystifed by Keith Jack, and you might have a different view. \$\endgroup\$
    – Justme
    May 7, 2020 at 22:08
  • 1
    \$\begingroup\$ I literally designed TV encoder IP for a living. And DVD players. And media chips with TV encoders (and decoders.) This is my wheelhouse. And I can tell you that no DVD player chip since the very beginning of DVD supports this mythical extended-bandwidth composite. That's what Y/C is for. \$\endgroup\$ May 7, 2020 at 22:16
  • \$\begingroup\$ In few applications I have used Analog Devices encoders that did allow for up to 6 MHz bandwidth, on NTSC too, not just PAL. That is my understanding from those times. Granted, it was long time ago, and I've been doing digital video like SDI and HDMI since. Certainly, DVD chipsets may only output compliant signals. \$\endgroup\$
    – Justme
    May 7, 2020 at 22:31
  • \$\begingroup\$ I had the Pioneer Laser Disk in 1981 and it had 425 H lines of resolution on NTSC. smokie and the Bandit and ABBA were burnt into my memory cells as I used the signal to test my Moog synth for TV scrambler descrambler demonstrations \$\endgroup\$ May 8, 2020 at 0:14

The DVD player composite encoder will low-pass filter the 720x480 YCbCr digital picture prior to encoding in NTSC or PAL. This is to comply with the spec for baseband composite, allowing the video signal to fit in the allotted RF bandwidth prior to modulation.

If the composite video exceeds that allotted 4.2MHz bandwidth, bad things happen to the sound carrier when they get modulated to RF, so that's why it's not supported or available on a DVD player.

Y/C (S-video) is the step-up alternative that provides better luma bandwidth (although chroma is still limited to that provided by composite).

So the number of lines on baseband composite is only as good as RF composite format allows - about 330 lines, as you state.

So why did (and do) DVD players limit composite video bandwidth, even though it's technically possible to send higher resolution signal if RF isn't in the signal chain?

It's not a technical issue, but a marketing one.

In 1997, when DVD was introduced, TV coax-in was the norm. Sets with A/V inputs were still a premium item. High end cinephiles adopted DVD component for home theater, migrating from (composite) laserdisc. They upgraded their whole signal chains because, well, they could.

Meanwhile, low-end consumers delayed their TV upgrade for HDTV (1999) and/or flatscreen+HDMI (2002-ish). In those several in-between years they soldiered along with coax-in RF or, if they were lucky, A/V-in until their TV upgrade cycle came, and with it, new-TV choices sporting a slew of better inputs including Y/C (S-video), component, and slightly later, HDMI.

We can divide the DVD player era into two distinct time periods:

  • Pre-HDMI (1997-2002): Component for high-end, RF for cheapskates and Grandma
  • Post-HDMI (2002 onward): Whoo-hoo! A new big ol' flatscreen from Costco! Forget RF!

In either pre- or post-HDMI era, there was no incentive for 'better', non-RF-compatible composite in a DVD player. Pre-HDMI it would have been an annoyance for the many, many RF users. Post-HDMI, buyers couldn't care less about juiced-up composite because there were so many better connection choices on the new flatscreen TVs whether they wanted HDMI or not.

And let's not forget about the content providers who were freaking out at the idea of any higher-quality format enhancing the ability to capture analog out to re-encode for 'piracy' (nevermind DeCSS rendering that point moot). The last thing they wanted was better composite (let alone component); they worked really hard to make it worse with Macrovision. That rights-freakout hellbeast carried over to HDMI, with its own fun set of problems and hassles.

Anyway. And now, storytime. My first trip to Costco, January 2005. Standing in the next checkout line, a younger, Santa Cruz Mountains looking dude with exactly two items on a flat cart:

  • 40" Vizio TV
  • handle of Tanqueray

And I thought to myself, "that cowboy is ready to par-tay!"

  • \$\begingroup\$ The DVD players might limit the bandwidth. However the output of these devices is not meant to be directly broadcasted with a compliant transmitter, so luminance information from composite sources can be present at up to 6-7 MHz to have better quality. \$\endgroup\$
    – Justme
    May 7, 2020 at 21:13
  • \$\begingroup\$ That doesn't follow industry practice for any video gear on the market. Why? CPST is limited to 4.5MHz to prevent intrusion into sound carrier when modulated on RF. It will actually flunk validation on a test set (e.g., VM700) if it doesn't. On the other hand, Y/C provides that extended luma bandwidth when the gear supports it. \$\endgroup\$ May 7, 2020 at 21:15
  • \$\begingroup\$ Ah, I wasn't sure whether composite video from DVD players/video game consoles was fully NTSC/PAL-compliant or just followed their colour encoding. How come S-Video/RGB can look sharper on old video game consoles with resolutions as low as 256x224 like the Super Nintendo though? Is this purely to do with the separation of luma and chroma? Will composite video always max out at ~330 TVL for NTSC video? \$\endgroup\$
    – pvmnerd999
    May 7, 2020 at 21:30
  • \$\begingroup\$ If it is broadcast equipment, then it has to follow broadcast standards. If it is home DVD equipment or professional studio equipment, the composite is not meant to be broadcast over RF anyway, not without further processing, only between equipment, so it does not have to pass tests that are meant for composite signals that will be broadcasted over RF. Even if it is modulated to RF for a TV set, the modulator would limit the composite bandwidth before adding sound carrier. Even then it can have higher chroma bandwidth than meant for RF broadcast. \$\endgroup\$
    – Justme
    May 7, 2020 at 21:33
  • 1
    \$\begingroup\$ Keith Jack notwithstanding, that's not what the DVD industry did, for the reasons I stated: to avoid compatibility issues with RF modulators. \$\endgroup\$ May 7, 2020 at 22:27

Welcome to Electrical Engineering.

Your video quality is likely to be limited by the fact that you are using composite video rather than anything the modulator is doing. The colour difference signal needs to be restricted in bandwidth before being added to the composite signal. The TV has various options for separating the luminance (black + white) and colour difference signals out of the composite signal. The cheaper options will restrict the luminance bandwidth.

I played around with the various connection options when I first got a DVD player (to use with a wide-screen CRT TV). I had a baseband composite video connection, without needing to use a modulator. It was still noticeably worst than the next best quality connection - S-Video.

My best connection was component video but this was only superior to S-Video for menus generated in the DVD player itself. The DVD format records a colour difference signal so there's no extra information for a component video connection to carry in addition to that available via S-Video.

(I'm in the UK so my TV offered RGB for component video rather than YPbPr. The DVD player had both.)


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.