Decoding PAL Video from a Nintendo with an Airspy SDR
Oona (also known as [Windytan] and @windyoona) was recently looking for a way to capture PAL composite video from her old 1980’s Nintendo Entertainment System (NES) without spending a bunch of money on what are often poor video capture cards. As she already owned an Airspy SDR she decided to receive the PAL signal with the Airspy and modify some software to act as a PAL decoder.
SDR-based PAL decoder is still black & white, but after a notch filter on the audio the picture quality is getting a lot better. pic.twitter.com/SUoBlnBZF3
— Oona (windytan) (@windyoona) August 16, 2019
PAL decoding was handled via some modifications to her private Tempest software. Normally Tempest type programs like TempestSDR that we covered in a [previous article] are used to spy on computer/TV monitors from signals that are unintentionally emitted in the surrounding area.
Oona has made the connection from the composite output directly to the SDR antenna input so it’s not unexpected that you’d have a strong signal. However, I have to admit that’s an incredibly clear image for a video being demodulated via a software radio.
What makes this an even more amazing feat is that the latency is low enough that it’s nearly playable using a computer and SDR in place of a television set.
I’ve been looking for ways to capture NES video on my Mac. No easy+cheap solutions, but with some changes to my Tempest tool I can use the Airspy to receive the analog video carrier. The latency is almost good enough for playing, though it’s not my goal 🙂 pic.twitter.com/B6x44NEuvK
— Oona (windytan) (@windyoona) August 11, 2019
We note that we’ve also seen SDRs used to decode standard PAL TV broadcasts before with an SDR# plugin called TVSharp.
That is very close to something i’m thinking about for a while: capture magnetic head’s raw RF signal of any kind of magnetic tape or disk device, and then do all the demodulation and signal treatment by software.
SDVCR, SDTapeDeck, SDTapeReader… for instance.
i believe it could be pretty useful and could provide much better image/sound/data quality, by bypassing all old and often badly aligned signal treatment’s electronic, and replacing it with much powerfull software (possibly adding some refined treatment unavailable on original device).
But my first wonder about it is how to connect, for example, a VHS VCR’s video head directly to an SDR’s input? I guess that some electronic would be needed for, at least, impedance matching?
From what I remember from analogue video recording on magnetic tapes there are similar things to observe as I wrote before. VHS VCR had an AM modulated RF and the analogue video out with a varying bandwidth 0 to 3.5 MHz or more, the bandwidth is higher the better the recorder and used magnetic tape material was.
Do you want to detect the combined video or Synch RGS and the BW picture seperately?
Is it a Single or Multi-Standard-, VHS or SHS support for a recorder?
Which Standards are supproted NTSC, PAL and SECAM, and equally important
which version of NTSC, PAL and SECAM, e.g. resolution and tone carrier placement, 50 Hz or 60 Hz?
Is ist a true norm compliant recording or somewhat quick and dirty solution stretching standards and what variations in bandwidth and (stereo) sound carrier offset to the carrier are supported.
How early do you want to pick up the signals before they are combined as combinded “sync, BW-, colour-, tone-carrier(s)-output or sync-, BW, RGB-signals provided by the VCR? Which signals and where do you want to pickup the signals?
As early as when the VCR provided the signals from the many heads they use or after the VCR had combined the signals from the various heads?
You still depend on the use of all the mechanics and control circuits involved to get the tape aligned and transported, tape speed controlled …..
Which enhancements do you want to implement e.g. a comb filter to avoid cross color flimmer ……
Unless you want to demodulate the RF AM modulated signal you would have to use a number of AD converter and not just impedance matching.
But still you have the limitations that tapes are stretched, have deformations at the edges …
Unless you are really into electronic SM and intend to reverse engineer the (S)VHS-VCR design you have in order to process the many internal signals that the analogue video signal is composed of, why don’t you try a a good but old USB or Firewire-Wire Grabber using a Windows version that supports the software and or Firewire (e.g. PCMCIA card if Firewire is not in the Notebook) and not some of the small 15€ quick and dirty grabber. I still have my Firewire and USB 2 and wou would be astonished how good they are in converting what is provided, meaning the signal left of tapes on stretched tapes with deformations at the edges after such a long time.
Receiving Black and White picture is elementary, while decoding analogue color requires knowledge of which of the many varients of analogue color TV were used.
Color information was “added” to the existing Black and White (BW)-TV signals using AM with a reduced carrier and one sideband with additional carrier(s) for sound. The color information is interleaved into the existing Black and White (BW) signals (HV-sync and picture) in such a way, that there is only minor degredation e.g. cross-color flimmer.
Every analogue TV receiver would receive color TV signals and display them as BW picture. Most TV-receiver were very tolerant for off spec signals like from the Aplle II. If you had a 50 Hz TV you would receive also 60 Hz based signals, but without having support for the correct sound carrier frequency you had no sound.
It a long time until the necessary chipsets were available that would allow to built Multi-Standard color TV receiver that supported not only all three variations used for color adding color information (NTSC, PAL and SECAM), but also supported the existing variations in bandwidth and (stereo) sound carrier offset to the carrier.
That is why I find it unlikely that color support is that easily to achieve unless you know which color encoding system and which parameters were used.