Oscilloscope Pt 2: Component Video Testing with Ste

Here's a continuation of Ste teaching us how to use an oscilloscope. This time we focus on testing component video testing, as well as brush up on our scope skills. For the full formulas used in voltage calculation, please see the main page: retrorgb.com/os...
Part 1: • Oscilloscope Basics: T...
Please consider supporting the channel in any way: www.retrorgb.c...

[@theseabass]

This video was very enlightening! Great work from both you and Ste! What I got from it was component video has a standard that you should try and meet, but each output device is likely going to give different results due to slight differences in how that device outputs its electrical pulses (variation could come from age, manufacturing differences, etc) and the quality of the cables. Poor quality cables (not shielded well so crosstalk from other lines happens or not thick enough of wire) are not equipped to handle either the rapid changes in the pulses or the super high/low ends of these pulses which is what causes video degradation, as the image is supposed to be outputting one thing, but the cable is stuck somewhere in between where its only outputting what it is able to. Color balance in cable making should be tuned to the white balance of the output itself (the baseline), and not necessarily the exact standard of component video. Although, you should aim to get the white balance in the cable to as close to spec as possible, but having a slightly lower than spec signal is better than having it higher, to avoid clipping. I jotted this down as my own type of notes. Did I understand everything correctly?

[@bowi1332]

Truly awesome!

[@merlingt1]

These videos are great!

[@wingsuitjames]

Please consider putting a link to part 1 in the description. :)

[@RetroRGB]

Wow, I really missed the mark on that one!!! I even forgot to put it in the blog post!!! Thanks very much for reminding me: kzbin.info/www/bejne/ZnSqpYJjjJ1_bbs

[@iCach0]

I love both of your bodies, deeply.

[@beefcurtainz69]

iCach0 testicles deep

[@electronash]

Bob!

[@electronash]

Which model of Samsung TV is that? It is a bit of a weird glitch, and obviously shouldn't been feeding back to the input, but it looks a bit like "ground bounce"? Or, some clamping gen in the TV is kicking in. It's also weird that it happens just BEFORE the rising edge of Hsync, as if it's part of the PLL that is interfering.

[@electronash]

One way to look at the colour bars pattern on the o'scope is that every odd bar on the image contains blue (but with slighlty more blue for every odd bar). I realise that they are colour difference channels too, but you can kind of see that if both the Pb and Pr channels signals are 0V (and Y still high), then it gives a white colour on the screen (ie. lack of any *difference* in colour from the B&W Luma signal). But, it also has to extra the "Green" info from those signals, so both Pb and Pr get subtracted from Y to do that, kind of. (or at least a certain proportion of each.) Pb = B-Y. Pr = R-Y. If you look at the Pr channel, you can see it is at a higher voltage only for the Yellow, Pink, and Red bars, which makes sense. It is a weird system, but kind of makes sense if you adjusts the colours while staring at the scope for longer. lol Most of the colour decoders in TVs used a very similar principle, so they could retain the original Black & White (Luma) signal, but add the colour difference channels with other techniques. With Composite video (essentially what gets broadcast via "RF" on analog TVs), the colour difference info gets modulated onto the Y (Luma) signal using a higher carrier frequency. The phase shift of that modulation determines the hue of the image, and the colour difference signals are derived from that. (the colour burst at the start of each line is used to keep the PLL reference inside the TV locked on to the colour carrier frequency, so it can then determine the phase shift of the actual carrier along each line.)

[@electronash]

I'm up to about the 38-minute mark in the vid now. The only minor thing I would say is that it may give the impression to some that the Genesis itself is actually outputting Component video "directly". It's actually the clever analog circuit inside the HD Retrovision cable that is converting from the Genesis' RGB output into Component. But yep, I see how Ste and Nick spent a LOT of time to essentially "calibrate" each type of cable for each different model of console, so their cables should effectively give the closest to "ideal" levels from the Component output side. (the DACs on many retro consoles often weren't that accurate to begin with, and generic Component converters don't really take those differences into account.) Oh, and I didn't mention earlier that the Component signals are basically going both positive and negative (with reference to the "0V" point, which is erm, what it means. lol) So, on a pure Green colour bar (which lacks any Blue or Red), both Pb and Pr signals actually go negative, but the Y (Luma) signal stays positive. (since the "green" channel is derived from all three signals, and using subtraction of Pb and Pr.)