How is "brightness" and "contrast" defined in rgb video at the signal level?

I know Luminance (Y) is "brightness" for rgb video, and that Luminance is defined by Y = 0.2126 R + 0.7152 G + 0.0722 B, but is the "brightness" of the video just determined by the peak to peak voltage of the Y signal?

Also, what is "contrast" as as the signal is concerned?

The overall brightness of the displayed signal is the average of the video signal levels. The peak brightness is the peak video level. The contrast is the peak-to-peak level of the video. The brightness control on a TV thus moves the displayed average (DC) up or down. The contrast control adjusts the gain of the video amp to change the displayed peak-to-peak video level.

crutschow said:

The overall brightness of the displayed signal is the average of the video signal levels. The peak brightness is the peak video level. The contrast is the peak-to-peak level of the video. The brightness control on a TV thus moves the displayed average (DC) up or down. The contrast control adjusts the gain of the video amp to change the displayed peak-to-peak video level.

Click to expand...

So in summary, brightness is the average level and contrast is the peak to peak... so does that mean if you want to change the brightness only, you have to attach a (litterally) dc bias to each of the signals? Because if that input was AC coupled, the dc bias (brightness) would disappear? Or am I not understanding properly?

As far as contrast, since it's the actual signal amplitude (peak to peak) is varied, that would still be presented if AC coupled.

The average brightness can not be easily changed in the video signal if it's an NTSC signal with the sync, since the sync level determines the black level and the video signal is referenced to that. Thus, even if the signal is AC coupled, a DC restore circuit (clamp) is used to generate the DC level for the video signal in the display. To change the average brightness you would need to change the level between the video signal and the sync levels, which requires some specific circuits.

That is why the brightness level is normally changed only in the display where the DC level has been restored, the sync signals have been removed, and you thus can change the average DC level of the video signal.

In all the RGB monitors that I've used, "brightness" is the minimum level displayed. Even the cheapest color monitors/televisions had DC restore circuits. When video goes away the raster went as dark as it could go. I set it by removing video and adjusting brightness so the raster just disappears. The black level is set relative to the "back porch" (a short period following the sync pulse) in the waveform, not the sync itself. Black level is supposed to be 7.5 IRE units (Back Porch being 0 and white being 100).

Only with old, old, B&W (before color TV) NTSC televisions without DC restore did "brightness" control the average. You could tell because when the signal went away the raster went gray not black. It was common to see the vertical retrace on the cheapest ones.

mneary's description is correct.

mneary said:

In all the RGB monitors that I've used, "brightness" is the minimum level displayed. Even the cheapest color monitors/televisions had DC restore circuits. When video goes away the raster went as dark as it could go. I set it by removing video and adjusting brightness so the raster just disappears. The black level is set relative to the "back porch" (a short period following the sync pulse) in the waveform, not the sync itself. Black level is supposed to be 7.5 IRE units (Back Porch being 0 and white being 100).

Only with old, old, B&W (before color TV) NTSC televisions without DC restore did "brightness" control the average. You could tell because when the signal went away the raster went gray not black. It was common to see the vertical retrace on the cheapest ones.

Click to expand...

When you say the "minimum" level displayed, what do you mean as far as the signal goes? Do you mean the lowest voltage value in the signal?

For example say you have waveforms (a) and (b) of a single scan line:

**broken link removed**

Would you say that waveform (b) is brighter than waveform (a)?

If so, if the monitor has the signal AC coupled, the DC information about that waveform would be lost, correct? Or did the DC restore actually restore the input DC information? Also, what if the output stage of the video is AC coupled? Then no DC information would get past anyway.

Also, can you give me more information about black level being set relative to the back porch?

The reason why I am asking all of this is because I made a RGBs / RGBHV (little s for composite sync) to component converter. The picture looks perfect when using console sources (15.7khz horizontal sync), but the picture is bright when using a computer monitor (640x480, 31.5KHz). Trying to see what I can ascertain from the actual input and output signal.

Most ordinary people set the brightness (black level) much too high so that black parts of the image are grey.
I remember old TVs used capacitor coupling without a DC restorer and the black level was fairly good when there was lots of contrast but dark scenes were washed with grey paint.

Another characteristic of AC video coupling without DC restore is that a block of white on a grey background would have a black horizontal streak on both sides of the block.

FusionITR said:

When you say the "minimum" level displayed, what do you mean as far as the signal goes? Do you mean the lowest voltage value in the signal?

For example say you have waveforms (a) and (b) of a single scan line:

**broken link removed**

Would you say that waveform (b) is brighter than waveform (a)?

If so, if the monitor has the signal AC coupled, the DC information about that waveform would be lost, correct? Or did the DC restore actually restore the input DC information? Also, what if the output stage of the video is AC coupled? Then no DC information would get past anyway.

Also, can you give me more information about black level being set relative to the back porch?

Click to expand...

Waveform b is brighter than A

The back porch voltage is defined as the black level. A proper operating DC restore circuit will clamp the back porch level of an AC coupled signal to 0V, thus re-establishing the proper DC level. From that point to the display, the video signal must be DC coupled to maintain the proper black level. Any signal above the black level is thus going toward white.

I have another general question about horizontal and vertical sync.

I can get some very specific information on horizontal sync for 15.7khz horizontal resolution from google:

4.7us sync pulse
5.9us back porch
51.5us color brightness
1.4us front porch

Repeated for 242 scan lines, then 20 scan lines of empty video while performing vertical sync

The problem is, I can't find any specifics on the vertical sync timing information. I also done know what the timing differences are for even and off fields.

I found this:

**broken link removed**

But the chart is confusing and there is no timing information.

Also, how do these numbers change with resolution? IE. 15.7KHz to 31.5KHz

The vertical sync is shown in the diagram you attached. It is detected by circuitry in the monitor that observes the pre-equalizing, serration, and post-equalizing pulses and reconstructs the bottom line shown in the diagram. The bottom signal is not broadcast; it is internal to the monitor.

There is no NTSC standard for 31.5 kHz, and to the best of my knowledge there aren't any devices that attempt to cram R, G, B, H, and V onto one wire at 31.5 kHz.

But the chart is confusing and there is no timing information

Click to expand...

The dotted red lines in the chart represent the beginning of each horizontal sync pulse, and are therefore about 64 us apart. (Some extra marks show half lines on the lowest traces.) See the Horizontal timing charts on the same web page for details.

Would you say that waveform (b) is brighter than waveform (a)?

Click to expand...

Please point out the back porch; it appears to me that it's the flat spot just to the right of the sync pulse, about the same length as the sync. The brightness is relative to the back porch. A DC restored video should rely entirely on the back porch level to set black. If those two signals are before the DC restore circuits, then the 0V lines are meaningless. The video in the two waves are different, but very similar. One has a lot larger sync, but the DC restore should ignore that. The one on the right looks like it may have a tiny false front porch at 0V which also should be ignored.