[OT] Video Standard Q: (Was Re: New Thread....about anything but Pulseaudio or yum)

Tim ignored_mailbox at yahoo.com.au
Thu Jan 17 12:52:57 UTC 2008


On Thu, 2008-01-17 at 10:17 +0800, Ed Greshko wrote:
> So, I wonder what "106% NTSC color capability" means.  If NTSC is a
> standard then how can you have over 100% capability?

Perhaps able to handle over-/under-driven colour signals?  On some TV
systems you allegedly couldn't transmit 100% pure colours from a
composite video source, it upset the carrier signals.  So the colours
were matrixed together in a way that always subdued them slightly.
Perhaps they're trying to undo the matrixing.

But it sounds more like some typographical error, or a nonsense claim
like how various manufacturers state power output using PMPO instead of
RMS.

Though I've just had fun trying to work out voltages inside an audio
mixer when the manual stated dBm or dBu, and darned if I could remember
the formula.  Then when I did, Gain(in dB) = 20 log (Vtest/Vref),
reversed it into Vtest = Vref x antilog (Gain/20), I couldn't find a
calculator that had antilog!  Some rummaging around for a
twenty-year-old calculator, and I was half way there, but the
calculations didn't match the measurements (what voltages the CRO said,
while the meters were showing +8dB more signal).  I can't help thinking
I missed something obvious.

What I was trying to do was figure out what voltage should have been
coming out of the +8dBu output terminals, when I couldn't remember what
it ought to be (I'm more a video guy than an audio guy, it's not a
figure I've committed to memory).  So I thought I'd *work* out what +8dB
should be compared to 0dB (0.775V either into 600 Ohms, or not,
depending whether it was dBm or dBu).

I'm not the only one confused, it seems.  While just about everyone
knows 0dBm or 0dBu is 0.775 Volts RMS, there's differing ideas on what
voltages you get for +4dB or +8dB of gain (just try a quick Google, and
you'll come across conflicting assertions).

Just to make my mathematical nightmare worse, I don't have a
millivoltmeter, so I'm using a CRO, and I have to convert from RMS to
peak or peak-to-peak, back and forth.  ;-)

Between the maths, and the tone generator going wonky (another thing to
repair), I gave up on trying to sort out the level problems, for the
while, and settled for working on getting all the panel lamps
operational again (ripping out blown globes, soldering LEDs into the old
bases because replacement globes cost 10 times as much, and replacing
lots of globe holders where I needed real light globes but can't get the
right models anywhere).  Timeconsuming, but not brain teasing work.

If only I could fix it with a hammer...

-- 

Don't send private replies to my address, the mailbox is ignored.
I read messages from the public lists.




More information about the fedora-list mailing list