Ditigal TV Signal Quality

Following on from Joergs DTV aerial discussion I did a few empirical measurments on the signal strength and quality metrics on my set - a Panasonic 32" 100HZ LCD. The DVD Tuner is an earlier Panasonic chipset and gives a similarly abrupt transition between Q=10 and Q=rubbish.

The results on UK DVT signals are as follows

Signal Quality Notes Level

10 10 perfect 8 10 perfect 7 10 perfect 6 10 perfect 5 10 OK some glitches per hour 4 7 just about OK 4 6 unwatchable

So there is a very sharp transition between quality 10 that is fully error corrected and Q=6 which is unwatchable. This is with natural signal degradation from a heavy rainstorm.

It all seems a bit too digital. I would expect the quality metric to fall to zero for an image that is fundamentally unwatchable with bit errors in every block rather than 6. Below signal level 4 was useless.

I used an attenuator to find out if with a steady lower level signal there was a different result. And not surprisingly there was.

Signal level 4 with a passive attenuator in fine conditions still gives Q=6 but the picture is basically OK with the odd rare glitch.

It seems like the quality measures used only work on lab attenuated signals rather than naturally degraded noisy ones in the real world during heavy rainfall :(

Anyone else care to share the stats off their set (wet and dry)?

Regards, Martin Brown

** Posted from
formatting link
**
Reply to
Martin Brown
Loading thread data ...

On a sunny day (Thu, 24 Jul 2008 12:31:48 +0100) it happened Martin Brown wrote in :

'Quality' display is probably simply a 'bit error' display. There is a lot of error correction algos involved. Above a certain number of faulty bits, the errors can no longer be

100% corrected, and that kills digital. The same may happen in case of extreme noise (weak signal).

I presume you tested DVB-T.

Reply to
Jan Panteltje

Yes. I had hoped that the "quality" was actually an indication more along the lines of 10 - A* - B*. It stays at 10 until there are multiple uncorrected errors and then it jumps quickly to 6. Hardly a useful metric for anything.

As diagnostic tuning tools go it leaves a lot be desired. It seems to be unreasonable that the picture falls apart totally at quality "6/10". Better than nothing though which seems to be what Joerg has to fight with.

Regards, Martin Brown

** Posted from
formatting link
**
Reply to
Martin Brown

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.