Can you see the difference between...

#1
broadcast pixelation and "live" pixelation

I was watching the Olympics with a friend the other day when we saw some pixelation. (Anyone watching TV with me expects to spend some time viewing the signal meter, and they know they'll have to listen to me babble about digital reception if anything changes.)

As we were watching the interview of the gold medal winner, we saw some interesting pixelation. I pulled up the signal meter and as we watched the rock solid signal, my friend remarked that there sure wasn't any signal problem.

I told him I suspected as much and asked him if he could see the difference between pixelation that is broadcast (incorporated into the stream before sending) and "live" pixelation that comes from home reception.

He said that he also had suspected as much as he was seeing what I see. Broadcast pixelation looks more like "pictures" of pixelation. It lacks the spontaneity of live multipath and spurious bursts of noise that cause pixelation.

So the question remains, "Who sees the difference between broadcast pixelation vs. reception pixelation?
 

Don_M

DTVUSA Member
#2
Pulling up the signal meter has long been the first thing I do when the kind of pixelation where part of the image freezes begins to occur. As you observe, Jeff, this can be from mutipath interference or a weak signal at the receive end -- and it can also be partial loss of signal at the station's satellite downlink. A steady meter eliminates the former cause as a reason. Downlink signal loss tends to result in macroblocking that's much more fleeting: It's almost always long over by the time one gets the signal meter on display.

The Olympics broadcasts, OTOH, have suffered from quite a bit of pixelation that seems to be occurring because NBC's Vancouver broadcast center isn't using a "pipe" sufficiently large enough to capture fast action shots that are a staple of any Olympic Games. This pixelation tends to be finer, smaller and blurrier than macroblocking caused by signal loss.

Our local OTA board over at AVS has had a running debate about this topic since the Games began. Is this occurring because Denver's NBC affiliate chooses to broadcast two sub-channnels (for which it gets tarred regularly), or is it the fault of the network? This poster seems to suggest it's the network. He's in a location where he's able to receive the affiliates in both Denver (KUSA, to his north) and Colorado Springs (KOAA, to his south), and reports no PQ difference between stations. (KOAA has only one sub-channel.)
 
#3
Pulling up the signal meter has long been the first thing I do when the kind of pixelation where part of the image freezes begins to occur. As you observe, Jeff, this can be from mutipath interference or a weak signal at the receive end -- and it can also be partial loss of signal at the station's satellite downlink. A steady meter eliminates the former cause as a reason. Downlink signal loss tends to result in macroblocking that's much more fleeting: It's almost always long over by the time one gets the signal meter on display.

The Olympics broadcasts, OTOH, have suffered from quite a bit of pixelation that seems to be occurring because NBC's Vancouver broadcast center isn't using a "pipe" sufficiently large enough to capture fast action shots that are a staple of any Olympic Games. This pixelation tends to be finer, smaller and blurrier than macroblocking caused by signal loss.

Our local OTA board over at AVS has had a running debate about this topic since the Games began. Is this occurring because Denver's NBC affiliate chooses to broadcast two sub-channnels (for which it gets tarred regularly), or is it the fault of the network? This poster seems to suggest it's the network. He's in a location where he's able to receive the affiliates in both Denver (KUSA, to his north) and Colorado Springs (KOAA, to his south), and reports no PQ difference between stations. (KOAA has only one sub-channel.)
Don,

I have heard many "accusations" of quality problems that are caused by broadcasters who are using sub-channels. Thus far, I haven't seen anything that I would regard as evidence of any problem with the using of those channels.

I think the problem is the fault of the network. It is possible that NBC Vancouver may be pushing the limits (just a little) in an attempt to minimize the turn around time to broadcast. But, from what little I've seen (I've only watched a couple hours), I think it is a mixture of cameraman error, and some "less than optimized" live-uplinks.
 

Trip

Moderator, Webmaster of Rabbit Ears
Staff member
#4
You're also in St. Louis (as per your Location), where subchannels are not prevalent on most of the stations. KETC probably doesn't look so great though.

- Trip
 
#5
You're right, Trip, we don't have many subs in STL. Wish we did, but aside from some broadcast pixelation lately, PBS is looking just fine to me.

The problem is this...it is very hard to see and perceive a picture that has lessened in quality. It isn't until you go the other way (to increased quality) that it is easiest to see a difference. (So, if PBS quality has decreased, I haven't seen it yet.)

It is very difficult to perceive the "graceful degradation" of digital signal. It's just not all-or-nothing.
 

Trip

Moderator, Webmaster of Rabbit Ears
Staff member
#6
KETC is also side-converted to 720p, which probably helps. My local PBS is doing 1080i at 11 Mbps and it looks just awful. Just the opening graphics in the Newshour devolves into a blur of blocks. To add insult to injury, they have an SD simulcast of it sucking up 3.3 Mbps. The HD would look a lot better at 14.3 Mbps...

- Trip
 

Similar threads

Top