(see sublink) about TWC and so-called switched broadcasting or video I posted in my local TWC thread. Google sent me an e-mail yesterday listing the link, a newer Google feature that monitors news items for you.
Regarding whether TWC 'compresses' network sources: Cable folks often deny it. As this 2003 Broadcast Engineering article
points out, cable firms certainly 'repackage' HD signals. They strip away some or all PSIP data (channel/program information), plus any subchannels within the 19.39-Mbps (OTA) channel signal not being relayed by cable. Then, varying amounts of rate shaping may be employed, such as removing null zeros--'filler' data sent to complete a transmission format--that doesn't influence PQ. Rate shaping can involve requantization, modifying MPEG-2 video without having to decode it. "This process achieves further bit-rate reductions. But requantization is a lossy process and, under stressful conditions, it may result in artifacts," the article notes. "Rate-shaping systems perform best when reducing rates by less than 25 percent," it adds. Also, programs delivered at constant bit rates may be changed to variable bit rates (VBR), then the VBR may be reduced.
Next, whether rate-shaped or not, video is combined with other programs (remultiplexed), often using statistical multiplexing (statmuxing), which allocates bit rates between a number of channels depending on detail and motion present at each instant. Statmuxing high bit rate channels such as sports with mostly talking-head sources helps ensure capacity is available when it's needed. "A disadvantage of this process is that the coincidence of high motion and high detail in two or more program streams can adversely affect the picture quality of a third, low-motion, low-detail program," notes the article.
Cable STBs can limit effective resolution
(resolvable detail) to <1300 lines. Using HDNet test patterns relayed by NYC's TWC, my SA8300HD measures
1290 lines. Another cable STB, from RCN, delivered 1335 line resolution. And remarkably, a briefly used SA8000HD from NYC's TWC only provided 890 resolvable lines. Measured effective resolution (~1280 resolvable lines from 1920X1080) differs from 1920X1080 reformatted to 1280X1080 (DirecTV's 'HDLite') since a delivered 1280X1080-re-formatted signal must still be MPEG decoded in homes, which reduces effective resolution.
Of course, using HDNet test patterns to estimate received video quality generally isn't necessarily valid. A cable company could easily 'tweak' the video quality of one or more channels. With so many variables involved, perhaps only unbiased professional organizations, using calibrated spectrum analyzers or similar gear, could pin down how much cable-delivered HD/SD video quality differs from that delivered to head ends by video sources. -- John