I think tvropro and I have basically agreed to disagree about 4DTV SD vs. pizza dish HD. I am no fan of pizza dish HD-lite and have been searching for better solutions for quite some time. In that regard I have spent considerable thought and effort in determining how to rank the possibilities, both objectively and subjectively.
I too have well-trained eyes and ears, having worked off-and-on in the broadcast and studio worlds since the mid-1970s. I have seen some of the finest SD and HD masters available, and have tried to get as close to this at home, although that is difficult. Over time one gets very good at detecting faults in the broadcast chain with only these tools, especially over-compression. However it can be difficult to objectively compare different systems only with one's senses, particularly when the source material is not the same, or not from the same source.
I believe it is also useful to quantitatively understand how a system works and how this affects the video and audio. When something changes over time, objective measurements often spot it first and give a quick assessment of the best methods for subjective comparison. Thus I evaluated several different forms of satellite delivery systems both with my eyes/ears and with direct measurements. They nearly always agreed.
As I indicated above, I'm nervous about drawing too many conclusions from subjective observations over time on different equipment and with different source material. I'm sure tvropro has seen many examples of pizza dish HD-lite, but is very proud that he doesn't subscribe to these services. On the other hand I have run my FTA and OTA systems in parallel with 4DTV, Dish and Shaw subscriptions at the same time. In addition I have collected the original transport streams from each of these services for both subjective and objective comparisons. FTA/OTA is easy, but I have installed R5000 mods in my 4DTV, Dish and Shaw receivers, making the original data available for these sources also.
The most interesting results are when I am able to collect transport streams from two or more delivery systems that are sourced from one or another. An example would be an OTA broadcast derived from a FTA network feed. This will show losses introduced by the broadcast station. Another example would be DN vs. the original 4DTV or FTA feed. I have spent a lot of time on this, particularly when HBO and STARZ had both HD and SD available on the 4DTV side. But occasionally a DVB-S or S2 master feed will go in the clear and I have been lucky to make a number of simultaneous recordings of the TS for both Dish and the originals. Comparing Shaw is harder because I haven't been able to trace back to their sources. However I do have a number of TS recordings of the same material that aired at a different time in the States, whether FTA, 4DTV or DN. This is useful, but not as ideal as the direct comparisons.
With the TS I can loop segments of video/audio that are exactly the same frames from different providers. I can go frame-to-frame and compare pixel-to-pixel, both with eyes and numerical comparisons. The differences are not small, and one doesn't need to spend much time making fine distinctions. Quality varies all over the map.
The FTA 30+ Mbps HD feeds are absolutely gorgeous, often with negligible artifacts. I found 4DTV HD in its heyday to be pretty good, but definitely a step down. One does have to recognize that what was available by subscription was hard-limited by 4DTV's obsolete receiver technology to less than 30 Mbps per channel. While I couldn't compare them directly on the same material, it was clear the 4DTV HBO HD was noticeably better than 4DTV STARZ HD. HBO's data rates were not coincidentally higher.
Shaw HD is a mixed bag. I have seen Shaw provide better HD and higher rates at times than any 4DTV HD TS from either HBO or STARZ, from the same source material. But normally Shaw is lower in their rates, and this leads to a slightly lower perception of quality. I do believe Shaw has their compressors adjusted to heap on a lot of bits when needed, and this can lead to fewer artifacts than I saw in the same scenes on 4DTV HD. In the end I would rank Shaw HD in between the no longer available HBO and STARZ on 4DTV.
There is no question that Dish HD-lite ranks lower overall. A fair amount of their HD material is bit-starved, but this is not universal. Some channels apparently receive a higher bit-budget than others, and when there is less load from the temporary channels, the rates and perceived quality can get very good. On average it is middling. One does have to account for the fact that Dish invariably delivers HD as anamorphic 1440x1080i. That and over-compression softens the image.
Out of morbid curiosity I recorded a few 4DTV SD streams on HBO, STARZ and a couple of others to compare against the same 4DTV and DN HD material. The 4DTV SD was very smooth looking, had nice colors and few artifacts. But it was TERRIBLY soft compared to any HD or HD-lite. The video seemed analogous to listening to an audio recording from the 1930s. One could enjoy the content, but technically it was simply inferior.
I have done no formal comparisons of Shaw or DN SD. Shaw SD seems a little lower in perceived quality than 4DTV SD, but not too bad. I'm set up for both the western and eastern arcs on DN, but mostly stay on the west. DN SD on the WA is simply awful.
I have not had a chance to compare DirecTV or Bell. Anecdotally the fights on Satguys never seem to yield a victory for either DN or DTV on PQ. They are likely equally mediocre. The same does not seem to be the case for Shaw vs. Bell. Bell uses DN technology and apparently squeezes even more channels per TP MHz than DN does. It is also my understanding that Bell HD is only 720p. I have found few Canadians that disagree with the assertion that if you want numbers you go to Bell, but if you want quality you go to Shaw.
To summarize I am providing a rough ranking order of satellite providers from my subjective eye/ear evaluations, that invariably correlate with objective measurements. No doubt this is entirely dependent on the quality of the source material and the effort placed by the provider. One can find examples that contradict the rankings, but these are more the exceptions than the rules. Most of the sources are MPEG2 and I have given very rough guidelines of the rates I have measured.
1. FTA HD, 30+ Mbps
2. FTA HD, 20-30 Mbps
3. 4DTV HBO HD, ~15-20 Mbps - no longer available by subscription
4. Shaw HD, 10-24 Mbps - always the same channels
5. FTA HD, 12-20 Mbps
6. 4DTV STARZ HD, ~10-15 Mbps - no longer available by subscription
7. DN HD, 6-10 Mbps (H.264) - not common and only on certain channels
8. Shaw HD, 6-10 Mpbs - always the same channels
9. DN HD, 3-6 Mbps (H.264) - the vast majority
10. FTA SD and 4DTV SD - virtual tie
11. Shaw SD
12. DN SD - the goon award
Based on comments from others, I would assume DTV is on par with DN and Bell HD would rank below the lowest DN HD. OTA is hard to place, partly because more and more stations here are adding sub-channels. I don't really watch it anymore.
.
