SD resolutions dropping on 110W

I still think their goal is to free up 2 more TPs for HD. The combination of down resing and I bet they move error correction a bit too will give them the space they need. Customers that complain get sold an HD package.
 
Here's an updated version of the comparison image posted the other night.

Left: Dish @ 480x480, no edge enhancement (HBO Comedy SD)
Middle: Bell @704x480, no edge enhancement (Channel Unknown)
Right: Dish @ 544x480, with edge enhancement (HBO East SD)




As with the Dish vs. Dish example with the picture of a woman earlier in this thread (with and without edge enhancement), shapes and details are retained more faithfully when edge enhancement (EE) is not used.

The 480x480 sample from HBO Comedy SD has more problems with spots in the image blending together than the other images due to its greatly reduced horizontal resolution, but the shapes and edges generally look much closer to how they should since no EE filtering is being used. The image is rather soft, but it looks much more natural and accurately detailed than the 544x480 sample using EE. That's how big the difference can be when aggressive edge enhancement is turned off, even at a lower resolution.

Just don't forget that some channels that do use the nasty EE filtering have also switched to 480x480, so that would look like a combination of the worst aspects of the samples on the left and the right sides. If the reduction from 544x480 to 480x480 were accompanied by the elimination of edge enhancement, many channels would have much better pictures and more details than before even at the reduced resolution. However, this isn't happening, so Dish is piling the resolution reductions on top of channels suffering from edge enhancement, just making a bad situation worse.

For reference, the 480x480 sample has a raw bitrate that is just 1.4% below that of the Bell sample at 704x480, once again showing that Dish is absolutely not using anything close to the latest and greatest tools for their video.
 
wow now that is a MAJOR FAIL! 480 x 480 @ 4034 KB/s... which includes the MPEG Audio stream, that is POOP. No wonder most of the SD channels look terrible, soft/murky and blocky. How can they even begin to think what they are broadcasting is acceptable. I wish FiOS had a better DVR I'd jump over in a heartbeat.

For SD thats not poop. Remember some of the IP video providers are reducing SD down to 1.5Mb/s including audio. HD as low as 6.5Mb/s.... As better hardware H.264 encoders come along I expect it to go lower. "Distribution"

Contribution video is going higher and is currently running MPEG2 @ 40Mb/s or some even as high as 105Mb/s.. I am even starting to see some HD-SDI uncompressed stuff go across the network at 1.485Gb/s...

Also keep in mind that most of the stuff today is VBR or Variable bit-rate. So the compression equipment can set channels bitrate higher or lower depending on content. "Within a preset range".
 
I still think their goal is to free up 2 more TPs for HD.

That very well may be their plan. I don't think I've tried to disagree with that point.

However, it doesn't mean they have any good excuse for using junk for processing their SD video. Most subscribers are still SD-only, and there are still quite a few very popular channels that aren't available in HD and aren't expected to be available in HD any time soon. A very, very large part of the service Dish Network provides is based around SD video, yet their technology for handling SD video is obviously not up-to-date compared to services like Bell. Bell is a much smaller operation than Dish Network, yet they really seem to have a much better handle on technical details that are key to providing a high-quality picture to their subscribers.

Dish needs to stop screwing around with garbage like edge enhancement on all channels rather than just some, and they need to be introduced to encoding technology from this decade before the next decade begins.
 
I don't understand. Are you talking about OTA or Dish locals?

My Dish locals in HD look very lightly compressed. We have a TP with 3 channels on it at the moment. We have TP3 spot on 61.5 and TP14 on 129 uplinked (not available). It is hard to tell the difference between OTA and Dish HD locals. There is a 4th station that should be added to our spot TPs soon. Eventually we might have to share our TPs, most likely with some secondary Dallas stations, but for now 3 stations/TP MPEG-4 looks great.
 
Which is precisely why the encoding to mpeg-4 should be happening at the content providers' facilities, and not at Dish uplink centers. Content providers might even adjust the timing, e.g. with shorter or longer commercials, so that they could stat mux their 10 channels optimally.

I would think the encoding has to be done at Dish's side since they are feeding different channels into the mux than the provider. The provider wouldn't know to back off a channel for a few seconds since another is showing a lot of action.

My Dish locals in HD look very lightly compressed. We have a TP with 3 channels on it at the moment. We have TP3 spot on 61.5 and TP14 on 129 uplinked (not available). It is hard to tell the difference between OTA and Dish HD locals. There is a 4th station that should be added to our spot TPs soon. Eventually we might have to share our TPs, most likely with some secondary Dallas stations, but for now 3 stations/TP MPEG-4 looks great.

On 61.5 you should have some MPEG4 SD feeds mixed in with the HD feeds too.
 
Dish needs to stop ...

It's amazing to me that you continue to spend hours posting here instead of emailing them directly about the situation, and then have the arrogance to say that they "need" to do various things that an amateur thinks needs to be done.

Each subscriber is definitely entitled to their own opinion of the PQ, but amateur opinions of how that should be obtained is a bit silly.

I don't see a Heart Surgery Guys Forum with people saying " if they would quit tying off this artery, and instead... " . :rolleyes:
 
I would think the encoding has to be done at Dish's side since they are feeding different channels into the mux than the provider. The provider wouldn't know to back off a channel for a few seconds since another is showing a lot of action.
Yes, that's how it's doubtlessly done now, with the occasional artifacts that marr our PQ when the bandwidth is insufficient for all channels simultaneously. But as Mike mentioned on the previous page of this thread, there is a better way: non-real-time multi-pass compression.

Content providers have an incentive to do the compression themselves, since they can cram more channels into the same bandwidth, while avoiding the artifacts we suffer today when there is not enough bandwidth for all multiplexed channels simultaneously. This has the added benefit of giving them more control of the PQ of their own channels than they currently enjoy.

And Dish also has that same incentive, so long as they kept blocks of channels from the same provider on the same transponder. Plus they wouldn't have to pay for (so many of ) those expensive real-time transcoders and statistical multiplexers.
 
Yes, that's how it's doubtlessly done now, with the occasional artifacts that marr our PQ when the bandwidth is insufficient for all channels simultaneously. But as Mike mentioned on the previous page of this thread, there is a better way: non-real-time multi-pass compression.

Content providers have an incentive to do the compression themselves, since they can cram more channels into the same bandwidth, while avoiding the artifacts we suffer today when there is not enough bandwidth for all multiplexed channels simultaneously. This has the added benefit of giving them more control of the PQ of their own channels than they currently enjoy.

And Dish also has that same incentive, so long as they kept blocks of channels from the same provider on the same transponder. Plus they wouldn't have to pay for (so many of ) those expensive real-time transcoders and statistical multiplexers.

I understand your theory but it wouldn't work with the providers feeding several feeds to each provider. Not to mention they have to take "apart" the stream anyways and running it through encryption. Just too many variables on Dish's side, they are the ones who needs to do the compression.
 
it wouldn't work with the providers feeding several feeds to each provider.
I beg to differ. All they have to do is agree on the same flavor of mpeg-4. Then, to take but one example, Viacom could multiplex let's say 10 channels into a bundle, and provide that bundle to all distributors. It would even give providers another excuse to bundle, other than simple greed. Wouldn't you like to get 10 HD channels onto one transponder, with the quality of each one of them better than they are today?
Not to mention they have to take "apart" the stream anyways and running it through encryption.
Yes, but that is easily done in real time, whereas the transcoding and statistical multiplexing is not.
 
I beg to differ. All they have to do is agree on the same flavor of mpeg-4. Then, to take but one example, Viacom could multiplex let's say 10 channels into a bundle, and provide that bundle to all distributors. It would even give providers another excuse to bundle, other than simple greed. Wouldn't you like to get 10 HD channels onto one transponder, with the quality of each one of them better than they are today?
Yes, but that is easily done in real time, whereas the transcoding and statistical multiplexing is not.

Still won't work out. DirecTV and DishNetwork are running different modulation schemes which ends up with different raw throughput per TP. They both run different tables in the stream, etc.

I'd never do something like this if I was in charge of things at any provider. You lose the flexibility of moving things around in the system as things change, uplink problems/satellite issues, etc.
 
It's amazing to me that you continue to spend hours posting here instead of emailing them directly about the situation, and then have the arrogance to say that they "need" to do various things that an amateur thinks needs to be done.

Posting on here is likely to be a better use of my time because it brings the problems I have pointed out to attention of other subscribers. The more people that know what is going on, the better they will understand what's wrong with their picture quality, and the more likely they will feel confident and comfortable reporting the problem on their own if they want to. Companies like Dish Network don't want to listen to individuals because they don't feel it is worth their time. They also very strongly avoid complaints filed by multiple people, as they have demonstrated recently with upsets over channels not being added to the Turbo HD packages. Still, lots of little complaints are at least more likely to be considered rather than ignored as one big complaint would be. All of that being said, I already stated that I plan to contact Dish after the dust settles a bit with the latest changes. I won't be expecting a response or any changes to be made in favor of their subscribers, but I will contact them all the same.

On some level you must be able to realize that Dish is the only amateur of concern in this situation. They "need" to keep up with technology because that is supposed to be part of their business and normal competition, not because I have so strongly suggested that they should. When customers are faced with regular price increases for the sake of added programming and infrastructure upgrades, the last thing they deserve in return is for Dish to limit programming additions to the highest tiers and skimp on technology, and the last thing they would expect is a worse picture rather than a better one. Your views clearly can't be swayed by facts and evidence demonstrating the mediocrity of their technology, especially since they directly contradict your previous and obviously false claim that:

kstuart said:
Dish Network gets the latest versions of professional encoding software costing quite a lot of money. It makes the PQ look better than what it would look like if they were still using older encoder versions.

Latest relative to when? Professional compared to what? It's clearly not professional compared to Bell, so what is it really? I made it quite clear that Dish has interior PQ to real professional-grade technology, yet you remain in denial. If you don't believe what I have said and shown, please feel free to verify it for yourself. If you're not willing to do your own studies on PQ then it would help if you didn't try to confuse others with baseless claims suggesting that Dish can't do any better.

Prices recently went up, resolution is now down 12% more then before, bandwidth cuts will follow soon enough, and subscribers are supposed to like it? Imagine if every subscriber tried to call Dish asking for 12% off of their bill, even for just one month. They would never, ever stand for it. They can take away 12% of the picture though, in addition to what they had already taken before (at least 15%), heavily filter what is left of the video, and encode it poorly, and for some reason we're supposed to shut up, pay the bill, and look forward to another price hike next year. Nice.

Just in case you're thinking about it, please don't suggest that I just go away, switch to DIRECTV, drop Dish, or anything else like that. I have both Dish and DIRECTV and enjoy some programming from each of them that the other doesn't carry. I know that both companies could do better and very much wish they would, but right now Dish is in need of some tech upgrades quite a bit more than DIRECTV.
 
Still won't work out. DirecTV and DishNetwork are running different modulation schemes which ends up with different raw throughput per TP. They both run different tables in the stream, etc.
Numbers please? My wag is that the modulation scheme or tables cannot make a huge difference in bandwidth, or else both Dish and Direct would use the same (better) scheme. Hmmm. I am vaguely remembering that Direct uses higher frequencies, though. So bandwidth might be dramatically greater on a Direct transponder than on a Dish transponder. No matter! If 10 channels is the wrong granularity, then 2 groups of 5 would do just as well. Maybe Mike can chime in here. We're not talking about a 1% effect, but rather a 10% or greater effect by using multi-pass, non-real-time encoding.
I'd never do something like this if I was in charge of things at any provider. You lose the flexibility of moving things around in the system as things change, uplink problems/satellite issues, etc.
I say "You would" if somebody proved that the $$$'s were sufficiently compelling. ;) As a distributor such as Dish, you want to cram as many channels onto a transponder as your customers can bear to watch. As a provider such as Viacom, you want to force the bundling of channels while keeping your PQ as high as possible. Plus content providers have their own bandwidth costs to conserve.

There is unexploited synergy here. Even if you (as Dish) wanted to keep the flexibility, and assuming you had the contractual flexibility, you could still demultiplex the channels and remultiplex them onto different transponders, while keeping most of the savings in bandwidth caused by the switch to a provider-optimized high-quality multipass compression scheme. You might still goof it up, though, and bit-starve some channels with your real-time stat muxer. :(
 
You've compared Dish to Bell. How about Dish to DirecTV?

If you'd like, I should be able to do add DIRECTV to the shirt close-up comparison this weekend. I focused on comparing Dish to Bell here since I needed to compare Dish to something that would show how much room for improvement there is. The SD on both Dish and DIRECTV is rather poor in comparison to Bell, and which of the two is better can vary from channel to channel, especially when it comes to whether or not the channel is EE-filtered on Dish.

I did a comparison here a couple of months ago between Dish with EE at 544x480 and DIRECTV without EE at 480x480. I wasn't familiar with edge enhancement when I made that post, so I didn't recognize some of the problems with the Dish image for what they were. That shouldn't change the conclusions I drew for the channel being considered, but Dish channels without EE were left out, so the comparison wasn't really a complete one.
 
If you'd like, I should be able to do add DIRECTV to the shirt close-up comparison this weekend. I focused on comparing Dish to Bell here since I needed to compare Dish to something that would show how much room for improvement there is. The SD on both Dish and DIRECTV is rather poor in comparison to Bell, and which of the two is better can vary from channel to channel, especially when it comes to whether or not the channel is EE-filtered on Dish.

I did a comparison here a couple of months ago between Dish with EE at 544x480 and DIRECTV without EE at 480x480. I wasn't familiar with edge enhancement when I made that post, so I didn't recognize some of the problems with the Dish image for what they were. That shouldn't change the conclusions I drew for the channel being considered, but Dish channels without EE were left out, so the comparison wasn't really a complete one.

Last night MSNBC looked pretty bad, and that's usually one of the better looking SD channels on Dish. Maybe it's because I read that the resolution was dropping so I focused on the flaws in the picture, or maybe it was worse. But it would be interesting to compare Dish and DirecTV now.

I saw some SD on Comcast recently here (100% digital) and I thought it looked pretty good compared to Dish. But it's hard to tell unless you can do an A/B comparison.
 
Last edited:
Last night MSNBC looked pretty bad, and that's usually one of the better looking MSNBC channels on Dish. Maybe it's because I read that the resolution was dropping so I focused on the flaws in the picture, or maybe it was worse. But it would be interesting to compare Dish and DirecTV now.

Just peeked at MSNBC on Dish. It's 480x480 with heavy edge enhancement. The bad picture is not in your imagination. :(
 
Posting on here is likely to be a better use of my time because it brings the problems I have pointed out to attention of other subscribers. The more people that know what is going on, the better they will understand what's wrong with their picture quality, and the more likely they will feel confident and comfortable reporting the problem on their own if they want to. Companies like Dish Network don't want to listen to individuals because they don't feel it is worth their time.

Hello ? Observation based on ZERO data points.

You already said that you have not emailed the address once.
 
Hello ? Observation based on ZERO data points.

You already said that you have not emailed the address once.

Personally, it's based on past experiences reporting problems to tech support for both Dish (who never responded to me) and DIRECTV (who always suggested that further upgrading my programming would fix technical issues). I always kindly suggested that the problem reports be forwarded to engineering if the normal tech support wasn't quite sure what I trying to report, but I wouldn't be surprised if that suggestion was ignored. More generally, I have no doubt that other forum members here have encountered similar problems in the past, reporting both to tech support and to the dishquality address, like this one regarding HD-Lite and Dish basically claiming that anything they can stretch to an HD resolution is a "true HD image." Seriously, do I really have to try to make the case that big corporations aren't interested in what one person, one outsider has to say? There have to be many thousands of reports online, dozens of which I've encountered on various occasions, of these kinds of large companies completely dismissing or denying their own problems to save face. If the company doesn't acknowledge a problem, then it must not exist, and the sham carries on. Dish and DIRECTV won't hesitate to communicate to subscribers via their statements or other means that some new service is being offered that they can pay more for, but where and when do they announce to customers that the poor SD quality is about to get worse, and ever so generously at no extra charge?

No, I haven't specifically e-mailed the dishquality address yet. I already acknowledged this twice and explained that I haven't yet, but I do plan to contact that address once the latest changes settle in. It makes more sense to me to know how far-reaching the latest changes will be before contacting them. For all I know, additional changes for better or worse may have begun before that time. I can't know yet it another wave of changes is planned for during or shortly after the continuing gradual resolution drops across the platform. Just as an example, I did not expect the resolution drops to involve both 110W and 119W since the the drops appeared to start only on 110W and most channels on 119W are much more heavily filtered than most channels on 110W. I hoped they were smart enough to not make their worst quality channels even worse. I was overly optimistic and wrong, learning a few days later that they were going after 119W as well. Gaining that extra knowledge from trying to wait and see how things work out should help when it comes time to state my case.
 

Users Who Are Viewing This Thread (Total: 0, Members: 0, Guests: 0)

Who Read This Thread (Total Members: 2)