622 MPEG4 HD Recording Time

[Smith P.]

Most HD locals (the only exceptions I am aware of are LA and NY) are transmitted as MPEG 4. The new Voom channels, UniversalHD, and ESPN2 are transmitted in MPEG2 with MPEG4 headers to fool the older HD receivers into not accepting them.

No such thing exists as "MPEG2 with MPEG4 headers" ; there is other service type assigned to the mentioned channels, not "to fool" but to limit reception of the channels for new receivers like 211, 411 or 622.

 

[Kirby Baker]

Yup, the "mpeg4" channels (non locals) are plain old mpeg2 streams.

 

[Tom Bombadil]

This is why in message #1, I asked for those who could record true MPEG4 channels to assist us in this quest.

Hadn't thought of the 720p v 1080i difference. 720p does use less bandwidth and could skew the results.

I would think the best comparison would be to use 1080i, as this is what E*'s timers seem to be based upon.

The best case scenario would be if someone had a clean 622 and was receiving a 1080i HD LiL. Then set up their timers to record HD programs until the disc was full and totaled up the hours. But even in this scenario, the total hours would be affected by the bitrates that Dish gave the channels.

So this isn't an easy answer to determine.

Some of the data posted so far does seem to indicate that we will be able to get more than 25-30 hours of HD recording time from the true MPEG4 channels.

 

[Kirby Baker]

If I had HD LiL I'd help...

I would also be able to capture the same show on the R5000 and give exact file size that E* is sending down, but alas, until I figure out a "move" or get Detroit HD LiL up on E10, I am helpless... (in more ways that I care to admit!)

 

[ronjohn]

Not knowing the source of your 24, OTA or mpeg4 sat, there is another factor that could be coming into play here. I highly doubt the 622 is using any special formula for time calculation, and probably was set to baseline off a 1920x1080i channel. 24 is a 720p channel, so you already have quite a difference in size/time comparisons to a 1920x1080i channel. Probably would be best that everyone uses the same resolution channel for making comparisons, at least when comparing mPEG2 and MPEG4. Just a thought... Or am I on crack?

I had posted earlier in the thread that I recorded it from the mpeg4 local HD station. The source bitrates for 1080 are about 12.5% higher than 720. I have not been able to find a chart that compares bitrates on the mpeg4 channels to see what the difference between networks and formats is currently running(I did see one at some point). I do not have enough time to record 40-5? hours to test the recording capability and was only posting what I could do. I make the assumption that the 30 hour estimate is based on a bitrate of 16-17mbs which HDNet and a few others run at. If the mpeg4 channels are running at 8-10mbs, which I believe I did see in a chart, then the 1.7 factor does make sense.

 

[The Big Wood]

I would think the best comparison would be to use 1080i, as this is what E*'s timers seem to be based upon.

Er, am I missing something? I thought I did a direct comparison of a 1080i when I recorded the simultaneous broadcast of a 1 hour CBS show (not sure if it was 1920x1080i or some other vertical # due to HDlite). Sure it was for only 1 hour, but to record 30+ hours of content of one format versus the other wouldn't give you as nearly accurate results as recording the same show at the same time for the same length of time.

I'm sure that recording a MPEG 2 HD stream with the 622 will not give you exactly 30 hours 0 minutes of record time...there's too many variables and different levels of compression in HD streams.

 

[The Big Wood]

No such thing exists as "MPEG2 with MPEG4 headers" ; there is other service type assigned to the mentioned channels, not "to fool" but to limit reception of the channels for new receivers like 211, 411 or 622.

Well, I'll be a little nit picky too. I got the "headers" statement from previous posts in other discussions. Sorry if it is the wrong terminology to use.

Also, what I meant by the term "to fool" was exactly what you said...to limit the reception of channels.

I'm just not as smart and sophisticated as you I guess!

 

[Kirby Baker]

Well, I'll be a little nit picky too. I got the "headers" statement from previous posts in other discussions. Sorry if it is the wrong terminology to use.

Also, what I meant by the term "to fool" was exactly what you said...to limit the reception of channels.

I'm just not as smart and sophisticated as you I guess!

I don tthink that was his point. In the beginning there was talk from people here and elsewhere, that the new channels would be some sort of MPEG4 stream (as defined by a header) with MPEG2 data inside. Thats what they are not. The new channels are pure MPEG2. Dish just decided to set those new channels into a different service mode or whatever, limiting them to MPEG4 receivers only.

 

[KKlare]

As mentioned by Tom Bombadil, the 720p OTA broadcasts seem to take up less space than the 1080i.

I though I saw a ratio of 4 hours of SD to 1 hour of HD on the 622--probably a Tech Chat. Is that what you see on your 622's?

The fixed displayed ratio for a 921 is 7:1. The actual rattio varies considerably depending on letterboxing, action level, broadcast resolution 720p vs 1080i and 1280 Voom vs 1920 HDNet. Sometimes more, sometimes less than real time. FWIW,

-Ken

 

[navychop]

As pointed out earlier, there's not a lot of difference between the data rate of 720p & 1080i - 12.5%

720*1280*60 = 55,296,000
1080*1920*30= 62,208,000

 

[Tom Bombadil]

So it looks like we have what we expected to have.

Recording times are dependent upon the bitrates of the programs being recorded. These bitrates can vary significantly from program to program, from channel to channel. While resolution is a factor, it is not necessarily the determining factor, as a 1920x1080i program might get only 9mb/s, while a 1280x1080i gets 14 mb/s, and a 1280x720p could be at 16mb/s.

The early info reported here does sound like E* is achieving some bandwidth savings via the use of MPEG4. That is the key. Can we get just as good a picture from a 1920x1080i MPEG4 channel at 9mb/s as from a MPEG2 channel at 12mb/s? As I haven't seen many complaints about MPEG4 picture quality (but I haven't read all of the threads), it looks promising.

What I feared was that E* hadn't yet been able to achieve good quality MPEG4 decoding, so that there were no real bitrate savings. I'm still not convinced that E* is getting impressive results here, but as this will likely improve over time, it is encouraging to hear that some modest gains are already being realized. It would be nice for a 622 to be able to record 40+ hrs of HD at some future time.

 

[abricko]

I'm not holding my breath on a updating the encoder / decoders... they'll continue to use what it spec'd to work... their mpeg4 settings will improve over time through beta users feedback (and eventually get over compressed to squeeze more channels) the only way we'll get more recording time is for dish to allow SAN storage or use the USB port... I hope one (the USB option will probably be first) happens soon so i can keep a season of HD programming on my unit.

 

[Tom Bombadil]

Well, we would get more recording time on the resident hard disk if the encoding were 50% better.

Obviously what we would all like is a 50% improvement in MPEG4 efficiency *AND* an active USB 2.0 port for external drives.