[lguvenoz]

I'm pulling this from a post of mine in a different thread to have a separate discussion and get input from other folks on it. I'm really curious for the people having problems, the following info:

Do you record and watch MPEG4 HD locals?
If you do, what are your signal strengths on 99 and 103?

I'd really be curious for those with multiple units to chime in. Here's my rationale from my previous post for this thought:

There is another thread here talking about a possible source of issues, and I think it might even explain the light use machine being buggier than the abused machine that Mike is talking about.

It is very possible that there are cascading failures that are started from poorly handled exceptions in the MPEG4 codec. From my following the forum most users with few or no problems have no MPEG4 channels. So we have three sources that I can think of for the errors in the MPEG4 stream:

1. At the Origin - Either the MPEG4 encoding at the local D* sites or in the backhaul to CA for uplinking to the satellite errors are being introduced in the signal.
2. At the Customer - Either poor cabling, faulty multiswitch or dish alignment is introducing errors in the data stream.
3. In the HR20 - Either through trickplay usage or data hiccups in the drive subsystem we are getting errors in the stream.

All of these sources will inevitably introduce some amount of errors. It is unavoidable. I would not be surprised to see a lower strength on buggy boxes. Even if it doesn't seem abnormally low it might be just enough to trigger some issues.

Of all the theories bantered about here, the MPEG4 codec issues look to be the most valid (D* reworking the codec, resulting the image pixelation in FF, in the prior release to resolve most BSOD issues is a real indicator). It is the one thing that could explain the complete lack of consistency between user experiences, and is the one thing that could easily be tripped up especially given the rapid deployment of HD locals that D* is on.

Thoughts???

 

[litzdog911]

There's no doubt that the MPEG4 data stream is one of the "buggier" areas in the HR20. Most of my HR20 problems usually involve MPEG4 local channels, but not always. My signal reading for my Seattle locals transponder on the 99-deg satellite is "100".

 

[mcl]

Is MPEG4 rallying the troops and launching sneak-attacks on MPEG2? Because these problems manifest themselves on non-MPEG4 content, too, sparky.

 

[lguvenoz]

There's no doubt that the MPEG4 data stream is one of the "buggier" areas in the HR20. Most of my HR20 problems usually involve MPEG4 local channels, but not always. My signal reading for my Seattle locals transponder on the 99-deg satellite is "100".

I just wonder if the non-MPEG4 issues are minor though??? Or they are agravated by a "ripple effect" from poorly handled MPEG4 problems.

 

[mcl]

I just wonder if the non-MPEG4 issues are minor though??? Or they are agravated by a "ripple effect" from poorly handled MPEG4 problems.

yeah, sure. My botched recordings on an MPEG2 channel in the 200's last night was caused by one of my MPEG4 locals.

You should run for office. You have the same sort of grasp on reality as politicians.

 

[matto]

Don't forget that in addition to errors introduced in encoding or transcoding, and those caused by bad dish aim or cabling, this is still RF technology. Noise and dropouts are to be expected in the datastream.

Since it is impossible to engineer these causes out of the system no matter what (unlike encoding or dish aim), the burden really is on DirecTV to roll out a robust decoder that can handle its expected operating conditions.

For the record, I avoid MPEG4 whenever possible, and I do not experience crashes; however, I do experience "Cancelled" or "Partial" missing recordings on a weekly basis.

 

[lguvenoz]

Is MPEG4 rallying the troops and launching sneak-attacks on MPEG2? Because these problems manifest themselves on non-MPEG4 content, too, sparky.

It's possible that issues with MPEG2 could be caused by MPEG4. In any complex system you can have cascading failures. One mishandled exception in the video stream corrupts the next thing you play type of scenario (the way you see old frames when switching between recordings could be an example of old data hanging around potentially causing issues).

Out of curiosity, do you have MPEG4 and MPEG2 content? And what types of problems are you having?

For me, we record and watch both. Inevitably though we will have problems start with an MPEG4 issue and go downhill from there. The only missed recordings we've ever had with the HR20 have all been MPEG4, and the only BSOD recordings we ever had were MPEG4.

 

[lguvenoz]

Don't forget that in addition to errors introduced in encoding or transcoding, and those caused by bad dish aim or cabling, this is still RF technology. Noise and dropouts are to be expected in the datastream.

Since it is impossible to engineer these causes out of the system no matter what (unlike encoding or dish aim), the burden really is on DirecTV to roll out a robust decoder that can handle its expected operating conditions.

For the record, I avoid MPEG4 whenever possible, and I do not experience crashes; however, I do experience "Cancelled" or "Partial" missing recordings on a weekly basis.

I agree whole heartedly with you that D* needs to get this thing tightened up. Our of curiosity, were you "Cancelled" and "Partial" recordings MPEG2 I assume?

 

[Vinny]

Do you record and watch MPEG4 HD locals?
If you do, what are your signal strengths on 99 and 103?

Thoughts???

I do 99% of my recordings from MPEG4 locals.

Sat 99
1-8...... 89 97 96 95 85 97 N/A N/A

Sat 103
Transponders
1-8...... 0 0 100 78 95 69 N/A N/A
Rest.....NA

Virtually the same readings on both tuners.

My HR20 has been fairly reliable. I never missed a recording, I've only had 2 unwatchable recordings. I've experienced the trick play bugs everyone has due to the update prior to the last update. Also experienced a couple of lock-ups.

 

[lguvenoz]

One other note.

Having a poorly handled exception could easily lead to cancelled and partial recordings. Again, this is a complex software system. The unhandled exception could kill processes that are handling the recording of shows or even managing the "To Do List". You can not underestimate the potential for cascading failure if D* does indeed have poor exception handling in their MPEG4 codec.

 

[lguvenoz]

I do 99% of my recordings from MPEG4 locals.

Sat 99
1-8...... 89 97 96 95 85 97 N/A N/A

Sat 103
Transponders
1-8...... 0 0 100 78 95 69 N/A N/A
Rest.....NA

Virtually the same readings on both tuners.

My HR20 has been fairly reliable. I never missed a recording, I've only had 2 unwatchable recordings. I've experienced the trick play bugs everyone has due to the update prior to the last update. Also experienced a couple of lock-ups.

I would be curious if there's anyone else in the New York area that could compare their experience, and signal strengths??

 

[Vinny]

"Since it is impossible to engineer these causes out of the system no matter what (unlike encoding or dish aim), the burden really is on DirecTV to roll out a robust decoder that can handle its expected operating conditions.

For the record, I avoid MPEG4 whenever possible, and I do not experience crashes; however, I do experience "Cancelled" or "Partial" missing recordings on a weekly basis."

I totally agree. D* is on the hook to roll out a decoder that can handle the expected operating conditions!

The cancelled recordings I've experienced were due to Guide glitches. (I think?)

I never had a partial; but have had freezes when attempting to watch a recording.

 

[matto]

I agree whole heartedly with you that D* needs to get this thing tightened up. Our of curiosity, were you "Cancelled" and "Partial" recordings MPEG2 I assume?

I honesty haven't been keeping detailed statistics, but the last missed recording was "cancelled" CSI, which I record off the MPEG2 west coast HD DNS feed.

 

[btmoore]

My HR20 has been fairly reliable. I never missed a recording, I've only had 2 unwatchable recordings. I've experienced the trick play bugs everyone has due to the update prior to the last update. Also experienced a couple of lock-ups.

I think this is a great example of different levels of expectations. You consider this fairly stable, I consider that very unstable. You have had 2 unwatchable recordings and a couple of lockups. For me that is unstable and not reliable. My benchmark is can I count on the DVR to record what I tell it to, over the past 2 weeks (since the last sw release) you have at least 4 times when your DVR failed or was incapable of doing its job, IMO that is not a stable consumer devise.

Now that being said, I have had 2 lockups, and maybe 6-7 unwatchables and about the same number of missing recordings with the Partial bug. I may use my box more, I have 47 programs in my prioritizers and I watch a broad spectrum of local mpeg4, HD mpeg2, and SD content, all of which have experienced these issues. On top of these problems I have also seen the 771 bug twice if I recall correctly

 

[Vinny]

I think this is a great example of different levels of expectations. You consider this fairly stable,

Just to clarify....I didn't say fairly stabile I said fairly reliable. I can say this only because I haven't had a missed recording. My stats are as follows:

To Do List: 63 items
History: 126 items
Prioritzer: 32 items.

As far as stability is concerned; I agree with you. I believe that the unit should be 100% stabile; and it's far from that right now. I do undertand that stability and reliability are very closely related. I can rely on the unit to record; but not so reliable while watching the recording, ie, trick play instability, freezes, etc.

 

[btmoore]

Just to clarify....I didn't say fairly stabile I said fairly reliable. I can say this only because I haven't had a missed recording. My stats are as follows:

To Do List: 63 items
History: 126 items
Prioritzer: 32 items.

As far as stability is concerned; I agree with you. I believe that the unit should be 100% stabile; and it's far from that right now. I do undertand that stability and reliability are very closely related. I can rely on the unit to record; but not so reliable while watching the recording, ie, trick play instability, freezes, etc.

Just to clear up some limitations of word usage. My usage of "Stability" is regarding stability of the core DVR functionality i.e. the HR20s ability to reliably execute a schedule of recordings, produce a viable recording, play, FF, RW, pause and other associated trick plays. So am saying, under the current software the HR20 is not functionally stable, and I agree the box HR20 is not technically stable either, which makes the HR20 unreliable as a DVR.

 

[Vinny]

Just to clear up some limitations of word usage. My usage of "Stability" is regarding stability of the core DVR functionality i.e. the HR20s ability to reliably execute a schedule of recordings, produce a viable recording, play, FF, RW, pause and other associated trick plays. So am saying, under the current software the HR20 is not functionally stable, and I agree the box HR20 is not technically stable either, which makes the HR20 unreliable as a DVR.

I think we agree. Although my expectations were high prior to obtaining the HR20; I would have to rate it a 3.5 out of 5; 5 being - meets all expectations.

I am hoping for the day when we can all feel confident that everything we ask of the box will actually take place. Meaning, watching live tv with no glitches in sound or picture; record fully what it's asked to record; and being able to watch this recording with full usage of every remote command. Then we can all say it is completely reliable and stable!

 

[mcl]

Odd. I wouldn't rate it that way at all, and my expectations are fairly low.


I expect an HR20 to:

1) Turn on.
2) Record what I tell it to record.
3) Let me watch what I've recorded.
4) Let me watch live tv.



On those points, I'd give it a 1 out of 4, because it fails in every aspect except turning on. There are days when I question whether it'll continue completing that simple task as well.

 

[matto]

On those points, I'd give it a 1 out of 4, because it fails in every aspect except turning on. There are days when I question whether it'll continue completing that simple task as well.

I've had it turn on without even being asked, so I guess it gets extra points for that

 

[btmoore]

Odd. I wouldn't rate it that way at all, and my expectations are fairly low.

I expect an HR20 to:

1) Turn on.
2) Record what I tell it to record.
3) Let me watch what I've recorded.
4) Let me watch live tv.

On those points, I'd give it a 1 out of 4, because it fails in every aspect except turning on. There are days when I question whether it'll continue completing that simple task as well.

I think this is where the disconnect is that creates the rub.

If we assessed the HR20 from a binary perspective on your criteria using a sample duration of the past 2 weeks under the current software. I would agree:

1 Pass
2. Fail
3. Fail
4. Fail

But if we are more analog in the assessment (based on my experience)

1. 100% of the time
2. 85% - 90% of the time
3. 85% - 90% of the time
4. 95% - 100% of the time

So the question is, is that analog assessment a passing grade for a DVR.

I would say no, it would fail based on what my expectations of what a DVR is suppose to do.

I think where the disconnect and frustration expressed by so many people comes from what their expectations are. My expectations for a DVR are that it needs to work with five 9's of reliability in the core DVR functionality and it is ok that there may be corner cases that have issues or glitches in the secondary functionality. Others may not have that same expectation and do not see it an issue that they may loose a few recording or there is a reboot, they would a DVR with my same assessment a passing grade. Do I think they are perhaps crazy, yes I do I do have a hard time understanding why it would be ok to have lost or failed recordings or a box that hangs even if it was once in window as small as 1 or 2 weeks.

For those who get angry that there are those of us who are very pissed off about the quality issues with the DVR just understand that we may have higher expectations of the quality of a specific purpose closed system device like this HR20 PVR