498 Discovery

[vertigo235]

OK so, I had CE loaded on my eSATA, works great, no problems what so ever, all channel tests fine. Slide is up, blah blah.

I decided to unplug the eSATA and restart on internal drive, then load the current national release.

Guess what? Black screen on 498, looks like it could indeed be a software issue.

All they have to do is update everyone to a new release and then turn the channels on!

[n2deep2bn]

OK so, I had CE loaded on my eSATA, works great, no problems what so ever, all channel tests fine. Slide is up, blah blah.

I decided to unplug the eSATA and restart on internal drive, then load the current national release.

Guess what? Black screen on 498, looks like it could indeed be a software issue.

All they have to do is update everyone to a new release and then turn the channels on!

wrong im running current update and still getting a grey screen on 498

[vertigo235]

Well that's a bummer, I wonder why my box works great with the new CE but the national release doesn't work.

[davring]

Both of my HR20's are fine on the NAtional release,

[vertigo235]

doh!, now it doesn't work on my CE release eSATA.

I guess they may really have a problem eh?

[LameLefty]

Well that's a bummer, I wonder why my box works great with the new CE but the national release doesn't work.

Many boxes on the national release are working just fine. It's almost certainly a timing issue in how long the box is signaling the switch and BBCs to do their things. THAT depends on the quality and length of cable runs and connectors, how well the switch is operating, and the efficiency of the tuners in the receivers (which will vary a little bit from box to box). So if you have a very "clean" system, reasonably up to spec across the board, no problem. If SOMETHING is off, problem. Isolating which thing it is system by system will take time, but I'm sure D* hopes they can simply tell the HR20 via a new software release to do what it takes to get signal locked in without requiring "re-dos" of potentially millions of systems.

[skyviewmark1]

Many boxes on the national release are working just fine. It's almost certainly a timing issue in how long the box is signaling the switch and BBCs to do their things. THAT depends on the quality and length of cable runs and connectors, how well the switch is operating, and the efficiency of the tuners in the receivers (which will vary a little bit from box to box). So if you have a very "clean" system, reasonably up to spec across the board, no problem. If SOMETHING is off, problem. Isolating which thing it is system by system will take time, but I'm sure D* hopes they can simply tell the HR20 via a new software release to do what it takes to get signal locked in without requiring "re-dos" of potentially millions of systems.

So, by what you say, there could be no channels until there is a new national software update for receivers. That make sense to me but it would sure screw up a bunch of CE'er's when the software overwrites their CE software. But if that is what it will take for new HD, I am sure most will make the sacrifice.

[hilmar2k]

So, by what you say, there could be no channels until there is a new national software update for receivers. That make sense to me but it would sure screw up a bunch of CE'er's when the software overwrites their CE software. But if that is what it will take for new HD, I am sure most will make the sacrifice.

Any new software will probably be based on the current CE. I doubt it would be a step back for those participating in the CE program.

I actually think the current CE is about ready for prime time anyway, so this could be a fairly easy solution.....if it's really the problem.

[vertigo235]

Well it's working now, but yeah it seems to be intermitent.

Weird.

[RunnerFL]

doh!, now it doesn't work on my CE release eSATA.

I guess they may really have a problem eh?


The OS isn't stored on the hard drive, it's stored in eeprom. Regardless of what drive you use you'll have the same OS version.

[LameLefty]

The OS isn't stored on the hard drive, it's stored in eeprom. Regardless of what drive you use you'll have the same OS version.

Good pickup - I missed that line in his post.

Another thing - other people are suggesting that a big part of the gray/black screen on 498 issue is that the slide is static, which confuses the MPEG4 decoders. Put up a few images or a small slowly moving graphic and most of those with issues would have no problem. So who knows? I'd love to be a fly on the wall at D* this weekend to know what they're thinking based on our feedback.

[vurbano]

Many boxes on the national release are working just fine. It's almost certainly a timing issue in how long the box is signaling the switch and BBCs to do their things. THAT depends on the quality and length of cable runs and connectors, how well the switch is operating, and the efficiency of the tuners in the receivers (which will vary a little bit from box to box). So if you have a very "clean" system, reasonably up to spec across the board, no problem.
I guess that is why 50 feet of brand new Rg6 run from the reciever out the door to the DIsh didnt work either? And I guess thats why if you make the right voodoo key presses on the remote it makes it work for some people?

[vertigo235]

The OS isn't stored on the hard drive, it's stored in eeprom. Regardless of what drive you use you'll have the same OS version.

Ahh I didn't know that.

At least that means I can still download the latest CE again tonight.

So maybe there is weight to the CE being better after all.

I swear it worked every time on the CE release, the national version is inconsistant.

[ezegoin]

I guess that is why 50 feet of brand new Rg6 run from the reciever out the door to the DIsh didnt work either? And I guess thats why if you make the right voodoo key presses on the remote it makes it work for some people?

Voodoo key presses have been working for me. I did notice that if I was coming from an OTA HD or one of the D* MPEG2 HD channels the congratulations slide would appear. I would sometimes have to press pause then play and it show up.

If I am coming from a SD channel it usually wouldn't appear.

[LameLefty]

I guess that is why 50 feet of brand new Rg6 run from the reciever out the door to the DIsh didnt work either? And I guess thats why if you make the right voodoo key presses on the remote it makes it work for some people?

That doesn't mean you don't have marginal tuners in your box, or that your brand new RG6 doesn't have bad connectors or a flaw in the shielding or center conductor (that happens).

And yes, it DOES explain why the "voodoo keypresses" work for some people, if you would take the time to read the posts rather than rant.

[RoundRockJohn]

That doesn't mean you don't have marginal tuners in your box, or that your brand new RG6 doesn't have bad connectors or a flaw in the shielding or center conductor (that happens).

And yes, it DOES explain why the "voodoo keypresses" work for some people, if you would take the time to read the posts rather than rant.

It's pretty hard to swallow, that the tolerances could be that tight. Timeouts in the software should be easy enough to adjust, but if the timeouts are related to hardware timers, that's some very bad juju.

[LameLefty]

It's pretty hard to swallow, that the tolerances could be that tight. Timeouts in the software should be easy enough to adjust, but if the timeouts are related to hardware timers, that's some very bad juju.

I doubt its timing hardware in the receiver - it's almost CERTAINLY issues with how long the entire OS in the receiver expects certain events to occur (e.g., the switch to cycle, the BBC to do its thing), and how that relates to the frame buffer in the video subsystem, possibly combined with the fact that the screen is blank until you do something to "nudge" it into assembling enough bits to refresh the screen. The fact that keypress gymnastics seem to work for some people is why other people have speculated that a moving picture won't suffer from this issue, as the data will be very dynamic and thus won't trigger a blank screen. I don't know and no one outside Directv will either until they start displaying a moving image.

[RoundRockJohn]

I doubt its timing hardware in the receiver - it's almost CERTAINLY issues with how long the entire OS in the receiver expects certain events to occur (e.g., the switch to cycle, the BBC to do its thing), and how that relates to the frame buffer in the video subsystem, possibly combined with the fact that the screen is blank until you do something to "nudge" it into assembling enough bits to refresh the screen. The fact that keypress gymnastics seem to work for some people is why other people have speculated that a moving picture won't suffer from this issue, as the data will be very dynamic and thus won't trigger a blank screen. I don't know and no one outside Directv will either until they start displaying a moving image.

Don't forget decrypting the mpeg stream. That's another piece of hardware that's gotta be crucial to getting the data to the screen.

[LameLefty]

Don't forget decrypting the mpeg stream. That's another piece of hardware that's gotta be crucial to getting the data to the screen.

That's true, too - decrypting the data stream (which is why you need an access card in the box) and decoding and displaying the data is basically what goes on in every single Directv transmission. MPEG4 is a different codec but locals have been using it, along with some RSNs, for over a year. Besides which, this is a very low-bandwidth signal - not much data to decrypt and display after all.

Of course BECAUSE it's so low bandwidth, it's possible there's just not enough data in the stream to makeup a valid image in the frame buffer unless your timing is just right and you catch the signal quickly enough to get an iframe before the video system just puts up a blank screen. This is, in essence, what some people believe to be the issue with grey/black screens some people are having.

[rmetcalf]

I would bet my salary on the issue being the HR20 - hardware or software. Excluding the people who have their dish pointed north and those that have diplexers/multiswitches that just don't get along, the constant is the DVR box, regardless of manufacturer. The people getting a good signal to a non-DVR box are not having ANY issue with 498. I said it another thread... and I will say it again here... I would have the product and engineering leads of the HR20 product set logging some overtime.

[vertigo235]

Updated to CE again, channel works flawlessly like before, I don't have to fanagle it to make it work.

Definetly something about this.

[lwilli201]

Check out this thread if you have not seen it.

http://www.dbstalk.com/showthread.php?t=100287

[Greg Alsobrook]

The OS isn't stored on the hard drive, it's stored in eeprom. Regardless of what drive you use you'll have the same OS version.

that's what i thought... so by uplugging the esata... the software version would remain the same..

[texasbrit]

I doubt its timing hardware in the receiver - it's almost CERTAINLY issues with how long the entire OS in the receiver expects certain events to occur (e.g., the switch to cycle, the BBC to do its thing), and how that relates to the frame buffer in the video subsystem, possibly combined with the fact that the screen is blank until you do something to "nudge" it into assembling enough bits to refresh the screen. The fact that keypress gymnastics seem to work for some people is why other people have speculated that a moving picture won't suffer from this issue, as the data will be very dynamic and thus won't trigger a blank screen. I don't know and no one outside Directv will either until they start displaying a moving image.

Something like this is definitely involved, the HR20 is receiving the signal just not displaying it. There was an apparently- related problem with a (Denver?) RSN earlier this year, where people with a DVR just got a blank screen - pausing and then play filled the buffer to the MPEG-4 codec in the DVR and the picture appeared.

[RunnerFL]

that's what i thought... so by uplugging the esata... the software version would remain the same..

That's correct

[TMar]

I can confirm that My issue isn't fixed by the latest CE. I'll be up on the roof tomorrow checking connections.

[Geekzilla]

I think it's an MPEG4 issue. I think they should put a duplicate channel on the same xponder in MPEG2 to narrow it down. I wonder if they have ever done a static slideshow before in MPEG4?

[skip22]

channel 76 is now working.

[vertigo235]

that's what i thought... so by uplugging the esata... the software version would remain the same..

Yes, that's right, but I never started up the internal hard drive on the CE release, I downloaded the national release when I restarted.

Thats' when I started having problems, then when I plugged the eSATA back in, I still had problems (national release), then when I redownloaded the CE at 11:15 last night, I had no problems anymore.

[LameLefty]

channel 76 is now working.

Um, what? Discovery HD Theater has been working. :confused:

[loudo]

Another thing - other people are suggesting that a big part of the gray/black screen on 498 issue is that the slide is static, which confuses the MPEG4 decoders.
On that theory, on the MPG4 HD RSNs, on 96 & 97, before and after the games they have static screens with the D* logo and they display fine.

[LameLefty]

On that theory, on the MPG4 HD RSNs, on 96 & 97, before and after the games they have static screens with the D* logo and they display fine.

They also have higher bit-rates than what is apparently assigned to 498, and there have been isolated reports of similar issues with RSNs in the past and I know from personal experience that it occurs (very infrequently) on MPEG4 locals. Changing channels or pausing and then FF'ing makes it go away. I've had it happen maybe 3 times in the past year, usually on a 1080i channel.

[Herdfan]

and the efficiency of the tuners in the receivers (which will vary a little bit from box to box). So if you have a very "clean" system, reasonably up to spec across the board, no problem.

Lets say I have 2 locations(A(SWM5) & B(BBC's) and 2 boxes(1(NR) & 2(19A)) .

I get the following permutations:

A1 - works
B2 - doesn't work.
A2 - works
B1 - works

So why will Box A work in both locations, but Box B only work in 1. Box 3 works fine in location C, but it is in a rack and I'm not ready to take the time to move it around.

All cable is high quality Belden. A is the shortest run, B is slightly longer and C is the longest.

I am fortunate that I can replace all my cable if it becomes necessary, but there are many subs that might not be able to do that.

[LameLefty]

I would suspect that Box B is having trouble cycling the BBCs or the multiswitch fast enough. They rely on signals from the tuners at specific voltages - if the tuner is putting out a weak signal or if a connector is loose or bad, or a cable somewhere along the way has a bad internal shield (doesn't matter the brand or the cost - this stuff is all made in industrial facilities by production workers and stuff happens) . . . if the tuner is just enough out of spec on its voltage or timing and you've got some kind of glitch with the switch or BBC or connector or cable, well . . . that's the kind of gremlin that haunts electrical engineers and drives them to drink.

[Cybercowboy]

I'm getting 498 just fine on both H20's and also my HR20-700 (running 0x18a software currently). I can't get it not to come in. And the cable run from the multiswitch to the HR20 is 75 feet or so at least.

[B Newt]

I see 498 shows up in the guide when you use the HD program filter. Didnt show up yesterday.

[Herdfan]

I would suspect that Box B is having trouble cycling the BBCs or the multiswitch fast enough.
So the big question is whether this "issue" can be corrected via a software update or if it is a hardware issue.

If it is a cabling issue, then I can correct it as almost all of my locations are fed by conduit. Others might not be as lucky. And faced with the prospect of redoing all the cabling in their house vs. a return to cable, D* might very well lose some subs.

If it is a hardware issue, then at least there will be solutions although they might take some time to implement.