Just A Thought - A Problem Source

[hasan]

Anticipating MPEG-4/HD-Locals in the next month here got me thinking (very dangerous, I know)

Many people have had all sorts of stability issues and also have MPEG-4/HD-Locals. Is it possible that errors in the MPEG-4 data stream that are not properly trapped by the HR20 cause a "lockup/crash/freeze" AND the result is a corrupted hard drive? This could cause all sorts of problems no matter what kind of recording/buffering/etc. might follow. Once the HD is corrupted, writing to or reading from it could cause new problems, unrelated to the original MPEG-4 problem.

Now, if this has any chance of being correct, we have an unstable box, with no cure until a reformat (or some sort of repair of the corrupted HD happens) is done.

This certainly doesn't explain everything, but it might account for some of the more apparently erratic and random failures. These failures happen again and again, but not necessarily after a specific, user identifiable action. They remain persistent, (but not consistent) because the root cause (hd corruption) has not been taken care of.

Anywho, just a thought from someone who has had very, very few problems and doesn't have MPEG-4/HD-Locals yet. No lockups/freezes means (according to my little theory) that my HD integrity is still there.

Obviously (per my other posting), there are tons of possible causes ...this is just one that occured to me.

[matto]

Many people have had all sorts of stability issues and also have MPEG-4/HD-Locals. Is it possible that errors in the MPEG-4 data stream that are not properly trapped by the HR20 cause a "lockup/crash/freeze" AND the result is a corrupted hard drive? This could cause all sorts of problems no matter what kind of recording/buffering/etc. might follow. Once the HD is corrupted, writing to or reading from it could cause new problems, unrelated to the original MPEG-4 problem.


A software engineering maxim concerning network code is to be conservative in what you send, and liberal in what you accept.

Attributed to the great Jon Postel, what this means is that your code must rigourously adhere to protocol specifications when sending data. This part applies to the MPEG encoders and transcoders which DirecTV operates.

This also means that the code responsible for decoding the stream in the HR20 must, must be able to deal gracefully with exceptions to this case, where either because of encoder flaws or garbage in the data stream (this is RF, interference must be expected).

I agree with your suspicion that a large amount of HR20 malfunctions are due to problems with data it receives- but really, considering the expected environment of operation, there should have been much more work done to ensure the box handles these conditions gracefully.

[hdtvfan0001]

MPEG4 here on 2 units - still no problems.

[matto]

MPEG4 here on 2 units - still no problems.

Thank you for your contribution. Your new, additional data is appreciated.

[lguvenoz]

A software engineering maxim concerning network code is to be conservative in what you send, and liberal in what you accept.

Attributed to the great Jon Postel, what this means is that your code must rigourously adhere to protocol specifications when sending data. This part applies to the MPEG encoders and transcoders which DirecTV operates.

This also means that the code responsible for decoding the stream in the HR20 must, must be able to deal gracefully with exceptions to this case, where either because of encoder flaws or garbage in the data stream (this is RF, interference must be expected).

I agree with your suspicion that a large amount of HR20 malfunctions are due to problems with data it receives- but really, considering the expected environment of operation, there should have been much more work done to ensure the box handles these conditions gracefully.


I agree that this code should be pretty much bullet proof, but we may be facing issues akin to things like 1st generation DVD players. I can remember how often my first DVD player couldn't play discs because of dirt, and now the new ones play just about anything with little if any difficulty. Heck, the first D* receivers had a lot more sound and picture problems than the ones they make today because the codecs all got better. I just hope they get this squared away quickly.