Site Survey 2.0 - Looking For Your Ideas

[Milominderbinder2]

Thank you all for your help with the first Site Survey. Click here for those results. (http://www.dbstalk.com/showthread.php?t=78576)

That first Site Survey helped establish a baseline of our history. It was long because we did not know what factors we could rule out. For this next Survey, the point is to get a snapshot of today. Plus we can get our number of quesitons down significantly.

Doug is making huge strides with the Survey website to allow more flxible questions and better answer labels. Thank you Doug!

Several requested that you log in. As before you can update your vote but only your last update will be counted. We want to insure that everyone gets just one vote.

Here is the reduced and revised set of quesitons. What do you think?

Installation
10. Who installed your HR20 cables & dish? (me - mix - Professional)
11. Were new cables installed with your HR20? (old - mix - new - don't know)
12. What type of dish do you have? (1LNB-2LNB-3LNB-5LNB-Don't Know)
13. What type of Zinwell multi-switch do you have? (6x8-6x16-Multiple-Other-none)
14. Do you use a UPS (battery backup)? (Yes - No)
15. Do you use a surge suppressor? (Yes - No)
16. Do you use the USB port to power an external device? (Yes - No)

Setup
20. What was your HR20's typical temperature over the past two weeks [Menu>Setup>Info]? (110-115-120-125-130+)
21. What is your typical Video Out to your TV? (HDMI-Component-S-Video-Composite-Other)
22. What is your typical Audio Out to your TV? (HDMI-Optical-L/R-Other-unknown)
23. How many of your HR20 satellite tuners are connected? (1 - 2 - don’t know)
24. Are your BBCs (thin grey boxes) connected to the back of your HR20? (Yes - No - don’t know)
25. Do you play MPG3s/JPGs from your PC on your HR20? (often - sometimes - never)

Usage/User
30. How many programs are in your Prioritizer? (10 or less -20-30- 40 -50)
31. How often do you watch live TV? (Usually - Sometimes - Never)
32. Do you turn off your HR20 [standby] when done? (Usually - Sometimes - Never)
33. How many Autorecords are you using? (0-5-10-15-20+)
34. How many months have you had an HR20? (0-2-4-6-8+)
35. How often do you listen to HR20 Dolby? (Usually - Sometimes - Never)
36. What % free hard drive space do you usually have? (5 - 10 - 20 - 30 - 40+)

Programming
40. How many hours a day do you actively record or watch your HR20? (0-4-8-12-16+)
41. How many programs do you record a week with your HR20? (0-5-10-15-20+)
42. What resolution do you use the most? (Varies-480i-480p-720p-1080i)
43. How much do you use Native Mode? (Usually - Sometimes - Never)
45. What is your typical satellite % signal strength? (don't know-75-80-85-90+)

In the Past Month...
50. In the past month, how many times has your HR20 been replaced? (0-1-2-3-4+)
51. In the past month, how many times have you reformatted your HR20 [reset everything]? (0-1-2-3-4+)

In the Past 2 Weeks How Many Serious Problems Have You Had?
60. Forced to reset? None - (1-2) - (3-5) - (6-10) - (11+)
61. Lost a recording? None - (1-2) - (3-5) - (6-10) - (11+)
62. Video problems? None - (1-2) - (3-5) - (6-10) - (11+)
63. Audio problems? None - (1-2) - (3-5) - (6-10) - (11+)
64. Serious Closed Captioning problems? Don't Use CC - None - (1-5) - (6-10) - (10+)
65. Other serious problems? None - (1-2) - (3-5) - (6-10) - (11+)

Please give any thoughts!

[Doug Brott]

Thanks for doing this Craig.

[nick1817]

11 and 12 I would have no idea, honestly. I didnt know they put up a new dish until I realized they should have put one up.

[Doug Brott]

In the Past 2 Weeks How Many Serious Problems Have You Had?
60. Forced to reset? (None-Once-Weekly-Daily-More Often)
61. Lost a recording? (None-Once-Weekly-Daily-More Often)
62. Video problems? (None-Once-Weekly-Daily-More Often)
63. Audio problems? (None-Once-Weekly-Daily-More Often)
64. Serious Closed Captioning problems? (None-Once-Weekly-Daily-More Often)
65. Other serious problems? (None-Once-Weekly-Daily-More Often)

Perhaps if you split these values by number instead:

For instance

None - (1-2) - (3-5) - (6-10) - (11+)

#60 should probably be "Forced to Reset (RBR or GUI)?"

[Milominderbinder2]

11 and 12 I would have no idea, honestly. I didnt know they put up a new dish until I realized they should have put one up.
We found no correlation between cable length and problems so we can remove the cable length question. As for old or new, that is very important. D* paid them to install four new RG-6 cables with all new connectors of up to 125' each. You probably signed the work order that they had.

To see if they are new cables, look at your ground plate near where the cables come into the house. The cables should say RG-6 and the connectors are pretty long and were not crimped on or screwed on.

Old cable were found to be the problem for some.

I did add a "don't know" answer for that one as I am sure you are not alone.

- Craig

[Milominderbinder2]

Perhaps if you split these values by number instead:

For instance

None - (1-2) - (3-5) - (6-10) - (11+)

#60 should probably be "Forced to Reset (RBR or GUI)?"

Great idea. I will update the original post now. Your new survey labels make that possible now!

As to...#60 should probably be "Forced to Reset (RBR or GUI)?"

Does it matter if it were RBR or GUI? I do not want to say RBR or GUI because some may hav eno idea what TLAs those are.

- Craig

[Tom Robertson]

We found no correlation between cable length and problems so we can remove the cable length question. As for old or new, that is very important. D* paid them to install four new RG-6 cables with all new connectors of up to 125' each. You probably signed the work order that they had.

To see if they are new calbes, look at your ground plate near where the cables come into the house. The cables should say RG-6 and the connectors are pretty long and were not crimped on or screwed on.

Old cable were found to be the problem for some.

- Craig

Isn't the actual wording says something like "install up to 4 wires as needed"?

Cheers,
Tom

[veryoldschool]

I would still like to see whether the 5 LNB dish is the AT or AU 9

[kenmoo]

I would like to see a 2nd question after "#64. Serious Closed Captioning problems? None - (1-2) - (3-5) - (6-10) - (11+)" along the lines of "How often do you use CC?" or even "Do you use CC?". It's been my experience that not a lot of folks use CC but when they do it's needed or very important. If out of 1000 responders only 50 use CC and 40 of them have problems that seems to be an issue for me but maybe not if the raw number is 40 of a 1000.

Is there any way to ask us what HD content we record, MPEG2 vs MPEG4? Almost all my recording/trick play/CC problems seem to be MPEG4 locals. The ABC feed is especially bad.

[Milominderbinder2]

I would like to see a 2nd question after "#64. Serious Closed Captioning problems? None - (1-2) - (3-5) - (6-10) - (11+)" along the lines of "How often do you use CC?" or even "Do you use CC?". It's been my experience that not a lot of folks use CC but when they do it's needed or very important. If out of 1000 responders only 50 use CC and 40 of them have problems that seems to be an issue for me but maybe not if the raw number is 40 of a 1000.
I added an option "Don't use" for question 64. I just forgot it this time.
Is there any way to ask us what HD content we record, MPEG2 vs MPEG4? Almost all my recording/trick play/CC problems seem to be MPEG4 locals. The ABC feed is especially bad.
We have talked about a linked 10 question Bug report. This would be one of those 10 questions. Fill out the site survey once and then fill out a quick bug report for each incident.

Then you would be able to find out things like:
+ How many DIFFERENT bugs per day for this release versus previous?
+ How long do you need to wait for bugs to start appearing?
+ What types of bugs for this release versus previous?
+ Are we seeing fewer bugs for this releasse?
+ For each type of bug, are their similarities in setup or usage?

First we need the Site Survey for the basic installation/Setup/Usage data.

- Craig

[SuperTech1]

Still no mention of IR or RF for remotes. I still believe this to be a big risk factor for erratic behavior.

[hasan]

Question #20:

Do you mean with no external cooling, or what temp with it? As an example, I was running a very predictable 127 without, and with it, I'm getting 102.

[hasan]

Question 22: Audio out?

Well, I use optical out for my HT, but I also have an HDMI connection, so if I turn the volume up on my TV (instead of muting it), I have audio via HDMI (and have to turn down the home theater amp or I get an echo, of course)

I'm assuming for this condition, I would call it optical, as it is my primary.

[hasan]

Question #40: Hrs/day watch or record.

I never turn mine off, so it's going to be 16+...is that what you intended for my usage pattern?

[Milominderbinder2]

Still no mention of IR or RF for remotes. I still believe this to be a big risk factor for erratic behavior.
We surveyed 777 people on RF vs IR remotes the first time.

We also surveyed brand of remote.

I was certain that one or both of these were causing problems. But I could find no correlation at all with any of the problems. None. I was so convinced that it had to be there that I started running covariant on it. Nothing.

You can see for yourself in the raw data in the first survey. It tells how to do any analysis you want from the data.

15 items that were no factor in the first survey are removed from the second. We really need to get under 30 questions but we are close now.

- Craig

[Milominderbinder2]

Thanks Hasan!
Question #20:

Do you mean with no external cooling, or what temp with it? As an example, I was running a very predictable 127 without, and with it, I'm getting 102.
How about this?
20. What was your HR20's typical temperature over the past two weeks?
Question 22: Audio out?

Well, I use optical out for my HT, but I also have an HDMI connection, so if I turn the volume up on my TV (instead of muting it), I have audio via HDMI (and have to turn down the home theater amp or I get an echo, of course)

I'm assuming for this condition, I would call it optical, as it is my primary.
How about this?
21. What is your typical Video Out to your TV?
22. What is your typical Audio Out to your TV?
Question #40: Hrs/day watch or record.
I never turn mine off, so it's going to be 16+...is that what you intended for my usage pattern?
This is broken into two questions:
32. Do you turn off your HR20 [standby] when done?
40. How many hours a day do you record with or watch your HR20?

Any thoughts on wording changes?

- Craig

[SuperTech1]

We surveyed 777 people on RF vs IR remotes the first time.

We also surveyed brand of remote.

I was certain that one or both of these were causing problems. But I could find no correlation at all with any of the problems. None. I was so convinced that it had to be there that I started running covariant on it. Nothing.

You can see for yourself in the raw data in the first survey. It tells how to do any analysis you want from the data.

15 items that were no factor in the first survey are removed from the second. We really need to get under 30 questions but we are close now.

- Craig


I guess I can't argue with the numbers. While using RF mode may not correlate with the problems tracked in the survey I still believe it adds an unnecessary variable in troubleshooting an already difficult situation. Best of luck with Survey II. I look forward to seeing the results.:)

[hasan]

Thanks Hasan!

How about this?
20. What was your HR20's typical temperature over the past two weeks?

How about this?
21. What is your typical Video Out to your TV?
22. What is your typical Audio Out to your TV?

This is broken into two questions:
32. Do you turn off your HR20 [standby] when done?
40. How many hours a day do you record with or watch your HR20?

Any thoughts on wording changes?

- Craig

The way it works, in reality, is if you say NO to 32, then 40 is automatically the MAX. As long as you can "deal" with that in the data analysis, you are fine. The problem comes if someone says no to 32, but then interprets 40 to mean "viewing" and they watch say, 4 hours per day, but never turn it off....you get my drift.

For troubleshooting problems never turning it off means several things.

How about this:

How many hours per day are you ACTIVELY involved using your HR20 including focused viewing of programs (not just letting it run or viewing it in the "background").

How many hours per day is your HR20 on, but you are not interacting with it?

This would distinguish between cases like mine, where it is never turned off, but I still "interact" with it an average of only say 4 hours per day. It sits for an hour or two at a time (many times per day) on a given channel, like MSNBC while I do my other responsibilities. In other words, it is ON, but I am not paying any attention to it, other than listening to the news as I go about my business. In this sense it is in "background" mode, but still on.

The two cases have different implications for possible fault conditions:

a. If actively viewing, you could be doing a lot of trick play or channel/guide surfing.
b. If actively viewing, you would be more likely to see a fault condition.

c. If just on, but not interacting, the stress on the HR20 is much lower, but it is still a different "condition" than being in standby.

The wording of my first question is terrible (too long), but you can see what I'm getting at.

[Milominderbinder2]

Hasan,

Your HR20 can be on with out you watching or recording. It has a secret life of it's own.

- Craig

[Stuart Sweet]

I think you need to factor in cutting edge testing. For example, users are currently reporting less stability with the latest non-national build.

There should be a question between 52 and 60, In the last month, how often have you been running a software version which was never released nationally (not at all - - - whenever it was available) This should index with all the questions in the 60s because the failure model may be different for those who are running cutting edge versions.

Otherwise, it looks pretty good! I'm going to keep on looking at it but I think it will be a very valuable diagnostic tool.

[Milominderbinder2]

I think you need to factor in cutting edge testing. For example, users are currently reporting less stability with the latest non-national build.

There should be a question between 52 and 60, In the last month, how often have you been running a software version which was never released nationally (not at all - - - whenever it was available) This should index with all the questions in the 60s because the failure model may be different for those who are running cutting edge versions.

Otherwise, it looks pretty good! I'm going to keep on looking at it but I think it will be a very valuable diagnostic tool.
Brott,
Is there a way we could ask what version they are on? 0x __ __ __?
- Craig

[Stuart Sweet]

Brott,
Is there a way we could ask what version they are on? 0x __ __ __?
- Craig

This doesn't completely trap out a condition where a user might currently be running a national release version but spent the better part of the last month on a cutting edge version.

[Doug Brott]

Brott,
Is there a way we could ask what version they are on? 0x __ __ __?
- Craig

Maybe ... I've got some ideas cooking, but no plans for putting them in place yet. The easy path may simply be to list the most current ones.

You can actually change the questions (to some degree) as the Survey progresses. If you are only looking at running it for a few days, then Version Overlap should not be too significant.

Personally, I don't think the actual information is that useful. It may be nice to know if the person strictly runs the National Release or if they participate in the CE releases, though.

[Stuart Sweet]

It may be nice to know if the person strictly runs the National Release or if they participate in the CE releases, though.

I know, I know, dead horse -- but I think this information is of critical importance. If you were asking someone how many RBRs in the last week, someone with the HDMI/Native Bug in 0x130 would more likely give a higher number and this has no correlation to the error rate for HDMI/Native users in 0x12a. This is now a documented bug, and will likely be addressed before the next national release, so hopefully has no bearing on the long-term stability of the system. Lumping CE users in with National-Release-Only users can, and will, skew the data toward false positives on CE-only bugs.

Man, that's one dead horse. Sorry.

[Doug Brott]

I'm not suggestion that some question be asked, I'm just wondering if collecting the actual version number is useful.

Also, one way to ask the question right now would be to have two possible answers 0x12A or 0x130. Craig could add version numbers to that question if additional versions are released. If the Site Survey cycle is long, then 5 possible answer might not be enough, but for a less than one-week run time could be facilitated by the current question format.

[leww37334]

I would like to add "Do you have MPEG-4 local channels available to you?"