S104 Very buggy

I was a beta tester for the 922. We had it in our showroom for a few months but didn't get to mess around with it much. I then brought it home and got to play around with it a lot. I would consider my a normal customer and I think the receiver is great. I have noticed a few issues in which I needed to reboot it. I think they should have waited a little longer before releasing it to tell the truth.
 
I was a beta tester for the 922. We had it in our showroom for a few months but didn't get to mess around with it much. I then brought it home and got to play around with it a lot. I would consider my a normal customer and I think the receiver is great. I have noticed a few issues in which I needed to reboot it. I think they should have waited a little longer before releasing it to tell the truth.

Here is the problem, in the software development world, your testing would have been called a late alpha or a closed beta test.

You have a vested interest in the product since you sell them. A true beta test is when people outside of the company (actual customers) test the hardware and software. Microsoft wizened up to this process with Windows 7 and held a public beta that lasted over a year for anyone who wanted to participate. As a result Win7 released as a viable product that has seen many different installation posibilities etc. With Vista they did internal testing and a very limited public beta.

Dish could do two things, first to test software/new features, have a way for the consumer to volunteer, and a way for them to revert the receiver to production factory settings. DirectTV offers this on certain receivers by entering a code that one can find by doing a little googleing.

Second program, test new hardware. Dish can come up with criteria to qualify as a hardware tests ( user for at least a year, already has specific receiver, etc.) then have them pay a deposit of $x, like $200 or so, sign the usual paperwork and let them use it as a second reciever in the early stages and then as primary later. when the test is over, they receive the production model for the deposit cost (not a lease.)

This type of testing is much more accurate than having a dealer test the new model.
 
Here is the problem, in the software development world, your testing would have been called a late alpha or a closed beta test.

You have a vested interest in the product since you sell them. A true beta test is when people outside of the company (actual customers) test the hardware and software. Microsoft wizened up to this process with Windows 7 and held a public beta that lasted over a year for anyone who wanted to participate. As a result Win7 released as a viable product that has seen many different installation posibilities etc. With Vista they did internal testing and a very limited public beta.

Dish could do two things, first to test software/new features, have a way for the consumer to volunteer, and a way for them to revert the receiver to production factory settings. DirectTV offers this on certain receivers by entering a code that one can find by doing a little googleing.

Second program, test new hardware. Dish can come up with criteria to qualify as a hardware tests ( user for at least a year, already has specific receiver, etc.) then have them pay a deposit of $x, like $200 or so, sign the usual paperwork and let them use it as a second reciever in the early stages and then as primary later. when the test is over, they receive the production model for the deposit cost (not a lease.)

This type of testing is much more accurate than having a dealer test the new model.

I made a comment about doing an open beta in another thread yesterday.

If anything is for certain, the early adopters will beta for you... All you have to do is upgrade them for free.

I am over a group of six radio stations and I beta software for our automation system. In return, each release, they put in a couple of features that I ask for. It works great. I get bug fixes (sometimes daily) until things are fixed. Since I have a station that pushes the envelope in using features, they love to use me, since I can break it fast.

Dish would have had NO problem doing a signup for an open beta on this site and picking say 100 people to beta test this box. Granted, they would have been out of $4000 in upgrade fees, but they would have gotten invaluable telemetry on the bugs. And they would have made 100 people very happy.

They will get with the times. Microsoft finally realized that open beta's work... Look what it did for them :)
 
Last edited:
wow...so yeah just got back from vacation. S104 is the buggiest piece of sh*t I've used in a long time. Recorded shows have recorded only half the show if not less, guide lock ups, had to reset 3 times now. A few timers showing up as being watched yet aren't and can't delete them. Awesome job Dish! Tools
 
Dish would have had NO problem doing a signup for an open beta on this site and picking say 100 people to beta test this box. Granted, they would have been out of $4000 in upgrade fees, but they would have gotten invaluable telemetry on the bugs. And they would have made 100 people very happy.

That's still a closed beta. There's no realistic way to do an open beta for a hardware product. They work well for a software product that there is already a large number of people with the hardware to run the software. I'm not sure that adding another 100 users would have necessarily prevented the problems we're seeing now.

Trying to compare the results of Microsoft allowing anyone to download Windows 7 and test it verses adding 100 users is silly. Microsoft was running 500k+ user closed betas since Windows 95. I'm pretty sure that there's no way Dish is using that many users in their betas (at $400 per box that'd be $200+ million dollars for the hardware).

If you think adding more users was the cause of the improved quality of Windows 7 (and I'm skeptical) then you're going to need a hell of a lot more than 100 users to get the same results. Otherwise, Microsoft would have already had the quality you're attributing to their beta program with their closed beta programs.
 
Actually, I only used Microsoft as an example of a company using a beta.

I would more parallel a Dish beta to what I do with our automation product.

My beta cycle works like this... <Software Company> uses maybe 10 sites, all doing things slightly differently. Then, they collect bug reports and telemetry from each site. They then fix major issues (ie, what would take us off the air) first. Then, operational bugs (unexpected GUI behavior, etc...). This process takes them 3 to 6 months. With just the info from those 10 sites, they can typically release a product that only contains minor issues. By minor, I mean will not cause an exception within the code, no memory leaks, no gpf's.

Dish could have done that with 100 people. It would have gotten them farther than they are... I can guarantee that you would not see the bugs you are looking at on this forum if this was done with power users. You can call it a hardware product all you want. It is a computer running code. They are not revising hardware to fix issues. They are changing the code.

I am sorry you took my end smiley face Microsoft comment as a 1:1 comparison. That was not my intention.

And while we are on Microsoft, the beta did produce a number of bug fixes in the final code. Windows 7 was feature complete before the beta. And, said beta, was as much for marketing as it was for bug fixes.
 
Last edited:
Dish could have done that with 100 people. It would have gotten them farther than they are... I can guarantee that you would not see the bugs you are looking at on this forum if this was done with power users. You can call it a hardware product all you want. It is a computer running code. They are not revising hardware to fix issues. They are changing the code.

It's still a combined hardware and software system. You can't have a useful beta test without the users having the hardware and software. On a new hardware platform, you can't beta test without providing the hardware to users.

We've already had several people post here saying they had the unit during the beta. Heck Scott posted his first look based on a beta unit. His first look made it clear that he worked with them to report bugs and other problems.

I'm not convinced that the problem is a lack of adequate testing. For all we know the software engineers knew about the problems we're having now and that the boxes were released to us by management even though they knew the software still needed work.

We just know too little about EchoStar's development process to make reasonable suggestions about what they should have done differently to avoid the problems we're seeing.
 
I have had no issues with S104 or any of the other ones for that matter, besides the known issue of EHD transfers. The only real issue I've had was that I missed one recording and after some digging, that was my own error.
 
Yes, I too have not had any problems with SW version 104. I'm still scared to try and transfer any recorded shows from my EHD. I did it before and it corrupted files on the 922 HDD. I will wait until there is a true fix.
 
It's still a combined hardware and software system. You can't have a useful beta test without the users having the hardware and software. On a new hardware platform, you can't beta test without providing the hardware to users.

We've already had several people post here saying they had the unit during the beta. Heck Scott posted his first look based on a beta unit. His first look made it clear that he worked with them to report bugs and other problems.

I'm not convinced that the problem is a lack of adequate testing. For all we know the software engineers knew about the problems we're having now and that the boxes were released to us by management even though they knew the software still needed work.

We just know too little about EchoStar's development process to make reasonable suggestions about what they should have done differently to avoid the problems we're seeing.

Suppose we will have to agree to disagree.
 
So tonight my 922 rebooted in the middle of 24. When it came up it was showing no timers. I tried to hit the record button to get it to record 24, but it said it couldn't get guide information. I had been using the backup feature to backup to my remote and so I tried the restore. After the restore it needed to reboot again. There were still no timers when it came back up.

I hadn't backed up the machine up since the previous night so some of my timers were not backed up. Fortunately, I have my TiVo still hooked up and setup to record shows I really care about as a backup. So I put the 922 in standby mode and left it alone. Since it was behaving like it did when it didn't seem to have an index for the guide.

After watching a DVD, I turned it back on and the timers were back where the backup was from.

So I've come to a couple of conclusions:

a) The 922 reboots may be related to a corrupted or damaged EPG. It's probably indexing and updating this in the background. It hits something that isn't formatted right and the receiver crashes. The index is then corrupted so when it comes back up it has no index again and has to build a new one from fresh.

b) When the index is not built it is unable to show the timers. So if you come back to the unit and find that there are no timers, don't panic. They may still be here. However, until the index is rebuilt the timers won't record anything either. So you may not lose any timers, but you will probably miss something.

This could explain pretty much all of the symptoms people have been complaining about lost recordings and timers going missing.

If you came back after this had happened and the index had been rebuilt you'd probably just see that programs that should have been recorded weren't.

If you can back while the index is being rebuilt you'd see all your timers missing.

The corruption in the EPG data may be due to interference with the satellite transmission and isn't appearing uniformly, which is why some people may not be having any problems and others may be having a really bad time.
 
Wouldn't you know it when I posted yesterday that I haven't had any issues with my 922 besides the EHD something happened last night. Two things: 1. I had all 10 episodes of the BBC Earth series Life recorded and somewhere between a week ago Sunday and last night 4 of the episodes disappeared. 2. When Chuck started recording OTA, I had two episodes already recorded. Once the timer fired last night, those two recordings disappeared and only the currently recording episode remained. I changed the sort to &quot;No Folders&quot; and there were empty blanks at the very end of the list showing 0 minute recordings. (I bet that's where my Chuck episodes went.) Luckily nothing too important, but annoying none the less.
 
I have not had any timer/recording issues with S104, it has just been guide issues.

I am on Western Arc. I wonder if there is a difference of behavior between the arcs. My current setup is 110/119/118.7/129. I was having stability problems, but it was 102 not 104 when I had my 61.5 dish hooked up, so I hooked up 118.7 instead.
 
I have not had any timer/recording issues with S104, it has just been guide issues.

I am on Western Arc. I wonder if there is a difference of behavior between the arcs. My current setup is 110/119/118.7/129. I was having stability problems, but it was 102 not 104 when I had my 61.5 dish hooked up, so I hooked up 118.7 instead.


western arc. 110, 119, 129 lots of problems
 
***

Users Who Are Viewing This Thread (Total: 1, Members: 0, Guests: 1)

Who Read This Thread (Total Members: 1)