Separate names with a comma.
Discussion in 'DIRECTV General Discussion' started by sswheeler, Mar 21, 2009.
Not mad at DTV, but we are giving them a ton of money to make sure the equipment works properly.
Ahh, “the march of technology”.
I was just thinking about how this kind of problem would have been the minorest of inconveniences in the pre-DVR days. I lived without a program guide when I had cable and OTA, so a few hours without one isn't such a big deal. But throw in a DVR and all heck breaks loose.
…and that's the problem. My DVR has simultaneously brought me more joy and anger than any other contraption in my life outside of my PC(s).
It's a joy to have the concept of being able to record two shows at once and watch a third, or record shows day and night and never change a tape. It's a joy to just press a button and have a show marked to record. Or an entire series. And it's super cool to be able to do it from the Opera browser on my (gag me—it's pre-paid!) phone.
Unfortunately, the concept and the execution don't always jibe. The system is slow and buggy, and (apparently) incredibly dependent on reliable guide data. The hard drives can fail, the processors crap out and who knows what else. Behind the scenes the software updates seem to cause as many problems as they fix and they sometimes come at the most inopportune times. And the day I commit suicide it will be after I hit the “30 second skip” button and it, for reasons only known to DirecTV and god himself, skips to the end of the dang program. Wasn't it the band Bush who said, “it's the little things that kill”? :lol:
If TiVo can get this stuff right, why can't D*?
I spent over two hours Thursday night adding manual recordings in search of new shows to watch. I am not even going to bother to see if any got cancelled or messed up; it's just not worth the heartache.
Definitely if you manually set a recording (even by pressing the (R) button) I'd double check. Seems those are the ones that are not always showing up. The series links seem to be doing fine, though.
All this conversation is why CEO's make millions. One mistake, probably a judgement call months or years back, and everybody gets fired up. Who wants that stress?
We do it to ourselves. People are human. Companies are made up of people. Bad things happen; burning them at the stake before you know what really went into this is crazy.
And why don't they share the cause? Probably because hindsight is 20/20, and folks would be raving about how they would have spent the $1,000,000 or whatever it took to avoid the problem, and what a "no brainer" that was.
Folks who have never had to make that kind of a decision, of course...
and live with the risks associated with it. If the failure never occurred, you're a fool for spending the money and making everybody's bill goes up.
And if the failure occurs, you're an idiot for not seeing it coming.
Now, if they decide to do nothing to fix it, long-term, I'd be upset. But as someone who does this kind of work (computer software) for a living, I can tell you -- a lot that you wouldn't see coming can go wrong. Trying to see everything can be outrageously ridicluous. So can testing every possiblity.
We have to keep in mind that costs need to be controlled -- or nobody can afford satellite TV, and nobody goes into business providing it as a result.
Is perspective an option? Let's say you spend $120 a month with D* or $4 a day. Let's further assume the outage lasted 6 hours or 25% of a day, meaning anyone impacted (and I was) has been screwed out of $1. And not even that, given you could still watch all the channels, just couldn't use the guide or execute recordings. Should D* do that to you? No, of course not. But do we really need the pitchforks and torches? How about we save those for the poor bastards from AIG? They managed to screw every man, woman and child in America out of $.50 (300 million Americans, $165 million in bonuses)
They seem to be over anxious to push new features out, which 95+% of the customer base probably doesn't even realize is there. Instead of making the box rock solid and capable of dealing with adverse conditions, they just keep adding new features.
Seems to me the numero uno priority should be without even a sliver of a doubt, dependability, these boxes should be juggernauts at this stage in their development. They should be fixing bugs, and stress testing them under as adverse an environment as they can dream up. The boxes really need to be a lot more fail safe than they are currently.
We, the customers are required to make expensive commitments, I expect the same commitment from the service provider.
This wasn't specific to any box :scratchin Even the TiVo's from many years ago were affected.
Well, my equipment came from D*, I would expect it to be more fault tolerant, they are continually updating the boxes, are the TIVOs still being maintained, and D* is trying to move everyone to the newest, I would expect the D* created equipment to be the one that works through thick and thin. Two wrongs don't make it right.
I think the fact that it affected all different types of boxes shows pretty clearly it was a problem with their timesync. Note none of the D* boxes let you set the date/time, it takes sync from the sats, which should be damn accurate. Somehow for whatever reason (since nobody has gotten an answer) they lost timesync, or pushed out some bad time info, and as such all the boxes that saw it just changed to the new time way in the future. At that date/time there was NO guide data, and as such no programming to queue. Also any programming from this current time, was expired and hence gone as well.
Now this is just my speculation, but I suspect it was a simple as a bad date/time got pushed out, and some units picked it up. Should this happen, NO, but it did, and once things got set back things continued on OK..
Just my .02 for what it's worth...
Based on the evidence in this thread I agree with your assessment. I don't think this particular problem can be blamed on the software running on the various STBs. The part that sends the date & time information over the satellite is to blame.
Doug's point was only that this didn't have anything to do with new features. The issue likely had nothing to do with the software in the boxes...
Maybe your point is that resources should be diverted from building new features to making things more fault tolerant. Fair -- but I think we're at the point where if we were to increase reliability any more, there'd be NO new features, and perhaps an (even higher) monthly bill to go with that.
In my opinion, the three things we're up against -- cost, reliability, and new features -- are fairly well-balanced.
New feature are cheap, compared to the cost of added reliabilty. Meanwhile, folks could care less about reliability once you get to the last 0.001% improvement. The best example is your cell phone -- it's way less reliable than your land line, but it's "reliabile enough" that you'll pay more for it than you do your land line -- because the feature (mobility) is something that's far more important than reliability, to the average consumer.
I understand that...so am I. But...getting mad just makes you feel worse. It's pretty clear that the level of quality they demand of themselves has been dropping pretty fast over the past few years.
Except that the STBs could have code to protect themselves from these types of errors. I'm pretty sure several people around here have told me that's the way it should be.
I lost a series recording that was supposed to record during that time. It did not record and erased the setup series recording from prioritizer. Not a big deal I just set up the series recording again. The only series affected was the one set to record during the time the issues were happening. People might want to check their prioritizer.
I was thinking about that. Something like:
IF time/date update is within 120 seconds (e.g.) of current time on box
THEN accept the updateELSE (if not a DST change), wait for the next update.
Problem is, what if there's an issue with the box's on-board clock (CMOS battery, e.g.) and the time is truly off, and needs to be corrected? Maybe just let the box fail, so the customer will call for a replacement? /steve
That approach makes alot of sense.
Same here. Rebooted both HR20-700 and is back to normal
I'd tend to agree. However with no protection plan, that approach would cost a less savvy customer at least $99 and more like $149-199 for a replacement. Is that really the right thing to do given the environment?
Now if the receiver let D* know what was wrong with it before the customer calls in so there is no equipment charge or commitment extension, that'd be a good solution.
But with what happened Saturday night, it was a lot more than a few minute change. It was nearly 4 months. That should have been a red flag right away.
Prime example of not reading a post I suppose. Where did I say it was the CE's fault? I merely reported my experience. The NR immediately fixed the problem on the HR20 probably an hour after the problem was first noticed. the problem remained on the HR21 until the next day when I forced a download of the NR. If it fixed it because D* fixed the problem then that was some pretty incredible timing. Your expalination doesnt make sense for the HR21 though, I guess D* didnt fix that until the next day.
I like the suggestions for a "reality" check on the clock setting algorithym. Due to playing around with overclocking my PC, I had to reset the bios jumper and clear the cmos, so the clock was off by quite a bit (months IIRC). When I rebooted I figured I would just auto update the time so I would not have to set it manually. When I clicked the button, Windows 7 told me that the time difference was too great for automatic settings and needed to be set manually.
When the HR20 is set up, you have to go through and do all the rest of the setup with zip codes, sat setup and whatnot. No reason a quick date and time section could not be put in, it would add all of 30 seconds to initial setup. Surely they user or installer would get teh time close enough and then the machine would be able to get the true time when it downloaded the sat info as part of the setup.