1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

FIOS TV pixellation fix - attenuate to SNR 31

Discussion in 'TiVo Series3 HDTV DVRs' started by AbMagFab, Nov 19, 2007.

  1. richsadams

    richsadams New Member

    Jan 4, 2003
    I know the gentleman is frustrated...understandably so. But those two paragraphs pretty much sum up how I suspect most anyone left reading this thread feels.
  2. wm2008

    wm2008 New Member

    Jan 2, 2009
    Webin, here are my responses to you...

    1) There are people on this specific thread who stated that these problems are easily and simply solved with attenuation. There are people on this thread who have indicated that attenuators are the magic silver bullet, and that I'm being unreasonable because they're that easy and reliable of a solution. That clear, and that blunt. And it is a lie. Please read the threads and you'll see that I have been consistent in acknowledging that attenuation sometimes does and sometimes does not solve the issue, and when solved sometimes the issue recurs. Further, I have attempted (and continue to attempt) to correct this. I "borrowed" 2 attenuators (that's all I could find on short notice), and I did schedule a service call with Verizon. It is actually scheduled for today - that's the first available date for this particular attempt.

    2) Tivo will NOT admit that there is catestrophic failure on the boxes. I have called Tivo corporate and they frankly are doing nothing. The only thing they offered to do was allow me to cancel my subscriptions. I can provide the dates, times of the calls, and the names of the individuals. This as stated includes calling Tivo Corporate, which frankly was nothing but the same help desk at a different number. Furthermore, you may recall that one of the 2 boxes in question is ALREADY a replacement for a "borked" box. Sending new boxes obviously does nothing. I'm frankly confused why you would even suggest such an obviously unhelpful suggestion.

    3) I am not a project manager - at least no more than anyone else in this business who is required to understand project management. Your assumption is incorrect. And, BTW, I have extremely low turnover and always have. No issues here, so I please remember the first 3 letters of the word "assume".

    4) I totally disagree with Lrhorer and his assessment of the tuners as well as the difficulty in engineering the tuners. It is a fact that the Moto units function (from a pixelation perspective and tuning) more reliably on both the Comcast and the Fios systems. They certainly may have their own disadvantages, but that's not one of them. That Moto box is used across multiple systems and does not have this deficiency. The Tivo unit does. The description of sensitivity is fine. However it changes nothing.

    5) I am most certainly saying that Tivo needs to address this issue with Tuner implementations on their design, otherwise they will continue to create the same issues as time passes. Furthermore, the issues are without question likely to only increase in frequency as Fios becomes more and more of a large scale standard. I'm not saying that they should recall every TivoHD. I am saying that they should "participate" in attempting to find a solution for known issue happening with paying customers. They are not.

    6) I only used the "31db" phrase to indicate that there is NO STANDARD AT ALL BEING IN QUESTION. When you read these threads, and see people making comments about how "oh, just attenuate down to 31db and you'll be fine" or "The Tivo doens't report signal strength - only S/N ratio", "Verizon is out of spec in their signal", or "the Verizon signal is not within industry standards" maybe you'll understand the sarcasm that was intended - especially since there has yet to be a single example explicitly showing the "standard" and in what way Verizon is failing to maintain it. Add to this that in this and other threads, when confronted with specific language about the "standards" and their implementation, "some" people here then just say that they don't "trust" the information.

    My relationships with my groups is actually very good. Frankly, the problem is an inability to recognize and admit to design issues - and to hold Tivo at least partially responsible for a known defect. I frankly would not consider an individual behaving this way to be an Engineer. They would simply be more of a techie who is unable to understand product lifecycle management, and quality design processes. Honestly, we don't run into that very much. It's actually more of a symptom of days gone by. The consumerization of many of these markets has pretty much killed that out of date philosophy. The differentiator is that we understand that we have customers - and we're all focused on those customers to the extreme. So it is unacceptable for us to simply foist our customers off to other vendors when our products experience issues. Here, Tivo will do nothing whatsoever. Their total answer and support is to tell me to call Verizon, and frankly that if that means me paying for a service call, etc, then it's my problem. Furthermore, as stated MANY times, I fail to understand the rationale of Tivo not spending the few pennies to try sending attenuators to customers experiencing this issue. Without question, this would go a long way in "working with the customer". The frankly honest truth is that I do not see Tivo making any effort whatsoever in these cases. I see no action. Except, that is, the action of taking service fees.
  3. Phantom Gremlin

    Phantom Gremlin New Member

    Jun 20, 2002
    Tualatin, Oregon
    I'm one of the people who have read most or all of this thread, and I've learned a lot. I don't agree that it's "beating a dead horse".

    Over the years TiVo has designed many boxes, some more reliable than others. They have also released good software and bad software. I can accept all that. What I can't accept is TiVo's invariably poor response to big problems and small problems, to hardware problems and software problems.

    For the most part, TiVo's answer to hardware problems is to ignore them. E.g. I owned one of TiVo's original S1 boxes, it had a very substandard modem. Yet TiVo continued to ship that same modem design for years, instead of improving it.

    This tuner problem with FiOS is, to me, much the same thing. TiVo certainly can't afford to spend $30 per unit to fix the problem, but maybe there's a more creative solution. Maybe there's a $3 solution that fixes the problem for 90% of the people. Heck, maybe there's a $0.30 solution that fixes the problem for 50% of the people. At any rate, I doubt that TiVo has even tried to attack this issue from a hardware perspective.

    Same with software. When TiVo releases new software it's invariably two steps forward, one step back. Sometimes three steps back.

    IMO TiVo does not have in place a good engineering culture.
  4. NJRonbo

    NJRonbo New Member

    Jun 3, 2005
    I thought perhaps I already posted this here already...

    I sent letters to Tivo corporate, specifically the CEO and VP of Distribution.

    Received a phone call a few evenings ago from a tech at the Executive desk.

    He explained that Tivo has been aware of the Fios problem for a long time
    now. They are unable to modify the tuners as in doing so they risk new
    problems with compatibility issues with cable systems.

    Now I don't particularly buy this excuse, but I am unqualified to argue it.

    He thought the attenuators were the best idea as that is what he has found
    works for most people. I told him I was using a -10db attenuator and that all
    the pixelation was gone except for a few seconds here and there. He suggested
    rather than adding another attenuator and risking losing signal, to go out and
    buy a short coax patch cable and barrel connector to bring the signal down just
    ever so slightly more.

    Tivo also extended my 30 day hardware and cancelation options to 60 days so
    that I could ensure that the system was working properly.

    So far, after adding -10db attentuation and a coax cable with barrel connector,
    I can say that there are no longer any pixelation issues....at least, for the moment.
  5. wm2008

    wm2008 New Member

    Jan 2, 2009
    OK, just to be open about the situation concerning Fios and the known TivoHD pixelation issues, here's the latest.

    I left work early to meet Verizon at my home. They showed up. Can't say they knew anything about Tivos, but at least the technician was easy to work with and open to discussion. Some findings....

    1) Sunrise meter showed SNR to be at around 28.
    2) TivoHD showed SNR to be at around 38
    3) VZ Moto box diags showed SNR to be at around 28.

    Question that we both had: Why do both TivoHDs show a significantly higher SNR reading than do either the sunrise meter or the Moto?

    4) Pixelation observed across multiple channels on both TivoHDs on both tuners on each TivoHD.
    5) No pixelation observed on the Moto unit.
    6) All readings and specs for signal strength, SNR, extremely well within "normal acceptable standards" according to Verizon tech.
    7) Applied -14db of attenuation on each TivoHD. (experimented a bit to get this, as we needed to do this through the Tivos since they are reporting completely different SNR than the sunrise meter.
    8) Now, SNR is at 31, sometimes 32.
    9) While pixelation is vastly improved, it is not gone from EITHER of the tuners on EITHER of the TivoHDs.
    10) Applied another -3db of attenuation to one of the TivoHDs
    11) Started losing signal on some channels.
    12) Went back to -14db.

    That's where it's at now. It appears as though the TivoHD tuner requires a VERY narrow range in which it can reliably operate, at least from my experience and with Verizon there using a meter and testing with attenuaters. So, the problem is not completely resolved to begin with, and I'll continue to observe to see how it plays out. Further, the attenuators had to be applied at each Tivo rather than prior to distribution, because otherwise it would have reduced the signal too much for reliable operation of the Moto unit which continues to function perfectly. I still currently have the Moto unit AND one of the TivoHDs connected to my "main" TV until I get a better handle on this.

    One other update: Verizon showed me yesterday that their new (and being installed in my area) HD DVRs do in fact do MRV in HD. So from a base functionality perspective, the only thing really really important to me that they still miss is expandable storage. Since the Netflix feature (which is cool) so frequently locks up and forces an unplug reboot of the Tivo I have to say that they are getting closer to really being technically competitive with Tivo. Their UI is still nowhere near as nice. Of course, I'm also seeing program guide errors on Tivo right now. I've got content that is recorded (repeatedly) as one thing when it's really another. The Moto program guide is correct and the Tivo (on channel 462 for example) continues to be wrong.
  6. bkdtv

    bkdtv New Member

    Jan 9, 2003
    DC Metro Area
    Why don't you indicate the channel names? For me, 462 is WJLA Weather (a local subchannel), which probably .1% of FiOS customers watch .1% of the time. Better to have 99% of the channels right on the TiVo than 99% of the channels periodically wrong on the FiOS DVR, imo.

    The FiOS DVR has information wrong on a regular basis for popular channels like ABC, CBS, FOX, NBC, TNT, ESPN, FX, Versus, etc. Most of the time, new episodes are flagged as repeats and repeats are flagged as new episodes. But it's not uncommon to see listed programs off by 1-hour to 24-hours; for example, many of the recent college football games were off by 24-hours in Verizon's guide. No need to take my word on this -- a quick look at the Verizon boards will show people periodically missing new episodes of popular shows like Heroes, House, 24, etc. Here's one thread of dozens with guide data complaints.

    These errors also assume Verizon's Motorola DVR is functioning normally. With the latest software, the Verizon DVR will occasionally "lose" its tuners, so no programs get recorded until a reboot. And its not uncommon for the file system on the Verizon DVR to become corrupt, resulting in 50-75% lost recording capacity until a reset (all recordings deleted). With just 25-50% usable capacity, more recordings get deleted due to limited space.

    The TiVo certainly has faults but guide information is not one of them. Verizon uses Tribune Media (same company used by DirecTV and Microsoft's Windows Media Center). It's not 100% perfect, but it's the best and most reliable information available in the country. Verizon's own Motorola boxes use guide data from FYI Television, which is the least reliable guide data provider.

    For an extra $4/mo ($19.99/mo plus tax), you can add "MRDVR" capability to the Verizon DVR that will allow you to stream SD and HD recordings to the QIP HD STBs ($9.99/mo) and SD recordings to the QIP SD STBs ($5.99/mo). Streaming to other DVRs is not supported. After tax pricing in my area:

    First box
    Verizon MRDVR ($21.59/mo) = $259.08/yr
    TiVo monthly ($12.99/mo) + M-CARD ($4.31/mo) = $17.30/mo = $207.60/yr
    TiVo prepaid yearly ($129/yr) + M-CARD ($4.31/mo) = $180.72/yr

    Second box for MRV
    Verizon HD STB ($10.79/mo) = $129.48/yr
    TiVo monthly ($9.99/mo) + M-CARD ($4.31/mo) = $14.30/mo = $171.60/yr
    TiVo prepaid yearly ($99/yr) + M-CARD ($4.31/mo) = $150.72/yr

    That's what the overwhelming majority of Verizon FiOS customers pay. Of course, there are some customers that have promotional discounts. Customers on the TiVo retention plan pay $6.99/mo on every box, which works out to $135.60/yr after the FiOS M-CARD and taxes. Verizon has promotions for its own DVR too; any new FiOS customer that signs up online can get a free Motorola DVR for the first six months.

    What does all this have to do with pixelization? Ooops...
  7. ah30k

    ah30k Active Member

    Jan 8, 2006
    You forgot that you need to pay upfront purchase price for TiVo though. Comparing dollars and cents is very difficult since they operate on slightly different models.

    When talking about MRV, you need to look at whole-home costs. In that case, Verizon will bill for a second HD client at the secondary sets at full price while TiVo will offer a MR discount for secondary sets.

    If you have an S3, then you need two CableCARDs for each set.

    In general, comparing costs is a very individual activity.

    edit - wow, just realized how off topic I've gotten here. Sorry.
  8. wm2008

    wm2008 New Member

    Jan 2, 2009
    Some edits and corrections....

    Not sure that the "overwhelming majority" of Verizon customers don't get discounts. Everyone in this region seems to be getting Moto DVRs at some sort of discount or special. But, that could stop at any time to be honest. It just hasn't as of the past year. I paid (and still do) $9.99 for my VZ Moto DVR. I pay 2x$3.99 for cablecards alone for each TivoHD - not including the Tivo cost or service. Those incentives may not be available everywhere I'm sure. It is a case by case basis.

    The Moto DVR (only IMHO and experience) has neither lost it's tuners or signal - nor has it ever experienced corruption. At one time I had 3 of them, I'm down to 1 (plus the 2 TivoHDs) now. None of them ever had any of these issues. Not saying that they don't exist.

    I'm not surprised that the info I got about MRV last night may not be entirely accurate. I'm going to check on this also, but I'm guessing that the VZ guy either misunderstood the technology, or he misunderstood my question.

    I also agree that the TivoHD scheduling is far superior to the Moto. But (to get back to pixelation) you need to be able to watch the content for that to mean anything. I'm still working on that. After attenuation, still experiencing pixelation.
  9. sinanju

    sinanju New Member

    Jan 3, 2005
    Assuming they left you with some attenuators of varying values, as I suggested you request, and you're still seeing significant pixellation, you might want to try removing a bit of attenuation. I'm sure the magic number for many is 31, but my experience is that a bit higher actually works better... For me, ~34 seems best.
  10. webin

    webin New Member

    Feb 13, 2008
    Hillsboro, Oregon
    Very good question indeed, and as best as I can remember, you're the first to report that variance. Is there anyone else reading with both boxes that can check on this?

    Within Verizon's standards, which I would expect, but still hotter than most CATV systems.

    How would you rate the pixellation now? Is it watchable, but not perfect now? Or is it still above your pain threshold? How often does the pixellation occur?

    Also, I wanted to thank you for the tone of your posts today. I like you better this way :)
  11. SeanC

    SeanC ECT

    Dec 30, 2003
    When I get home I'll check it out.
  12. SeanC

    SeanC ECT

    Dec 30, 2003
    Verizon STB on Channel 505 ABCHD 26dB
    Series 3 tuned to same with no attentuation 35-36dB signal strength 95-100
    With 20dB attentuation 31-32dB signal strength 68-72

    If I don't use the 20dB attentuator I'l get random severe pixelation. At least lasting several minutes in duration sometimes whole programs.

    With the attentuator I occasionally see very brief pixelation on DiscoveryHD and ScifiHD. Very brief meaning I can tell that I saw the pixelation but it was probably less than .2 seconds in duration.

    I consider that to be 99.9% pixelation free, and I'm happy with that.
  13. lrhorer

    lrhorer New Member

    Please think about that statement for a moment. You are speaking with a professional engineer whose very professional existence and livelihood are entirely dependent upon his ability to design, build, and troubleshoot systems like the one about which we are speaking. He's not a young, green engineer, either. He is a 30 year veteran who has seen many, many engineers of inferior talent fall by the wayside. He is damned good at what he does, and both his position and his paycheck reflect that competency. (To the rest of the forum, I apologize that this sounds like self aggranizement, but he has called my professional qualifications into question, and I feel the need to publicly clairify what those qualifications are.)

    You aren't even bothering to read my posts. READ MY LIPS: a Signal/Noise ratio of 31 dB has nothing to do with this issue. It's an unintentional red herring the person who started this thread and others have perpetuated.

    This has nothing to do with the market, and I understand the situation completely. It has everything to do with this: DO YOU WANT YOUR TIVOS TO FUNCTION PROPERLY, OR NOT?

    No, it isn't. How can I make this simple for you? It is impossible - hammer that into your head: IMPOSSIBLE IMPOSSIBLE IMPOSSIBLE to design at any cost a unit capable of an infinite range. It is Impossible: IMPOSSIBLE IMPOSSIBLE IMPOSSIBLE to design a receiver with a reasonable price tag which can handle the entire range of signals found in the marketplace under all conditions. Do you understand that? If not, go back and hammer it into your head, again. Now that you understand it, realize that Tivo, for a reasonable price, can either design their units to cover 80% of the customers out there, or they can design their unit to cover 98% of the customers out there. Which should they do?

    If you answered, "80%", then you are a fool. If you answered, "98%", then guess, what? You are left in the cold, because you are one of the 2% excluded by the practical, ethical, economic, and market savvy approach. If you truly believe it is better to kiss off 20% of the marketplace as opposed to 2%, then you don't know anything about marketing, either.

    It's irrelevant because it is non-diagnostic. Either you want to get your problem fixed, or you don't. NO MATTER WHAT ELSE, YOUR PROBLEM HAS NOTHING TO DO WITH YOUR MOTOROLA DVR, SO FORGET ABOUT IT!!!!! It won't help you, TiVo, FIOS, or the friggin' tooth fairy to fix your porblem. It certainly is not helping anyone else.

    If you don't want your problem fixed, then you ARE a troll, your actual possession or frustration with any number of TiVo units notwithstanding. I am not saying you are a troll, but if you persist in blathering on about marketing and failure modes, then clearly you have no interest whatsoever in fixing your problem, and consequently that makes you a troll, QED.

    Then go get a salesman or a hairdresser to fix your problem. Let us know when he has it fixed.

    No, it is not. They have a responsibility to provide a unit which at a minimum meets industry standard specifications and functions properly with other related systems which meet industry standard specifications. They also have a responsibility to try to meet their customer's needs and expectations within reasonable limits. Expecting that they must function with some system which fails to meet industry standard specs is just stupid. That some other device can function in that environment is irrelevant. If you purchase an automobile with a gasoline engine, the fact your neighbor's vehicle runs on diesel is irrelevant. Don't expect yours to. Some diesels are even capable of running on gasoline, but if you buy an automobile with a diesel engine and put gasoline in it, don't be surprised when the engine blows large holes in the cylinder walls and cracks the heads. The fact SOME engines can get away with it, does not mean a Mercedes 450 SL can or should.

    I'll allow that this could easily be a misapprehension on your part of what I have said, but it is based upon not reading my posts carefully. Let me try to correct it. If we just for for the moment assume the TiVo has a "flaw" in that it cannot handle high enough levels, and that the problem you and others are experiencing stems solely from the presence of higher levels than the TiVo can handle, then with 100% certainty we know that an attenuator of the proper size will fix the problem. In the event, the solution costs less than $3.

    You with me so far?

    Now let me introduce you to the concept of reducta absurdum. In a formal proof, we can show that an assertion is false by showing if it is assumed it leads to a contradiction, a paradox. The paradox in this case is, some of the members of the conference have reported adding attenuation does not fix the problem. Assuming their reports are acurate - and I have no reason to believe they are not - it follows that high levels are not the issue, or at least not the entire issue. QED

    That is what I have been saying in this thread from day 1. How you missed it is frankly beyond me.

    I never once said they were effective. Not once. It is a lie to say I did. They are a simple and inexpensive solution. That they are effective at all is not entirely clear, although the fact they have been reported to be at least somewhat effective is not surprising at all in view of the likely culprits. Any 3rd order distortion product in a receiver can be reduced significantly by reducing the overall input power. This does not mean the source of the unwanted products is high levels, it just means the problem can be masked by attenuating the signal. Conversely, any OOB carriers, particularly ones close to the receivers 1st stage IF frequency can produce high levels of 3rd order distortion. Reducing the input levels will greatly reduce the distortion products, but getting rid of the carrier will eliminate them altogether. To simplify: an attenuator may help, and in some cases may be called for, but a filter is the only thing that will remove the element that is most likely to be the primary proximate cause in many, most, or possibly even all these situations. (Note: assuming all these reports the same root causes is not appropriate in any case.)

    Do you understand, yet? If so, then great. If not, then you just need to do one of 2 things:

    1. Forget about what you think you know and take the word of an acknowledged expert.

    2. Remain frustrated.

    Consideration is not the issue. I read very carefully. What's vastly more important, I understand the IMPLICATIONS of what you said, as I am afraid you clearly do not.

    Yes, you did, or at least your statements unambiguously lead to the situation where your statements require they should ship amplifiers to people who have problems resulting from the re-design.

    Try and follow again.

    You are adamant that TiVo should fix your problem. Unless I am to assume you are just an arrgant, overweaning @ss, then I must infer you think this to be the case for any customer suffering from a result of TiVo's design parameters. Again, assuming high levels are the only current issue, and that the solution as you suggest is to design the TiVos with less sensitive receivers, then the very large number of customers with low levels who previously would have had good PQ will now be suffering with poor PQ, the solution of which is for TiVo to ship out amplifiers. QED

    You said it. You just did not think deeply enough about what you were saying to see the inevitable consequences of your statements.

    First of all, the statement is untrue. High levels are rare. They are especially rare in the absence of other, quite serious artifacts produced by the high levels in the CATV plant. High levels (high enough to cause problems, anyway) from an antenna are also rare, especially since the antennas produce only a handful of carriers. FIOS is an exception. Because of their network design, they can and do deliver signals that are not just high, but outrageously high. That said, and being a fact, I do not believe that high levels, or high levels alone are responsible for all the issues FIOS customers are having. Nonetheless, the simple fact is, 30% or better of all RF feeds found in households are at the low end of the spectrum, well below optimum specifications. Less than 10% experience high levels, and a large fraction of those experience artifacts in the video regardless of whose receiver they use, the distortion products having been produced in the CATV plant. I tell you I know this because not only have I read the industry's engineering reports, I have measured them myself. Now you have to either believe me, or else you think I am a liar. If the latter, I have to insist you point to the peer reviewed documentation which supports that belief.

    No, it wouldn't. That's the whole point. It would solve problems for 2% or less of all TiVo customers while creating problems for 10 - 20%. That is not a reasonable trade-off. What's more, the cost per customer to alleviate the 10% - 20% group would be more than 20 times higher than those required to alleviate the problems for the 2%. Add it all together, and the result is it could cost 100 times more to fix. Again, that's assuming levels are the problem, which I doubt.

    My understanding is FIOS is in fewer than 5% of homes serviced by CATV or CATV-like services, while Time Warner, Comcast, and Cox service more than 80% of such customers. Correct me if I am wrong. Of the 5% who have FIOS, and once again assuming levels are the only issue, then less than half of FIOS customers would be likely to have the problem. Being very generous, that leaves 2.5%.

    Although my background as a successful businessman does provide me with just such credentials, one needn't be a genius to know that no vendor can possibly meet the needs of every potential customer. Even if it were possible, it is never practical. Instead, the wise businessman makes it a point to invest his resources in such a way to best meet the demands of the demographic for whom he is best suited to deliver the highest volume of margin, whether it be a small number of high margin customers or a large volume of low margin customers. Catering only to a small number of low margin customers is a great way to go bankrupt in a hurry, and that is precisely what you are suggesting TiVo should do.

    I posted them. See to it Verizon's signals meet them, and you will find yourself without issues. Continue to blather on about things about which you haven't the faintest clue, and we will all see you are the very thing I did not think you were: a troll.
  14. bicker

    bicker Gruff

    Nov 9, 2003
    Your insistence on ignoring the reality of the Motorola boxes undercuts your whole argument. If Motorola can do something, it is not unreasonable for TiVo to have been able to do the same thing. And that's regardless of how long your message is, how much you bold your comments, and how much you make your comments in all-caps.
  15. lrhorer

    lrhorer New Member

    It's possible, but not overly likely. The ONT should be very stable.
  16. lrhorer

    lrhorer New Member

    That is patently HORRIBLE. FCC specifications for an analog carrier are no less than 55 dB. Even if Verizon is putting their digital carriers 15 dB below analog levels - much lower than normal - that would still comeup with a 40 dB S/N. Given FIOS' system design, they should be easily hitting 60 or 70 dB. Nonetheless, QAM transport systems can usually live with S/N ratios of 20.

    A more sensitive tuner will always have a higher internal S/N than a less sensitive one for a given input signal, provided the input signal itself has a better S/N than the receivers. Otherwise, the noise of the input signal overwhelms the noise floor of the receiver. Having a lower S/N floor is by definition what makes a more sensitive receiver more sensitive. If your tech didn't know that, then he's got a lot to learn.

    Is this supposed to be something we didn't know? Please stop posting irrelevancies. What were the signal levels? What were the levels and frequencies of OOB carriers?

    That is meaningless. What were those figures?

    <a bunch of blather deleted>
    Please stop posting irrelevant nonsense.

    Your experience is zero, and your alleged attempt to fix your issue seems all but deliberately designed to fail. Your deductions are so laughably ludicrous as to be beyond belief.

    Out of curiosity and a desire to inject a little sanity into this situation, I did a little testing today. Since I am no longer a CATV engineer, I do not have access to all the tools I once did, and thus my testing is a bit limited, but the results are quite telling.

    The only 1000 MHz amplifier I could find lying around was a little 10dB number, and while I used to have some pads lying around somewhere, I couldn't find any. The best I have at the moment is a 4-way splitter, but even that gives me better than 17 dB of range. Note that FCC specs call for a 10 dB window between 0 and +10 dBmV for optimal RF inputs to an analog TV set. Most do a bit better than this, allowing even an 800MHz broadband signal to be as high as +12 dB or so, and many still showing very decent pictures at -6 or in rare cases even -10 dBmV. Most average sets begin to show slightly noticeable noise and snow around -4 to -6 dBmV. Tomorrow or Thursday I will find some pads around here or go get some so I can take these guys all the way down until they start to fail. They did not do so today.

    The signal levels behind the set here at the house are just a bit high, although nothing out of spec. The spectrum is quite nicely flat, with a fairly ordinary negative slope. TWC San Antonio runs their QAM carriers 6 dB below their analog carriers. Normalizing for all analog video, the spectrum would run from +10 dBmV on channel 2 to +3.5 dB on the highest carrier at 783 MHz. Normalizing for all digital carriers, that woudl be +4 dBmV at 55MHz and -2.5 dBmV at 783 MHz. The analog carriers stop at channel 77 (541 MHz).

    Adding the 10dB amplifier, the levels become +20 on channel 2, +16 on channel 77, +10.5 at 591 MHz, and +7.5 at 783 MHz. This far exceeds FCC specs. The pictures are perfect. Most channels show 0 uncorrected errors and about 1 corrected error every 2 minutes or so. Removing the 10dB amp and replacing it with a 4-way splitter results in +3 dBmV on channel 2, -1 dBmV on channel 77, and -11 at 783 MHz. This is below FCC specs on all channels above channel 50. Pictures remain perfect, and I only found a couple of channels with any errors, corrected or not, over a 2 minute period of observation of each channel.

    'No great surprise there.

    What did I tell you? The Motorola is considerably less sensitive than the TiVos.

    No, it doesn't, at least not if the other members of this forum are to be believed. According to them, FIOS signals may run as high as +20 dBmV. Since you failed to post your levels, either before or after, it's hard to tell for certain what effect 14 dB of attenuation had, but if the Motorola is failing with 14 dB of attenuation on a signal that was anything near +10 dB at the outset, then it is truly a piece of crap. If it's really true (I'm skeptical), then no wonder FIOS blasts out signals so high with their ONT. If what you say is even remotely true, then part of what is happening is FIOS is producing incredibly high signal levels to make up for the turdware they are delivering as a DVR. I actually don't think it is the case - for the millionth time I think it is almost certainly OOB carriers and maybe the spectrum tilt more than the signal levels that are out of whack, but it's the conclusion with which you have left us from the information you have given. It follows then that the TiVo is failing because of the outrageous levels FIOS is delivering to make up for their POS DVR. So by your logic, it is up to Motorola to redesign their unit and then FIOS to correct their network to follow suit.

    The simple truth, however, is that you did little that was of any value and gave us no information of any value at all.
  17. webin

    webin New Member

    Feb 13, 2008
    Hillsboro, Oregon
    For those playing the home game, these out of band (OOB) carriers lrhorer is referring to are the MoCa signal Verizon uses to deliver Internet and program guide information. It operates in the 1150Mhz range, according to reports. You can prevent signals at this frequency from reaching the TiVo using a low pass filter, or a diplexer that passes through signals in the 5-900 Mhz range. Based on what I have on hand, it seems many splitters also limit to this range, though I'm sure there's a technical reason why they are ineffective (maybe that's their spec range of proper operation, but also allow other frequencies through?).

    As a point of reference, which I've discussed previously, when we were first looking into MoCa interference as a culprit, I took the step to remove MoCa signals from my system. I wired my Internet connection to run on an existing Ethernet cable, disconnected the coax from my router, and called FiOS Solution Center and had them disable MoCa on my ONT. I can't verify that there are no MoCa signals being generated, but the appropriate light on the ONT is off. While the improvements aren't an on-and-off type result (I never had strong pixellation to begin with), I can say that since doing so, I experience VERY little pixellation. This would indicate (inconclusively) that the MoCa signal was causing some amount of pixellation on my Tivo.
  18. lrhorer

    lrhorer New Member

    Oh, I nearly forgot. For purposes of comparison and perhaps to aid a bit in understanding, here are some signal quality results. I picked 5 channels more or less at random, plus the OOB carrier at 75MHz, and recorded their S/N and level specs for each attenuation.

    Ch	Freq	-7		0		+10
    2	669	30/63		34/87		35/100
    50	741	31/68		34/81		36/100
    164	657	30/63		35/87		35/100
    180	705	31/72		35/87		36/100
    OOB	75	23		26		28
  19. lrhorer

    lrhorer New Member

    Thanks, webin. It is even entirely possible there is something else going on here none of us has considered, and since I am not physically on site in front of any of these problems, my ability to troubleshoot is very limited, and also distressingly prone to be ineffective, like trying to perform surgery with a blindfold on. That said, in order of what I believe to be the major contributor to these problems and the best solution there to to the most minor contributor, we have:

    MoCa or other spurious carriers : Install 850 MHz Low Pass Filter
    High levels : Attenuate the signal
    Skewed levels : Equalize the signal

    Whether in your case an attenuator is also indicated, I don't know. If all the signal levels are in the high 90s or 100, then it may not be a bad idea, although it is preferable someone with a Signal Level Meter check out the levels. If the signals at one end of the spectrum are all 100s and the signals at the other end of the spectrum are all 50s, then it may be a good idea to equalize the signals with either a tilt compensator or a long piece of cable, depending on the slope. Once equalization has been achieved, it may be necessary to remove some attenuation. In some cases, it may even be necessary to add an amplifier, but this situation should be rare, especially on FIOS systems.

    Note no transmission system is perfect, and an occasional pixellation event should not be considered cause for consternation. By "occasional", I mean something on the order of once every two or three days, at most once a day. Also, if one channel is experiencing the problem more or less continuously, but only that one channel has the problem, then they are probably having some sort of issue at the source. This is one instance where having two devices can be diagnostic. Obviously, if both the Motorola and the TiVo are chugging out garbage on one channel and no other, the problem is unlikely to be with the DVRs. The same is true if the problem persists on a handful of channels. If something like 11 SD channels or 2 HD channels and 1 SD channel are spitting out garbage, the CATV company may have a QAM modulator going out.

    Here, for example, I copied a movie to the video server three or four days ago that had a couple of very brief, small pixellation events in it, right during the opening credits. The rest of the movie was pristine, so I left it on the server, and I'm not going to worry about it. One program on TNTHD the other day had quite a few drop-outs and pixellation events in it, but it was the only channel that had the problem. It quit after an hour or so.
  20. lrhorer

    lrhorer New Member

    The bandwidth spec for a splitter or other flat-loss device is the range of frequencies throughout which the device is guaranteed to operate within the given tolerance. Firstly, this does not mean a 900 MHz splitter does not operate beyond 900 MHz, just that it is not guaranteed by the manufacturer to operate beyond that region. If it does, it does. If it doesn't, you can't return it (or box-loads of them) claiming they do not work. Secondly, a properly operating splitter or other flat-loss device will typically pass frequencies well above and below its operating range, but the loss at those frequencie may be higher than the guaranteed flat-loss of the device. It may not be much higher, possibly only a couple of dB or so, but it still fails spec, and as such cannot be considered operational in those frequencies. To be effective against a high level interference carrier, the device may need to attenuate the signal at the offending frequency by 10, 20, even 30 or 40 dB. If the level of the MoCa carrier is no higher than the lowest QAM carrier, then dropping its level by 15 - 20 dB is probably fine, but most 900 MHz splitters probably attenuate signals at 1150 MHz by only 3 or 4 dB more than at 800 MHz, if that.

Share This Page