Interesting Comcast Visit

[barnabas1969]

I'm going to run new RG6 next week, replacing any RG59 I put in over the last couple of years. Wanna bet my USB box still won't work?

I bought 500 feet for $33.

All RG-6 is NOT created equal. Hopefully you've got yourself some "QS" or quad-shield RG-6 on the way.

edit: Completely missed the "not" in the first sentence!

Quad shield RG6 is not necessary unless you have a very long run, or if you are near some very strong sources of interference. Also, the center conductor does not need to be solid copper, due to the skin effect at higher frequencies. Copper clad steel is OK (unless you need to run quite a bit of DC current through the cable to an amplifier).

[mldenison]

You want your signal strength as close to zero as possible. It varies by frequency, so you'll need to check the strength on a range of frequencies. Since you're using a Ceton tuner, there is already a tool for that here:
http://experts.windows.com/frms/windows ... 03442.aspx

Your S/N (aka SNR) should be in the mid 30's. 35 is great. 33 is OK. The higher the better.

I've been using this tool. It's pretty handy.

Currently, on the RG59 run to both tuners, my SNR averages around 35 dB with a few channels/frequency around 38 dB. I suspect my signal levels are low. On the USB tuner, they range from -3 to -10 dBmV. On the PCIe tuner they range from -4 to -8 dBmV.

It'll be interesting to see what the readings show after the cable is replaced.

What's curious is that the tool reports the channel list count for the PCIe tuner as 463 and the USB tuner as 488. I probably should run the tool several times to see if the count changes.

[haydongreenbutton]

Quad shield RG6 is not necessary unless you have a very long run, or if you are near some very strong sources of interference. Also, the center conductor does not need to be solid copper, due to the skin effect at higher frequencies. Copper clad steel is OK (unless you need to run quite a bit of DC current through the cable to an amplifier).

True, and RG59 works okay for many installations as well. Even the pigtail abomination on the InfiniTV works... for very short distances! But RG6 QS with solid copper does provide the best performance and the incremental price increase is worth it to have peace of mind and not have to re-do things when you learn you've got a borderline case. It's what the cable companies have historically used for several reasons even though the cable costs more. Because sending trucks out to troubleshoot problems costs even more money.

Copper clad steel is a bear to work with, particularly hard on cutting tools, and isn't great at cornering. Higher break resistance though and makes it a bit harder to accidentally violate minimum bend radius.

Lots of stuff works but I'd rather spend a bit more to eliminate gremlins. Unless I maybe had easy to access conduit run everywhere... Then I wouldn't care quite as much.

[Paw Paw]

Since channel numbers have no bearing on the frequency and are not sequential

John, does the tool that I linked not list high/med/low FREQUENCIES? I've never used it, but that's what I suggested to the person who wrote it. I assumed that's what he did. Did he pick high/med/low CHANNELS instead of FREQUENCIES?

There are two versions of the script available. One version selects a few frequencies in the low mid and high range for the output. the second, "All Ceton Signals" checks every frequencies that have a TV signal on them. It takes a minute or two longer to execue but gives you a complete listing of all of the used frequencies and their signal strength and SNR. I use the All Ceton Signals script about once a week to be sure the Comcast does not have any issues with my feed.

Look at the second post on page two for the All Ceton Signals script

[barnabas1969]

Currently, on the RG59 run to both tuners, my SNR averages around 35 dB with a few channels/frequency around 38 dB. I suspect my signal levels are low. On the USB tuner, they range from -3 to -10 dBmV. On the PCIe tuner they range from -4 to -8 dBmV.

-10dBmV is on the border of being too low.

What's curious is that the tool reports the channel list count for the PCIe tuner as 463 and the USB tuner as 488. I probably should run the tool several times to see if the count changes.

The channel list comes from the CableCARD. Both cards should have the same list. You may have a problem with one of the cards. It's probably not something wrong with the card itself... it's probably something the cable company screwed up at their end.