I'd recommend avoiding the ETH6

Ask fellow members about Ceton's infiniTV tuners here.
Forum rules
Ceton no longer participate in this forum. Official support may still be handled via the Ceton Ticket system.
ZPrime

Posts: 19
Joined: Mon Jul 15, 2013 8:45 pm
Location: Mayfield Hts, OH

HTPC Specs: Show details

I'd recommend avoiding the ETH6

#1

Post by ZPrime » Wed Jul 24, 2013 12:13 am

I purchased an Eth6 at Microcenter about a month ago now, and it has been chock full of trouble.

Background: I'm an experienced IT guy with ~15 years under my belt working on mostly Windows stuff. I've used WinXP MCE in the past with ATSC tuner cards so I have a general idea of how it all works. The Eth6 looked like it was the answer to my cruddy cable-box purgatory with Time Warner... but it has been worse in its own way.

Macroblocking / pixelation abounds. I'm able to reproduce it at will by using the web diagnostic page on the device to manually tune multiple tuners... which Ceton support then told me NOT to do. :eh: (If the manufacturer-included webpage is unable to correctly work with the device, why should anything?) To appease support, I've done all manners of test recordings - everything from 4 to 6 simultaneous recordings, plus just trying to use the system for live/buffered TV. Macroblocking randomly seems to come and go on the ITV, all the while my TWC boxes are fine. I've seen WMC randomly "lose" all of the tuners, requiring a restart of the eHome Receiver service, or occasionally I have to instead reboot the Eth6. I've seen the Ceton Diagnostics app tell me that the CableCard isn't authorized, and then have that error disappear and everything seem to be OK, but nothing tunes... and then WMC will decide that I have to go through the CableCard Wizard again.

I've also had trouble once with the ITV's status pages responding glacially slow - rather than being "snappy" and loading up all of the details on the main status page, I've seen it take upwards of a minute to display all of the information there. Again, this leads me to think there is either a hardware problem with the unit, or something crazy in the firmware... but my last contact with support had them saying "everything looks fine" and telling me to install some un-related USB hotfixes (remember, this is an Eth6, not a USB4).

According to the status page the signal level is normally fine during the macroblocking incidents... except when the status page goes stupid and shows me levels of -48dbmv or "higher" (lower signal) for a tuner that is actively displaying live TV and sometimes even NOT macroblocking at the time. :roll: Temps were originally hovering in the mid-60s when I got all tuners going, which is supposed to be OK... but on a whim I moved the unit onto a wire shelf (it had been sitting on top of a plastic-cased NAS that did not have any appreciable heat on the top) and that dropped temps about 10 degrees into the lower 50s/upper 40s... so temp shouldn't be an issue either. The unit really should have higher feet on it to improve airflow around it though.

Neither of the two TW boxes in the house suffer from any significant macroblocking. (There's the very occasional hiccup that I am tolerant of - I'd tolerate the same from WMC, but not what it is giving me now.) I don't want to disconnect them completely because they are acting as my "backup" for when WMC records a show full of macroblocking and audio-dropouts. They do happen to be whole-home boxes that speak MoCA, but I have a MoCA filter on the ITV.

The network is Cat5e and both the WMC7 system and the Eth6 are directly connected to a switch port at gigabit full-duplex speed. The switch is a small-business class HP 1810G-24, and there are zero errors on the ports for the Eth6 or the WMC box. I only have a single WMC box in the house; other TVs are fed with Xbox 360s (Extenders).

I'd be happy to test beta firmware or anything else Ceton would like from me, but the techs keep asking me to try basic stuff like "make sure all Windows updates are installed" - I understand the need to ask this, but it was asked and answered at the beginning of the case and 2-3 weeks later I'm being told the same thing... :problem: :crazy:

I'm fed up with the run-around. Either give me something else to try, or I'm going back to Microcenter and exchanging this crap for a SiliconDust. Unfortunately I'll end up having to purchase two of them and have two CableCards which will cost me more per month... but I'd rather pay an extra $50/year and have the equipment work as advertised

I was ready to return the Eth6 to the retailer last week thinking that it seemed like a hardware problem, but Ceton support actually called me on the phone and told me not to do this, so I kept providing them info and diagnostic reports. Now I'm outside of my return window, and I still have a basically unusable device (if I told my wife that TV was going to look like this from now on, she'd probably file for divorce :D).

The Case # with Ceton is MIJ-556-37709 if anyone cares.

blueiedgod

Posts: 726
Joined: Fri Jul 15, 2011 3:02 pm
Location: Amherst, NY

HTPC Specs: Show details

#2

Post by blueiedgod » Wed Jul 24, 2013 5:01 pm

Sorry to hear about the problems. I am pretty certain that if it is indeed a hardware problem, Ceton will let you return it to them, and then send you a replacement on.

If you are local to me, you can try InifniTV4, since I have been wanting to move up to ETH6. We can do an even swap, since I paid $430 for my InfiniTV4 few years ago. It has been rock solid.

ZPrime

Posts: 19
Joined: Mon Jul 15, 2013 8:45 pm
Location: Mayfield Hts, OH

HTPC Specs: Show details

#3

Post by ZPrime » Wed Jul 24, 2013 5:11 pm

Yeah, I'm not going to trade a 6-tuner device for a 4 tuner one, sorry. ;) I'm also pretty far from you (just updated my profile to include location and PC details).

They indicated during my phonecall that if it was hardware-related it could be swapped, and they also mentioned that if we couldn't find an answer to the problem they would refund me themselves. However, the most recent update to the case indicated that (a) they thought the hardware was fine and (b) implied that there was nothing else they could do... which isn't really acceptable to me, especially after they explicitly told me not to return the pile of junk to the store.

I am thinking that my next step will be to try exchanging it at Microcenter... I'm outside my return window, but they had another on the shelf and if I explain the situation it's possible they'll allow me to swap hardware anyway (I'm a pretty regular customer there). If it is indeed a hardware problem with the Eth6, this swap should resolve everything, but I'm not sure why Ceton hasn't offered an RMA yet at this point themselves.

The fact that the device's own internal webpage is behaving "funky" indicates to me that something is drastically wrong with it, but Ceton support apparently doesn't feel that this is relevant (or has ignored my notes about it).

shortcut3d

Posts: 317
Joined: Sat Jan 19, 2013 7:33 pm
Location:

HTPC Specs: Show details

#4

Post by shortcut3d » Wed Jul 24, 2013 6:25 pm

I agree 100%. Ceton support is of no help. They just want to ignore the problem. I also spent months troubleshooting the very same issue with Ceton. The entire infrastructure was changed including, HTPC, routers, switches, cables, and CableCARD(s) [I have the receipts to prove it.]. The problem still exist and Ceton politely told me the InfiniTV6 ETH is not for me then. I then asked why a hardware swap was never attempted if everything else was replaced because it only makes sense the last component being the InfiniTV6 ETH is the problem. Pam kindly called me and said the InfiniTV6 ETH was not for me.

In addition to the pixelation, the InfiniTV6 ETH also cause the wait cursor to display in WMC at first load. The InfiniTV6 ETH also has tuner not available issues randomly that Ceton seems to want to sweep under the rug. These problems do not exist on SiliconDust HDHomeRun Prime(s). I suggest Ceton purchase a Prime and see how it functions in a clean environment. They might be surprised to find the Prime works much smoother and faster.

ZPrime

Posts: 19
Joined: Mon Jul 15, 2013 8:45 pm
Location: Mayfield Hts, OH

HTPC Specs: Show details

#5

Post by ZPrime » Wed Jul 24, 2013 6:28 pm

shortcut3d wrote:I agree 100%. Ceton support is of no help. They just want to ignore the problem. I also spent months troubleshooting the very same issue with Ceton. The entire infrastructure was changed including, HTPC, routers, switches, cables, and CableCARD(s) [I have the receipts to prove it.]. The problem still exist and Ceton politely told me the InfiniTV6 ETH is not for me then. I then asked why a hardware swap was never attempted if everything else was replaced because it only makes sense the last component being the InfiniTV6 ETH is the problem. Pam kindly called me and said the InfiniTV6 ETH was not for me.

In addition to the pixelation, the InfiniTV6 ETH also cause the wait cursor to display in WMC at first load. The InfiniTV6 ETH also has tuner not available issues randomly that Ceton seems to want to sweep under the rug. These problems do not exist on SiliconDust HDHomeRun Prime(s). I suggest Ceton purchase a Prime and see how it functions in a clean environment. They might be surprised to find the Prime works much smoother and faster.
Thank you for the reply. I appreciate knowing that someone else has been through the same crap and that I'm not just insane. I was about ready to start shopping for a new switch myself, even though there are zero indications that it is the problem. :/

I'm going to see if Microcenter will take this POS back for a SiliconDust then. I really don't want to do it, because I'd rather have a single tuner box, single CableCard, and single Tuning Resolver (since TWC charges for each CC)... but if the only way to make it work correctly is with a pair of everything, so be it.

shortcut3d

Posts: 317
Joined: Sat Jan 19, 2013 7:33 pm
Location:

HTPC Specs: Show details

#6

Post by shortcut3d » Wed Jul 24, 2013 6:50 pm

I'm waiting for the 4-tuner SiliconDust HDHomeRun Prime with hardware compression. We definitely use all 6-tuners beause I came from the no longer available 6CC. However, we would appreciate the reliability of the Primes and would be willing to sacrifice a couple tuners for it.

ZPrime

Posts: 19
Joined: Mon Jul 15, 2013 8:45 pm
Location: Mayfield Hts, OH

HTPC Specs: Show details

#7

Post by ZPrime » Wed Jul 24, 2013 7:21 pm

shortcut3d wrote:I'm waiting for the 4-tuner SiliconDust HDHomeRun Prime with hardware compression.
What's the hardware compression for? The stream is already MPEG2 when it comes from the cableco, is this to convert it to H.264 and stream to portables or something? (Irrelevant to me because I'm only using a single WMC box and don't care about mobile access or anything.)

A 4 tuner would just barely cover my needs, but I liked the flexibility of having 6 which is why I jumped on the Eth6. (Kind of like jumping on a grenade, at this point... :wtf: )

To be honest, I wanted to do Tivos again because the wife is familiar with them and we like the UI, but the prospect of having near unlimited and redundant TV storage was too juicy to pass up, so I built a HTPC instead. Plus, Tivo's Minis don't do tuner pooling, so with 3 TVs you can't really do one Tivo + two Minis if you need to have 4 active recordings at times. Add to this the fact that I already have an XBox 360 at every TV and WMC just made more sense.

barnabas1969

Posts: 5738
Joined: Tue Jun 21, 2011 7:23 pm
Location: Titusville, Florida, USA

HTPC Specs: Show details

#8

Post by barnabas1969 » Wed Jul 24, 2013 7:49 pm

If you read some of my posts on this site, you'll definitely see that I'm not necessarily a fan of Ceton. In fact, after testing a HD HomeRun PRIME, I sold my InfiniTV4, bought a 2nd HD HomeRun PRIME (for a total of 6 tuners) and have never looked back. I was also a beta tester of the Ceton Echo, and was totally dissatisfied with that sorry excuse for a retail product. I returned it at the end of the extended warrantee period.

However, I will say that your post didn't really answer the question of whether or not your signals are in fact "good". I've found that both the InfiniTV4 and the HDHRP are very picky about signal strength and quality, as are the tuning adapters.

Since you're on TWC, I'm going to assume that you have a tuning adapter, correct?

So...

I recommend that you download the tool at this thread (I'm assuming that it works with the ETH. It was written for the ITV4):
http://experts.windows.com/frms/windows ... 03442.aspx

Follow the instructions in the thread, run it, and report the results back here.

Also, browse around in your tuner's web pages, and find the one that reports the signal information for your tuning adapter (aka tuning resolver). Report the FDC and RDC signal levels (aka "power") and SNR (aka "S/N").

RyC

Posts: 724
Joined: Tue Aug 21, 2012 10:21 pm
Location:

HTPC Specs: Show details

#9

Post by RyC » Wed Jul 24, 2013 7:54 pm

barnabas1969 wrote:If you read some of my posts on this site, you'll definitely see that I'm not necessarily a fan of Ceton.
I would go further and say you definitely are not a fan of Ceton period :P


Sent from my iPhone using Tapatalk 2

shortcut3d

Posts: 317
Joined: Sat Jan 19, 2013 7:33 pm
Location:

HTPC Specs: Show details

#10

Post by shortcut3d » Wed Jul 24, 2013 8:48 pm

I will say my signal is about as perfect as possible. When I originally received the InfiniTV6 ETH, I had just moved into a brand new development. Comcast actually had to use an 8-way splitter to step down my signal strength (properly terminated unused ports). This is were support gets me. I actually told them connecting the InfiniTV6 directly to the drop may damage the unit. They would not go forward without testing it at the drop, even though my signal was perfect at the living room feed. So I can't speak for the OP, but my signal is perfect and the problem exist.

ZPrime

Posts: 19
Joined: Mon Jul 15, 2013 8:45 pm
Location: Mayfield Hts, OH

HTPC Specs: Show details

#11

Post by ZPrime » Wed Jul 24, 2013 11:57 pm

Image

Image

Image

Tuning Resolver info:

Code: Select all

CURRENT FDC
Freq: 75.000 MHz
DAVIC: Connected
Status: Locked
Level: -6 dBmV
Seconds: 1355271
Corr Bytes: 298
Uncor Blks: 183
Errs Avg/Inst: 0 / 0
Total Bytes: 3210481700
S/N: 31 dB
CURRENT QAM
Freq: 705.000 MHz
Tuning Mode: QAM-256
Status: Locked
Level: 4 dBmV
S/N: 39 dB
Seconds: 73898
Corr Bytes: 0
Uncor Blks: 0
Errs Avg/Inst: 0 / 0
EQ Gain: 1.0
CURRENT RDC
Freq: 21.000 MHz
Power: 45 dBmV
Delay: 670 uSec
Retrans: 5
I have the feed from TWC coming into a 4-way. 1 Leg to cable modem, 2 legs to two TWC boxes, and last leg goes into the Cisco TR. I have the Eth6 connected to the output of the TR because it is amplified and doesn't lose as much signal as using a 2-way for the pair of devices.

If I trusted the Eth6 to not be a pile of crap, I'd return the TWC boxes and go down to either a 3-way (each device on its own leg) or a 2-way (with the Eth6 still behind the TR)... but given how buggy the Eth6 has been I can't do that yet.

I guess the signal levels are a little marginal here on the lower end of the spectrum... But they aren't horrible. The crappy TWC boxes also don't have any pixelation problems to speak of (there is the very rare little hiccup here and there but nothing like what the InfiniTV does). When I've had pixelation actively happening on the Eth6, I've captured diagnostic reports from the Ceton utility, and Ceton didn't mention anything about signal problems. I've also pulled up the web diag page while a channel is going to crap and witnessed numbers better than the stuff in yellow (although I didn't make note of the frequency, perhaps I should've). For example, I regularly see problems on 1151 (FoodHD), but according to this table it is fine.

How exactly does SDV work as far as frequency choice goes? Does the same channel always land on the same frequency? Or does the TR work with the headend to pick whatever happens to be available at the time? If it is pseudo-random, I could see how I could get pixelation on various channels due to SDV putting them on different frequencies and one of those happening to be "bad."

All of the above doesn't excuse the wacky behavior that I've seen from the InfiniTV. Things like me using the diagnostic page to force all tuners active (even if they are all on the same channel) will cause near-instant pixelation on the one I'm watching. After some indefinite period of uptime (7+ days or so) the diagnostic page runs slow as molasses. The Ceton Diags will occasionally tell me that they can't find a CableCard, but then magically find it a second later. Fairly regular "no tuners available" messages when there are no active recordings and nothing watching anything live, requiring me to either kick eHomeRcvr or reboot the Eth6 or both. This does not feel like a fully-baked product to me, especially when others talk about how the SiliconDust doesn't suffer from them.

Also, if the above output is helpful for diagnostic reasons, why the crap has nobody at Ceton support asked me to provide it? :crazy: :eh:

shortcut3d

Posts: 317
Joined: Sat Jan 19, 2013 7:33 pm
Location:

HTPC Specs: Show details

#12

Post by shortcut3d » Thu Jul 25, 2013 12:43 am

For the pixelation issue through the device webpage, I would think Ceton would release a firmware fix that changes that behavior. Maybe release the current tuner when you go to the next tuner and manually check the strength. I think Ceton is holding the tuner in that state to improve tuning performance. Its not a surprise to me the InfiniTV6 ETH is super slow tuning compared to HDHomeRun Prime. The InfiniTV4 PCIe was slower and that's internal without network overhead. The seems like Ceton responded to SiliconDust by making an "adapter" for a PCIe product and not designing a device from the ground up. SiliconDust is also much more responsive to fix issues in the beta channel.

I also get the random no CableCARD warning in the Ceton Diagnostics. I have tried a couple CableCARDs and they all have the same issue. I was more annoyed with the pixelation and wait cursor, so my tickets focused on those. Ceton's official word was there is not a problem and the InfiniTV6 ETH doesn't sound like its for me. No more troubleshooting.

JohnW248

Posts: 786
Joined: Fri Jul 20, 2012 7:23 pm
Location:

HTPC Specs: Show details

#13

Post by JohnW248 » Thu Jul 25, 2013 3:05 am

There is a signal issue with your TA

CURRENT RDC
Freq: 21.000 MHz
Power: 45 dBmV
Delay: 670 uSec
Retrans: 5

Retrans are bad news, it means that the TA didn't get a reply from the node or headend and had to ask again, this usually shows up as 'no video" or SDV1 errors. On recording it tries again, on live tv it tries once. Your upstream power is also a bit high (usually look to be under 40 dBmV). If the tuner is after this device, you are feeding it a less than ideal signal. Amplifying a bad signal just makes a stronger bad signal. IF TWC scanned your drop and said it was "ok" you need to get the lead or a supervisor out.

barnabas1969

Posts: 5738
Joined: Tue Jun 21, 2011 7:23 pm
Location: Titusville, Florida, USA

HTPC Specs: Show details

#14

Post by barnabas1969 » Thu Jul 25, 2013 4:02 am

I agree with JohnW248 (he's our resident tuning adapter expert). The up-stream (RDC) power is a little high. The cable company will tell you that it's fine as long as it's under 52dB, but that's just not true.

Also, it really couldn't hurt to bring up all of your other signals (including the FDC on the tuning adapter) by about 5 dB. Consider a unity-gain amplifier instead of a 4-way splitter. You can test this by temporarily removing the 4-way splitter (essentially connecting your TA and ETH directly to the feed coming into your house).

Also, you might be surprised at what you will see if you use a 2-way splitter before your TA and tuner, instead of feeding the tuner from the TA's output. It's not a bad idea to try it both ways. In my situation with the HDHRP, it works better to feed the tuner from the output of the TA... but when I had an InfiniTV4, it worked better with a splitter. I know this sounds goofy, but the amplifier inside the TA can add noise to the signal... and the InfiniTV tuners seem to be very sensitive to noise.

barnabas1969

Posts: 5738
Joined: Tue Jun 21, 2011 7:23 pm
Location: Titusville, Florida, USA

HTPC Specs: Show details

#15

Post by barnabas1969 » Thu Jul 25, 2013 4:06 am

Oh... and more about John's last paragraph...

When I finally got my cable company to send someone who actually knew what they were doing, he disconnected my house from the feed and found that I had a signal ingress problem. It turned out to be a cheap store-bought cable in one of the bedrooms (between the wall and the TV) that was picking up some stray signals.

His scanner showed an intermittent spike right at the frequency that is used by my TA (and cable modem). When we removed that cable, the problem went away. He happily made a new cable for that bedroom, and then scanned again. The problem was gone... and so were my TA problems.

JohnW248

Posts: 786
Joined: Fri Jul 20, 2012 7:23 pm
Location:

HTPC Specs: Show details

#16

Post by JohnW248 » Thu Jul 25, 2013 2:11 pm

How exactly does SDV work as far as frequency choice goes? Does the same channel always land on the same frequency? Or does the TR work with the headend to pick whatever happens to be available at the time? If it is pseudo-random, I could see how I could get pixelation on various channels due to SDV putting them on different frequencies and one of those happening to be "bad."
The TA gets a channel map with channel layout every morning around 3AM (it's on Serial Link Anomalies).
SERIAL LINK ANOMALIES
DecodeErr Tx/Rx: 0/0
LastDecErrTxTime: n/a
LastDecErrRxTime: n/a
TxFlowCtrlToggled: 0
Overflow Tx/Rx: 0/0
LastOvFlowTxTime: n/a
LastOvFlowRxTime: n/a
ApduTxEr/OtherEr: 0/0
CHANNEL TABLE HISTORY
Rsp/Upd/Err/Rtx: 1/2/0/0
07/24@10:08:03 Rsp v000 Blks=00
07/24@10:08:43 Upd v001 Blks=07
07/25@03:01:15 Upd v002 Blks=07

The current configuration file is shown on the Tuning Resolver tab:
TUNING ADAPTER
Authorized: Yes
OpStatus: Ready
ChanTblLoadTime: 07/25@03:01:10
HmacKeyGenTime: 07/24@10:08:01
TA CONFIG FILE
LoadTime: 07/25@03:01:06
Version: 07/24@04:45:04
Size: 9KB
NumChan/Split/Desc: 778/0/3
LastLoadErr: NoErr
LastErrTime: n/a
APDU STATISTICS
Total Tx/Rx: 151/164
Status Tx/Rx: 16/23

And the current status of your TA and the Node so that the TA knows what's running on your node right now is on the mini-carousel tab:
MINI CAROUSEL INFO
Status: CacheReady
Def Freq: 555 MHz
Tvp/Tv Id: n/a
Load Time: 07/25@06:14:52
Version: 6
Size: 1808 bytes
Num Entries: 109
Cache Hits: 5
Cache Misses: 0
Cache Overrides: 0
Load Count: 4
Load Failures: 0
Last Load Err: NoErr
Err Time: n/a
Last Load Attempt:

This information tells the TA whether the channel you want is already on your node or if it has to ask the headend to send the channel to the node. When this works properly with proper signal levels, etc. the process is totally transparent to the user and they should see no difference between an "always on" channel, one on the node or one requested.

The TA also keeps track of the number of session for each tuner and what activity is being performed, i.e. watching or recording. If you join a channel on the node that someone else was watching, your TA will keep the channel from timing out. Most systems will stop a SDV channel after 4-5 hours of watching but will not stop a tuner if it is recording. The channel location will change with the daily channel map although the same map may appear if there are no changes from the digital control network supervisor.

ZPrime

Posts: 19
Joined: Mon Jul 15, 2013 8:45 pm
Location: Mayfield Hts, OH

HTPC Specs: Show details

#17

Post by ZPrime » Thu Jul 25, 2013 7:35 pm

JohnW248 wrote:There is a signal issue with your TA

CURRENT RDC
Freq: 21.000 MHz
Power: 45 dBmV
Delay: 670 uSec
Retrans: 5

Retrans are bad news, it means that the TA didn't get a reply from the node or headend and had to ask again, this usually shows up as 'no video" or SDV1 errors. On recording it tries again, on live tv it tries once. Your upstream power is also a bit high (usually look to be under 40 dBmV). If the tuner is after this device, you are feeding it a less than ideal signal. Amplifying a bad signal just makes a stronger bad signal. IF TWC scanned your drop and said it was "ok" you need to get the lead or a supervisor out.
Well, TWC hasn't touched the drop since it was re-run last fall due to not being properly grounded outside the house. What are the magic words to say to them to get them to look at it? The TWC boxes all perform just fine. The first line techs act like you have 5 heads when you start talking about CableCard & TA's... Should I just try going directly to the CableCard activation line and explain the problem there? Or maybe TWC's Twitter team?
barnabas1969 wrote:I agree with JohnW248 (he's our resident tuning adapter expert). The up-stream (RDC) power is a little high. The cable company will tell you that it's fine as long as it's under 52dB, but that's just not true.

Also, it really couldn't hurt to bring up all of your other signals (including the FDC on the tuning adapter) by about 5 dB. Consider a unity-gain amplifier instead of a 4-way splitter. You can test this by temporarily removing the 4-way splitter (essentially connecting your TA and ETH directly to the feed coming into your house).
This isn't really an option, because I lose internet access then... The problem is that the macroblocking / pixelation is very sporadic, so it can be difficult to leave everything in a testing configuration for long enough to catch problems without irritating the wife. ;)

I do have an 8-port, 2-way Antronix amp (FRA8-0400) that I purchased myself when I first wired the house (everything is RG6 quadshield with compression fittings). I may be able to do a 2-way split at the drop, run one leg to the internet and the other to the Amp, and then hang the TWC boxes off of another 2-way and put the Eth6 and TA on their own ports. (The TWC boxes can't go directly on the amp because it isn't MoCA-capable, but if I use a 2-way between them and the amp that should be OK.)
Also, you might be surprised at what you will see if you use a 2-way splitter before your TA and tuner, instead of feeding the tuner from the TA's output. It's not a bad idea to try it both ways. In my situation with the HDHRP, it works better to feed the tuner from the output of the TA... but when I had an InfiniTV4, it worked better with a splitter. I know this sounds goofy, but the amplifier inside the TA can add noise to the signal... and the InfiniTV tuners seem to be very sensitive to noise.
I already tried using a 2-way (hanging off the existing 4-way) to feed the TA one leg and the Eth6 the other... signal levels were even worse on the Eth6, which is why I went back to the pass-through setup.

I picked up a SD Prime last night and am going to try giving it a shot, but I will also rework the wiring to see if that helps the Ceton. I would certainly prefer to keep the Ceton just because it means paying TWC less money (fewer Cablecards).

Prior to having TV, TWC actually put a 3db or 6db pad on the line at the back of the modem, saying it was a little hot, but I suppose that things could've changed outside and levels could be lower now (I took that pad off when they put in the 4-way to run the 3 whole-home boxes I originally had - one of those has been retired and the Eth6/TA put in its place, but the two WH-DVRs are still there). Wish I had my own testing gear because it is hard to find the time to sit around waiting for techs who show up an hour after their appointment window. :thumbdown:

JohnW248

Posts: 786
Joined: Fri Jul 20, 2012 7:23 pm
Location:

HTPC Specs: Show details

#18

Post by JohnW248 » Thu Jul 25, 2013 8:23 pm

Level isn't constant over the spectrum to begin with and it changes based on temperature and time of day. Depending on the age of some of the line amps you can have quite a swing and if the eq is off on some amps you have other problems.

I don't worry about how to master the RF from the cable company. They have the tools, the amps and the splitters. They have special splitters to use if you have telephone & Internet and then the split to tv services runs through a special upstream/downstream amp which they have to measure the signal to get the right one. Also if you have any user installed wiring in the house, you should have them scan the entire drop for any problems. Remember they can find shorts in coax and locate where they are with their equipment. Also a nail or sharp bend might not produce a short but will change the impedance and then a "filter" at that point much like a RC filter because the copper is the R and the diaelectric is the C and if it is compromised it'll provide a change in over all EQ.

If your system passes, and you still have trouble then you have eliminated the RF side of the problem. The other side is the network side and you might drop over to the SD forum because they've had their product out longer and there are other packet loss tests, etc. Don't forget that FW or AVS other than the Windows FW and MS Security Essentials can add their own fun to the puzzle. Not to say you have to run those two, but if you choose other ways of handling it you have another set of problems to deal with.

Back in "the beginning" when the only tuner was the ADM/ATI DCT and you could only run it on Vista OEM machines with a cableCARD bios and PID, the machines shipped with Norton or McAfee and we went though hell trying to get all the ports right, etc. AND ATI denied they even made the tuner--there was NO SUPPORT and of course Dell, Sony and HP used off shore phone support that didn't even know about cable let alone cableCARD.

Also, while Ceton might not say this is so, we've found over at the SD forum that running the tuner, PC, and other devices off the ports on the Router can cause problems as well. Best results come from running the tuner PC and extenders on an unmanaged switch with just a jumper to the router for IP services.

You really should shoot for a 0dBmV signal level on the tuner coming off whatever TWC puts in. Let them scan the drop at the end of the coax that you're going to attach the tuner and I would recommend that you don't use the loop through on the 1520. The moto is worse since those units have (or had if they've corrected them) a 6 dBmV signal drop from the loop through.

ZPrime

Posts: 19
Joined: Mon Jul 15, 2013 8:45 pm
Location: Mayfield Hts, OH

HTPC Specs: Show details

#19

Post by ZPrime » Thu Jul 25, 2013 8:43 pm

JohnW248 wrote:Level isn't constant over the spectrum to begin with and it changes based on temperature and time of day. Depending on the age of some of the line amps you can have quite a swing and if the eq is off on some amps you have other problems.

I don't worry about how to master the RF from the cable company. They have the tools, the amps and the splitters. They have special splitters to use if you have telephone & Internet and then the split to tv services runs through a special upstream/downstream amp which they have to measure the signal to get the right one. Also if you have any user installed wiring in the house, you should have them scan the entire drop for any problems. Remember they can find shorts in coax and locate where they are with their equipment. Also a nail or sharp bend might not produce a short but will change the impedance and then a "filter" at that point much like a RC filter because the copper is the R and the diaelectric is the C and if it is compromised it'll provide a change in over all EQ.
The problem with TW is that (a) they will charge me for a service call, even if the problem is outside my house, and (b) the techs that I've had in the past don't seem to want to do anything beyond replace splitters or put new ends on cables.
If your system passes, and you still have trouble then you have eliminated the RF side of the problem.
See, what I don't understand here is why everyone feels that it is acceptable for these devices (Ceton or SD) to behave any differently than the cableco's boxes. The TW boxes work perfectly fine with minimal or no macroblocking.
The other side is the network side and you might drop over to the SD forum because they've had their product out longer and there are other packet loss tests, etc. Don't forget that FW or AVS other than the Windows FW and MS Security Essentials can add their own fun to the puzzle. Not to say you have to run those two, but if you choose other ways of handling it you have another set of problems to deal with.
The network side is as optimal as it can possibly be. I'm using a business-class switch (HP Procurve 1810G-24), and the Eth6 is directly plugged into the switch, with the HTPC on a different port. This is a gigabit switch, everything is set at auto and is linking at 1000/full. The only "security" software on the HTPC is Windows firewall and MS Security Essentials. Windows was just installed about a month ago from scratch, so it's not a 3-year old "scratch and dent" system.
Also, while Ceton might not say this is so, we've found over at the SD forum that running the tuner, PC, and other devices off the ports on the Router can cause problems as well. Best results come from running the tuner PC and extenders on an unmanaged switch with just a jumper to the router for IP services.
An unmanaged switch and a managed switch are the same thing at layer 2 (unless you start playing with VLANs, which I do occasionally, but not with the tuner or HTPC). My "router" is a Soekris industrial PC running pfSense and it connects to the switch with a single port, everything else is on the same 24p switch, or in the case of one XBox, connected to an Airport Extreme (gigabit) that is in bridge mode and connected to the switch. There is also a "special" Netgear "router" that is actually in bridge mode between my pfSense system and the HP switch, it's a SamKnows broadband monitoring program box that just passes traffic unmolested through it. The cabling is Cat5e throughout the house and I show zero errors on the ports for the Eth6 or the HTPC, so I don't really see how Ethernet could be at fault here.
You really should shoot for a 0dBmV signal level on the tuner coming off whatever TWC puts in. Let them scan the drop at the end of the coax that you're going to attach the tuner and I would recommend that you don't use the loop through on the 1520. The moto is worse since those units have (or had if they've corrected them) a 6 dBmV signal drop from the loop through.
There isn't much of a "drop" to the Eth6. The main run from TWC comes into my crawlspace, where I have what is effectively a "wiring closet" under the stairs. This is an older picture (Eth6 isn't in here yet) and some stuff has changed, but it's a rough idea of the situation (click to embiggen):
Image
The Eth6 has a little 2 ft ethernet cable into the switch. There's maybe a 3ft coax from the 1520 into the 4-way split and another 2 or 3 ft coax from the 1520 into the Eth6. These were all pieces of RG6 quad from TWC or possibly leftovers from when I was with DirecTV - they should be good. Didn't see any braid touching the copper when I hooked it all up.

barnabas1969

Posts: 5738
Joined: Tue Jun 21, 2011 7:23 pm
Location: Titusville, Florida, USA

HTPC Specs: Show details

#20

Post by barnabas1969 » Thu Jul 25, 2013 9:05 pm

You could try that amplifier that you already have. It has a 4.5dB gain. If you do it this way, put a splitter before the TA, and don't feed the tuner off the output from the TA.

If you only have one coax running to your wiring closet, then use a 3-way splitter. Use one output to feed your cable modem, one output to feed your tuner, and one output to feed the TA.

Just give it a try. No need to make this too complicated (yet).

As for the techs from the cableco... in my experience, the sub-contractors are about as worthless as tits on a bull. The guys who actually work for the cableco (Mine is Brighthouse, but was formerly TWC in my area) carry better testing equipment and are trained wayyyy better.

Post Reply