Exactly the same problem. It actually was a rolling failing with more and more functionality disappearing over a couple of days. First it was failure to access TV and recordings on Google Chrome, then Firefox wouldn’t work with Tablo. Then, today, my recordings became inaccessible (although the list still shows) and finally live TV is falling apart despite rescanning the channels and updating the guide twice… only 2 of the 11 channels function.
I don’t know if even a rollback to 2.2.2 can restore everything at this point.
Very few problems before upgrade to 2.2.6, now takes about 1 hour to watch a 30 minute recording, regular “loading please wait”. Live TV is not watchable due same problem.
Cable internet 60/6 Mbps
Hard drive WD elements 2TB
Max recording quality HD 720 5 mbps
Router TP Link Archer C9
Roku 2 XD 6.2 build 6003
Cannot hard wire, antenna and cable inlet at opposite sides of house.
No problems with Chrome on laptop, Android on phone or Kindle HDX.
OK, sharing a few testing results (all pure wired except for line 1):
On 3 mbps/wired, watched about an hour of the world series with no issues whatsoever on the Roku 3 (wired), and about 30 minutes on Roku Stick (5 ghz wi-fi). Not a single ‘loading’ item.
Changed settings to 8 mbps, rebooted Tablo. On Roku 3/wired setup, seeing minor, but noticeable ‘loading’ items. Watched 10 minutes and had 2 occurrences of ‘loading’ errors. To be specific, it did not get caught in the ‘infinite loop’ that I observed Sunday night, just a 1-2 second pause in the play.
Changed settings to 5 mbps, rebooted Tablo. On roku 3/wired setup, observed a brief ‘loading’ pause within a few seconds of selecting the live broadcast, but, no other issues for following 30 minutes watching.
Fox, meanwhile, had a power outage in their truck and lost the World Series broadcast for 3-5 minutes…gotta love the irony, even a multi-billion dollar company, in one of their prime events of the year, has challenges.
Haven’t tried further wi-fi testing, but encouraging results. Note that I’m still having severe delays while watching recordings after fast forwarding. As a work around I’m going to use Tablo Ripper and ship my recordings to Plex to watch.
Just tried the restart everything re-install everything for the roku 2 on wifi. still getting long load times after ff. there were no problems just like everyone else said before the 2.2.6 update. We moved our roku to the main tv because it just worked better then our fire stick with the tablo. Now the fire tv stick has no buffer times and is working awesomely. but the roku is doing this long load times for recordings. I have noticed that if i fast forward 10 to 20 seconds it is fast but then as i increase the time the time for the buffer increases. Anyone else notice this? so maybe something strange is going on with buffering it seems to be different depending on how long i let the show play.
Tablo is on Ethernet with power line adapter. Roku 2 2013 on os 6.2
Here’s an observation, take it for what it’s worth.
It appears that the “Loading please wait” issue may be associated with higher bandwidth (1080i) broadcasts with lower signal quality (in other words, more receive errors).
I have a powered antenna with a 32dB LNA. I think I maybe be giving the receiver excessive signal on the closer more powerful channels. When I first set this system up I had this antenna sitting on a table in an upstairs bedroom. I got all the main local channels, but some of the more distant channels had poor signal quality (2 or 3 bars). This weekend I moved the antenna to the very top of my attic. Now, the channel scan reports 5 green bars on all channels, but perversely my strongest channels now have occasional error artifacts in the picture that were not there before. This makes me suspicious that those channels are seeing excess signal which is saturating the receiver. These are also the very channels that have the worse “Loading please wait problem”.
I wonder if when the Tablo is having to do error correction/handling that you’re getting a processor loading issue. It might also explain how you guys missed this issue in testing if you’ve got very clean signals in your lab. Signals that are too weak or too strong may be causing errors that are contributing to the LPW issue.
I’ve ordered some 3,6,and 10 dB 75 ohm attenuators to see if I really do have an issue with excessive signal. Should know something as soon as they show up. Needless to say, it would be really nice if the Tablo gave us more insight into OTA signal quality. RSSI, SNR, and Error Rate would be highly useful information.
As you have demonstrated empirically, signal strength has much less to do with proper OTA reception for digital signals compared to other metrics. Multipath and fading due to weather, foliage, aircraft, etc. make an enormous difference. It’s actually difficultt/ relatively expensive to build a useful signal quality meter for ATSC OTA signals. Google “Eye Diagrams digital signal quality”.
Not sure what your “32 dB LNA” actually delivers, but low noise figure rather than pure gain of 32dB is what you really need. Overloading the Tablo RF front end could be a real issue on strong local stations.
Distorted OTA signals push the error rates up and the transcoder in Tablo can and will fail, albeit gracefully. The Roku decoder codec may not be nearly as fault tolerant however, so reloading and stalling may be the consequential result. The Roku struggles but other players may handle the error-corrected stream without issues.
I’m suspecting that the signal out of my LNA is overloading the RF front end causing it to enter compression. If so the harmonic distortion that results will definitely close the eye pattern and lead to increased error rates. Have no idea what the NF is on the LNA integrated into this antenna - not advertised.
I agree it is difficult to quantify signal quality on a digital link. I have no idea what Tablo is displaying on their “signal dots”. Could be RSSI, could be “Link Quality” which tends to be derived from some equation involving Signal Strength (RSSI) and SNR. Who knows. If RSSI only that can be very deceptive indeed.
Of course, the final word is bit or packet error rate. If a person could look at RSSI, SNR, and Error Rates on a per channel basis, you could make some reasonable deductions about what is going on.
Working with Tablo support they were able to see that my Tablo was having network issues on the LAN. I rebooted the switch my Tablo was connected to, then rebooted the Tablo and all started working normally again.
I have multiple Samsung Smart TVs and a Samsung BluRay Player where I access Tablo through the Plex App. These all now, following the 2.2.6 maint., fail when trying to access Live TV or Recordings on the Tablo. On the TVs I get an error messaage that 'Player can not render this content. Media might be corrupt or media type is not supported."
It appears the update has rendered the streaming content incompatible with the Samsung Plex App for Tablo in some way.
So, I dropped the signal out of my amplified antenna by about 4.8 dB (3 way splitter with 75 ohm terminators in two of the ports). Much better picture, almost no error artifacts on the screen and the “Loading please wait” problem is greatly reduced, almost gone. I’ve got some real 6 and 10 dB attenuators on the way, I’ll probably drop the 10 dB in and see how that works when they arrive.
Also, I’ve gained some additional insight. As I say, with the OTA signal quality improved, the “Loading please wait” issue is almost completely gone - except I noticed that when my microwave oven is on I do see it sometimes.
Looking at the secrete WiFi menu on the Roku, it looks like I’ve only got about 10 ~ 12 dB of SNR. Not a huge amount of margin there. It seems that when the microwave oven is on, my SNR is low enough that I start dropping more WiFi packets.
It looks to me like the video decoder in the Roku is somewhat brittle and any anomalous packets can cause issues. Probably my OTA problems on HD interface were causing transcoding errors or weird timing. Likewise the decoder is unhappy with packets being dropped over the wifi interface. This problem I can address. I think I’ll put a little Netgear access point hardwired back to my router right behind the TV.
Things look much better tonight. If they stay that way for awhile I think I can call this progress.
Microwave ovens operate at 2.4GHz just like 802.11b/g/n/ac WiFi. (802.11n/ac also work at 5GHz. In general, 5GHz has better range but 2.4GHz is better at going through obstacles like walls.) The inside of a microwave is basically a 2.4GHz transmitter with a kiloWatt or more of power output. The oven part of the microwave oven is basically a Faraday cage to contain the power, but they commonly leak a wee bit. 5GHz is much, much less crowded in most environments (but don’t tell my neighbors!).
Sounds like overload from strong local stations causes the non-linear front end / 1st RF stage of the Tablo to go into distortion. An inline attenuator will help, but I am still concerned your antenna and 32dB amp are a very bad idea, considering the total reception environment. Although your strongest channels will no longer drive your tuners into as much amplitude and phase distortion once you pad the amplitude down by 5 or 10 dB, your weakest channels may disappear entirely. It would be great if you had enough antenna directivity to put the weak channels on the peak azimuthal lobe and have your strong signals sit down on the skirt 10 dB or lower, thus ‘leveling’ the spectral power distribution rather than bulk attenuation. Perhaps your location or antenna will not cooperate.
The Roku decoder / playback codec for h.264 is weak, and even non-OTA streams with FECC make it stutter. Thankfully Tablo makes pretty good transcodes from ATSC, but highest bitrate 1080 can push it past the tipping point when FECC can’t recover enough bits.
Also, FWIW, I would categorically say that hard wired Ethernet is so much better than WiFi for feeding a Roku at high bitrates that I always run STP Cat-5e or now Cat-6 for streaming video. I do 4K UHD on my local network at 100 Mbits/sec and WiFi would be truly pathetic. If you can go router to Roku and router to Tablo with copper, just do it ! And particularly if you have a microwave oven, cordless headphones, baby monitor, cordless phones, etx., why burden your Tablo or Roku with EMI when you already are coping with video streaming errors?
I think if I can pad everything down by 10dB without seeing the weaker stations drop out too much I should be fine. I’m padding down just under 5dB right now and the weaker stations look great, I just thought I’d buy myself some more margin on the high end. I guess the real test will be how the weaker stations look in sub optimal conditions, i.e. rain, etc.
Yes, wired ethernet is going to be more reliable than wifi. I’ve got a stick in this TV, but I’m planning on moving that to a small TV in the kitchen anyway, so maybe I’ll put in a wired connection when I replace it with a Roku 2/3. I’ve already got Cat5e in the wall behind this set.
And this is where the PowerLine adaptors are great… they dont rely on wifi and use your home wiring for ethernet connections… and are immune to microwave interference… I first used one on our family room TV which would get knocked out anytime someone was making popcorn…
The pad is a good way to confirm the underlying problem as you very nicely demonstrated. Getting the dynamic range of your antenna output to best ‘fit’ the linear region of the Tablo front-end when rain fade or other attenuation occurs needs to be decided based on your local factors and antenna constraints as well as your desire to spend time tweaking. I actually wound up with 2 OTA antennas and previously used 2 separate PC-based tuner cards in my prior DVR, and had every intention of buying a 2nd Tablo 4 tuner box until I encountered the never-ending saga of absolutely pathetic firmware which makes these entertainment devices a total PITA.
)