Networking woes: understandable ignorance exposed

-May 07, 2013

The other night, while I was over at a friend's house, he asked me to help him diagnose a problem he was having with his Roku streaming player. One of his Rokus, to be exact. Both are connected to his router over 802.11g Wi-Fi. The router, an oldie-but-goodie Linksys WRT54G, is located in his downstairs office. Two rooms over, in the living room on the same (ground) floor, is the newer Roku unit that I bought him and his wife for Christmas. And in the upstairs bedroom, directly above the office, is an older Roku model, the problematic one.


At first, I thought my friend might be having video streaming issues, which could theoretically be fixed by upgrading both the router and (older) Roku to higher-speed 802.11n-supportive variants. But his problem, as it turns out, involved far lower bitrate (therefore more bandwidth-friendly) audio, specifically sourced from Pandora. When accessing Pandora on the first-floor living room Roku, playback began near-immediately. On the upstairs unit, on the other hand, simply pulling up the Pandora UI was glacially sluggish; buffering prior to playback initiation was slower still.

My next thought was that the Pandora experience difference might be simply due to the model and age variance between the two units. But I own the same version as his older one, and it handles Pandora with aplomb. Admittedly, his Roku might be broken, specifically within its Wi-Fi subsystem. But any remaining doubts as to its functionality were dashed when he told me that the unit used to work just fine ... when it was downstairs, in the exact location that its successor currently inhabited.

Obviously, the Roku itself wasn't the problem; I was likely instead dealing with a wireless signal strength issue. To confirm my hunch, I pulled out my iPhone 4 and fired up its WiFiFoFum app (yes, my handset is jailbroken). I measured the signal strength in the office next to the router, then in the living room and bedroom. Then I scratched my head, because the Wi-Fi beacon was slightly stronger nearby the upstairs Roku than next to its downstairs peer. Why was the bedroom unit having connectivity issues, when it was being presented with a few-dB superior signal?

I looked closer ... and I had my answer. In-between the router and upstairs Roku were the office ceiling and bedroom floor (along with whatever was in-between them), of course. But the Roku was also sitting on top of my friend's audio receiver. The receiver contained pre-amp and power amp modules, along with a hefty power supply and other analog circuits, all of which were likely radiating all sorts of spectrum interference. And the receiver was constructed from a grounded thick metal chassis, which I suspected was acting as an effective Faraday shield. Indeed, moving the Roku away from the receiver, thereby giving its antennas direct exposure to the floor below them (and the router below it) produced immediate network-responsiveness benefits.

Often times, in situations (and blog posts) like this, I point blame fingers at one or multiple culprits. This won't be one of those times. Had my friend configured the Roku from scratch, he would have likely noticed the sub-par signal strength as reported by the device's setup UI. And as a non-techie, he can't be expected to understand the attenuating effects of a close-proximity device such as the audio receiver between a Wi-Fi source and destination. Had my friend perused a resource such as SmallNetBuilder's recently published "How to Improve Network Performance" series (here's Part 2 and Part 3), he might have also realized what he'd done (or perhaps not even done it in the first place).

But he moved an already-configured Roku from one location to another in the same house, connected to the same wireless LAN, and therefore reasonably skipped the setup sequence. And SmallNetBuilder (and its peers) are, I suspect, predominantly frequented by the already-techie and therefore of little use to someone like him. I don't have any easy answers to a situation like his. But I wonder how many folks just like my friend are out there. And I wonder how many unnecessary product returns and replacements occur every year, due to situations similar to his.

Next week, I'll share another recent networking stumble, this one that I personally experienced, and which this week's story reminded me of. Until then, I as always, welcome your comments.

Also see:

Loading comments...

Write a Comment

To comment please Log In