"dgrace" wrote:
Just ran the speedtest. First thing it said was the Netflix test was unavailable.
Apparently Netflix took down their speed test. I was hoping it was just temporary and it would come back, but it doesn't look like it's going to. I know they used to tell customers to run the test and give them the results when the customer had a problem. I wonder what they do now.
"dgrace" wrote:
I assume it's testing against the nearest server it finds, which is Missoula, about 12 miles from me.
Yes. I may add an option to pick the server in the future.
"dgrace" wrote:
The initial latency is 50.8ms, which is about what I see (midrange) when testing from a browser (38 to 80ms or there about). The other tests look as if they test with increasing sized packets, and the latency goes 450, 737, 1401, 2454, 4535 and 8434ms. The kbytes/sec range from 600 to 973.
Yes, it tests with increasing sized files. The numbers for the tests given in milliseconds aren't the latency but the total time it took to transfer the file (minus the initial latency). The rest is just math. If you take the size of the file and divide it by the time it took you'll get the kbytes/sec number. Multiply that by 8 and you'll get the kbits/sec number.
"dgrace" wrote:
Also, a question on "too many channels". I only have Netflix, Amazon, Pandora, MyMedia, MediaFly and TuneIn Radio. Wouldn't think that were too many.
MyMedia is needlessly huge. I sent el.wubo an update before Christmas that stripped out all (well, most) of the unnecessary stuff and it's less than 1/10th the size of the current channel. I guess I'm going to have to contact him and see if he's still interested in this project.
"dgrace" wrote:
Would the saved favs in Pandora, TuneIn and MediaFly also count against this, or are they retrieved when selected? For instance, I have about four "stations" set up in Pandora, but I can delete those via a browser so think they reside on their site, maybe I'm wrong.
I believe you're right about that, so they wouldn't count against the number of channels.
"dgrace" wrote:
As for the size of my Flip videos I've been streaming, they range too from very short 1 or 2 minute, to probably over the 64meg buffer memory.
I can't imagine the bottleneck is my router, as I've also been researching getting the best settings in it for my usage. I still want to think it's something in the roku, as I've never had a rebuffering with my Wii and Netflix, but have had on occasion with the roku. Same network, different performance.
Keep in mind that the Wii uses different Netflix streams than the Roku, so the comparison isn't necessarily relevant.
"dgrace" wrote:
I may play around with video codecs on the Flip mp4 files and see what I get. If the picture quality suffers then I'll give that up, but would like to find out why MediaFly vids often rebuffer.
It's worth a shot. I haven't tried MediaFly, but I still think it would be worth trying to bypass the router to see if it makes a difference, especially if you find a video that always rebuffers.
"dgrace" wrote:
Thanks for the tip on the speedtest, but in the interest of fewer channels I suppose it aught to come out.
The speedtest channel is tiny (14kB), so you wouldn't gain much by removing it.
"dgrace" wrote:
Updated: More questions. When I ping the roku from a laptop on my LAN (50 pings) I get min of 2ms and a max of 150ms and an average of 21ms. Should there be that much of a range internally here? I ping (from same laptop) to other computer, or a printer or the router and they vary a little but seem much lower.
That seems kind of strange. When I ping my Rokus I generally get 1 to 8ms. I know latency plays some part in the operation, but I don't really know how much. It still makes me think it's a router issue.
-JT
