<div dir="ltr">><span style="font-size:13px">This got an A+ rating, which I would not have given it, given the</span><br style="font-size:13px"><span style="font-size:13px">enormous load spike.</span><div><span style="font-size:13px"><br></span></div><div>I think there will always be the occasional incorrectly graded test,</div><div>this one is simply because the median of the downstream latency </div><div>ignores the spike. If I used average(), then it would not ignore</div><div>the spike, however one very high outlier could also ruin a good result.</div><div>After all, pinging anything on the internet can always get the odd</div><div>bad response now and again.</div><div><br></div><div>If neither average nor median is any good, then there needs to be </div><div>a filter function. But what filter? ignoring spikes that are hugely higher</div><div>than neighbouring ones? that would fail if there was a spike every 3rd</div><div>sample. Open to ideas..</div><div><br></div><div>Here is a result from the australian telco free public hotspot:</div><div> <a href="http://www.dslreports.com/speedtest/399962">http://www.dslreports.com/speedtest/399962</a><br></div><div><br></div><div>On the side of the hotspot it says 'send us your thoughts about this</div><div>free service'. Well my thought is that if one person posted a picture</div><div>to Instagram, the whole hotspot would be unusable for as long as it</div><div>took to upload. 6 seconds of buffer in there somewhere.</div><div><br></div><div>cheers,</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, May 1, 2015 at 4:05 PM, Dave Taht <span dir="ltr"><<a href="mailto:dave.taht@gmail.com" target="_blank">dave.taht@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">This got an A+ rating, which I would not have given it, given the<br>
enormous load spike.<br>
<br>
<a href="http://www.dslreports.com/speedtest/400387" target="_blank">http://www.dslreports.com/speedtest/400387</a><br>
<br>
Imagine if your steering wheel behaved like this.<br>
<span class=""><br>
On Thu, Apr 30, 2015 at 8:10 PM, jb <<a href="mailto:justin@dslr.net">justin@dslr.net</a>> wrote:<br>
> Already users are like "how can i fix this!".<br>
<br>
</span>The FAQ can be improved.<br>
<span class=""><br>
> I've just replied to one who has lower speeds on the surfboard SB6141 which<br>
> is a modem designed for crazy cable speeds. He has an "F" and his downstream<br>
> bloat is terrible, and upstream not much better.<br>
><br>
> I imagine a LOT of people on slower plans have a "recommended" modem like<br>
> this one.<br>
<br>
</span>I have not found a cable modem with less than 250ms bloat at 50mbit/5.<br>
The docsis 3 ones<br>
are often in the 800 ms range.<br>
<span class=""><br>
><br>
> However most of them will hear that the problems from bloat only happen when<br>
> you reach maximum upload or download speed and will think, well, I can live<br>
> with that, I never run my connection to capacity and I don't upload to<br>
> offsite backups..<br>
<br>
</span>Latency spikes are annoying no matter how they are inflicted, and happen<br>
all the time on nearly any workload. Your test is testing tcp in steady state,<br>
most web transactions are bursts of dozens to a hundred flows in slow<br>
start.<br>
<br>
It is the business class customers that feel it most often. I have never<br>
visited a business class cable customer that had reasonable amounts of delay<br>
and jitter during business hours.<br>
<br>
After living in bloat-free universe for quite some time now, annoying<br>
issues with things like netflix are decreased, voip and videoconferencing<br>
work all the time, same for games...<br>
<br>
it would be hard to create a metric<br>
for user satisfaction, but every before/after comparison someone<br>
implementing a solution is quite overjoyed.<br>
<br>
<a href="https://twitter.com/mnot/status/575581792650018816" target="_blank">https://twitter.com/mnot/status/575581792650018816</a><br>
<span class="im HOEnZb"><br>
><br>
> On Fri, May 1, 2015 at 10:48 AM, Rich Brown <<a href="mailto:richb.hanover@gmail.com">richb.hanover@gmail.com</a>> wrote:<br>
>><br>
>><br>
>> > On Wed, Apr 29, 2015 at 9:33 PM, jb <<a href="mailto:justin@dslr.net">justin@dslr.net</a>> wrote:<br>
>> > ...<br>
>> >> if it did get a rating it would be an "D" or "F"..<br>
>> ><br>
>> > How about "E" for error? That can be further explained in the text<br>
>> > "Sometimes the bloat is so bad that we cannot adaquately test for it -<br>
>> > and other times there is something else badly wrong with the link that<br>
>> > we cannot identify."<br>
>><br>
>> I would stay away from a letter grade for that state, since it could<br>
>> appear to be on the continuum of A+, A, B, C, D, E (?) F...<br>
>><br>
>> Better to give it a "-" or "?" mark. And if they hover over the "?", let<br>
>> the text show: "Sometimes the bloat is so bad that we cannot adaquately test<br>
>> for it - and other times there is something else badly wrong with the link<br>
>> that we cannot identify."<br>
>><br>
>> Rich<br>
><br>
><br>
<br>
<br>
<br>
</span><div class="HOEnZb"><div class="h5">--<br>
Dave Täht<br>
Open Networking needs **Open Source Hardware**<br>
<br>
<a href="https://plus.google.com/u/0/+EricRaymond/posts/JqxCe2pFr67" target="_blank">https://plus.google.com/u/0/+EricRaymond/posts/JqxCe2pFr67</a><br>
</div></div></blockquote></div><br></div>