Oh well, at least it's

different
Everything here is my opinion. I do not speak for your employer.
August 2015
October 2015

2015-09-03 »

Wifi marketing lies reach newer, crazier heights!

http://www.engadget.com/2015/09/02/asus-router-worlds-fastest-wifi/

Interestingly, this one claims 1 Gbps over 2.4 GHz.  How's that possible?  Is it the world's first 8x8 consumer wifi AP?

Nope.  It turns out that while I wasn't looking, 802.11ac seems to have grown two more even-more-unachievable MCS levels: 10 and 11:
https://en.wikipedia.org/wiki/IEEE_802.11ac#Theoretical

At 40 MHz, 4x4, you'd get 4x the 40 MHz value for MCS11 in that channel, so the "marketing rate" would be almost exactly 1000 Mbps.  To deconstruct that a little using my formulas:
- 1000 Mbps advertised
- real clients only get 2x2: 500 Mbps
- real clients only get mcs7 or less: 300 Mbps
- PHY rate to UDP rate: 2/3 -> 200 Mbps
- UDP rate to TCP rate: 90% -> 180 Mbps
- Realistically 2.4 GHz networks can only use 20 MHz channels: 50% -> 90 Mbps.

Thus achieving a world record lie factor of (1000-90)/90 = 1011%.

2015-09-12 »

It's a relatively well-known thing that wifi speeds decrease if you're too close to the access point: the "screaming in your ear" effect.  But we now have stats that demonstrate it clearly.

I zoomed into just Chromecasts.  In the top graph, notice the speed decline starting at -30 dB or so RSSI.  In the second graph, notice how there's actually a spike in number of devices starting at around -30 dB.

My guess is the bimodal distribution happens because there are two places to put your chromecast: right next to your AP which is right next to your TV, or not.

We talk to the chromecast on 2.4 GHz, which means we're using ath9k.  The ath9k driver currently sends all packets at maximum transmit power, even to devices right next door.  I wonder if a minstrel-like algorithm (such as minstrel-blues) which experiments with different power levels would actually fix this problem and achieve full speeds (downstream toward the chromecast anyway, albeit not back upstream) even when gratuitously close by.

2015-09-20 »

OMG creepy.  I think this is what we call "turning a negative into a positive" in marketing.  Not sure it landed right, though.

https://stream-digitallife.att.com/2015/05/27/teenager-house-party/

2015-09-22 »

So I thought I was doing pretty well against impostor syndrome, until last week we released a wifi speed dashboard.  What I see in the dashboard is a long list of wifi performance points we still want to improve.  Everybody else seems to think this thing is the best thing ever.

Definitely the most literal feeling of "waiting for people to catch on at any moment" that I've had since I got here.

2015-09-23 »

"Every day is a bad hair day; every moment is one of those awkward moments between haircuts."

2015-09-25 »

   // We only test the success case because the failure cases shouldn't occur

2015-09-30 »

Something is wrong with this picture, and I don't mean the part on the right where I clipped out "The Meatball Shop" as a match for my search for "cupcakes."

August 2015
October 2015

I'm CEO at Tailscale, where we make network problems disappear.

Why would you follow me on twitter? Use RSS.

apenwarr on gmail.com