Upcoming Maintenance Alert:

The UBNT Community will be upgraded at 5pm MDT on April 25th. During this time the community forums will be set to read-only status.

Learn more

×
Reply
Emerging Member
Posts: 73
Registered: ‎01-28-2010
Kudos: 34

5.3.2 Improved CCQ calculation

Not sure if this is an improvement in the actual CCQ or just the calculation, but you can clearly see where I upgraded from 5.3 to 5.3.2...
b2h-53to532.jpg
Highlighted
Veteran Member
Posts: 4,992
Registered: ‎08-26-2009
Kudos: 1410
Solutions: 42

Re: 5.3.2 Improved CCQ calculation

The CCQ no longer takes into account the data rate, just the error rate now

CCQ (pre-5.3.2):

Keba - "Equation is simple:
CCQ = (100 - Error_Rate) * Current_Rate / Highest_Rate"

Now CCQ, is probably just 100 - Error_Rate, which is really what it should have always been. Taking the data rates into account seems useless, as the data rates are already reflected in AMC, that's what AMC is for!

So we went from having three different values that all reflect data rates, down to two values:

AMC/AMQ (didn't change):

Sriram - "Capacity is related to the device's :

1) RX and TX transmit rates
2) RX and TX packet error rates.
3) Maximum possible RX and TX rates for the mode


So its the average of
((TX_RATE * 100)/MAX_TX_RATE) * (100 - TX_PER)/100
and
((RX_RATE * 100)/MAX_RX_RATE) * (100 - RX_PER)/100

The capacity for 1x1 devices is capped to 50%, so the value is divided by 2.

Quality is weighted average computed using TX/RX rates, TX/RX signal RSSI, TX/RX PER "
Established Member
Posts: 1,149
Registered: ‎02-19-2010
Kudos: 63
Solutions: 1

Re: 5.3.2 Improved CCQ calculation

Wich software did u use to take out these graphs? I use cacti, what do u use?
Dott. Elia Spadoni
---
Network Administrator
MTCNA, MTCRE, MTCTCE, MTCINE, MTCWE
Spadhausen Internet Provider
www.spadhausen.com
Established Member
Posts: 2,058
Registered: ‎08-02-2008
Kudos: 129
Solutions: 3

Re: 5.3.2 Improved CCQ calculation

Yes, it is strange, now my CCQ readings are pretty much meaningless, they all rocketed way up with the new firmware. Not sure what to believe, the old readings or the new readings.
Emerging Member
Posts: 73
Registered: ‎01-28-2010
Kudos: 34

Re: 5.3.2 Improved CCQ calculation

I use Nagios with the plugin I posted here to collect the stats and pnp4nagios to generate the graphs.
Ubiquiti Employee
Posts: 7,391
Registered: ‎11-27-2007
Kudos: 4210
Solutions: 167
Contributions: 45

Re: 5.3.2 Improved CCQ calculation

Yes, it is strange, now my CCQ readings are pretty much meaningless, they all rocketed way up with the new firmware. Not sure what to believe, the old readings or the new readings.


You can believe them both if you want.

The old CCQ was more of a number that contained a tx rate over theoretical max rate multiplier.

The new CCQ is simply a quality value at the current rate (errors/retries, etc), more of a PER.

Now when datarate drops, the CCQ doesn't also drop, like it was. Logical thinking was when the rate drops, quality improves, but CCQ was showing the opposite.

Plus CCQ at static rate never used this rate multiplier, so everyone thought AutoRate had bugs because CCQ at Auto Rate was always way less than Static Rate.

If you want to know a capacity based quality value you can still look at the AirMax Capacity, or just take a look at the rate itself.
Regular Member
Posts: 444
Registered: ‎08-31-2010
Kudos: 49

Re: 5.3.2 Improved CCQ calculation

I will be deploying 5.3.2 this weekend, but the new CCQ seems to be more intuitive to me. I want CCQ to show me how well the link is running at it's current air rate. If I want to see what rate it's running, I look at the rate. If I want to know how close it is to the ideal perfect link, I look at AMC
Reply