What is expect throughput of an MR18

[Paul03]

MR18 was on port 33 of Cisco switch and getting 60 Mbps throughput

MR33 was on port 21 of the same switch and getting 260 Mbps throughput, both ports configured the same way

All our MR18’s in different buildings and cities are getting 60MBps and MR33’s get around 260 Mbps

Switched ports 21 and 33 and rebooted both AP’s, MR18 still getting 60 Mbps and MR33 still getting 260 Mbps so not a switch issue but is a Meraki MR18 issue.

Meraki tech support open a case for this and after 6 weeks decided it was normal behaviour and expecte throughput for the MR18, even though they had my run iperf rtests and update merakis and many other test sduring the 6 weeks, so if it was expected how could they not know that 6 weeks ago?

From the web 

Dual enterprise class 802.11n radios, up to 600 Mbps
The MR18 features two powerful radios and advanced RF design for
enhanced receive sensitivity. Combined with 802.11n technologies
including 2x2 MIMO and transmit beamforming, the MR18 delivers
data rates of up to 600 Mbps and enhanced coverage, meaning
fewer access points are required for a given deployment. In addition,
the MR18 uses band steering to automatically serve 5 GHz-capable
clients with the 5 GHz radio, maximizing capacity in the 2.4 GHz
range for older 802.11b/g and 2.4 GHz-only clients

[PhilipDAth]

The MR18 is 802.11n.  The 600 Mbps is the "wire" rate (or more specifically the access rate).

The MR33 is 802.11ac.  It is also much newer hardware and should be far superior.  It has an advertised wire rate of 1.3 Gbps.

I would be disappointed if the MR33 did not outperform the MR18.

[Paul03]

I agree the MR33 should be faster than the MR18 but the MR18 should be much faster than it is, it advertises600 MBps and Meraki didn't know for 6 weeks that was the best I could expect? Won't be purchasing any more merakis

[MerakiDave]

Hi @Paul03 let me know your case number in a private message and I can look into it.  While it's true that 600Mbps is the theoretical maximum PHY rate, that is not achievable in any actual deployment even with the most pristine RF conditions, highest data rate, short guard interval and using all 4 spatial streams (which no client devices have).  Those are all the things that get 11n to the 600Mbps number.  But that said, even though the MR18 is a much older AP with a different chipset, you should certainly be getting better than 60Mbps.  I have seen plenty of MR18 APs pass hundreds or Mbps consistently all day long, so there must be some other variables in play.  Sorry to hear it sounds like a case was lingering for several weeks, but I can have a look into it and maybe provide some suggestions via private message. 

I also wanted to clarify you are not using the "Dashboard Throughput" test on the live tools page for the AP, since that is only meant to tell you if you have a reasonably good connection to the Dashboard for management, definitely not for AP performance testing or Internet speed testing.  I was assuming you're using some 3rd party speed test tool, from the same device like a laptop, connected to the MR18 and then repeat with the only difference being associated to the MR33.  Also wanted to confirm if the MR18 and MR33 are part of the same Meraki Dashboard network, with the same configuration and same radio settings.  

[Paul03]

Sorry for slow delay case number is Case 05452657

Initially test was with dasboard but then ran with iperf and still same slow results.  "Tech" would say one thing on phone and contradict himself in e-mail.

Thank you

[TBHPTL]

A 2x2:2 802.11n can only hit a maximum of 300Mbps PHY at 40MHz 5GHz PERIOD.  

• That's HT MCS rate 15 / VHT MCS rate 7.
• 802.11n uses a 20MHz or 40MHz wide channel only.   
• 802.11n 3x3:3 can hit 450Mbps @40MHz 
• 802.11n 4x4:4  can hit 600Mbps @40MHz 

WiFi 101 / WiFi 10 Commandments  -

Commandment 1. Cut through the sales BS by understanding the MCS tables and be able to read a data sheet

Commandment 2. Thou shall never, ever, ever run 2.4GHz past 20MHz wide or you will be stoned in the biblical sense.

Anyone claiming that a 2x2:2 can hit 600Mbps  is just flat wrong an attempting to mislead you either on purpose or due to not knowing how it works. How they are getting the BS sales number is by adding the 2.4GHz @40MHz wide max PHY rate  with w/ a 40MHz wide 5GHz  max PHY of 300Mbps. 

In practicality it is super easy to hit this PHY rate, the problem is people too often think that is the throughput. IT IS NOT. This is the maximum theoretical speed the device transmits and receives the data over the air when it gets to "talk or "listen" to  a client. You throughput will be substantially lower than that. A good median speed for 2x2:2 802.11n on 40MHz wide 5GHz is between 240 to 300Mbps divided by 2, because wireless is half-duplex.   For 20MHz wide N on 2.4GHz  it is between 115 and 144Mbps, again divided by 2. If you are connecting using any modulation below QAM64 you have other issues in your environment that you must correct and/or account for.  If you aren't getting these PHY rates at QAM64 you have to look for the interfering devices on the airwaves... something is wrong. If you are getting PHY RATES using  QAM64 and it still sucks.... check your cabling and interconnects. Also, make sure you don't have any rate limiting GP's applied. Note A GP will not change your MCS but it can limit your throughput.