hAP ac^2 Problems---Extremely Poor Performance found in 2.4G and 5G WiFi

RouterOS Wireless Networking
1

Someone has removed the word “SOLVED” from the thread title. 2 May 2018

As requested by one of Forum member, I add “Not” solved in the heading.
https://forum.mikrotik.com/viewtopic.php?f=7&t=132648&start=200#p658634
YH 1st May 2018

The WiFi issues on hAP ac^2 for both 2.4GHz and 5GHz were completely resolved. If you are not able to achieve the targets such as 400Mbps (DL/UL) for 5GHz and 100Mbps (DL/UL) for 2.4GHz, the possible cause of the problem(s) may be due to the incompatiblity between hAP ac^2 WiFi chip and your device (such as PC, tablet). You may try to find another device to test the speed. (Please note that you have at least upgrade your hAP ac^2 to v6.43r5 or above).

For the last message, I found that using Intel-based WiFi chip (CHUWI notebook) can give us the least errors and stable speeds.

Latest Firmware Downloading (DL) Performance Short Summary:

Firmware v6.43rc5
File server → network switch → WAN of hAP ac^2 → WLAN 2.4GHz → PC (MacBook Pro Retina 2015), 55.9 Mbps, Retr=1298 (This can be resolved by selecting 802.11n and 20MHz of bandwidth), please see http://forum.mikrotik.com/t/radius-server-not-working-in-2-8-11/127/1
File server → network switch → WAN of hAP ac^2 → WLAN 5GHz → PC (MacBook Pro Retina 2015), 388 Mbps, Retr=286 (Based on the test results, the 5GHz is resolved)

Firmware: v6.43rc3
File server → network switch → WAN of hAP ac^2 → WLAN 2.4GHz → PC (MacBook Pro Retina 2015), 38.9 Mbps, Retr=1566
File server → network switch → WAN of hAP ac^2 → WLAN 5GHz → PC (MacBook Pro Retina 2015), 163 Mbps, Retr=5709
File server → network switch → WAN of hAP ac^2 → LAN → PC (MacBook Pro Retina 2015), 941 Mbps, Retr=29

Firmware: v6.42
2.4GHz DL Performance: 55Mbps, Retr=1481
5GHz DL Performance: 124Mbps, Retr=6418
LAN DL performance: 941Mbps, Retr=0

Firmware （Stock/Factory Firmware: v6.40.5)
2.4GHz DL Performance: 40Mbps, Retr=3541
5GHz DL Performance: 16Mbps, Retr=3628
LAN DL performance: 941Mbps, Retr=0

After having done a number of tests for hAP ac^2 including MikroTik btest.exe, actual file test, iperf3, based on member’s experiences and the results obtained. In the sub-sequent illustrations, 5-minute iperf3 -R test is employed. The data path direction is iperf3 sever → network bridge → WAN of hAP ac^2 → WLAN (2.4G and 5G) → Laptop computer.

For comments on Listings 1 to 6, please click https://forum.mikrotik.com/viewtopic.php?f=7&t=132648&start=50#p653894

Listing 1: hAP ac^2 v6.40.5 (stock firmware) (Server->WAN->LAN->Laptop PC)

[ ID] Interval           Transfer     Bandwidth       Retr
[  4]   0.00-300.00 sec  32.9 GBytes   941 Mbits/sec  324             sender
[  4]   0.00-300.00 sec  32.9 GBytes   941 Mbits/sec                  receiver

Listing 2: hAP ac^2 v6.40.5 (stock firmware) 5G (Server->WAN->WLAN 5G->Laptop PC)

[ ID] Interval           Transfer     Bandwidth       Retr
[  4]   0.00-300.00 sec   580 MBytes  16.2 Mbits/sec  3541             sender
[  4]   0.00-300.00 sec   578 MBytes  16.2 Mbits/sec                  receiver

Listing 3: hAP ac^2 v6.40.5 (stock firmware) 2.4G (Server->WAN->WLAN 2.4G->Laptop PC)

[ ID] Interval           Transfer     Bandwidth       Retr
[  4]   0.00-300.00 sec  1.46 GBytes  41.9 Mbits/sec  3628             sender
[  4]   0.00-300.00 sec  1.46 GBytes  41.9 Mbits/sec                  receiver

Listing 4: hAP ac^2 v6.42rc56 (Server->WAN->LAN->Laptop PC)

[ ID] Interval           Transfer     Bandwidth       Retr
[  4]   0.00-300.00 sec  32.9 GBytes   941 Mbits/sec    0             sender
[  4]   0.00-300.00 sec  32.9 GBytes   941 Mbits/sec                  receiver

Listing 5: hAP ac^2 v6.42rc56 5G (Server->WAN->WLAN 5G->Laptop PC)

[ ID] Interval           Transfer     Bandwidth       Retr
[  4]   0.00-300.00 sec  4.64 GBytes   133 Mbits/sec  7390             sender
[  4]   0.00-300.00 sec  4.64 GBytes   133 Mbits/sec                  receiver

Listing 6: hAP ac^2 v6.42rc56 2.4G (Server->WAN->WLAN 2.4G->Laptop PC)

[ ID] Interval           Transfer     Bandwidth       Retr
[  4]   0.00-300.00 sec  1.42 GBytes  40.6 Mbits/sec  3025             sender
[  4]   0.00-300.00 sec  1.42 GBytes  40.5 Mbits/sec                  receiver

Recently I bought four pieces of hAP ac^2, I found the performance of WiFi section including both 2.4G and 5G are extremely poor in performance. Based on my past experience of using MikroTik products, the stock firmware seems not prepared for the hAP ac^2 since 1) the Quick Set is “CPE” instead of “Home Dual AP”. I believe the stock firmware is used in other products like wAP 60G; and 2) other issues such as memory size, CPU type (IPQ4019), etc were reported. Also there are tones of threads in this forum mentioning the actual file transfer rate is extremely bad.

How comes the WiFi perofmance of hAP ac^2 are so poor?

Let me show you some figures obtained by using MikroTik btest.exe.

Figures 1 and 2 show the 2.4G and 5G WiFi performance. The throughput can be estimated by the area under the curve in blue. In Figure 1, the average file transfer is around 40Mbps for 2G WiFi. In Figure 2, the area under the curve in blue is around 10-20Mbps. Many users of hAP ac^2 have already reported this issue.

As there are tons of complaints reported in this forum, MikroTik started to face this issue by providing a number RC’s. Although the recent RC (V6.42RC52) can increase the transfer rate (area under the curve is estimated to be around 90-100Mbps), the file transfer rate is still not able to meet our expectations. We expect the WiFi transfer should be very table at around 400Mbps as we found in 5G ac mode of hAP ac lite tower (max 100Mbps for all ports), as shown in Figure 5.

Does anyone know whether MikroTik are still working on this poor WiFi performance? Since this issue has already driven many MikroTik users to purchase other brand’s AC products due to the unexpected and extremely poor WiFi performance.

============
Other problems besides the WiFi issues
1、hAP ac² noisy when using WiFi http://forum.mikrotik.com/t/hap-ac-noisy-when-using-wifi/118025/20

Figure 1: Stock Firmware V6.40.5 2.4G Performance of hAP ac^2

Figure 2: Stock Firmware V6.40.5 5G Performance of hAP ac^2

Figure 3: Firmware V6.42RC52 2.4G Performance of hAP ac^2

Figure 4: Firmware V6.42RC52 5G Performance of hAP ac^2

Figure 5: Firmware V6.40.6 5G Performance of hAP ac lite tower

2

This is 5Ghz on my hAP AC2

http://www.speedtest.net/result/7188352780

3

We cannot see the image and your speed test information is useless without knowing the expected speed of the connection.

4

Does anyone know whether MikroTik are still working on this poor WiFi performance? Since this issue has already driven many MikroTik users to purchase other brand’s AC products due to the unexpected and extremely poor WiFi performance.

Mikrotik has decided to develop its own chipset driver, and thereby turned into a one-way street

that is not even possible fixed, wait for mum this week, but not so much hope

5

Max speed is that is get from Router. 150/5

6

Many thanks for letting me know the MikroTik planning to resolve this performance issue by redesign their own chipset driver. However this should be provided by the chipset manufacturer Qualcomm https://www.qualcomm.com/products/ipq4018. I do not know whether MikroTik has this ability or not as they are application engineers not chipset driver development engineers.

My point is how comes the performance is so poor that is much below the user’s performance expectations—low latency, high throughput and stable transfer rate of at least 300Mbps.

7

Perhaps the problems stem from RouterOS using an extremely outdated kernel (v3.3.5, May 2012). The latest ath10k driver for the IPQ401x wireless chipset requires a newer kernel, so maybe Mikrotik decided to write their own driver for that reason. But there exists a backport project too for older kernels, see https://wireless.wiki.kernel.org/en/users/drivers/ath10k/backports - using the official driver will surely provide better results.

There are plenty of other devices using IPQ401x chipset without the terrible wifi performance, presumably they are using newer kernels with the proper ath10k driver. This is hopefully a software problem, Mikrotik need to get a working driver and ideally an updated kernel in place ASAP.

8

mistry7,

Could you please help me to ask the Technical Support Team in MUM Berlin 2018 for this issue if you go to attend it?

My question is "Do MikroTik have any solid plan to rectify the WiFi issue of hap AC^2?

BTW, how did you know they have been working on developing a new device driver by themselves for the hap AC^2 WiFi issue?

Many thanks in advance.

9

Would you like to share any test results using other IPQ401x chipset?

After having done a number of experiments, I obtained the following results for the discussion of software stability issue.

Figures 1-3 show the bandwidth graphs for different initial power starts. Figures 1 and 2 show the similar results which are consistent over the recored period. However the result will be different as shown in Figure 3, which represents the most typical graph for other period of time. What are the information that we can obtain. As seen in Figure 3, the maximum bandwidth is around 360Mbps, it indicate that the WiFi power amplifier performance will not be degraded with respect to time. However the resultant throughput (area under the blue line) will be much lower than the maximum rate 360Mbps. It is around 100Mbps in average. Based on the results of the three figures, it gives us a very clear indicator, the problem comes from the software (V6.42rc52) used in the hAP ac^2. I believe your observation is correct. There are some incompatibility between RouterOS and the chipset driver. The WiFi transmission will be stopped or even corrupted by them (RouterOS and chipset device driver). If the results are flat, the ping time will be very stable. Otherwise, the ping time will be extremely high due to the system is not able to perform the required duty (to provide a very stable data transmission using WiFi). I guess there are some software counter overflows, the system needs to restart the data transmission program to resume the transmission. That is why you see choppy graphs.

The current RC firmware used in hAP ac^2 is definitely required to be modified or rewritten in order to combine the power of RouterOS and IPQ4018 device driver to provide a stable and high throughput at over 400Mbps. Otherwise more and more MikroTik users would not consider to buy hAP ac^2 anymore due to the scrappy firmware.

Figure 1: hAP ac^2 5G performance after powering up for a short period time

Figure 2: hAP ac^2 5G performance for different power cycling

Figure 3: hAP ac^2 5G performance for other power cycling

10

In this morning, I checked the performance of my hAP ac^2 WiFi router using MT btest.exe software. Initially it performed very well as shown in Figure 1. It just ran for a few seconds and disconnected suddenly. I re-started the test again and got the results as shown in Figure 2. Based on the results, it indicates if after few hours without any data transmission, hAP ac^2 may become normal. However whenever there are large amount of data traffics, the RouterOS may stop for a while and return to his current normal condition—choppy transmission as shown in Figure 2.

MikroTik support must respond this issue quickly otherwise more and more WiFi will buy other’s AP due to unstable and power 5G WiFi transmission. Please not the ping time must be very stable since some game players require the ping time to be very short and very stable.


Figure 1: Very good results but disconnected in a few seconds.

Figure 2: Back to the general condition for another attempt after disconnection.

11

I second the problematic tx performace of RBD52G. It is not able to keep higher moduation rates or two tx channels for a while, fastly jumping down. Even with 6.42rc52 it provides floating tx with real data transfer between 20-80Mbps. Looks like really not finished yet. I am starting to regret…

12

Hey Mikrotik!

WHEN will this issue be fixed?
Or should I initiate a refund process?

13

Has anyone managed to get OpenWRT booted on it to compare wifi performance?

14

All those tests are made with the same wireless frequency?

15

I am sure but they are preset to auto mode. I can use fixed frequency if you want. I use WiFi analyzer to confirm there are no interference among other 5G frequency.

You meant I need to choose different frequency to seek for higher stability, right?

16

Another tests with different frequencies: http://forum.mikrotik.com/t/hap-ac-lan-wifi-5ghz-performance-issue/118005/1
As required here: http://forum.mikrotik.com/t/hap-ac-lan-wifi-5ghz-performance-issue/118005/1

There are NO ANY OTHER 5 GHz network neighbors!

17

Well as MikroTik told us, *.rc releases are for testing purposes, not for production. One of my colleagues owns a Huawei Mate phone and was able to connect to wAP ac without any problems with earlier 6.42rc releases. ~3 weeks ago he showed me that he cannot connect anymore. Also with current rc52 I cannot find any information about his logging attempts. Tested with WPA2 PSK, WPA enterprise and unprotected SSID.

Back to topic: It would be great, If you post your benchmarks with 6.40.7 and 6.41.3.

18

Since the factory version is v6.40.5, I expected the bug fixed version of v6.40.7 should clean the bugs in the previous version. As seen in the following figure, the 5G WiFi results are terribly bad. The average speed is around 25Mbps, which is much below our expectation of 500Mbps for ac products. As a result, it is no point to test any bug fixed version of firmware until the official release v6.42.

19

Is this the constant problem on this model? Plan to buy one but see no point in it if it performs so bad.

Really bad news for me!

20

CAP AC is no better for me either. I need to pick up a different access point in the short term.