[Network] UniFi vs. Ruckus Unleashed for Home WiFi | Part I: Performance Comparison

Update (11/16/2023): 2.4 GHz test added

Background

After pondering for potential alternative to my current home’s full UniFi WiFi setup (ref), I’ve concluded to try out Ruckus. 

Ubiquiti UniFi is poster child of prosumer graded network gear. They have many features and system design that are otherwise limited to enterprise grade systems. There are many home users who have UniFi gears as their network setup. I personally consider UniFi can be viewed as the gold standard for, prosumer grade home WiFi setup. I have been using UniFi system for the last 3 and half years. Ever since the day I tested UniFi HD access point, I immediately converted myself from consumer graded gears (back then full WiFi 6 Asus AiMesh setup), and never looked back until recently.

For those of you know who have never heard of the brand. Ruckus is enterprise grade gear brand that competes with the brands like Cisco Meraki, HPE Aruba. Schools or cooperation around you maybe using these products. 

Ruckus is often acclaimed for having premium hardware quality. As in other true enterprise setup, Ruckus have subscription based controller and service for full operation, which include AI analytics. But as a home user, we are looking at license-free firmware version, Ruckus “Unleashed”.

Despite not subscribing or paying for license, these are real enterprise grade hardwares. They just use different firmware. So if someone ever wants to use this for the full blown real enterprise setup, they can use the exact same hardware but install different firmware.

Setup

Initially, as I was not certain if I’d be truly converting to Ruckus or not, I’ve decided to only deploy two Ruckus AP as the 4/5 UniFi AP setup. I did not expect full corner to corner performance improvement over UniFi with half number of APs with Ruckus, but I thought this should give me enough data points to compare two setup, and make fairly accurate educated guess, how many more Ruckus AP I would need to achieve similar or better coverage and ranged performance to my current UniFi setup.

For the actual placement of APs, one AP was set in ground level Kitchen as its the center of our home. This is one of most WiFi demanding area in our household. This was where recently broken U6 Mesh was sitting.

Since our home is two stories (3800 sqft) with walkout basement, there are areas that are quite WiFi unfriendly in the lower/basement level. 

In order cover these areas better, I have tried a few different placement of the second Ruckus AP. First, I placed it in basement hall-way in place of U6 Pro AP. I’ve also tried basement entertainment area where UniFi SHD AP was located. Both resulted in creation of dead spot.

So I eventually placed in ground level family room, corresponding to the one of U6 Pro AP location.  This achieved entire 5 GHz WiFi coverage with 2 AP.

Test Methods

All tests other wise specified used iPhone 14 Pro Max, which is 2×2 WiFi 6 capable client.  Throughput were measured using local network 30 seconds continuous iPerf to the local NAS server.  To reduce WiFi variation, I have ran 2-3 times of iPerf at the same spot and recorded more reflective data result. For example, if one test result were much slower than the another, then repeat and use the one of test that looked similar. This is because rest of home network as well as WiFi were in use during these test, so I they could certainly affect some numbers, or some were actually during roaming. Whenever I was testing one system, I had the other system’s all AP completely powered off to ensure no interference from them.

Comparison

Limitations

Since I am just a casual network hobbyist, I can’t just buy full system to match two multi-access point system setup to give true apple to apple level of comparison.  So current comparison have couple limitations.

First, the major difference is the Ruckus setup used in this comparison has only 2 APs while UniFi has 4. 

Second, UniFi AP positions were optimized for original 5 AP setup. The one that died was located in the middle of home, Kitchen, where one of Ruckus AP is now placed. So if I were to truly optimize UniFi setup, I should have re-position 4 APs. 

Throughout this write up, I will try to pick data points where these limitations have minimal to no effect when comparing two system setups.

For the comparison, I have looked at following 4 categories.

  1. Performance 
  2. Stability
  3. Features
  4. Simplicity
  5. Cosmetic
  6. Scalability
  7. Cost

Performance

Speed

First number we will look at is the throughput. Perhaps, everyone’s favorite when comparing networks because it looks like easy objective numbers to look at.

In general, I consider there are two big category of throughput. 

One is close proximity throughput, which is basically the max throughput when you have the client device right next to the access point. 

The other is probably more important for most home users in a practical sense, the ranged throughput. 

The range of WiFi coverage and ranged throughput are closely correlated but not necessary the same. One may have WiFi connection up to let’s say 3 rooms away. But throughput there may be too low for video streaming. In such case, there is a range of coverage, but range throughput may be too low for the WiFi use you need. 

The most important throughput number for home user is the throughput at your typical device using area.

Max throughput test

First, let’s start with max throughput. In order to measure this,  I had my iPhone about 5 ft away from the each target access point with full line of sight.

Left 2 data are from UniFi U6 Pro and right 2 data are from Ruckus R-750.

These test results showed 4 facts.

  1. Average throughput were moderately (~100 Mbps/~15%) higher on R750.
  2. Max throughput was slightly (~30 Mbps/~4%) higher on U6 Pro.
  3. Min throughput was significantly (~480 Mbps/4.8x) higher on R750.
  4. WiFi efficiency of UniFi is 50-57% while Ruckus is constant 63%.

Now let’s try to interpret what each of these result could mean.

Average Throughput

The difference in average throughput depicts the real world benefit where continuous large file transfer taking place such as downloading or uploading files where WiFi is acting as the bottleneck. If you are doing file transfer within the local network like NAS server, you will likely to see the difference. However, if you have internet service that’s slower than these numbers, you are unlikely to see the benefit.

Max Throughput

Max throughput is a single instant of burst throughput and unpredictable. So this number should not have any effect on the practical end user experience.

Min Throughput

Personally, most interesting test result was this number.  

On U6 Pro, this happened almost always at the very first measure point. My recollection of prior iPerf test on other UniFi APs, I think this has always been the case. 

For this reason, I had always thought this is analogous acceleration time in vehicles. In WiFi, I would call it rise time as used in electronic circuit. With R750, first data point is always near its peak and almost never minimum. So R750 has instant rise time.

R750 has instant rise time.

How could this result potentially affect end user’s daily experience? This is just my hypothesis/theory, but I believe this is the throughput most of us experience when browsing internet or performing a small data transfer tasks, especially for the data size that are less than these throughput values.

With minimum throughput of 573 Mbps, R750 could in theory transfer over 70 MB of data in the first 1 second. In contrast, UniFi will be 12.5 MB. In reality, there are other factors including data overheads, and also R750 have higher than minimum on its first data point, so the difference could possibly be more significant.

So if you are loading some webpage, 5-6x difference in this number could potentially affect our end user experience as something like 2-3 seconds difference in webpage loading. It turns out there are various websites talks about this, and many users consider optimal time being under 2 seconds.

Now in many cases, web page loading may not be related to our WiFi. It could be server or even client device rendering related, latency or something else. However, regardless of the source, I think many of us would appreciate instant webpage loading over a few seconds of loading on every click.

Practically, I have noticed the difference between UniFi and Ruckus setup. I occasionally do multiple 50-100 MB file transfer from my NAS server to WiFi device. With UniFi setup, this always take a few seconds before the initiation of the initial data transfer. In contrast, Ruckus, this starts immediately one after another.

Then, subjectively I think I am feeling improvement in web browsing experience. Ever since the deployment of Ruckus APs, I feel many web sites loads instantaneously where UniFi setup only provided such experience right after reboot/new setup. Although this maybe a honeymoon period/placebo effect, if this is real then rise time may be the potential reason.

WiFi Efficiency

This is simply calculated based on relative to max PHY link speed for WiFi 6 2×2 client at 80 MHz channel width (1200 Mbps).

WiFi efficiency = Average throughput/PHY (=1200)

This has been believed to be in the range between 50-70%. UniFi’s official site says 50-60%. I believe having consistent, high WiFi efficiency is one of good proof for the quality of hardware. In this regard, R750 not only outperformed efficiency number itself but was very consistent across the two data points.

Ranged throughput test

Using iPerf, I have measured throughput in every room at the same spot between two setups.

Certain areas of throughput difference are simply related to difference in access point placement between two setup. So those numbers are not fair ranged throughput comparison points.

For ranged throughput comparison, following 3 areas of throughput are relevant.

  • Family room
  • Storage
  • Dining/Kitchen

Family Room

 Since both set up has AP located in the same spot and throughputs were measured in the same room about 20 ft away with line of sight to the AP, this number represents same room 20 ft ranged throughput comparison. Average throughput of Ruckus R750 at this range was higher than that of U6 Pro by 6%.

Storage Room

Storage room is located in corner of the walkout basement and partly surrounded by concrete walls. So despite UniFi having a basement access point facing toward this way, I believe the primary AP supplied this room was from ground level, family room APs.

*I did not confirm this with UniFi setting.

The difference throughput here is even more significant than the 20ft with R750 having over 60% gain in throughput. This data point actually have more interesting detail, which will be discussing in range comparison section.

Dining/Kitchen

Between the Ruckus and UniFi setup, Dining and Kitchen AP and clients positions were simply reversed. So two set up had essentially the same distance and obstacles. There were walls but the door between two rooms were open.

R-750 at this distance essentially achieved max throughput equivalent of U6 Pro. In contrast, U6 Pro went down to 228 Mbps. So at this particular distance, there was almost 3x throughput difference between two systems.

M room’s mystery?

At a first glance, one peculiar data point is M bedroom (bedroom 2 in ground level). Despite being next room to the AP, there was a significant throughput drop, especially on R750 and UniFi outperforming Ruckus.

There are two potential explanations for this. 

One is directional nature of APs. I had both Ruckus and UniFi AP in Family room facing away from M bedroom in wall mount position. This means iPhone was connected to these AP at their rear or side lobes, which are not recommended WiFi connection by manufactures. They are essentially using leaking/spilled WiFi signal coverage area.

Alternatively and most likely explanation is, for UniFi set up, there was a U6 Pro in the basement hallway facing towards M bedroom. So I think with UniFi setup, the client were most likely associated with the AP instead of Family room one. Although I could have easily confirmed this if I had noticed the difference before taking down the UniFi setup, but I didn’t see it until start analyzing so this was not verified.

2.4 GHz band throughput

Since UniFi seems to put less emphasis on 2.4 GHz radio compared to Ruckus, I have decided to do a quick additional test on this particular radio.

Setup

  • AP to client device: full line of sight < 5ft
  • For Ruckus, I used R350 ($200 brand new on Ebay) instead of R750.
  • Two AP tests were performed one after another to try keep internet speed test server load/traffic similar
  • For iPerf tests, default iPerf -c [server ip] and iPerf -R -c [server ip] were used.
  • Each AP showed significant variability, so I have ran test 2-3x times to confirm relative consistent output achieved at least consecutively.

Discussion

On iPerf, there is consistent average 15% throughput gain of R350 over U6 Pro.

However, the difference was much more drastic on internet speed test. Personally, I do not trust internet Speed test. The first and second tests timing was done just 15-20 minutes a part and you can see relatively significant difference even on the same AP. In fact, one time I tested on R350, it actually got upload of 120 Mbps. So I’d think internet throughput here has significantly variation due to internet traffic load especially the number being relatively smaller, the average value gets more affected, or possibly my testing server vs. time of the day might have been issue.

However, 2.4 GHz radio has intrinsic expected significant variation. So I would probably not take these as absolute number of either unit, but relatively speaking, R350 appear to outperform U6 Pro on this radio band most of time.

Also, I did not match testing load. For U6 Pro, I ensured there was no other client connected besides the single iPhone while R350 had 10 other devices connected with 5 of them being 2.4 GHz band. So if anything, U6 Pro should have had an advantage.

The importance of this is many IoT connects to this radio band by default and they are often placed in suboptimal location. So this isn’t necessary for the throughput, but it is important for range and stability.

Practically speaking, I can say either UniFi or Ruckus (R750), if I try to stream video on iMac and then cast that to Apple TV, both will stutter on 2.4 GHz band.

Range

Next, let’s compare range of WiFi coverage. The actual mileage and even end-user experience will vary significantly for this part. This is because the range of coverage highly depends on your home environment, and what you do at certain distance would be totally different from mine. 

Specification comparison

Let’s take a look at specifications first.

Area of coverage specification

Area of coverage reported by manufacture’s specification sheet are highly unreliable for home use purpose. This is because home environment are so variable including number of walls, construction material and various other factors. As a manufacture, I am sure they have tested with relatively the best condition such as wide open, high ceiling environment with line of sight everywhere.

For example, our 3800 sqft ranch-style home with walk-out basement creates some WiFi unfriendly areas, particularly in the basement.

This has consistently required me to use significantly more number of mesh nodes or access points on all my previous setups that are well over each manufacture’s listed area of coverage number. Each of these setup, I have always started with small number of units but ended up purchasing more to eliminate dead spot to attain required area of coverage and sometimes desired amount of throughput in various areas. These brands include Netgear Orbi, Eero Pro, Asus AiMesh and UniFi.

So instead, I think we should look and compare more objective 2 parameters when evaluating range of coverage an access point could provide:

  1. Transmission power
  2. Receiver sensitivity. 

Transmission Power

Transmission power indicates how far signal can be pushed from the AP. The standard unit used here is decibel milliwatts (dBm). This is a logarithmic scale unit, so difference in 3 dBm means twice the power consumption. In case of AP coverage, higher the number louder, longer range it can cover. 

With R750 having higher transmission power than U6 Pro on both 5GHz and 2.4 GHz bands, we expect better coverage by R750. 

In fact, the only U6 model that have compatible transmission power to R750 is U6LR and that’s only on 2.4 GHz band. No U6 model matches transmission power of R750 in 5 GHz band.

Minimum Receiver Sensitivity

As in real life’s communication, WiFi communication requires both talking as well as listening. In WiFi, the ability to listen is represented by reception sensitivity. This indicates how subtle of signal the AP can hear. All Ruckus AP has these values clearly depicted for each band; however, I could not find these numbers for UniFi APs.

So do we really care about this?

I believe the mismatch between AP’s transmission power and receiver sensitivity could potentially explain what seems peculiar situation where on our client device show full or 2 WiFi bars yet we keep disconnecting or not even able to join the network when too far away from the access point.

In these cases, client device can see and receive data from APs as AP has high enough transmission power; however, the client device cannot talk back or send data back to AP because client device can never talk as loud APs. So AP need to be able to have high enough reception sensitivity to capture these whispering radio back; otherwise, it becomes one way communication.

2.4 GHz Band

If someone is looking for the longest range of WiFi coverage from a given access point, 2.4 GHz band related parameters become very important. Although 5 GHz or 6 GHz band provider more stable, higher throughput, higher frequency radio have less range. So for the absolutely longest range of WiFi coverage, you will be using 2.4 GHz. This is fact is well depicted by UniFi’s AP naming convention. 

UniFi calls one of its WiFi 6 access point, U6 Long Range. When we look at its spec, it’s the 2.4 GHz band that’s different from other U6 lines while 5GHz transmission power remains the equivalent to the U6 Pro or U6E series. Also, many IoT devices still remain to only supports 2.4 GHz and not infrequently these devices may be placed in the suboptimal areas from Access Points. So 2.4 GHz band is still very relevant for home use.

Currently, there are two primary WiFi version of 2.4 GHz band. WiFi 6 and WiFi 4. During WiFi 5 era, 2.4 GHz standard were untouched. 

R750 has WiFi 6 4×4 for 2.4GHz while U6 Pro has only 2×2. Currently UniFi does not have any 4×4 WiFi 6 compatible 2.4 GHz access point.  If this really matters or not comes down to one’s use scenario. 

Small Net Builder performed early WiFi 6 testing using consumer gears a while back and one of conclusion said following:

In general, 2.4 GHz AX performance was better behaved than 5 GHz. Most products produced around 2X throughput when used with an AX STA vs. an AC STA. Right now, this is likely to be the thing that most buyers would notice from upgrading both router and device to Wi-Fi 6

So if you have WiFi 6 capable client that requires to use 2.4 GHz, you could actually see the gain when coming from WiFi 5 setup.

So based on all of the above specifications, I expected R-750 to have better WiFi coverage than either U6 Pro or SHD. 

As for expected overall coverage area between two R750 and 3 U6 Pro + SHD setup, I used sum of all APs transmission power for each setup. Basically, converted each dBm and added since we can’t just add dBm.

For UniFi setup this calculated as 1510 mW for 5 GHz band while 2 R-750 was 1260 mW. So I expect better overall coverage by 4 UniFi AP setup over the 2 R-750. With my original optimal UniFi setup for attaining high throughput almost every corner of my home consisted 5 APs, which calculated to be 1908 mW. Using this calculation, I estimated 3 R-750 maybe able to achieve that, which has total 1890 mW.

Real world comparison

Now let’s see how all these numbers translated to the real world.

Overall coverage

As expected from the specification difference, Ruckus had overall less area of high throughput coverage in comparison to UniFi setup.

However, to my surprise, I was still able to get actual 5 GHz band connections throughout my entire home with 2 R-750 with the lowest average iPerf measuring, 150 Mbps. In fact, the two areas, basement bathroom and guest bedroom were exact spot I had struggle with my previous mesh or UniFi setups, requiring extra mesh node or access points. 

Single AP coverage

For individual AP rangescomparison, the best spot to look is the Storage room. This area is located in the walkout basement and partly surrounded by concrete walls. This is where I have network rack and hardwired NAS servers. This is one of the most WiFi hostile environment in my house. Both UniFi and Ruckus setup have an AP placed just above the area but these APs are facing west, which is opposite direction to storage due to my mounting preference/limitation. 

For the advantage of UniFi, it has additional AP in the basement facing towards the storage room, but there is a concrete wall the basement AP likely has to penetrate through. So my guess was the client device still used family room AP rather than the basement one, but I did not confirm this.

The result as mentioned earlier for ranged throughput comaprison, average throughput was significantly higher with R-750. However, the most important fact was that the UniFi AP setup kept losing connection at this spot. This is shown by iPerf test number having minimum throughput value of 0. This kept occurring despite multiple iPerf test runs. This demonstrates superior WiFi coverage with single Ruckus R750 over the U6 Pro on 5 GHz band.

Interestingly, despite the fact UniFi AP kept losing the connection to the iPhone, it still got very respectable 251 Mbps max and average 146 Mbps throughput. I think this can be explained by my previous hypothesis of transmission power/reception sensitivity mismatch.

Summary

When comparing Ruckus R750 (WiFi 6 4×4 AP) to UniFi U6 Pro (WiFi 6 4×4 AP):

  • R750 has better max throughput than UniFi 6 Pro
  • R750 has significantly higher instant/initial throughput than UniFi 6 Pro
  • R750 has better throughput at equidistance on 5 GHz band
  • R750 has better WiFi coverage on 5 GHz band

So overall, throughput and range performance is consistently better with R750 than the UniFi U6 Pro.

2 Trackbacks / Pingbacks

  1. [Network] Ruckus Unleashed | Access Point comparison: R350 vs. R750 – Game & Tech Focus
  2. [Network] Premium Home WiFi 6 Setup 2023 | Comparison Part I: Performance – Game & Tech Focus

Comments are closed.