Asus RT-AC88U - roughly what speed may I expect on 5GHz

[enwe65]

Bought a new Asus RT-AC88U.
Have a 500Mbps/s (fiber) subscription. Wired I get about 525 up and down.
Wireless:
Linkspeed is 866Mbps
Router is set at 5GHz, using 80MHz and channel 'auto'.
Signal is very strong, distance to router is about 1 meter.
Using Ookla, mobile test I am getting roughly 250/280.
Samsung Galaxy Tab S 10.5 (supports ac)

Someone else, having the same provider, the same subscription, but using the AC87U, also set to 5GHz, 80MHz and channel set to auto, also using Ookla mobile test (but from is phone) is getting
well over 400Mbit/s and 1x even over 500Mbit/s.
8 out of 10 435-450Mbit/s.

Is there so much of a difference between the 87 and 88 ?

What is a fair speed that I might expect?
My Asus RT-N66U that I had until recently (linkspeed 300Mbps) wireless results were around 150-170Mbps. Improvement is rather moderate.

Should I better try to get the AC3200 then ??

Thanks
=

 

[Martin - SNBuser]

You measured both wireless and wired speed? If not, try to see if there's a difference. If not: Then keep what you have, IMHO.

 

[enwe65]

I measured both wired and wireless. As said, wired: no issue, at least 525Mbps (a little more than the subscription of 500Mbps). Both up and down, steady.

Wireless is the issue. I am using just one device. It is all relatively simple. 1 pc, 1 tablet.

 

[L&LD]

Testing at 1 metre distance is not recommended. Neither is using auto channels. I'm not sure why you're using 80MHz bandwidth settings either for 5GHz? Did you try with 20/40/80 MHz setting?

I think you're also not using the same device to test with (you're using the Galaxy Tab S and your friend is using his phone)? An additional problem here is that handheld devices are not configured for maximum performance, rather, they are optimized for long battery life and those two extremes are not in the same balance either between the two different devices.

For a better comparison test, use a manually selected channel (they don't have to be the same on both routers, and test a few too, to find where the best throughput and lowest latency is for each environment and router/client combinations). Test at more than a single location within the desired coverage area (both close, medium and far distances and with varying obstacles between the router and the test computer too), making sure you locate and orientate the laptop the same at each location between testing too.

On the same client and with the routers setup similarly (not using 'auto' settings'), the top speeds should be very similar (but the optimum range may vary).

At the very least, use the same device on a fixed channel at a more realistic distance (about a dozen feet away, even if an interior wall is in between at that distance).

 

[SoCalReviews]

I agree that you should try L&LD's suggestions. The way MIMO works is to differentiate the slight signal delays between the antennas. More differentiation can actually increase the throughput. You want the signal strength to be good but don't test the speed with the devices right next to the router. For best results I would recommend testing a distance away in the same room or one room away if you are testing for maximum throughput. You can keep moving farther away from the router for distance tests. Check the signal strength on your devices and see if you are still getting maximum signal or whether the signal has dropped off. That gives you an idea about range vs. throughput as it relates to signal strength detected by the device. You can also review the signal strength the Asus router sees from the device in the admin settings.

When you do your wireless testing you could try auto mode first. Record your results and then try fixed settings or fixed channel(s) next. Because the 88u has technology that is beyond that of the clients on current devices the clients you are using will most likely be the limiting factors on your throughput... not the router. So while you might see differences they probably won't be night and day differences. AC clients should show more improvement over N clients. You can't simply determine throughput differences for your router vs. another user's router on the same internet provider if you are using different clients in different environments.

A more advanced router can add a level of reliability, stability and responsiveness compared to older less advanced router models...(or sometimes it doesn't). That isn't something you will necessarily notice right away in range and throughput tests. You need to use your devices at different distances under normal use for at least several days. Check your devices for consistent connectivity, any loss of connectivity, responsiveness, signal drops, etc.. The maximum wireless throughput is only one factor involved.

The 88u/3100 has been a one of the more popular tried and tested newer Asus models over this past year. I would lean towards keeping it if you have already purchased it and want the latest technology... unless you purchased it purely for evaluation and want to test other router models. I wouldn't recommend the 87u over the 88u especially if you are using 5Ghz. The 87u 5Ghz radio has been reported by many users to be less reliable than other Asus models.

 

[sfx2000]

Someone else, having the same provider, the same subscription, but using the AC87U, also set to 5GHz, 80MHz and channel set to auto, also using Ookla mobile test (but from is phone) is getting
well over 400Mbit/s and 1x even over 500Mbit/s.
8 out of 10 435-450Mbit/s.

Is there so much of a difference between the 87 and 88 ?

Because somebody is BS'ing you...

Your numbers are fine... don't worry about it

 

[enwe65]

Thank you all for the feedback.

Truly appreciated, because I was seriously wondering whether I actually bought the right router…

Read some in depth magazine tests comparing a lot of routers at various, various frequencies, etc. and saw the AC3200 performing much, much better. There was another post from me asking for advice which one to buy (titled: "Asus RT-AC88U vs Asus RT-AC3200 vs The Netgear R8000 Nighthawk X6") Someone else suggested the AC88U, being a powerful router, so I decided to buy that one.


Anyway…
As for testing, (L&LD and SoCalReviews), I tried 20/40/80MHz (default) as well, tried in another room (distance to router about 7 meter plus a thick wall), as well as on the balcony and in the hallway, tried with changing the position of the antennas. up, down and 45 degr.

This all doesn't make any differences. The 'linkspeed' remained 866Mbps, signal 'very strong' and the Ookla Mobile speedtests roughly remained the same. This is the only free tool I know of to test with.

As for the differences between phone and tablet and phone being configured for maximum performance vs tablet being configured for battery save (L&LD): frankly, I didn't know.

As for choice of channel: I wasn't sure which one to select there, this whole WiFi stuff was new to me, I never went into those router settings before. "Wifi Analyzer" showed -27dBM on channel 104 at 5GHz (2.4GHz, channel 6, -32dBM). Should I select that one then?

Current settings: Wireless Mode: Auto, Channel bandwidth: 20/40/80MHz, Control Channel: auto (current control channel 100)

Maybe it has also something to do with the Galaxy Tab S 10.5 as well. This model seems to be performing just average as far as WiFI is concerned.
http://www.anandtech.com/show/8197/samsung-galaxy-tab-s-review-105-84inch/5
chapter Wifi and Camera

Well, I guess I have to accept that it isn't an outperformer with regards to WiFi but a stable router.

@sfx2000: I am not sure about this, I got screenshots of the performances. There is no reason to tamper with the figures on the screenshot. That aside, one would notice this immediately.

 

[L&LD]

Thank you all for the feedback.

Truly appreciated, because I was seriously wondering whether I actually bought the right router…

Read some in depth magazine tests comparing a lot of routers at various, various frequencies, etc. and saw the AC3200 performing much, much better. There was another post from me asking for advice which one to buy (titled: "Asus RT-AC88U vs Asus RT-AC3200 vs The Netgear R8000 Nighthawk X6") Someone else suggested the AC88U, being a powerful router, so I decided to buy that one.


Anyway…
As for testing, (L&LD and SoCalReviews), I tried 20/40/80MHz (default) as well, tried in another room (distance to router about 7 meter plus a thick wall), as well as on the balcony and in the hallway, tried with changing the position of the antennas. up, down and 45 degr.

This all doesn't make any differences. The 'linkspeed' remained 866Mbps, signal 'very strong' and the Ookla Mobile speedtests roughly remained the same. This is the only free tool I know of to test with.

As for the differences between phone and tablet and phone being configured for maximum performance vs tablet being configured for battery save (L&LD): frankly, I didn't know.

As for choice of channel: I wasn't sure which one to select there, this whole WiFi stuff was new to me, I never went into those router settings before. "Wifi Analyzer" showed -27dBM on channel 104 at 5GHz (2.4GHz, channel 6, -32dBM). Should I select that one then?

Current settings: Wireless Mode: Auto, Channel bandwidth: 20/40/80MHz, Control Channel: auto (current control channel 100)

Maybe it has also something to do with the Galaxy Tab S 10.5 as well. This model seems to be performing just average as far as WiFI is concerned.
http://www.anandtech.com/show/8197/samsung-galaxy-tab-s-review-105-84inch/5
chapter Wifi and Camera

Well, I guess I have to accept that it isn't an outperformer with regards to WiFi but a stable router.

@sfx2000: I am not sure about this, I got screenshots of the performances. There is no reason to tamper with the figures on the screenshot. That aside, one would notice this immediately.

If Auto Channel select is working for you (and it seems it is, right now, with the additional information you have provided), then continue using it. But if you get a better WiFi client, you may want to test each channel in the 5GHz band by taking the Control Channel off of Auto and manually selecting one (start systematically at the lowest (numerical) channel possible and test each one, working up and rebooting the router after every change). With your great ISP speeds, using the Ookla speed test should be enough to narrow it down to the best one or two channels for your environment.

You could do the above for your current Galaxy Tab S, but it seems you are way above it's limits and I'd be worried you'll actually pick a worse channel, overall. However, if the responsiveness of the network is slow for you (the latency is too high), do the above procedure will still help you pick the most responsive channel (lowest latency) for your router, environment and device combination.

Both Galaxy Tab S devices support 2-stream 802.11ac and can negotiate up to an 866Mbps link speed with a similarly capable router. Peak performance in practice is pretty shy of what we saw with the Galaxy S5 however. While the GS5 was capable of around 430Mbps under best conditions, the Tab S averages around 240Mbps. Compared to the immediate competition it’s not bad, but against everything else we’ve seen it’s less impressive.

The quote from the anandtech link you provided shows the real reason for you low speeds. The Tab S is capped to 240Mbps wireless performance. Even when it connects with an 866Mbps link speed.

Your 'issue' today is not the router. It is your client (wholly). Unless and until you begin using a different wireless client/device, there is very little you can do to increase the throughput. This is not a limitation of your router (once again), it has the hardware, firmware and the ISP speeds behind it to provide much more than the Tab S is able to use.

 

[enwe65]

Thank you very much indeed for your elaborate reply!
(Sorry for the delay in getting back on this)

It is good to know that it isn't the router then.That said, it is a bit shamefull of Samsung to limit the the WiFi speed of this fairly sophisticated tablet. There were many cheaper ones around, the time I bought this.
Therefore one would expect maximum performance.
But so be it, something I should pay extra attention to in future, should I buy a new device.

Anyway, thanks again.