Intel AX200 / NETGEAR RAX80, RAX120 Test Results

[avtella]

I tested the AX200 against the 9260AC (in a Dell Inspiron 7577), using the RAX 80 and RAX120 . Test was done one floor below the routers' location (about 12-15ft away.).
I noticed about a 10-15% gain in performance on 5 GHz between ac and ax modes. Between n and ax on 2.4 Ghz @ HT20 there was a pretty big delta. Between n and ax on HT40 difference was much lower.

If you have an ac router though, the AX200 will perform practically the same as the 9260ac.

 

[thiggins]

I'm not seeing the "big delta" between AX200 and AC9260 2.4 GHz results in the table.

 

[avtella]

I'm not seeing the "big delta" between AX200 and AC9260 2.4 GHz results in the table.

I apologize I wasn’t very clear. I meant in terms of “ax” over “n”. The 2.4 GHz - HT20 “ax” results were almost double the HT20 “n” results. 40 MHz “n” vs “ax” however didn’t show much gain, probably due to interference as there are quite a few APs nearby including my Arlo base station.

The AX200 is no different than the 9260ac otherwise. That’s why I said “n” vs “ax”. Without an “ax” router it’s literally just like the 9260ac both in 2.4/5 GHz as expected.

 

[thiggins]

The 2.4 GHz - HT20 “ax” results were almost double the HT20 “n” results.

Sorry, I still don't see the data in the table. I see, for example:
RAX120 2.4 GHz down AX200 9-11 MB/s; 9260 10-11MB/s

 

[avtella]

Yes those are the 2.4 GHz HT20 “n” results that you quoted and then in the yellow boxes are the AX results. Only the AX200 has the “ax” results; 18-19 MB/s on HT20 as the 9260ac can’t do ax. That’s nearly double the “N” results at HT20. Even if the 9260ac in “ax” mode it wouldn’t really make a difference as the card doesn’t support the protocol hence it’s “ax” results are blank.

To be honest those “n” results on the 9260ac are with “ax” turned on, on the RAX120 results.

I guess my communication skills are worse than I thought . I hope this post was clearer.

I highlighted what I meant in the table below.

 

[L&LD]

Yes those are the 2.4 GHz HT20 “n” results that you quoted and then in the yellow boxes are the AX results. Only the AX200 has the “ax” results; 18-19 MB/s on HT20 as the 9260ac can’t do ax. That’s nearly double the “N” results at HT20. Even if tested the 9260ac in “ax” mode it wouldn’t really make a difference as the card doesn’t support the protocol hence it’s “ax” results are blank.

I guess my communication skills are worse than I thought . I hope this post was clearer.

I highlighted what I meant in the table below.

I think the 9260AC results should be included even when in 'AX' mode. Otherwise, the testing looks incomplete and more importantly, it doesn't let us see if there are any benefits or pitfalls of using an 'AX' router with our AC clients.

 

[thiggins]

I highlighted what I meant in the table below.

Ah! Now I see it. Sorry for being so dense.
Did you see higher link rate for ax vs. n? n would be 173 Mbps max. ax would be 287 Mbps (MCS 11).

 

[avtella]

Ah! Now I see it. Sorry for being so dense.
Did you see higher link rate for ax vs. n? n would be 173 Mbps max. ax would be 287 Mbps (MCS 11).

Don’t worry I’m a terrible communicator. I saw a fairly consistent link rate of 576 Mbps (HT40) on 2.4 GHz and 2.4 Gbps (HT160) & 1.2 Gbps (HT80) on 5 GHz on “AX” mode.

I’m fairly certain on HT20 ax, the AX200 was at max link rate but can’t recall for sure.

In “n” & “ac” mode both the AX200 and 9260ac saw max link rates.

I think the 9260AC results should be included even when in 'AX' mode. Otherwise, the testing looks incomplete and more importantly, it doesn't let us see if there are any benefits or pitfalls of using an 'AX' router with our AC clients.

Here’s a secret: those “n” results on the RAX120 for the 9260ac were with ax turned on. But you are right I should have done a better job lol.

 

[thiggins]

I saw a fairly consistent link rate of 576 Mbps on 2.4 GHz and 2.4 Gbps on 5 GHz.

576 Mbps would be the 40 MHz link rate. Are you comparing 20 MHz B/W in n to 40 MHz B/W in ax?

 

[avtella]

576 Mbps would be the 40 MHz link rate. Are you comparing 20 MHz B/W in n to 40 MHz B/W in ax?

Updated my post above right as you posted lol.

I was comparing stats at same widths I was just trying to remember the link rates I saw, as I didn’t record them. I recall all but the HT20 ax link rate.

 

[thiggins]

Updated my post above right as you posted lol.

I don't think you answered my question. It would be clearer if you posted link rates for each measurement. That would answer my bandwidth/MCS questions.

It sure looks like your conclusion of 2.4 GHz 2x throughput for ax vs. n is based on 40 MHz vs. 20 MHz bandwidth.

 

[avtella]

I don't think you answered my question. It would be clearer if you posted link rates for each measurement. That would answer my bandwidth/MCS questions.

It sure looks like your conclusion of 2.4 GHz 2x throughput for ax vs. n is based on 40 MHz vs. 20 MHz bandwidth.

Every test was at max link rate for each width/band. The table is accurate. I “mistakenly” mentioned 576 Mbps as that’s what first came to my mind for 2.4 GHz but yes that was the link for HT40 my mistake. HT20 link was indeed lower. Will check when I get home.

 

[thiggins]

HT20 on AX did give close 2x results of HT20 on n. AX HT40 gave very little gain over AX HT20.

I'm not trying to give you a hard time, but things don't make sense unless I know the observed link rates along with throughput.

You can't see a link rate > 173 Mbps for a 2x2 ac STA @ 20 MHz B/W or > 287 Mbps for 2x2 ax STA @ 20 MHz. 576 Mbps is the max (MCS11) 2x2 ax 40 MHz link rate.

 

[avtella]

I'm not trying to give you a hard time, but things don't make sense unless I know the observed link rates along with throughput.

You can't see a link rate > 173 Mbps for a 2x2 ac STA @ 20 MHz B/W or > 287 Mbps for 2x2 ax STA @ 20 MHz. 576 Mbps is the max (MCS11) 2x2 ax 40 MHz link rate.

No no, I totally understand, you are just being thorough. I’ll update you when I get home and will update the table with link rates shown in Windows. I will do a quick retest with the AX200 to reconfirm.

 

[avtella]

I'm not trying to give you a hard time, but things don't make sense unless I know the observed link rates along with throughput.

You can't see a link rate > 173 Mbps for a 2x2 ac STA @ 20 MHz B/W or > 287 Mbps for 2x2 ax STA @ 20 MHz. 576 Mbps is the max (MCS11) 2x2 ax 40 MHz link rate.

I updated the table with link rates as seen in Windows during transfers. I attached one of the transfers to show that HT20 AX indeed does give about 18-19 MB/s and in some cases 20-21 MB/s transfer rate.

Edit: also attached a clipped image as it uploads blurry for some reason.

There is a person at notebook review forums with an Asus AX11000 who claimed he was getting 165-170 MB/s down on HT160 AX and about 140 MB/s downlink on HT160 AC. He was using the AX1650 in his Dell Precision 5530. They didn't provide any details such as distance etc.

 

[thiggins]

Thanks for including the link rates. I agree with your conclusion about 2.4 GHz 20 MHz throughput doubling. This is due to link rate difference (144 > 287). The 40 MHz B/W gain isn't as dramatic.

Thanks for testing and sharing the data.

 

[RSengine]

I updated the table with link rates as seen in Windows during transfers. I attached one of the transfers to show that HT20 AX indeed does give about 18-19 MB/s and in some cases 20-21 MB/s transfer rate.

Edit: also attached a clipped image as it uploads blurry for some reason.

There is a person at notebook review forums with an Asus AX11000 who claimed he was getting 165-170 MB/s down on HT160 AX and about 140 MB/s downlink on HT160 AC. He was using the AX1650 in his Dell Precision 5530. They didn't provide any details such as distance etc.

Hey avtella thanks for uploading your results. I just want to confirm that AX200 is not cnvio - is your laptop cpu a 7th generation intel cpu?

 

[avtella]

Hey avtella thanks for uploading your results. I just want to confirm that AX200 is not cnvio - is your laptop cpu a 7th generation intel cpu?

The AX200 is a standard full fledged WiFi card, not cnvio, cnvio variant is the AX201. Also yes my Dell 7577 uses a 7th gen CPU (I7 7700HQ)

 

[Razor512]

Has anyone found a trusted place where to purchase an AX200?

 

[pege63]

Has anyone found a trusted place where to purchase an AX200?

Yes there is a place but its in Denmark.