Any considerations on a multi-gigabit (1, 2.5, 5, 10) switch?

madbrain

Active Member
Jan 5, 2019
208
38
28
I'm not so sure Realtek is the problem at all. I just tested two machines

Machine A: Intel NUC 8i7 BEH, connected to a Pluggable USB 2.5
Machine B: MacBook Pro 2017 connected to an Asustor USB 2.5

And I tested with two switches: ICX 6610 and Netgear MS510TX

Code:
MBP:$ ./iperf3 -c machine_a
Connecting to host machine_a, port 5201
[  4] local machine_b port 63491 connected to machine_a port 5201
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-1.00   sec   282 MBytes  2.36 Gbits/sec             
[  4]   1.00-2.00   sec   283 MBytes  2.37 Gbits/sec             
[  4]   2.00-3.00   sec   282 MBytes  2.37 Gbits/sec             
[  4]   3.00-4.00   sec   283 MBytes  2.37 Gbits/sec             
[  4]   4.00-5.00   sec   283 MBytes  2.37 Gbits/sec             
[  4]   5.00-6.00   sec   283 MBytes  2.37 Gbits/sec             
[  4]   6.00-7.00   sec   283 MBytes  2.37 Gbits/sec             
[  4]   7.00-8.00   sec   282 MBytes  2.37 Gbits/sec             
[  4]   8.00-9.00   sec   282 MBytes  2.37 Gbits/sec             
[  4]   9.00-10.00  sec   283 MBytes  2.37 Gbits/sec             
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-10.00  sec  2.76 GBytes  2.37 Gbits/sec                  sender
[  4]   0.00-10.00  sec  2.76 GBytes  2.37 Gbits/sec                  receiver
And in reverse

Code:
MBP:$ ./iperf3 -c machine_a -R
Connecting to host machine_a, port 5201
Reverse mode, remote host machine_a is sending
[  4] local machine_b port 63675 connected to machine_a port 5201
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-1.00   sec   279 MBytes  2.34 Gbits/sec             
[  4]   1.00-2.00   sec   282 MBytes  2.37 Gbits/sec             
[  4]   2.00-3.00   sec   282 MBytes  2.37 Gbits/sec             
[  4]   3.00-4.00   sec   282 MBytes  2.37 Gbits/sec             
[  4]   4.00-5.00   sec   279 MBytes  2.34 Gbits/sec             
[  4]   5.00-6.00   sec   281 MBytes  2.36 Gbits/sec             
[  4]   6.00-7.00   sec   281 MBytes  2.36 Gbits/sec             
[  4]   7.00-8.00   sec   282 MBytes  2.37 Gbits/sec             
[  4]   8.00-9.00   sec   281 MBytes  2.36 Gbits/sec             
[  4]   9.00-10.00  sec   280 MBytes  2.35 Gbits/sec             
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-10.00  sec  2.74 GBytes  2.36 Gbits/sec                  sender
[  4]   0.00-10.00  sec  2.74 GBytes  2.36 Gbits/sec                  receiver
This is with the ICX6610 (with Ipolex 10GbaseT transceivers). Even with the Netgear, the results are the same. I would suggest you get a proper switch and also make sure your cabling and termination is up to spec. I've also had the Asustor connected to a Windows server 24/7 for 2 months and it's been fine, never had drop outs or anything. The USB adapters do get considerably warm hence need to be cooled or provided with adequate ventilation.

btw - I am using very good quality Cat5E (and can get 10 GbE on it, with one run being 80 feet) and recently redid my patch panel termination to make sure the wires were twisted right upto the insertion points. I'm also redoing the wall jacks slowly - you'd be surprised when you look at the way some of the wiring is done. When you get to 2.5 GbE, you need just that little bit better quality.
Cabling definitely not the issue here. I'm using brand new CAT8 . 10 Gbps works perfectly and I get full line rate with 2+ TCP stream between two machines running Aquantia NICs connected to the TEG-7080ES. Full line rate with 1 Gbps too.
Things go bad at 2.5 Gbps and 5 Gbps, apparently regardless of NIC.

Here are some single-stream results from today. This is info I sent to Trendnet also.

In the tests below, client is Win10 / r2004 . i7-5820k OC at 4.3 GHz and 32 GB of RAM.
Server is Ubuntu 18.02 / i5-6600k OC at 4.4 GHz and 32GB RAM.
Both with Aquantia AQN-107, hooked up to Trendnet TEG-7080ES with CAT8 of 6ft each.

Client NIC left on autonegotiation, at 10 Gbps:

C:\Users\Julien Pierre\Desktop\iperf3>iperf3.exe -N -c server10g -R
Connecting to host server10g, port 5201
Reverse mode, remote host server10g is sending
[ 4] local 192.168.1.26 port 63330 connected to 192.168.1.27 port 5201
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-1.00 sec 874 MBytes 7.33 Gbits/sec
[ 4] 1.00-2.00 sec 892 MBytes 7.48 Gbits/sec
[ 4] 2.00-3.00 sec 914 MBytes 7.66 Gbits/sec
[ 4] 3.00-4.00 sec 942 MBytes 7.90 Gbits/sec
[ 4] 4.00-5.00 sec 943 MBytes 7.91 Gbits/sec
[ 4] 5.00-6.00 sec 940 MBytes 7.89 Gbits/sec
[ 4] 6.00-7.00 sec 955 MBytes 8.01 Gbits/sec
[ 4] 7.00-8.00 sec 955 MBytes 8.01 Gbits/sec
[ 4] 8.00-9.00 sec 958 MBytes 8.03 Gbits/sec
[ 4] 9.00-10.00 sec 958 MBytes 8.04 Gbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bandwidth Retr
[ 4] 0.00-10.00 sec 9.11 GBytes 7.83 Gbits/sec 0 sender
[ 4] 0.00-10.00 sec 9.11 GBytes 7.83 Gbits/sec receiver

iperf Done.

Client NIC forced down to 5 Gbps in device manager :

C:\Users\Julien Pierre\Desktop\iperf3>iperf3.exe -N -c server10g -R
Connecting to host server10g, port 5201
Reverse mode, remote host server10g is sending
[ 4] local 192.168.1.26 port 63365 connected to 192.168.1.27 port 5201
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-1.00 sec 185 MBytes 1.55 Gbits/sec
[ 4] 1.00-2.00 sec 190 MBytes 1.60 Gbits/sec
[ 4] 2.00-3.00 sec 191 MBytes 1.61 Gbits/sec
[ 4] 3.00-4.00 sec 193 MBytes 1.62 Gbits/sec
[ 4] 4.00-5.00 sec 193 MBytes 1.62 Gbits/sec
[ 4] 5.00-6.00 sec 195 MBytes 1.63 Gbits/sec
[ 4] 6.00-7.00 sec 191 MBytes 1.60 Gbits/sec
[ 4] 7.00-8.00 sec 193 MBytes 1.62 Gbits/sec
[ 4] 8.00-9.00 sec 194 MBytes 1.62 Gbits/sec
[ 4] 9.00-10.00 sec 194 MBytes 1.63 Gbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bandwidth Retr
[ 4] 0.00-10.00 sec 1.87 GBytes 1.61 Gbits/sec 12364 sender
[ 4] 0.00-10.00 sec 1.87 GBytes 1.61 Gbits/sec receiver

iperf Done.

Client NIC forced down to 2.5 Gbps in device manager :

C:\Users\Julien Pierre\Desktop\iperf3>iperf3.exe -N -c server10g -R
Connecting to host server10g, port 5201
Reverse mode, remote host server10g is sending
[ 4] local 192.168.1.26 port 63496 connected to 192.168.1.27 port 5201
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-1.00 sec 107 MBytes 895 Mbits/sec
[ 4] 1.00-2.00 sec 105 MBytes 882 Mbits/sec
[ 4] 2.00-3.00 sec 106 MBytes 887 Mbits/sec
[ 4] 3.00-4.00 sec 105 MBytes 878 Mbits/sec
[ 4] 4.00-5.00 sec 106 MBytes 885 Mbits/sec
[ 4] 5.00-6.00 sec 105 MBytes 881 Mbits/sec
[ 4] 6.00-7.00 sec 107 MBytes 894 Mbits/sec
[ 4] 7.00-8.00 sec 105 MBytes 884 Mbits/sec
[ 4] 8.00-9.00 sec 107 MBytes 898 Mbits/sec
[ 4] 9.00-10.00 sec 105 MBytes 879 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bandwidth Retr
[ 4] 0.00-10.00 sec 1.03 GBytes 887 Mbits/sec 5409 sender
[ 4] 0.00-10.00 sec 1.03 GBytes 886 Mbits/sec receiver

iperf Done.

Client NIC forced down to 1 Gbps in device manager :

C:\Users\Julien Pierre\Desktop\iperf3>iperf3.exe -N -c server10g -R
Connecting to host server10g, port 5201
Reverse mode, remote host server10g is sending
[ 4] local 192.168.1.26 port 63577 connected to 192.168.1.27 port 5201
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-1.00 sec 113 MBytes 950 Mbits/sec
[ 4] 1.00-2.00 sec 113 MBytes 947 Mbits/sec
[ 4] 2.00-3.00 sec 113 MBytes 948 Mbits/sec
[ 4] 3.00-4.00 sec 113 MBytes 948 Mbits/sec
[ 4] 4.00-5.00 sec 113 MBytes 949 Mbits/sec
[ 4] 5.00-6.00 sec 113 MBytes 949 Mbits/sec
[ 4] 6.00-7.00 sec 113 MBytes 949 Mbits/sec
[ 4] 7.00-8.00 sec 113 MBytes 949 Mbits/sec
[ 4] 8.00-9.00 sec 113 MBytes 949 Mbits/sec
[ 4] 9.00-10.00 sec 113 MBytes 949 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bandwidth Retr
[ 4] 0.00-10.00 sec 1.11 GBytes 950 Mbits/sec 0 sender
[ 4] 0.00-10.00 sec 1.10 GBytes 949 Mbits/sec receiver

As you can see, everything is fine when the client NIC is set to 1 or 10 Gbps.
But there are problems when it is set to 2.5 Gbps or 5 Gbps.
2.5 Gbps actually performs slower than 1 Gbps ...

So, the problem is not limited to the Realtek RTL8156 NIC. It appears there are some issues in the NBASE-T implementation at 2.5 and 5 Gbps in the Trendnet TEG-7080ES switch.

Tests I did previously with the Netgear GS110MX showed it working fine at 2.5 Gbps with the RTL8156 NIC, which I no longer have.
Didn't try yet to force the Aquantia NIC on my HTPC down to 5 or 2.5 Gbps. But I expect it would work properly.

You can see a more complete description of all my network and cabling at :

All is perfectly fine at 10 Gbps with this whole infrastructure. Things only fail if something negotiated at 2.5 or 5 Gbps on the Trendnet switch.
With the Cable Creation RTL8156 NIC, 2.5 is the max it could negotiate. That NIC worked fine at 2.5 Gbps worked fine when plugged to my HTPC and attached to the Netgear switch. Just didn't work in my home office on the Trendnet switch ...
 

madbrain

Active Member
Jan 5, 2019
208
38
28
Trendnet support wrote this back today about the issue with 2.5 / 5 Gbps and their switch :

"We received an update and they identified the compatibility issue. We are in contact with the chipset manufacturers involved to resolve the issue. Unfortunately we do not have a hard ETA on a resolution. They are committed to resolving it as quickly as possible. Hopefully with the identification of the issue they can produce a quick resolution. "

I wouldn't be surprised if some of the other devices that have had problems reported at 2.5 Gbps, such as the iPolex SFP+, are using chipsets from the same source.
 

hmw

Active Member
Apr 29, 2019
337
110
43
I wouldn't be surprised if some of the other devices that have had problems reported at 2.5 Gbps, such as the iPolex SFP+, are using chipsets from the same source.
Ipolex probably uses the Marvell controller Marvell 88x3310. Marvell plans to use the newer Aquantia PHY especially for their own SFP modules and the like, but it’s too new for Wiitek. Ipolex etc


do you know what ASIC the TEG-7080 uses?
 

madbrain

Active Member
Jan 5, 2019
208
38
28
Ipolex probably uses the Marvell controller Marvell 88x3310. Marvell plans to use the newer Aquantia PHY especially for their own SFP modules and the like, but it’s too new for Wiitek. Ipolex etc


do you know what ASIC the TEG-7080 uses?
No idea. They didn't tell me more. And I didn't open it up. The Trendnet has been out a while though. And it has problems against Aquantia. So it's got to be something old.
 

randman

Member
May 3, 2020
67
12
8
One caveat about the ZyXEL that I noticed. If it's warmer than about 75 degrees F ambient temperature, I noticed its fan is silent for 5 to 10 seconds, and then the fan ramps up for a second or so, then is quiet for another 5 to 10 seconds, then it ramps up for a second or so, then is quiet, and so on and on... I sit about 2 feet from the switch, so I do hear it. In a way, it might be better if it were just running on a very low RPM constantly, rather than ramping up, being quiet, ramping up, etc. My other ZyXEL is in an equipment closet, which is in a guest room. You won't really hear it when 10' away even if the equipment closet door is open.
ZyXEL came out with a new firmware for this switch (XS1930-10) about 5 days ago. It's the first firmware update available since the switch was released. It has one fix. Per release notes:

""Optimization on intermittent fan speed operation.""

I'm happy to report that this firmware update resolved the fan noise issue I mentioned above. The switch is fairly quiet now.
 
  • Like
Reactions: DeltaQ and PigLover

rafale77

Member
Sep 28, 2020
73
28
18
Hmm wish I had found this thread sooner as I would have gotten a Zyxel XS1930-12HP instead of a combination of a Mikrotik CRS312 and Netgear MS510TXPP as I need POE an multigig. Appreciate all the reports of success with the zyxel.
 

gr8form

New Member
Jan 23, 2021
2
0
1
Wow, really nice content here; respect! I've been looking to upgrade my home LAN and was looking at multi-gigabit switches as I've Cat6A in the walls. I'm curious though why no mention of Ubiquiti's US-XG-6PoE product which looks to me to be about nearly the same specs as the ZyXel XS1920-12HP just half the ports which is really all I need though it's not at half the cost of the ZyXel at $1Kish unfortunately, but reasonable at $570ish. I've Unifi AC-Pro APs and their lower end managed switches currently which are OK and yeah I've some issues with the quality of the Unifi software 'releases'; so, I just avoid auto-update and watch for a 'safe stable' release based on the community reports. I didn't see any reviews for the Ubiquiti product and given the bad experience of the edge switch that was reviewed, I can understand why not. So, given the specs and my reduced port needs, I wanted to sense check to see whether you folks know anything else about this unit? Thanks in advance!
 

rafale77

Member
Sep 28, 2020
73
28
18
@gr8form,

I came from a full unifi setup(11 switches, 4 APs) and got rid of it. The US-XG-6POE is one of the better elements I was hoping to keep but eventually weird bugs and the constant firmware/controller song and dance adding more features I don't need along with restrictions and bugs convinced me to part ways. Overall I even saved half the cost of my unifi setup by upgrading and selling my unifi gear on ebay and now I have more than double the bandwidth and 100% less time spent maintaining, testing and fixing my network. If you only need half the ports, the new QNAP offering or the Netgear ones are definitely much better and cheaper choices. Actually you could even buy 2 Netgear MX510TXPP and have money to spare. I did end up getting the zyxel replacing the combination of the Mikrotik+Netgear and am very happy with it. The unifi is overall an overpriced and underpowered (and hardware-wise obsolete) setup with a very pretty interface... A bit like very nice makeup applied on an ugly pig.
 
  • Like
Reactions: gr8form

gr8form

New Member
Jan 23, 2021
2
0
1
Thanks Rafale77. I’m well aware of the UniFi software pitfall and lipstick comment is quite appropriate. I‘ll likely go with the ZyXel for future nBaseT expansion anyways as new APs start rolling out. I realize this is off-topic, but out of curiosity and consideration, what did you replace your UniFi APs with; I’ve several of the ups-ac-pro currently and was looking at their HD line due to density issues and hoping to get better 2.4Ghz performance. Seems the Ruckus R series are well-regarded on this forum.
 

rafale77

Member
Sep 28, 2020
73
28
18
After a few tests of various units, I settled on the Engenius EWS 377AP. They are 4x4 AX units with a 2.5G uplink. Other considerations for the newer zyxel NWA201AX. I think if I was to get another one, that's what I would get. Funny thing with the engenius is that I benefitted from a sale which got me the skykey, equivalent of the cloudkey for free and it 10x faster than unifi's original. Both cost me significantly less than the UAP AC HD they replaced and... yeah they perform way better.

There are other equivalent products from Aruba (instant on) and Netgear (WAX610) which, if you look closely is an engenius EWS357AP with a Netgear firmware.
 
  • Like
Reactions: gr8form

DinoRS

New Member
Jan 17, 2020
6
4
3
While I can't speak for Netgear I don't see "equivalent products" on Aruba side. The EWS377AP uses 4x4 40MHz 1024 QAM while on the Aruba Side most equivalent (but doesn't really matter which of the 500 Series AP really) would be an AP555 which additionally to 4x4 40MHz 1024 QAM for 2,4 GHz aswell does either 8x8 80MHz 1024 QAM or 4x4 160 MHz 1024 QAM which frankly is quite different from what EnGenius offers. I have been looking for ax APs for a while and I went for AP-515 which does 4x4 160 MHz QAM. Not sure if you would notice a difference for most use cases between an EnGenius or Aruba AP but we did notice huge differences between a Mikrotik cAP (granted, they're crap on 5 GHz anyways) and AP-515.

I don't work for either companies mentioned above but from my research the Arubas (most likely followed by Ruckus) seem to be those APs who are designed by maximum ax specifications at least on the Wifi Alliance side. IEEE 802.11ax - Wikipedia
 
  • Like
Reactions: gr8form