WAVLINK USB C naar 5G Ethernet-adapter, USB C-adapter met 5000 Mbit/s, drivervrije ethernet-netwerkadapter voor laptops, computers, compatibel met Windows 10/11, Mac OS 11 of hoger, Linux en meer

I tested on my Intel MacBook with OSX 15.4.1 and on an ARM Windows 11 computer. Both computers recognized the this device immediately without needing any drivers. All the routers in my house are 2.5Gbps, so I can't test anything faster than that. I used to iperf3 and got 2.36Gbps throughput on Mac and 1.95Gbps on the Windows PC. 2.36Gbps is pretty much full speed for a 2.5Gbps router. For Windows, it's a really sluggish ARM device, so not totally surprised it came in slower. Nonetheless, my Mac shows the hardware is capable of doing at least 2.5Gbps Ethernet.

I use this on my laptop. and I actually let it run for several hours to check speeds and to also check the temperature of the unit. It will generate heat because of the speed that it's dealing with, but it doesn't get very hot. It gets warm, but the center of my laptop keyboard where the CPU is definitely gets warmer. This is a good brand. I have several of their other items, so if you'd need fast wired Ethernet, then this adapter is the way to go.

This Ethernet adapter works well on my M3 MacBook Air. No driver installation was needed. I just plugged it in and it was ready to go. It holds a good, reliable connection.

I can rant about Ethernet standard for hours, but it would bore 99.99% of you. I'll summarize by saying that the IEEE 802.3an standard for 10GBASE-T (10 Gigabit Ethernet over copper) was ratified in July 2006 (nearly 20 years ago).More recently, the IEEE 802.3bz standard, which allows for 2.5 & 5Gbps Ethernet was approved on September 23, 2016. (not quite 9 year ago). Why did we need the 2.5 and 5Gbps standard? Industry wanted to soak consumers by selling gobs of hardware that would soon become obsolete, so they could spend again on the upgrade. (Just like they always do.)I've been planning a 10Gbps copper Ethernet upgrade for a while, and I'm still evaluating products. I ordered this adapter to give it a try, even though it supposedly caps out at 5Gbps.The product description section on "Compatible Devices" says; ..."Compatible: Linux kernel 3.2 and above (basic support, install driver or update to 5.x kernel for full support)"... Well, I'm running a 6.11 Linux kernel with Ubuntu 24.04, so I should be okay, right? Wrong!I connected this adapter to a 20Gbps USB-C port, and saw the kernel enumerate the new Ethernet device:[ 350.565587] usb 6-1: new SuperSpeed Plus Gen 2x1 USB device number 3 using xhci_hcd[ 350.579580] usb 6-1: New USB device found, idVendor=0bda, idProduct=8157, bcdDevice=30.00[ 350.579595] usb 6-1: New USB device strings: Mfr=1, Product=2, SerialNumber=7[ 350.579601] usb 6-1: Product: USB 10/100/1G/2.5G/5G LAN[ 350.579606] usb 6-1: Manufacturer: Realtek[ 350.579610] usb 6-1: SerialNumber: 0003803F5DD09777[ 350.653879] cdc_ncm 6-1:2.0: MAC-Address: 80:3f:5d:d0:97:77[ 350.653936] cdc_ncm 6-1:2.0: setting rx_max = 16384[ 350.654198] cdc_ncm 6-1:2.0: setting tx_max = 16384[ 350.655001] cdc_ncm 6-1:2.0 eth0: register 'cdc_ncm' at usb-0000:66:00.3-1, CDC NCM (NO ZLP), 80:3f:5d:d0:97:77[ 350.710619] usbcore: registered new interface driver cdc_wdm[ 350.713455] usbcore: registered new interface driver cdc_mbim[ 350.716388] cdc_ncm 6-1:2.0 enx803f5dd09777: renamed from eth0I gave the new interface a private IP address:# ifconfig enx803f5dd09777 192.168.144.10/24 upI connected an Ethernet cable between the WAVLINK and one of my systems with a compatible interface (1/2.5/5/10), then ran a speed test:# dd if=/dev/zero bs=1G count=10 | ssh 192.168.144.2 pv -f > /dev/null10737418240 bytes (11 GB, 10 GiB) copied, 57.5595 s, 187 MB/s10.0GiB 0:00:54 [ 189MiB/s] [ ]The test took about a minute, but throughput was poor. I expected something closer to 500MB/s, but got only ~187MB/s.I ran the test in the opposite direction (from the remote system):# dd if=/dev/zero bs=1G count=10 | ssh 192.168.144.10 pv -f > /dev/null10737418240 bytes (11 GB, 10 GiB) copied, 59.3061 s, 181 MB/s10.0GiB 0:00:53 [ 190MiB/s] [ ]The results were equally dismal.Looking at the device settings from Linux revealed an absence of useful information:# ethtool enx803f5dd09777Settings for enx803f5dd09777:Supported ports: [ ]Supported link modes: Not reportedSupported pause frame use: NoSupports auto-negotiation: NoSupported FEC modes: Not reportedAdvertised link modes: Not reportedAdvertised pause frame use: NoAdvertised auto-negotiation: NoAdvertised FEC modes: Not reportedSpeed: 705Mb/sDuplex: HalfAuto-negotiation: offPort: Twisted PairPHYAD: 0Transceiver: internalMDI-X: UnknownCurrent message level: 0x00000007 (7)drv probe linkLink detected: yesRunning the same command with a different brand of (2.5Gbe) USB-C adapter gave me a more informative answer:# ethtool enx0050d6a91d49Settings for enx0050d6a91d49:Supported ports: [ TP MII ]Supported link modes: 10baseT/Half 10baseT/Full100baseT/Half 100baseT/Full1000baseT/Half 1000baseT/Full2500baseT/FullSupported pause frame use: NoSupports auto-negotiation: YesSupported FEC modes: Not reportedAdvertised link modes: 10baseT/Half 10baseT/Full100baseT/Half 100baseT/Full1000baseT/Full2500baseT/FullAdvertised pause frame use: NoAdvertised auto-negotiation: YesAdvertised FEC modes: Not reportedLink partner advertised link modes: 10baseT/Full100baseT/Full1000baseT/Full2500baseT/FullLink partner advertised pause frame use: NoLink partner advertised auto-negotiation: YesLink partner advertised FEC modes: Not reportedSpeed: 2500Mb/sDuplex: FullAuto-negotiation: onPort: MIIPHYAD: 32Transceiver: internalSupports Wake-on: pumbgWake-on: gCurrent message level: 0x00007fff (32767)drv probe link timer ifdown ifup rx_err tx_err tx_queued intr tx_done rx_status pktdata hw wolLink detected: yesRunning the same throughput tests with the other brand of adapter gave me slightly lower throughput (181/174MB/s vs. 187/181MB/s).I figured I'd try a bigger MTU (9000 vs. 1500), but when I tried to change the MTU on the WAVLINK device I saw:# ifconfig enx803f5dd09777 mtu 9000SIOCSIFMTU: Invalid argumentThe above command worked fine on the other branded (2.5) device, and throughput rose to 196/199MB/s.My conclusion is that there's either poor Linux support for this device, or the device is not conformant to the driver requirements. Either way, it's not good.So I can get about 2Gbps with this 5Gbps adapter while running Linux.Edit:I wasn't happy with some of the variables that I had ignored, so I did a little bit more testing.I used a shorter category-8 Ethernet cable, and I booted an even more modern Linux distro (Ubuntu 25.05, kernel version 6.14). The ethtool output was identical, so apparently there's no difference in the (poor) performance of this product between Linux kernel versions 6.11 and 6.14.There was some noticeable improvement in the throughput, benefiting either from a better Ethernet cable and/or a newer Linux kernel. Throughput was better than my previous tests above, so now I'm seeing about 380MB/s, which is at least beyond the 2.5Gbps boundary, but it's still pretty far short of the 5Gbps boundary.Note: I went to the provided website to get the latest driver, which redirected me to the Realtek site. I downloaded the latest Linux driver module, which was last updated on May 13th, 2025 (17 days ago). I compiled and installed the module and it now provides better information to 'ethtool', and lets me set the mtu from 'ifconfig'. Note that my Linux kernel is now 'tainted', because I loaded an out-of-tree module. I doubt that most people would be willing or able to do this.[ 8.367652] r8152: loading out-of-tree module taints kernel.[ 8.367669] r8152: module verification failed: signature and/or required key missing - tainting kernel