One of the things that I found most frustrating about USB-C hubs is how hard it is to find one that actually gives you multiple USB-C ports. I have several USB-C devices but most hubs just give you one USB-C port and a bunch of USB-A ports. At most it’s 2 USB-C ports but only with the hub that plugs into both USB-C ports on my MacBook Pro (so I’m never able to get more ports than I started with). The result is I end up having to keep swapping devices. For a connector that was supposed to be the "one universal port," it's weird that most hubs assume you only need one USB-C connection. Has anyone found a decent hub with multiple USB-C data outputs?
Can anyone tell me why I have several devices in my home that demand a certain USB-C cord in order to charge? They are mostly cheap Chinese devices that won’t acknowledge a more expensive (e.g., Apple) uSB-C cord plugged into them. Even when plugged into the same transformer. They only charge with the cheap USB-C cord they came with. What gives?
Because the USB Consortium made a terrible mistake. Instead of speccing USB-PD power supplies to default to 5V <3A when there are no resistors in the port of the other device, the default is do nothing. So in order to be in spec, you have to refuse to charge non-compliant ports. This means the compliant power supplies are worse, in a way. So you need to use a "dumb" USB-A power supply and USB-A to C cable, which does default to 5V <3A to matter what. As for why some devices choose to use non-compliant ports - I assume it's extreme cheapness. They save a penny on a resistor.
It's not a terrible mistake. A terrible mistake would have been having such power available on ports that even a reasonable person might short out by innocently connecting a USB C cable between them.
A couple 5.1k resistors add about $0.00001 to the BOM cost. The terrible mistake is on the designers of devices who try to forego these.
Except that connecting 5V to 5V does not cause a short circuit. No current will flow without a voltage difference. If there is a difference, the capacitors in one of the power supplies will charge up to the same voltage and then current stops flowing again.
That would be true if both sides were exactly 5.0V, but they're not. There's a 5% tolerance, from 4.75V to 5.25V, and in practice you will see chargers often run "5V" at 5.1V intentionally, to account for resistive loss in the charging cable. If you accidentally connect your "5V" device to the host's "5V" you may find that the host simply disables the port, which has happened to me more than once. So no, you can't just blindly connect them together.
No need to have a separate USB-A brick - simply have a USB-C brick plus C-to-A adapter. An adapter will force 5V down the port no matter what. But afaik you still need USB-A cable (or another adapter?), which kinda defeats the whole idea of having just one cable.
I guess this is only partially true, as I have a A-to-C charger cable from Huawei that works with everything except my Pixel 4A phone. And my Pixel 4A phone works with everything except that specific cable.
Maybe the cable is missing the CC pin resistors (all USB-A to C cables are supposed to have them to identify themselves as such), and maybe only the phone cares.
At this point I'm even surprised that compliant cables and chargers exist so the GP can have that problem.
But I believe the specs are that way to avoid problems with OTG devices. If both devices decide to just push a voltage into the cable at the same time, you risk spending a great deal of energy or even start a fire. That said, there are better ways to deal with this issue; those are just slightly more expensive.
In order to get anything from a USB-C power supply, a device needs to have 5.1kΩ resistors from the CC1 and CC2 pins of the USB-C port to ground. Devices that cheap out on these two resistors (which cost well under a cent each) will not get any power from the power supply.
Some badly designed USB-C devices don’t properly negotiate power supply, and as a result, only USB-A (since these always output 5V without any digital or electrical negotiation) or other non-compliant USB-C devices will actually charge them.
I've always ignored instructions that say to only use that product's USB cord (things like my kitchen scale and water flossed) and have never had an issue. Sounds like I've just gotten lucky though, based on your experience.
I was under the impression that the USB protocol just fell back to 1a 5v when power negotiation was unsure.
USB-C is 0v by default and you have to signal to get anything at all. A lot of junky devices are non compliant and aren’t set up to signal 5v so they get 0 when plugged in to a C-C cable.
With resistors on the CC pins. In particular, there is resistor value that indicates legacy USB charging. This is in the USB-A to USB-C adapters and cables.
The manufacturers cheaped out in not including the right resistors.
I would also guess that some of these cases are designs that were adapted from previous USB mini- or micro-b configurations. Like an intern got the assignment, switched the connector, and wiped hands on pants, not realizing that an electrical change was required as well.
And if you spin the new board and it works with the A->C cable sitting on your desk, then what could possibly be different about plugging it into a C<->C cable, right?
I have 2 powerbanks that cannot be charged by USB-C port when at 0%. The signaling circuitry simply doesn't work. No idea who designed this. It is silly beyond belief. I have to charge it with normal 5V A-to-C cable for 30 seconds, then unplug, then all the PD stuff will start working and I can fastcharge it with USB-C port again. I'm screwed without A-to-C cable.
USB-C hosts and power adapters are only allowed to provide 5V if they can sense a downstream device (either via a network of resistors or via explicit PD negotiation).
Out-of-spec USB-C devices sometimes skip that, and out-of-spec USB-C chargers often (somewhat dangerously) always supply 5V, so the two mistakes sort of cancel out.
It’s not really PD. It’s just they aren’t usb c spec compliant at all. USB-C has the power pins at 0v by default, and you have to signal there is a connected device to activate 5v. While usb-a has 5v hot all the time.
Since there aren’t any active chips in these cables, an A to C cable happens to have 5V hot on the usb c side, but this should not be relied on as it isn’t true for C to C
PD is optional for USB-C devices, but these out of spec devices don’t even support the basic USB-C resistor-based identification scheme (which is mandatory).
I’ve experienced this too and it’s not just no-names. I have a wireless gaming keyboard from SteelSeries, certainly a very legit brand. I lost the original USB-C cord. Tried every USB-C cord I could find, and they power the keyboard and charge it to exactly 1%, but no more.
Found plenty of people online with the same issue but no resolution.
Finally just paid the $25 to get the OEM SteelSeries replacement cable and it charges fully again. wtf… I guess the replacement cable was USB-A to C and I’ve only tried USB-C to C cables?
Adding SteelSeries to my never buy list, along with Unicomp (Unicomp's literally died on me weeks after the 1 year warranty ended. Got told to buy another at full price, went to Ellipse instead at modelfkeyboards dot com for 4x the price and never been happier).
Actually, in most situations with this problem it is possible to solder 2 additional resistors inside the offending USB-C device. I have done that on a flashlight and can confirm that it fixed the problem.
I have purchased multiple devices like this over the years. In all cases, it is that it doesn't have whatever circuitry is required to have a USB-C PD charger send 5v down the line. Using a USB A to C cable works every time. Ironically, using a C to A then A to C then makes it work with a USB-C charger.
They are likely not following the USB spec correctly. Things like pulling certain pins high or low or having a set resistance between certain pins or communications between the host and device will all affect what goes over the wire and whether the host or the device will accept this. Cables will also have some pins entirely unconnected.
Cheap, bad, shortcuts, etc. will result in an out of spec cable being necessary for an out of spec device to work correctly with an in or out of spec hub. It's terrifically frustrating but a fact of the world.
And this isn't just random no name knockoffs. The Raspberry Pi in certain versions didn't correctly follow the power spec. Both the Nintendo Switch and Switch 2 either incompletely, incorrectly, or intentionally abused the USB spec. The Lumen metabolism monitoring device doesn't follow the USB spec. This is one of those things where you want a bit of a walled garden to force users of a technology to adhere to certain rules. Especially when power and charging is involved which can cause fires.
I slowly replaced my home network piece by piece trying to find the bottleneck that was causing my gigabit internet to top out at ~300kbps in my office on the other side of the house from the modem.
After replacing the Ethernet cable run from the second floor to the basement with fiber optic... And the switches in between... And seeing no improvement... I tried a different computer with a built-in ethernet port on the same cable, and pulled 920kbps.
The problem... Was my Caldigit Thunderbolt Dock. I replaced it with an OWC one from Facebook marketplace for cheap and it solved the problem... Roughly $500 in. I'm still angry I didn't check that early earlier.
USB-PD hubs are very annoying. Devices with no battery on a hub (Raspberry Pi etc) will just get hard rebooted if anything else gets plugged into the same hub. I looked at a lot of hubs and they all behaved this way. They all cut power to everything, then re-negotiate the power allowance each device gets from the shared wattage, every time anything new connects. I could not find a hub that honored existing power contracts and gave new devices the remainder. My guess is the average customer expect the newest plugged in device to get max power (at the expense of everything else) or they return it to the store thinking it's broken.
I’m not sure if it’s a USB-PD hub but Jeff Geerling posted a video about a USB-C charger that doesn’t suffer from the renegotiation issue https://m.youtube.com/watch?v=dG2v4FHwJjE
Unfortunately there is no real solution to this, that would work in general case. With renegotiation, the power gets cut off, but most likely every device will get the allowance and all of them will still charge. With no renegotiation, a newly plugged in device might not charge at all. Not sure what's worse.
You need a tester like the FNIRSI FNB58 (not affilate link: https://www.amazon.com/gp/product/B0BJ253W31). This is just an example and not a personal recommendation, as I've just started looking into these myself.
They have fixed this, the modern spec has speed and power rating logos that (good) manufacturers can put on the cables. Just assume anything without a logo on it is USB2/5v only and buy some new cables you know the specs of.
Thunderbolt cables have always been marked either on the ends or with a sticker wrapped around the cable. Everything else can be assumed to be a 2.0 cable at 60w/140w
Colored electrical tape or heat shrink labels at both ends of each cable with a simple coding system (P=Power delivery wattage, D=Data speed, T=Thunderbolt) solves this problem permanently.
The USB IF really should have specified some labeling for cables. The icons used on USB-A connectors are too complicated. What I think would work well is colored rings, with some for the different USB3 speeds, and some for the 100W and 240W charging.
For example: Usb 5Gbps is the same regardless of usb 3.0, 3.1 or 3.2 gen 1. In fact, customers dont need to know that "3.0" number. they just need their ports support 5gbps
Man that is a lot of computer to put into that product for as little money as possible.
I'm not intending to excuse the products with likely bad firmware causing most of these issues, especially Ethernet PHYs. Though, in my professional experience doing embedded device firmware, Ethernet PHYs are always the biggest pain, hands down. The firmware included with them has many many ways to configure it wrong, and the defaults are typically just random settings, not a reasonable configuration. Just getting the drivers even running sometimes involves just knowing a few tricks.
Anyways, it doesn't surprise me many have trouble working right, especially when they indicate they are all running OEM firmware essentially.
Not strictly related but I just bought a USB4 USB-C cable which is rated at 40 Gbps. I still can't really believe it. (I still remember saving to cassette tape)
I use one of these for two 4k monitors, sound, ethernet, mouse, webcam, AND charging. It's amazing having one cable to plug in when I work from my desk. Unfortunately requires one of those $400 docks though.
I just have the cable. I dont have a computer nor a device that can transfer that quickly. Wont be long though! Actually looked like those 6k monitors @60Hz will fill the pipe.
More accurate to say it’s a dock than a hub, but I’m using a Dell 2427DEB monitor[0] with my Dell work notebook and a second monitor daisy chained from it on the DP out port.
My work laptop has just a single USB-C cable plugged into the monitor for everything making it super trivial to plug it back in when I use it away from my desk (which I do regularly).
My personal desktop has a single DP and also a USB A to B cable. The monitor has KVM capability so I can super conveniently switch between them even with the two screens.
Cables are completely minimized with this set up, I’m very happy.
The only thing that’s unfortunate is that I occasionally work on a MacBook Pro 16” M4 and it can’t drive the second monitor over the same USB-C cable as Apple CBA to have support for DP MST on even their premium priced hardware. So I have to also plug in an HDMI cable to the second monitor.
Also unfortunate with the MacBook Pro is that macOS UI scaling doesn’t allow ratios like 125% meaning the UI elements aren’t quite at my preferred size. My Windows 11 handles this perfectly.
When I was tasked with acquiring desk setups for our whole company (small, new startup) back when everyone was using just whatever I decided to go with a really similar setup, Dell P2425HE monitor with integrated USB-C hub, and a Dell P2425H daisy chained from it.
I'm really glad we went with that, so far they've been great (for software devs, no color sensitive stuff).
However, I can’t help but feel a little bit cheated by companies just buying off-the-shelf products, slightly modifying the case layout, and then quadruple the price because it’s “from a reputable company”.
LOL. Welcome to the world of OEM/ODM. As a conservative estimate I'd guess >95% of all consumer electronics is done this way. Even the big names like Apple, Dell, Lenovo, etc. do it.
However, if you are - according to Wikipedia - a two-billion-dollar company like Realtek, then I expect you to get your shit together. There are exactly zero excuses for Realtek to not have a driver release ready almost a year after Big Sur has been announced. Zero.
Mac users are in the minority. It's worth noting that the RTL8153 is a native RNDIS device, which has its history on the Windows side, and Realtek has only started contributing drivers to Linux relatively recently.
FWIW I've had great luck with Realtek NICs, although I don't specifically recall using their USB NICs.
> I've had great luck with Realtek NICs, although I don't specifically recall using their USB NICs.
I envy you. Realtek NICs (especially over USB) are tantamount to trash in my mind after 2 decades of fighting their frequent failures. Be it failure to initialize at all to driver crashes to pisspoor feature sets (or claiming to have features that don't work at all), and a myriad of other problems. Granted, they usually work in Windows, but I don't work in windows (I live and work in linux/BSD). It has become my personally policy to avoid/disable and realtek NICs and replace them with something actually functional whenever possible.
Hopefully their work on linux-side drivers will change this given their proliferation.
To be honest I've yet to find a reliable USB based network interface regardless of chipset/brand/manufacturer, outside of the ones that do PCIe passthrough via USB4/Thunderbolt and those tend to be quite expensive (though they are starting to come down in price).
The problem with USB NICs is now you have two flaky failure points - the chipset driving the USB port and the chipset driving the USB network interface.
I had a reliability issues using a Realtek 2.5 Gbps USB network interface. Kept locking up, or having latency issues. Until I switched which USB port I plugged it into (one that used a different chipset), and after that it was solid.
I suspect a lot of the flakiness is not the chip itself but the fact that, because it's cheap, the bottom-of-the-barrel manufacturers will always use it instead of the significantly more expensive alternatives, and then further cut corners with the board layout and design.
Ironically, the only problems I've had with NICs were on an Intel and a Broadcom.
Unfortunately I can't say I'm surprised that the common thread was Realtek network chips. I've found their NICs to be fairly flaky over the years - the majority work, but a solid minority don't. In contrast, Intel NICs have been bulletproof for me and I seek them out whenever I have any choice in the matter.
Looks like he only bought cheaper things, so no wonder they all eventually died. My USB-C hub is an HP Thunderbolt dock. It's beefy as heck, lasted for years with no issues. It has a tiny fan inside, which I assume helps with the longevity. I hear good things about CalDigit docks too. Those also are very expensive.
Expensive doesn't equal great either. My ThinkPad thunderbolt hub, which is not cheap by any standards, can't route HDMI without randomly restarting itself every few minutes. Connecting the same cable directly to my ThinkPad laptop works perfectly fine. Sort of defeats the whole point of buying a hub. I have sort of given up on hopes of getting a high quality hub - it's a money sink
Yep that one. I guess they don’t have everything, but the everything hubs tend to come with a lot of confusing pitfalls. I use that one in particularly on my steam deck mostly.
Back at one of my previous employers we had a long internal briefing about why our latest device did not have USB-C when other solutions on the market by then had.
The connector is solid but my god have there been disasters because of USB-C.
1. Power distribution upto high wattage, not always with auto sensing,
2. Wites rated to different data transmission speeds.
3. USB standard data transfers and Thunderbolt over the same connector and wire but most accessories are not rated for Thunderbolt.
I’ve also given up on USB hubs and I’m using a Thunderbolt 4 dock to get more IO out of my Mac Studio. It feels crazy to spend that much $’s, but it solved my problems.
I’m so glad to hear other people have had issues with these stupid cheap docks. I’ve burned through three of them over the last few years. It seems that not routing laptop charging power helps, but doesn’t solve the issue. Stupid cheap products.
recently went through the process of picking a new hub and it took hours to locate one actually appropriate to my use case. Helpfully enough, chatgpt/claude were both good at locating ones with specific needs (x USB-C ports, y USB-A, no HDMI, no SD card readers), would probably have taken a lot longer without.
LLMs really are fantastic tools for shopping and product comparison. It's going to suck once the marketers manage to successfully do the equivalent of SEO on them
Years ago, I bought a hub similar to that Satechi one. It worked great until I unplugged my laptop. Then my home network would die. After some sleuthing, I realized the RJ45 interface repeated any packets received, creating a network loop and confusing my Ethernet switch. I contacted Satechi about this obvious defect. Their support team insisted this was by design and told me to unplug USB-C power from the hub whenever I disconnected my laptop, which … kinda defeats the point?
Wow me too. This was very confusing and annoying as my wife's cheap dock would take down the network while we were both working from home. It took several annoying incidents before we connected the incidents to the dock.
I've been on a Macbook M1 Air for the last few years and wanted multiple screens, so I got a USB 3 hub (Dell D6000) which does DisplayLink. I had almost everything hooked in there, but still connected one screen direct via usb-c. Displaylink is good for an M1 as you can add loads of screens if you want, but you can't watch streaming services on them as MacOs thinks your screen is being 'watched'.
I did want a thunderbolt hub but as far as I could tell at the time Thunderbolt and Displaylink just didn't come in the same package, so I was stuck with two cables.
Three years on, I picked up an M4 machine that can do two external screens natively, great, I can reach my goal of only plugging one cable into my macbook. But the Dell can't do that on a Mac because of something or other meaning it would work as mirrored-only.
Time to grab an actual thunderbolt dock. I picked up an HP TB3 refurb (HP P5Q58AA) which was cheap (30 AUD) and seemed great on paper, only to find it needed a type of power adaptor I didn't have that put me back another 60 bucks, and when I got it all hooked up it didn't always work and couldn't support decent refresh rates on my monitors, with one always being stuck at 30Hz. There was a software update available, but for that you need a windows machine with a TB3 port, not something I have.
So then I grabbed a Kensington SD5750T, which was another 140 bucks, but I am pleased to report that it just works and supports dual 4k60 over thunderbolt/USB-C. There is no HDMI or Displayport on this thing, but my monitors both have USB-C in so... Unfortunately, now that I've read the article, I can also confirm it contains a Realtek 0x8153, and is an OEM'd 'goodway' dock.
I'm convinced the only way to get a quality piece of hardware is to spend a couple hundred bucks on a Thunderbolt 4 or 5 hub. I got a TB4 hub from CalDigit and it works great. And no misbehaving network chip to be found either.
I just bought a monitor arm that has a USB dock at the base. The dock is actually just a USB extension cord that is enclosed by the base of the arm. Think of it like putting a $10 USB hub into a box and poking holes in it to stick wires through. Great teardown, makes you wonder what's truly inside these boxes.
In other news, I do think desk makers should start incorporating the USB dock inside the top board of a desk. People go through a lot of money and bullshit to keep their setup clean, especially those who swap computers.
> In other news, I do think desk makers should start incorporating the USB dock inside the top board of a desk. People go through a lot of money and bullshit to keep their setup clean, especially those who swap computers
Humanscale the docks on the monitor arms look integrated but they're actually a really slick attachment that goes around the base nice and snug, so it just looks like one piece. It can be replaced if you take off the monitor, I really liked that. Humanscale is also extremely expensive as far as monitor arms go.
The selection isn't great yet, but you can get a USB charger or USB hub desk grommet (or one that does both).
Hopefully there will be a matching sized one (60mm) doing something useful, but if not you can just put a normal grommet cover in and do the stuff under the desk.
A couple 5.1k resistors add about $0.00001 to the BOM cost. The terrible mistake is on the designers of devices who try to forego these.
But I believe the specs are that way to avoid problems with OTG devices. If both devices decide to just push a voltage into the cable at the same time, you risk spending a great deal of energy or even start a fire. That said, there are better ways to deal with this issue; those are just slightly more expensive.
I was under the impression that the USB protocol just fell back to 1a 5v when power negotiation was unsure.
What kinds of devices?
The manufacturers cheaped out in not including the right resistors.
And if you spin the new board and it works with the A->C cable sitting on your desk, then what could possibly be different about plugging it into a C<->C cable, right?
I have 2 powerbanks that cannot be charged by USB-C port when at 0%. The signaling circuitry simply doesn't work. No idea who designed this. It is silly beyond belief. I have to charge it with normal 5V A-to-C cable for 30 seconds, then unplug, then all the PD stuff will start working and I can fastcharge it with USB-C port again. I'm screwed without A-to-C cable.
The spec was written by a SW engineer, this explains some things. /s
Out-of-spec USB-C devices sometimes skip that, and out-of-spec USB-C chargers often (somewhat dangerously) always supply 5V, so the two mistakes sort of cancel out.
Since there aren’t any active chips in these cables, an A to C cable happens to have 5V hot on the usb c side, but this should not be relied on as it isn’t true for C to C
We can't have nice things.
Found plenty of people online with the same issue but no resolution.
Finally just paid the $25 to get the OEM SteelSeries replacement cable and it charges fully again. wtf… I guess the replacement cable was USB-A to C and I’ve only tried USB-C to C cables?
Cheap, bad, shortcuts, etc. will result in an out of spec cable being necessary for an out of spec device to work correctly with an in or out of spec hub. It's terrifically frustrating but a fact of the world.
And this isn't just random no name knockoffs. The Raspberry Pi in certain versions didn't correctly follow the power spec. Both the Nintendo Switch and Switch 2 either incompletely, incorrectly, or intentionally abused the USB spec. The Lumen metabolism monitoring device doesn't follow the USB spec. This is one of those things where you want a bit of a walled garden to force users of a technology to adhere to certain rules. Especially when power and charging is involved which can cause fires.
I slowly replaced my home network piece by piece trying to find the bottleneck that was causing my gigabit internet to top out at ~300kbps in my office on the other side of the house from the modem.
After replacing the Ethernet cable run from the second floor to the basement with fiber optic... And the switches in between... And seeing no improvement... I tried a different computer with a built-in ethernet port on the same cable, and pulled 920kbps.
The problem... Was my Caldigit Thunderbolt Dock. I replaced it with an OWC one from Facebook marketplace for cheap and it solved the problem... Roughly $500 in. I'm still angry I didn't check that early earlier.
My network is 10 gigabit now though.
USB-C hubs and my slow descent into madness (2021) - https://news.ycombinator.com/item?id=30911598 - April 2022 (513 comments)
Is it a charge only cable (no data)? Is it usb3 5gbps ? Is it 100 watt power delivery? Is it thunderbolt 3/4/5?
[0]: https://www.usb.org/sites/default/files/usb_type-c_cable_log...
For example: Usb 5Gbps is the same regardless of usb 3.0, 3.1 or 3.2 gen 1. In fact, customers dont need to know that "3.0" number. they just need their ports support 5gbps
My work laptop has just a single USB-C cable plugged into the monitor for everything making it super trivial to plug it back in when I use it away from my desk (which I do regularly).
My personal desktop has a single DP and also a USB A to B cable. The monitor has KVM capability so I can super conveniently switch between them even with the two screens.
Cables are completely minimized with this set up, I’m very happy.
The only thing that’s unfortunate is that I occasionally work on a MacBook Pro 16” M4 and it can’t drive the second monitor over the same USB-C cable as Apple CBA to have support for DP MST on even their premium priced hardware. So I have to also plug in an HDMI cable to the second monitor.
Also unfortunate with the MacBook Pro is that macOS UI scaling doesn’t allow ratios like 125% meaning the UI elements aren’t quite at my preferred size. My Windows 11 handles this perfectly.
[0] https://www.dell.com/en-us/shop/dell-pro-27-plus-video-confe...
I'm really glad we went with that, so far they've been great (for software devs, no color sensitive stuff).
LOL. Welcome to the world of OEM/ODM. As a conservative estimate I'd guess >95% of all consumer electronics is done this way. Even the big names like Apple, Dell, Lenovo, etc. do it.
However, if you are - according to Wikipedia - a two-billion-dollar company like Realtek, then I expect you to get your shit together. There are exactly zero excuses for Realtek to not have a driver release ready almost a year after Big Sur has been announced. Zero.
Mac users are in the minority. It's worth noting that the RTL8153 is a native RNDIS device, which has its history on the Windows side, and Realtek has only started contributing drivers to Linux relatively recently.
FWIW I've had great luck with Realtek NICs, although I don't specifically recall using their USB NICs.
I envy you. Realtek NICs (especially over USB) are tantamount to trash in my mind after 2 decades of fighting their frequent failures. Be it failure to initialize at all to driver crashes to pisspoor feature sets (or claiming to have features that don't work at all), and a myriad of other problems. Granted, they usually work in Windows, but I don't work in windows (I live and work in linux/BSD). It has become my personally policy to avoid/disable and realtek NICs and replace them with something actually functional whenever possible.
Hopefully their work on linux-side drivers will change this given their proliferation.
To be honest I've yet to find a reliable USB based network interface regardless of chipset/brand/manufacturer, outside of the ones that do PCIe passthrough via USB4/Thunderbolt and those tend to be quite expensive (though they are starting to come down in price).
I had a reliability issues using a Realtek 2.5 Gbps USB network interface. Kept locking up, or having latency issues. Until I switched which USB port I plugged it into (one that used a different chipset), and after that it was solid.
Ironically, the only problems I've had with NICs were on an Intel and a Broadcom.
Yeah they cost more but they actually work properly.
The connector is solid but my god have there been disasters because of USB-C.
1. Power distribution upto high wattage, not always with auto sensing, 2. Wites rated to different data transmission speeds. 3. USB standard data transfers and Thunderbolt over the same connector and wire but most accessories are not rated for Thunderbolt.
Omg I love it and I hate it.
....like?
It’s called GEO and it’s becoming a thing.
(although.... looks like it's a realtek with the r8169 driver)
The main gist was that almost every hub used the same board.
I've been on a Macbook M1 Air for the last few years and wanted multiple screens, so I got a USB 3 hub (Dell D6000) which does DisplayLink. I had almost everything hooked in there, but still connected one screen direct via usb-c. Displaylink is good for an M1 as you can add loads of screens if you want, but you can't watch streaming services on them as MacOs thinks your screen is being 'watched'.
I did want a thunderbolt hub but as far as I could tell at the time Thunderbolt and Displaylink just didn't come in the same package, so I was stuck with two cables.
Three years on, I picked up an M4 machine that can do two external screens natively, great, I can reach my goal of only plugging one cable into my macbook. But the Dell can't do that on a Mac because of something or other meaning it would work as mirrored-only.
Time to grab an actual thunderbolt dock. I picked up an HP TB3 refurb (HP P5Q58AA) which was cheap (30 AUD) and seemed great on paper, only to find it needed a type of power adaptor I didn't have that put me back another 60 bucks, and when I got it all hooked up it didn't always work and couldn't support decent refresh rates on my monitors, with one always being stuck at 30Hz. There was a software update available, but for that you need a windows machine with a TB3 port, not something I have.
So then I grabbed a Kensington SD5750T, which was another 140 bucks, but I am pleased to report that it just works and supports dual 4k60 over thunderbolt/USB-C. There is no HDMI or Displayport on this thing, but my monitors both have USB-C in so... Unfortunately, now that I've read the article, I can also confirm it contains a Realtek 0x8153, and is an OEM'd 'goodway' dock.
Just as well I'm happy with wireless networking!
In other news, I do think desk makers should start incorporating the USB dock inside the top board of a desk. People go through a lot of money and bullshit to keep their setup clean, especially those who swap computers.
Will be obsolete in 5 years every time
Hopefully there will be a matching sized one (60mm) doing something useful, but if not you can just put a normal grommet cover in and do the stuff under the desk.
I'm using these things a bunch.