USB ethernet adapter versus PCI-E card

[follower]

You're doing it wrong if they're not similar.

Usb 2.0 adapters will only be limited by the 480Mbs speed. I know because I have one in operation for nearly a decade. If your usb 3.0 is affecting your 2.4Ghz, your system is messed up because there's no way it should be doing that unless its FCC radiation is way out of spec (FCC Class A vs B).

I think you don't have a lot of experience with USB ethernet adapters under various environment. I understand that you have only one USB ethernet adapter. USB ethernet adapters can't replace PCIe NIC or on-board NIC unless there is no choice.
Case 1: USB ethernet adapter can kill USB port.
Case 2: USB ethernet adapter can be overheated, network can be disconnected or slowed.
Case 3: USB ethernet adapter can make USB port overheated, network can be disconnected or slowed.
Case 4: USB ethernet adapter can give you high latency.
Case 5: USB 3.0 ethernet adapter can make a lot of issues for wireless devices.
and more.
Sometimes you may use aluminum foil for those issues.

 

[Tech9]

I know what bad things USB-to-Ethernet adapters can do in theory, but I also know I was running pfSense on an HP 800 Mini PC for quite a while using ASIX chip USB-to-Ethernet adapter. It was very stable and never damaged anything. The key is to select the right adapter, I guess. Internal card is going to be more reliable, no doubt about it. I don't use this PC for firewall anymore. It was just a successful experiment.

 

[Samir]

I think you don't have a lot of experience with USB ethernet adapters under various environment. I understand that you have only one USB ethernet adapter. USB ethernet adapters can't replace PCIe NIC or on-board NIC unless there is no choice.
Case 1: USB ethernet adapter can kill USB port.
Case 2: USB ethernet adapter can be overheated, network can be disconnected or slowed.
Case 3: USB ethernet adapter can make USB port overheated, network can be disconnected or slowed.
Case 4: USB ethernet adapter can give you high latency.
Case 5: USB 3.0 ethernet adapter can make a lot of issues for wireless devices.
and more.
Sometimes you may use aluminum foil for those issues.

Because of your lack of the basic understanding of usb as well as ethernet, I think you know far less about usb and ethernet than you are stating you do. I don't have just one usb nic, nor even just usb nics, and have worked with nics since the 1990s.

1. This is false. A USB adapter cannot kill a usb port unless the device is malfunctioning with a serious hardware issue or has an outside power source applied to it.
2. USB nics are very low powered due to the power limitations of USB ports in general. A pcie nic will almost always be warmer than a usb adapter, and heat itself is not an indicator on any level of performance or lack of performance--any overheating condition by any component will lead to odd behavior, not just a nic. And with the limited power available to a USB nic, you will be hard pressed to find a USB nic hot at all.
3. See #2
4. This is false. A wired usb nic will have nearly the same latency as a wired adapter to most applications. Ping can't tell the difference.
5. This is false (and the weirdest thing I've ever heard). USB 3.0 cannot affect anything else unless it is emitting RF outside of its FCC regulated design specs. All devices emit RF so this is not something specific to USB 3.0.

There's nothing magical about a USB nic versus a pcie nic. The pcie bus is 'closer' to the cpu hence it in theory can have better performance due to less overhead than another auxiliary bus like usb, firewire, scsi, etc. Will the average consumer notice? Nope. But if one has a choice, pcie will always be better, and usually cheaper.

 

[Samir]

I know what bad things USB-to-Ethernet adapters can do in theory, but I also know I was running pfSense on an HP 800 Mini PC for quite a while using ASIX chip USB-to-Ethernet adapter. It was very stable and never damaged anything. The key is to select the right adapter, I guess. Internal card is going to be more reliable, no doubt about it. I don't use this PC for firewall anymore. It was just a successful experiment.

I have one of those ASIX chipset based adapters too and it was stable too. But it was only 100Mb which I thought was enough for the application, but an upgrade changed that. So I replaced it with a Lenovo docking station using the DisplayLink combo chipset with a gigabit adapter. Works great.

 

[Tech9]

I just did search for what works with pfSense and one of the recommended adapters was Amazon Basics version with ASIX chip. It is surprisingly well made with metal housing and doesn't heat up in use. Worked on the WAN side for like 6 months without any issues. Recognized in setup, plug and play.

 

[Samir]

I just did search for what works with pfSense and one of the recommended adapters was Amazon Basics version with ASIX chip. It is surprisingly well made with metal housing and doesn't heat up in use. Worked on the WAN side for like 6 months without any issues. Recognized in setup, plug and play.

Good to know. I know most people don't recommend any usb nics for pfsense because the realtek based ones aren't great.

 

[follower]

Because of your lack of the basic understanding of usb as well as ethernet, I think you know far less about usb and ethernet than you are stating you do. I don't have just one usb nic, nor even just usb nics, and have worked with nics since the 1990s.

1. This is false. A USB adapter cannot kill a usb port unless the device is malfunctioning with a serious hardware issue or has an outside power source applied to it.
2. USB nics are very low powered due to the power limitations of USB ports in general. A pcie nic will almost always be warmer than a usb adapter, and heat itself is not an indicator on any level of performance or lack of performance--any overheating condition by any component will lead to odd behavior, not just a nic. And with the limited power available to a USB nic, you will be hard pressed to find a USB nic hot at all.
3. See #2
4. This is false. A wired usb nic will have nearly the same latency as a wired adapter to most applications. Ping can't tell the difference.
5. This is false (and the weirdest thing I've ever heard). USB 3.0 cannot affect anything else unless it is emitting RF outside of its FCC regulated design specs. All devices emit RF so this is not something specific to USB 3.0.

There's nothing magical about a USB nic versus a pcie nic. The pcie bus is 'closer' to the cpu hence it in theory can have better performance due to less overhead than another auxiliary bus like usb, firewire, scsi, etc. Will the average consumer notice? Nope. But if one has a choice, pcie will always be better, and usually cheaper.

I'm pretty sure you know nothing. All of your answers are wrong. You know what? You are just talking about it with your own experience and wrong thought. That's not even your knowledge. It's just a wrong thought. Your answers are so funny.

 

[Samir]

I'm pretty sure you know nothing. All of your answers are wrong. You know what? You are just talking about it with your own experience and wrong thought. That's not even your knowledge. It's just a wrong thought. Your answers are so funny.

At this point, I think you're just a troll. Post reported.

 

[thiggins]

@Samir and @follower Reasonable people can disagree without accusing each other of being ignorant.
Find another way to respectfully discuss the topic or stop the discussion.