4 ms·

> Up to 2 Gbps downlink > 2.5 Gbit/s PoE to upstream switch Can anybody explain to me why these supposedly premier networking devices are lacking so much in b

by tucnak 11mo ago

> Up to 2 Gbps downlink > 2.5 Gbit/s PoE to upstream switch Can anybody explain to me why these supposedly premier networking devices are lacking so much in bandwidth? I get it that mmWave is really only ever realistically going to hit 2.5G over the air, but is there any reason why they're not willing to provide at least 10G copper, or an actual SFP port? Hell, even Macs support 10G these days. I never understood this. Do they mean 2 Gbps downlink per client, or per device in total? If it's the former, 2.5G wired seems like a major bottleneck to any serious consumption. If a single client at 2 Gbps is all the promise of 5G amounted to, well, it would be disappointing to say the least.

fulafel 11mo ago
The whole 2.5 G spec is a weird step for ethernet speeds too. It's unfortunate it took off.
- tucnak 11mo ago
  They said the same thing about 40G but hey, I've loved it for bridging the gap between my two (10G and 100G, respectively) Mikrotik switches. You can have a dozen Gigabit ports, as well as up to four true 10G devices on your aggregation switch, and neither would be bottlenecked by traffic to and from the backside. This has been a massive boon. However, when it comes to 2.5G, I struggle to find one good reason to use it; such a tiny step-up in bandwidth, and for what?
  - JoshTriplett 11mo ago
    > However, when it comes to 2.5G, I struggle to find one good reason to use it; such a tiny step-up in bandwidth, and for what? Portability and heat. You can get a small USB 2.5G adapter that produces negligible heat, but a Thunderbolt 10G adapter is large and produces a substantial amount of heat. I use 10G at home, but the adapter I throw into my laptop bag is a tiny 2.5G adapter.
    - johncolanduoni 11mo ago
      I’m sure it depends on the model, but in my experience if you force a 10G copper transceiver to 2.5G the insane heat generation goes away. I don’t have any Thunderbolt 10G adapters, but I’m kind of surprised they’re much larger. A SFP+ transceiver is the same size as a SFP one.
      - JoshTriplett 11mo ago
        I think a major reason for the size is for heat dissipation, because it has to be prepared to handle the heat of a full 10G copper connection. Mine runs hot.
        0x457 11mo ago
        Most of my cables coming out of the aggregation switch are DAC and fiber, but there is no 10G copper because my PC came with 10G copper NIC integrated. Anyway, the difference in heat between this transceiver is shockingly large. I knew it runs hot before I deployed it, but I wasn't aware that you have to wait for it to cooldown before unplugging, or you get burnt.
    - fulafel 11mo ago
      The power has come down a lot and drops in shorter runs, from ~15w per port at the max 100M cable length back in 2006, to ~2w with todays tech and short cable run. There's of course fiber too...
  - johncolanduoni 11mo ago
    40G on Mikrotik is just channel bonding of 4 10G links at layer 2. It’s not like the vast majority of 100G that does layer 1 bonding. I really don’t know why they did it other than to have a bigger number on the spec sheet - I can’t imagine they save any money having a weird MAC setup almost nobody else uses on a few low-volume models.
  - u8080 11mo ago
    1x PCIE 3.0 has 8 Gbps raw speed - for 2.5Gbps duplex Ethernet you'll need 6~7 Gbps of raw link to CPU. For 5Gbps and higher, you'll need another PCIE line - and SOHO motherboards are usually already pretty tight on PCIE lanes. 10GbE will require 4x3.0 lanes
    - johncolanduoni 11mo ago
      Are motherboards commonly using PCIe 3.0 for onboard peripherals these days? I wouldn’t expect it to save them much money, but my PCIe knowledge is constrained to the application layer - I know next to nothing about the PHY or associated costs.
      - u8080 11mo ago
        Yes, i.e. https://tech4gamers.com/wp-content/uploads/2022/10/B650-AORUS-ELITE-AX-Block-Diagram-896x1024.jpg https://tech4gamers.com/wp-content/uploads/2022/10/B650-AORU...
    - tucnak 11mo ago
      This is got to be it!
    - fulafel 11mo ago
      PCIe is full duplex. And there's no requirement for ethernet ports to be able to do full tilt. Even with a 1x PCIe 3.0, a 10G port will be much much better than a 2.5G one. (But PCIe 3.0 of course is from 2010 and isn't too relevant today - 4.0, 5.0, 6.0 and 7.0 have 16/32/64/128 Gbps per lane respectively)
      - u8080 11mo ago
        It is relevant, SOHO boards still use PCIE 3.0 to attach NICs, i.e. https://tech4gamers.com/wp-content/uploads/2022/10/B650-AORUS-ELITE-AX-Block-Diagram-896x1024.jpg https://tech4gamers.com/wp-content/uploads/2022/10/B650-AORU... Agree that 10G would be still better, but OEMs couldn't write 10G on their hw while it is capped to 7Gb
    - 11mo ago
jasoncartwright 11mo ago
I think you answered your own question - also the places where mmWave is available, there is also often other better internet connection options.
johncolanduoni 11mo ago
This is a modem, it itself is the client of a cell tower/base station. So unless you put it in a faraday cage with the base station next to it, 2G is almost certainly enough. The better reason to put a 10G transceiver in this would be that some (cheap, honestly garbage) SFP+ transceivers can’t negotiate anything between 1G and 10G. But I’ve only seen that on bargain-bin hardware so I don’t know that they should be designing products around it.
hedora 11mo ago
This device is PoE. I’d guess peak wattage has a lot to do with it.
tecleandor 11mo ago
Probably because of the PoE. That discards SFP+, and makes difficult PoE over copper, as you'd probably need 802.3bt PoE++ (that probably most of the Unifi devices aren't compatible with), or a very short cable to avoid interference. 10Gb interfaces also tend to run quite hot and be a bit power hungry. This is a device that needs to be in a location with good 5G reception, so it makes sense to be PoE powered so you can put it near a window or in the location that gets the best reception, and only run a long ethernet cable. And, although I don't like it too much, 2.5G or 5G NBASE-T is the nearest thing that covers 5G speeds. The 2Gb downlink speed is the 5G downlink, the max for the whole 5G connection, so 2.5Gb ethernet is enough for that.

by tucnak 11mo ago

fulafel 11mo ago

The whole 2.5 G spec is a weird step for ethernet speeds too. It's unfortunate it took off.

tucnak 11mo ago
They said the same thing about 40G but hey, I've loved it for bridging the gap between my two (10G and 100G, respectively) Mikrotik switches. You can have a dozen Gigabit ports, as well as up to four true 10G devices on your aggregation switch, and neither would be bottlenecked by traffic to and from the backside. This has been a massive boon. However, when it comes to 2.5G, I struggle to find one good reason to use it; such a tiny step-up in bandwidth, and for what?
- JoshTriplett 11mo ago
  > However, when it comes to 2.5G, I struggle to find one good reason to use it; such a tiny step-up in bandwidth, and for what? Portability and heat. You can get a small USB 2.5G adapter that produces negligible heat, but a Thunderbolt 10G adapter is large and produces a substantial amount of heat. I use 10G at home, but the adapter I throw into my laptop bag is a tiny 2.5G adapter.
  - johncolanduoni 11mo ago
    I’m sure it depends on the model, but in my experience if you force a 10G copper transceiver to 2.5G the insane heat generation goes away. I don’t have any Thunderbolt 10G adapters, but I’m kind of surprised they’re much larger. A SFP+ transceiver is the same size as a SFP one.
    - JoshTriplett 11mo ago
      I think a major reason for the size is for heat dissipation, because it has to be prepared to handle the heat of a full 10G copper connection. Mine runs hot.
      - 0x457 11mo ago
        Most of my cables coming out of the aggregation switch are DAC and fiber, but there is no 10G copper because my PC came with 10G copper NIC integrated. Anyway, the difference in heat between this transceiver is shockingly large. I knew it runs hot before I deployed it, but I wasn't aware that you have to wait for it to cooldown before unplugging, or you get burnt.
  - fulafel 11mo ago
    The power has come down a lot and drops in shorter runs, from ~15w per port at the max 100M cable length back in 2006, to ~2w with todays tech and short cable run. There's of course fiber too...
- johncolanduoni 11mo ago
  40G on Mikrotik is just channel bonding of 4 10G links at layer 2. It’s not like the vast majority of 100G that does layer 1 bonding. I really don’t know why they did it other than to have a bigger number on the spec sheet - I can’t imagine they save any money having a weird MAC setup almost nobody else uses on a few low-volume models.
- u8080 11mo ago
  1x PCIE 3.0 has 8 Gbps raw speed - for 2.5Gbps duplex Ethernet you'll need 6~7 Gbps of raw link to CPU. For 5Gbps and higher, you'll need another PCIE line - and SOHO motherboards are usually already pretty tight on PCIE lanes. 10GbE will require 4x3.0 lanes
  - johncolanduoni 11mo ago
    Are motherboards commonly using PCIe 3.0 for onboard peripherals these days? I wouldn’t expect it to save them much money, but my PCIe knowledge is constrained to the application layer - I know next to nothing about the PHY or associated costs.
    - u8080 11mo ago
      Yes, i.e. https://tech4gamers.com/wp-content/uploads/2022/10/B650-AORUS-ELITE-AX-Block-Diagram-896x1024.jpg https://tech4gamers.com/wp-content/uploads/2022/10/B650-AORU...
  - tucnak 11mo ago
    This is got to be it!
  - fulafel 11mo ago
    PCIe is full duplex. And there's no requirement for ethernet ports to be able to do full tilt. Even with a 1x PCIe 3.0, a 10G port will be much much better than a 2.5G one. (But PCIe 3.0 of course is from 2010 and isn't too relevant today - 4.0, 5.0, 6.0 and 7.0 have 16/32/64/128 Gbps per lane respectively)
    - u8080 11mo ago
      It is relevant, SOHO boards still use PCIE 3.0 to attach NICs, i.e. https://tech4gamers.com/wp-content/uploads/2022/10/B650-AORUS-ELITE-AX-Block-Diagram-896x1024.jpg https://tech4gamers.com/wp-content/uploads/2022/10/B650-AORU... Agree that 10G would be still better, but OEMs couldn't write 10G on their hw while it is capped to 7Gb
  - 11mo ago

jasoncartwright 11mo ago

I think you answered your own question - also the places where mmWave is available, there is also often other better internet connection options.

johncolanduoni 11mo ago

This is a modem, it itself is the client of a cell tower/base station. So unless you put it in a faraday cage with the base station next to it, 2G is almost certainly enough. The better reason to put a 10G transceiver in this would be that some (cheap, honestly garbage) SFP+ transceivers can’t negotiate anything between 1G and 10G. But I’ve only seen that on bargain-bin hardware so I don’t know that they should be designing products around it.

hedora 11mo ago

This device is PoE. I’d guess peak wattage has a lot to do with it.

tecleandor 11mo ago

Probably because of the PoE. That discards SFP+, and makes difficult PoE over copper, as you'd probably need 802.3bt PoE++ (that probably most of the Unifi devices aren't compatible with), or a very short cable to avoid interference. 10Gb interfaces also tend to run quite hot and be a bit power hungry. This is a device that needs to be in a location with good 5G reception, so it makes sense to be PoE powered so you can put it near a window or in the location that gets the best reception, and only run a long ethernet cable. And, although I don't like it too much, 2.5G or 5G NBASE-T is the nearest thing that covers 5G speeds. The 2Gb downlink speed is the 5G downlink, the max for the whole 5G connection, so 2.5Gb ethernet is enough for that.