Kitz Forum
Chat => Tech Chat => Topic started by: chenks on January 17, 2021, 05:01:45 PM
-
anyone find that USB ethernet adaptors are generally rubbish and/or bordering on false advertisement.
i have a "gigabit usb ethernet adaptor" made by Plugable.
however as it only has a USB2 connection (as it states on the adaptor) then it absolute best it could ever achieve is 480Mbps.
when i run a speedtest using the connection it tops out at around 60MB/s which is pretty close to the 480Mbps limit of the USB connection.
how do they get away with this?
Plugable tend to be one of the more "trusted" brands but not with branding it seems!
-
I would regard a “Gigabit ethernet” device as one that is capable of transferring data over an Ethernet link with a negotiated datalink speed of 1Gbps.
I would not expect that to guarantee any particular level of performance throughput. If throughput was important to me I’d probably be looking for more detailed specifications.
-
the device is not physically capable of anything more than 480Mbps regardless of what the negotiated datalink speed is.
to advertise is as a "gigabit ethernet adaptor" is quite misleading.
-
I think you miss the point I was trying to make.
Gigabit ethernet is a technical standard to which it should conform if wants to describe itself as Gigabit Ethernet. It sounds like it does.
There’s nothing in the Gigabit ethernet standard to dictate the transfer speed of applications, other than (obviously) you won’t exceed 1Gbps.
In practical terms, assuming a gigabit device had a price premium, I’d be disappointed if such a device failed to exceed 100Mbps over a gigabit link. Else I might as well have bought a cheaper one. But anything decently above 100Mbps, say 300 Mbps, would leave me thinking the money was well spent - achieving speeds far in excess of the lesser 100Mbps device.
-
i didn't miss any point you tried to make.
however, we clearly have different thoughts on the matter, yours have been heard.
thanks for your input though.
Admin - Post edited to remove personalised attack remark.
Also deleted a following post
-
Enough!
Please everybody, spare us the personal attacks.
-
Enough!
Please everybody, spare us the personal attacks.
Everybody?
7LM could take offence at that....
-
I’ve got a USB-C Gigabit Ethernet adaptor, and an older USB-A 3.0 Gigabit adaptor.
Both provide the full Ethernet bandwidth to the host.
I too would expect a ‘Gigabit Ethernet Adaptor’ to provide the full transfer rate to the host. Labelling a USB-2 device with the term ‘gigabit’ is disingenuous at best. But good ones do exist. And work well. It’s just that these random-brand devices need their specs checked thoroughly before purchase. The headline description is often best read with a “yes, but” at the end!
-
As mentioned above, they ARE Gigabit Ethernet, that is perfectly true. The only thing they did wrong was to not mention in big letters on the front that the speed will be limited by the USB bus.
The thing is, if you're stuck with USB 2.0 and a built-in 100-BASE-T adapter or nothing, then even getting 200Mbit would be a huge improvement.
The same way I currently use a 5Gbit USB 3.0 adapter. Sure it can't hit 5Gbit, but it can transfer files at 400MB/s vs 120MB/s of the on-board Gigabit.
-
i need to make a correction.
when i was doing the tests i wrote down the results wrong.
the "gigabit" adaptor is, in fact, running at 67Mbps, not 67MB/s.
-
i need to make a correction.
when i was doing the tests i wrote down the results wrong.
the "gigabit" adaptor is, in fact, running at 67Mbps, not 67MB/s.
That certainly doesn't sound right at all -it should run faster than that.
With respect to the gigabit ethernet part, it's not dissimilar to the number of routers available that support 5GHz ac wifi but only have 10/100 ethernet ports. They cannot possible run at full wifi speed across the ethernet link. It's a case of buyer beware unfortunately.
:)
-
it's not dissimilar to the number of routers available that support 5GHz ac wifi but only have 10/100 ethernet ports. They cannot possible run at full wifi speed across the ethernet link. It's a case of buyer beware unfortunately.
:)
The WiFi speed is irrelevant over the Ethernet link?
You can still get full 5Ghz AC WiFi speeds on a router with 100Mb Ethernet ports.
It's only wired traffic that would be limited by the 100Mb/s ports.
-
...
You can still get full 5Ghz AC WiFi speeds on a router with 100Mb Ethernet ports.
It's only wired traffic that would be limited by the 100Mb/s ports.
I highlighted it because I've seen people buy these devices to use on Virgin media fibre connections for example and then complain that they can't download faster than 100Mbps.
:)
-
>>> it's not dissimilar to the number of routers available that support 5GHz ac wifi but only have 10/100 ethernet ports.
Even before you typed that, it struck me that labelling of wi-fi speeds are also misleading to the consumer.
Take for example wireless-n (802.11n) which according to the standards is up to 600Mbps. In real life the maximum is more like 45 Mbps after overheads .... and less if you aren't stood right next to the router. Even 802.11ac (3.46Gbps) tops out at around 100Mbps.
It's a common complaint when people do speed tests and cant understand why they only get ~45Mbps when their router is 802.11n but will say on the box up to 600Mbps.
-
Take for example wireless-n (802.11n) which according to the standards is up to 600Mbps. In real life the maximum is more like 45 Mbps after overheads .... and less if you aren't stood right next to the router. Even 802.11ac (3.46Gbps) tops out at around 100Mbps.
Don't forget n can support 5Ghz too on older devices, so you can get over 100Mbit.
Also 802.11ac tops out around 800Mbit in my experience. My Galaxy S10 can easily do that on an 80Mhz channel width (though that's a synthetic test), strangely my laptops need 160Mhz to do it despite all devices being 2x2 MIMO so should perform the same.
I typically can pull around 40-70MB/s on my laptops from my NAS over WiFi 5/ac. But that highlights the point, there are a LOT of supposed AC routers that can't get close to that because their CPUs are too weak. The Home Hub 5a for example struggles to do 25MB/s which is super annoying as all the hardware is capable, except the CPU.
-
You can still get full 5Ghz AC WiFi speeds on a router with 100Mb Ethernet ports.
It's only wired traffic that would be limited by the 100Mb/s ports.
I'd argue you can't, as typically they will have skimped on the CPU too and it will hugely bottleneck.
-
It seems to be the general case when it comes to these subjects that technical information is written down with a certain level of experience assumption.
I can understand it when we were back in the old days and the main people using computers were those who took an interest. But these days I've met a lot of people who use computers, usb, wifi, and routers with no clue, or interest, in how these items work.
I wonder how many complaints come from people not understanding what the real life expectation should be.
-
>>> so you can get over 100Mbit.
On dual band yes, but ~5/6 years ago there were an awful lot of devices that weren't able to I dont think I ever got above 40Mbps on my old ipad and even that was after tweaking settings. At around that time few phones/tablets had 2 wifi antennas. It's also not really that long ago that few fttc modem routers had dual band 802.11ac and at first even those were in short supply and carried a hefty price premium. Most of the ISP supplied ones were rubbish.
I think ISPs must have eventually cottoned on that a large amount of speed complaints were down to wireless limitations and not the actual connection so it was in their benefit to supply 802.11ac which should at least support the fttc connection speed.
>>> But that highlights the point, there are a LOT of supposed AC routers that can't get close
Agreed. Speedguide net (https://www.speedguide.net/faq/what-is-the-actual-real-life-speed-of-wireless-374) says
Below is a breakdown of actual real-life average speeds you can expect from wireless routers within a reasonable distance, with low interference and small number of simultaneous clients:
802.11n - 40-50 Mbps typical, varying greatly depending on configuration, whether it is mixed or N-only network, the number of bonded channels, etc. Specifying a channel, and using 40MHz channels can help achieve 70-80Mbps with some newer routers. Up to 100 Mbps achievable with more expensive commercial equipment with 8x8 arrays, gigabit ports, etc.
802.11ac - 100+ Mbps typical, higher speeds (300+ Mbps) possible over short distances without many obstacles, with newer generation 802.11ac routers, and client adapters capable of multiple streams.
These real life figures are a long way from "up to 600 Mbps & 1300 Mbps" defined in the standards which some manufacturers were printing on the boxes. The average consumer wouldn't have a clue and as Chenks says it is misleading when you purchase something that doesn't really have a chance of hitting the speed advertised. I guess its reminiscent of "up to" DSL speeds.
I'm conscious that Chenk's thread is about USB Ethernet and I don't really want to veer any further off topic other than to say that wireless also suffers unclear and speeds where Ive seen on the box it would say eg "up to 600 Mbps" but in real life is incapable of anything more than 50Mbps.
Possibly the industry needs an overhaul of showing more realistic achievable speeds when it comes to network speed through put too - be that Ethernet, USB or wifi ?
-
When it comes to the iPad on 802.11n, mobile devices sucked for a long time because they only had one antenna so couldn't do MIMO.
Possibly the industry needs an overhaul of showing more realistic achievable speeds when it comes to network speed through put too - be that Ethernet, USB or wifi ?
Trouble is you'd need both, as you need to know the technical ability and real-world realistic throughput. Which is basically what I was saying they should tell you with ethernet adapters that are bottlenecked by the USB bus, give a realistic real-world speed on the front to set expectation. eg My 5Gbit adapter can do 3.5Gbit.
WiFi is a whole different ball game though as they quote completely unrealistic borderline lies with 802.11n 2.4Ghz particularly bad as they would claim 300Mbit when that required 40Mhz channel width which in 99% of cases you can't use due to neighbouring networks. Although, technically, it was true - in early days of WiFi I DID used to get 100Mbit on 2.4Ghz with 40Mhz channel width, as nobody else had WiFi nearby.