Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

New 10 GbE USB adapters are cooler, smaller, cheaper

jeffgeerling.com

Post reply on HN
Print view View original

351–360 of 384 posts

Re: New 10 GbE USB adapters are cooler, smaller, cheaper

#351

Post by fulafel »

stonegray wrote: post #299
fulafel wrote: post #258

Earlier quoted context omitted.

A single threaded benchmark better represents real performance, I'd argue. 10 Gbps is only 1.2 GB/s after all and few applications use parallel streams.

I think the intention is to measure the adapter itself independent from the CPU/overall system. Besides, I can’t think of a typical single threaded application that would use those data rates, can you?

File transfer and storage (Dropbox, browser download, rsync, scp, NFS/SAN etc) is a classic use case that can utilize all the bandwidth you have and typically uses single streams between client and server.

Re: New 10 GbE USB adapters are cooler, smaller, cheaper

#352

Post by gsnedders »

ben-schaaf wrote: post #332
reaperducer wrote: post #186

Earlier quoted context omitted.

USB is just a complete mess. You have to go out of your way to make Apple's Lightning connector look sensible, but somehow the USB consortium has managed to do it.

To be fair, lightning only looks sensible because it never did anything other than USB2 and power delivery.

A few devices do support USB 5Gbps over Lightning!

Re: New 10 GbE USB adapters are cooler, smaller, cheaper

#353

Post by eqvinox »

teleforce wrote: post #202
eqvinox wrote: post #59

Earlier quoted context omitted.

It's inherently worse than anything fibre, or even DAC cables (which are kinda cheating.) It needs a shitton of analog "magic" to work with the bandwidth limitations of copper cabling.

Just wondering why you considered DAC cables cheating, is the analog magic mainly the impedance matching or I'm missing something?

DAC cables are cheating because due to the extremely short range limits (5m, 7m if you're very lucky) they can just put the 10Gbase-R/SFI signal straight on a pair of Copper at 10.3125 Gbaud.

10Gbase-T, to try to get to 100m, throws FEC on it and converts the signal to 4x PAM-16/THP at 800 Mbd, and then uses 4 copper pairs *bidirectionally*. That's the analog magic.

Re: New 10 GbE USB adapters are cooler, smaller, cheaper

#354

Post by eqvinox »

spockz wrote: post #243
eqvinox wrote: post #59

Earlier quoted context omitted.

It's inherently worse than anything fibre, or even DAC cables (which are kinda cheating.) It needs a shitton of analog "magic" to work with the bandwidth limitations of copper cabling.

Okay. Sure. But why do we notice that on 10GbaseT and on 1? Is there some signal processing which is exponentially expensive at faster speeds? I’ve seen cards using 25W per port.

cf. https://news.ycombinator.com/item?id=47908287

Yes, that signal processing is massively more expensive. A 10Gbase-T PHY is a sophisticated DSP. Not sure if the power needs are exponential, given we only have a few data points, but it's in the ballpark.

(1000base-T PHYs are already DSPs, but nowhere near as sophisticated)

Re: New 10 GbE USB adapters are cooler, smaller, cheaper

#355

Post by eqvinox »

marcosdumay wrote: post #308
eqvinox wrote: post #67

Earlier quoted context omitted.

Coincidentally, the USB-C spec is written such that wattage implies a minimum set of supported voltages: * ≤15W charger: must have 5V * ≤27W charger: must have 5V & 9V * ≤45W charger: must have 5V & 9V & 15V * (OT but worth noting: >60W: requires "chipped" cable.) * ≤100W charger: must have 5V & 9V & 15V & 20V (levels above this starting to become relevant for the new 240W stuff) (36W/12V doesn't exist anymore in PD…

It's a 3A supply up to the 100W one, that gets upped to 5A at higher voltages. Varying voltage power supplies are usually capped by current, not power. That's because many of the components, set maximum current and voltage that you must obey independently. At higher voltages people start accepting higher loses in stuff like cables, because fire-safety becomes a more important concern than efficiency. So the standard…

You're correct but it's irrelevant. My point was that these requirements are in the standard and if you want to put the USB logo on a power brick you need to meet them. And the consumer is intended to be able to rely on them - which was & still is a pretty good idea considering the USB-C cable carnage.

I wish they did something like this for USB-C cables, but it's probably too late.

Re: New 10 GbE USB adapters are cooler, smaller, cheaper

#356

Post by eqvinox »

adrian_b wrote: post #235
eqvinox wrote: post #207

Earlier quoted context omitted.

> Moreover, "5 Gb/s" is a marketing lie. It's not a lie, the b just stands for baud not bit ;-)

That is technically correct, but "b" has never been an accepted abbreviation for baud (which was Bd) and the naming of the first versions of the PCIe, USB 3 and SATA speeds, which were done by Intel, were obviously in contradiction with the industry standards and intended to confuse the customers. Previously to these standards promoted by Intel, the 1 Gb/s Ethernet used the same encoding and it was rightly called by…

Yeah, it was a tongue-in-cheek comment. It's a shitshow, and I wish it'd backfire, but it won't. Maybe the EU will come up with some (better) "true labelling" laws, if not I see no chance for this to get better.

(Why the current laws don't cover this, I have no idea. It's technically false advertising.)

Re: New 10 GbE USB adapters are cooler, smaller, cheaper

#357

Post by eqvinox »

adrian_b wrote: post #232
riobard wrote: post #217

Earlier quoted context omitted.

That's exactly the issue. I'm just pointing out that it's a fantasy to hope for simple numbering of max supported speeds will simplify the current USB mess. It will not. Consumers would expect plugging a 20Gbps device into a 40Gbps port should result in 20Gbps negotiated speed. In reality it will mostly likely end up at 10Gbps (or less) because of the mess.

Older Thunderbolt devices were not compatible with USB, so plugging them into an USB Type C port would not work. Newer Thunderbolt/USB 4 devices do not have any technical reason for preventing them to work as USB 3.2 2x2, i.e. to work at 20 Gb/s when plugged into a 20 Gb/s host port, and vice-versa for 20 Gb/s devices plugged into a USB 4/Thunderbolt host port, because both Thunderbolt and 20 Gb/s USB need the same w…

USB4/TB4 devices doing (only) PCIe tunneling will absolutely not work on a USB3.2 port, or even on an USB4 port without PCIe support (which you can find on some very recent smartphones I believe. It's spec compliant, PCIe is optional.)

Re: New 10 GbE USB adapters are cooler, smaller, cheaper

#358

Post by Gigachad »

mort96 wrote: post #333
Gigachad wrote: post #188

Earlier quoted context omitted.

The modern usb naming is to just list the speed or power output of the port. Rather than some absurd version number it’s now just “USB 20 Gbits”

Then why do I still see USB 3.2 generation 2x2

I'm not sure I've ever seen that on a product description. But at any rate, USB IF doesn't have any ability to enforce branding guidelines unless the product uses the official USB logo.

Re: New 10 GbE USB adapters are cooler, smaller, cheaper

#359

Post by rincebrain »

My question would be, I suppose, how well do they work for extended intervals (e.g. 30+ minutes of saturated traffic)?

I've tried buying several USB3 2.5/5/10GbE copper adapters, apparently mostly Aquantia under the hood, and all but one of them would, even with fans actively pointed at them cooling them, rapidly reach a temperature at which they would stop operating entirely, which has turned me off of trying to explore more.

Re: New 10 GbE USB adapters are cooler, smaller, cheaper

#360

Post by rincebrain »

jech wrote: post #97
hsbauauvhabzb wrote: post #37

Earlier quoted context omitted.

For those of us who don’t know, how does it save power vs a 1gbe running at low throughput?

> how does [100BASE-TX] save power vs [1000BASE-T] running at low throughput? 100BASE-TX uses just two pairs (lanes), one for sending and one for receiving. 1000BASE-T uses all four pairs, for both sending and receiving. Therefore, a 100BASE-TX interface that's only receiving needs to power up one pair. A 1000BASE-T interface needs to power all four pairs all the time. I recall reading about some extensions that allo…

My only annoyance with "Green Ethernet" things is that often they seem to work poorly.

The dedicated machine I still keep around for Windows things has two onboard 2.5GbE ports. It will apparently sometimes, even with all power saving features turned off, randomly negotiate down to 100 mbit if I leave the machine alone for a bit, and then stay at that speed forever unless I manually reset the link after wondering why transferring large amounts of data is bottlenecking severely.

Post reply on HN