Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Quick summary, HDMI 2.0 no longer "exists", and HDMI 2.1 only features are now optional according to the certifying body. Manufactures are supposed to indicate which features they support.

Whelp, I guess we should just stick to Display Port 1.4



I see they went to the USB school of standardisation


USB 2 Full Speed = USB 1 speed

This has been going on for 20 years


That was already called full speed in the USB 1 days though. There's a mode called low speed too which was 1.5mbit and meant for low throughput devices. Full speed was 12Mbit. This way stuff like keyboards and mice didn't need to include the highest end controller chips.

Calling it Full Speed was more a lack of vision than a deliberate attempt to confuse consumers as this HDMI thing seems to me.

So then when 2 and 3 came out they were forced to find superlatives. Hi-Speed, Superspeed.

The same happened with radio. High Frequency was up to 30Mhz. Then they found they could go higher. Very High Frequency, VHF. Then came more advances and Ultra-High Frequency.

Eventually they gave up after XHF and SHF and started using band letters :)

Using relative terms for constantly changing technologies is just a bad idea :)


The same people are the ones that decide to put _final/_latest/_etc in file names


The newest version of USB seems to have solved this by labeling ports with the supported speed (5, 10 or 20). Assuming they start doing the same with the cables, at least that part of the problem becomes tractable.


There is a reserved seat in hell for whoever allowed usb-c cables to have a max speed of usb2 (480Mbit/s).


USB SuperSpeed requires extra wires and much better shielding, which means a thicker, more expensive, less durable, cable.

You also can't make SuperSpeed cables that actually meet the spec once they get past a couple meters.

If I just want to charge my phone or laptop, a High Speed cable is actually better.


USB-C itself has nothing to do with data rate at all. It's really the combination of USB-C and USB 3.x standards made things like hell. It was supposed to have one single cable to do everything but the fact is that I have far more cables which look almost exactly the same but with dramatically different capabilities and I have no idea which can do what at all.

But nonetheless, USB-C indeed introduced one issue that I never imagined: sometimes my phone decided that it should charge the charger instead of being charged...


I don't know, sometimes I just want a very thin cable that is flexible - for charging mostly. I bought a proper USB-C anker cable for my phone early on, and replaced it literally within few days as it was proper big shielded cable that had zero flexibility, horrible for charging. I didn't care that technically it could do 10GB/s, it just wasn't necessary.


In theory, the choice of USB-C or -{,mini,micro}A/B is orthogonal to the choice of version: there are USB-A and even USB-B connectors (with additional pins) that support USB 3.0 as well. There is a logic to this.

In practice, it is confusing.


micro-USB requires a different extended connector for 3.0 [1], and I don't think there is a mini-USB 3.0. USB-A is generally coded with a different color and the extra pins are clearly visible [2], especially when you hold a 2.0 cable side-by-side. So it is all quite clear except for USB-C.

[1] https://www.startech.com/en-us/cables/usb3aub15cms

[2] https://www.howtogeek.com/wp-content/uploads/2014/01/usb-2.0...


There was logic to it: you could have a completely passive cable with a type c connector that would interoperate.

Agree or disagree, it wasn’t completely absurd.


There are "charge only" micro usb cables too; I hate them.


I'd be fine with charge-only cables if there was a distinctive, mandatory, universally-honored way of indicating that.

I deliberately carry charge-only cables when I anticipate encountering untrusted chargers, and I've designated all mine with a band of red heatshrink on both ends.


These are useful for charging in no security situations. You can connect your device to an untrusted port in a wall with a charge-only cable.


If they have an A connector at the other end they are easily identified by just looking how many contacts it has.

These were made to save money but I found them handy for security reasons (no need for a USB condom).


USB condom. And now I know. Where can I buy a trustable one of these?


Pretty easy to validate— you plug your device into a trusted computer first and it doesn't recognize it.


For Type A chargers they are readily available online and can be testing using the approach mikepurvis wrote in a comment parallel to this. Or you can make yourself a clunky one as it’s just two wires.

I have seen them called “data blocker” or “secure charging cable” but what you want is one that is a female-to-male device so you can attach it to charging cables not just bricks.

Type C is harder. I have one from the early days of Type C that doesn’t do PD so in the end it’s only useful for phones. I haven’t seen one that does Power Delivery.


"full" was bad planning but I've never seen anyone advertise USB "full speed".

And I've never seen a USB 1 port or device labeled as USB 2. Is that even allowed?

So the problem hasn't been going on for 20 years.


I really wonder what is up with that. These standards become increasingly frustratingly complex even for people who deal with them daily.


It’s just design by committee and long-standing efforts for backwards compatibility. Also the people writing the standards are far too familiar with them and thus a bit lost when it comes to making practical decisions.

Whenever you make changes there will be compromises and someone will have reason to be unhappy.


There is no way that eliminating 2.0, in such a way that everything that used to rate as 2.0 now qualifies as conforming to some subset of 2.1, can be justified as backwards compatibility.


Cable manufacturers probably realized that the secret to profits lay in resisting commoditization, beat a path to the table, and made it happen.


I don't think its particularly odd that the specifications are supersets of old versions; indeed that feels pretty common in the standards world. IETF specs are maybe the odd ones out where you typically have to read like ten different RFCs to get good picture of some standard.


Free market of fraud is going on, there should be a cost to lying


LOL. Why don't you understand superduperspeed USB 3.4.6 2x2x6? It's so eeeasy.

And, coming soon to an Amazon app near you:

(Pack of 3) USB4 Thunderbolt 4, 98 ft / 30m, 240W charging, gold plated! $19.99! Free Shipping!

It's not like anyone is checking products are what they say they are.


You can't legally call your food a hamburger if it's 90% meat glue and 10% cow.

Can we not do the same with cables?

Whoever is responsible for setting standards must be getting some good kickbacks from all this...


The people writing the standards are also the ones implementing it. That’s the kickback. That’s why every USB 3.whatever device suddenly became USB4 ones.


Same is done with the bluetooth standard. Manufacturers proudly claim their product uses bluetooth 5.0, reviewers assume that means higher speed and lower power, but the products don't actually have to support all sections of the standard and can pretty much just be bluetooth classic with 5.0 slapped on the label


Seems kind of reasonable. DisplayPort is exactly the same - just because your display supported DisplayPort 1.4 doesn't mean it is required to support 10 bit colour, VRR, etc.


I replaced my old Dell XPS laptop with a newer Dell laptop. The old one had been happily driving an external monitor through a thunderbolt cable (USB-C on the laptop side; DisplayPort on the monitor side.)

The new laptop still has a thunderbolt port proudly advertised in the specs. But I'm not allowed to use it. It won't send video data out that way. And when I called tech support, the most they would offer me for my newly-purchased laptop was "Don't use the thunderbolt port. Use the HDMI port."


I thought that the primary feature of Thunderbolt was DisplayPort signaling. What makes it Thunderbolt vs USB-C if it doesn't have DP support?


So Thunderbolt 2 included a DisplayPort signaling layer but (AFAIK; may be wrong) Thunderbolt 3 does not -- instead, the hardware must implement the USB-C Alternate Mode for DisplayPort (and in turn is also implementing the USB-C Alternate Mode for Thunderbolt; it's not USB, at least not in USB3).

Thunderbolt's main draw is being "PCIe on a wire".


My port is advertised as "Thunderbolt 3/USB 3.2 Gen 1 Type-C port with DisplayPort with alt mode". I guess the problem is that it can't speak whatever version the cable understands? The cable says "thunderbolt 3 compatible". And like I said, the same cable worked fine with the same monitor and a previous Dell laptop. :-/


That's crazy, I just got a new laptop from work, a lenovo instead of a dell, the dock is still an older dell thunderbolt dock, and it refuses to correctly work with the laptop. My (3) monitors show up but ethernet and usbs all cycle in and out. I kinda expected it bec the dock had a lot of problems even with dell laptops I eventually got it to work with mine but I know it got so bad dell stopped selling it pretty quickly. Now I need to shell out $400 to lenovo for a new dock that will (hopefully) support 3 monitors at 1080p60hz.

I feel like thunderbolt has gotten to the point that you could legitimately make a great side hustle on affiliate links from in-depth usb-c and thunderbolt accessory reviews.


That sucks - but I can attest that the Lenovo Thunderbolt dock is very solid (I have one of each generation).


That sounds like it should work, yeah. USB-C alternate modes should work over any TB3 cable (as far as I know). That's weird.


They won’t work over active thunderbolt 3 cables


Is that true? Hunh, I was unaware. So a TB3 dock with DisplayPort is handling the display protocol? Is there a standard driver on the OS side?


Not true, thunderbolt 3 can carry up to 8 DisplayPort lanes, something usb-c alternate mode cannot do. Both can coexist


Interesting - I must have misunderstood. Is there a good protocol breakdown of the USB-C to Thunderbolt changes and switch-over? (I assume there has to be, as USB4 brings them closer together?)


What model it is? To know what to avoid. Thanks.


G15.


Well that makes sense. Your 1080p display shouldn't be forced to accept an 8k signal just because the interface supports it.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: