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

USB-C is about to go from 100W to 240W, enough to power beefier laptops

theverge.com

Post reply on HN
Print view View original

211–220 of 498 posts

Re: USB-C is about to go from 100W to 240W, enough to power beefier laptops

#211

Post by wearywanderer »

burntwater wrote: post #127

Apple and Microsoft's Surface had (does the Surface still have?) the perfect connector - the Magsafe. The more we dig into this power over USB-C garbage, the farther away we go from that perfection. The patent expiration cannot come soon enough, I seriously hope all portable device vendors bring it back.

Magsafe were the perfect connectors, but it was a huge shame that Apple botched the cables attached to those connectors. I went through three of them in three years; they yellow within months then start to fall apart. But worse than the cables were the fanboys who were always eager to blame me for the cables failing. I've never had such problems with cables before or since. Despite liking the connector itself, I am glad to see these gone.

Re: USB-C is about to go from 100W to 240W, enough to power beefier laptops

#212

Post by h8hawk »

smoldesu wrote: post #192
jeffrallen wrote: post #7

Plus, if you thread the cable through your hot dog, it will be warm in time for lunch. Meanwhile, Apple's M1 chip is showing us that 240 watt laptops are the problem, not the solution...

For the record, the M1's GPU performance doesn't even start to compete with the 2021 laptop market, much less the 2014 one. GPUs have always been the biggest power draw in these laptops, and it's honestly no surprise that Apple can cut their power consumption down to 10w when their GPU is as pathetic as it is.

Which kind of laptops are you speaking for? M1 GPu performance Blow any integrated gpu and all middle class separate GPUs (mx250, etc).

Re: USB-C is about to go from 100W to 240W, enough to power beefier laptops

#213

Post by kiwijamo »

Filligree wrote: post #19
ulfw wrote: post #12

Laptops shouldn't ever need 240W. That's just getting stupid. Buy a desktop if this is what you need!

I can't take a desktop with me on an airplane.

You can! Many years ago I took a desktop machine (minus monitor, keyboard, etc as my friends already had spares of these at the destination) on not one but two flights (going there there and the going back back). Surprisingly the people checking in my bag didn't seem to surprised--the only unusual thing was that they asked me to sign a declaration that the airline wouldn't be responsible for any damage. The bag tag went all around the machine (I didn't have the original box so I took a gamble to see if the airline would accept the machine itself) and off it went on the belt at departure. And on arrival it came on the carousel with all the other bags. It worked fine after both trips. All so I could join a gaming session with my friends in another city! Though it should be said I can see why it might be more tricky once you factor in a monitor and other things. And I probably wouldn't do it again now I have a laptop that can run some pretty decent games--but yes good memories... :)

Re: USB-C is about to go from 100W to 240W, enough to power beefier laptops

#214

Post by gspr »

cletus wrote: post #11

I look forward to USB 4.1a Type C.2 Phase iii.b Gen 3E and the 482646282 different cable capability combinations that all plug in and have no distinction of those capabilities whatsoever. I also look forward to all the necessary and helpful posts telling me how the identical ports on my laptop are in fact diffferent despite their identical appearance. But at least everything uses the same plug, so that’s nice.

Ooh! Will some of them unexpectedly fry the host, the client, or both, unless you spend hours perusing through spreadsheets curated by some random hero on the internet too?

Re: USB-C is about to go from 100W to 240W, enough to power beefier laptops

#215

Post by Klinky »

madsbuch wrote: post #162
grlass wrote: post #134

Earlier quoted context omitted.

I wonder if the the committee have looked at integrating colour or some other indicator into the standard so that cable/port capabilities are clear visually. Though ofc the design challenge is for users to feel comfortable that putting red into blue won't break anything, it just might not give the expected features.

Or! We could go one step further, and change the connector to make sure that people use them correctly. Wouldn't that be clever!?

People already incorrectly buy USB A when they should have bought USB C or visa versa. Changing the physical form-factor doesn't solve this problem.

"I thought I needed a USB cable"?

Also it's a huge waste if say an HDMI 2.1 cable cannot be used on an HDMI 2.0/1.4 device. People would complain about that too.

Re: USB-C is about to go from 100W to 240W, enough to power beefier laptops

#216

Post by alerighi »

skybrian wrote: post #174
cletus wrote: post #11

I look forward to USB 4.1a Type C.2 Phase iii.b Gen 3E and the 482646282 different cable capability combinations that all plug in and have no distinction of those capabilities whatsoever. I also look forward to all the necessary and helpful posts telling me how the identical ports on my laptop are in fact diffferent despite their identical appearance. But at least everything uses the same plug, so that’s nice.

It actually is sort of nice, because "works" isn't binary. It's nice to be able to charge a laptop with what you have, even if it doesn't charge quite as fast. You can transfer data over a cable at lower speeds even if it doesn't run at top speed. Using the "right" cable is performance optimization. This isn't like the old days when plugging things in the wrong way might damage your machine.

Not using the proper cable could lead also to fires, for example. We are talking about a significant amount of current. If you don't use a cable rated for that current, the cable or the connector will overheat, and possibly start a fire.

And it's not a remote possibility. It happened to me with a fast charging phone power supply and the phone, and the cable was the one provided by the manufacturer! The type-C connector at the phone side was red hot and started to melt, probably caused by a bad connection caused by dirt in the phone connector. Fortunately I was there, and smelled the burnt plastic and disconnected the cable, but what if that happened at night?

And it's not something trivial that power supplies and phones can detect, there is not a way to determinate the voltage drop in the cable built into the standard (basically all they needed to do was to add a voltage sensing pin on the connector, to be shorted with VCC at the load side, so the power supply could sense the voltage at the other side and determinate if there too much drop, but as far as I know it doesn't have it).

Re: USB-C is about to go from 100W to 240W, enough to power beefier laptops

#217

Post by OldGoodNewBad »

smoldesu wrote: post #192
jeffrallen wrote: post #7

Plus, if you thread the cable through your hot dog, it will be warm in time for lunch. Meanwhile, Apple's M1 chip is showing us that 240 watt laptops are the problem, not the solution...

For the record, the M1's GPU performance doesn't even start to compete with the 2021 laptop market, much less the 2014 one. GPUs have always been the biggest power draw in these laptops, and it's honestly no surprise that Apple can cut their power consumption down to 10w when their GPU is as pathetic as it is.

Yikes dude you work for Intel?

Re: USB-C is about to go from 100W to 240W, enough to power beefier laptops

#218

Post by wearywanderer »

Klinky wrote: post #215
madsbuch wrote: post #162

Earlier quoted context omitted.

Or! We could go one step further, and change the connector to make sure that people use them correctly. Wouldn't that be clever!?

People already incorrectly buy USB A when they should have bought USB C or visa versa. Changing the physical form-factor doesn't solve this problem. "I thought I needed a USB cable"? Also it's a huge waste if say an HDMI 2.1 cable cannot be used on an HDMI 2.0/1.4 device. People would complain about that too.

> People already incorrectly buy USB A when they should have bought USB C or visa versa. Changing the physical form-factor doesn't solve this problem.

At least the nature of the problem is immediately apparent to such people the moment they attempt to plug the cable in. If the cable won't fit, they know there is nothing wrong with their computer; they didn't misconfigure anything or click the wrong button. The problem is unambiguous.

Some USB-C cables not working with some USB-C sockets leaves users feeling gaslit.

Re: USB-C is about to go from 100W to 240W, enough to power beefier laptops

#219

Post by dragontamer »

alerighi wrote: post #216
skybrian wrote: post #174

Earlier quoted context omitted.

It actually is sort of nice, because "works" isn't binary. It's nice to be able to charge a laptop with what you have, even if it doesn't charge quite as fast. You can transfer data over a cable at lower speeds even if it doesn't run at top speed. Using the "right" cable is performance optimization. This isn't like the old days when plugging things in the wrong way might damage your machine.

Not using the proper cable could lead also to fires, for example. We are talking about a significant amount of current. If you don't use a cable rated for that current, the cable or the connector will overheat, and possibly start a fire. And it's not a remote possibility. It happened to me with a fast charging phone power supply and the phone, and the cable was the one provided by the manufacturer! The type-C connect…

> We are talking about a significant amount of current.

Are we?

I'm not sure how they deliver more power in this specification. But traditionally, the additional power (from 5W to 20W to 100W) was through additional *voltage*, not current.

> And it's not something trivial that power supplies and phones can detect, there is not a way to determinate the voltage drop in the cable built into the standard (basically all they needed to do was to add a voltage sensing pin on the connector, to be shorted with VCC at the load side, so the power supply could sense the voltage at the other side and determinate if there too much drop, but as far as I know it doesn't have it).

Surely we just set a current-limit, written into the specification. Then we choose AWG wires / connectors as appropriate to support that current.

Voltage can't go up arbitrarily, but voltage can safely be increased to ~48V or so in most applications. After 48V, humans start to get shocked / hurt, so that's probably the reasonable limit.

-----

And I'm pretty sure you can sense the voltage drop across a cable. USB-3 delivers 5V by default, and then you send protocol commands to increase your voltage to 12V or whatever. If you detect that the power-supply is only supplying 11V (after the 12V command), then its either a PSU-error or a cable-error. And I'm not sure if it even matters which is which (either way, you're not getting enough power, so your device needs to probably shut off)

You can then disconnect / reset your device and maybe stick to 5V default specs.

Post reply on HN