Nintendo Switch consoles keep frying | Hacker News Reader

Nintendo Switch consoles keep frying

arstechnica.com

76 pointsTomte36 comments

Ars literally copy pasted the reddit thread, the quality of writing is almost blogspam at this point.

https://www.reddit.com/r/NintendoSwitch/comments/ckaiiv/an_e...

Long are gone the days of Jon Stokes

So what? My RSS subscriptions don't include random subreddits, but they do include Ars Technica, so their reporting absolutely provided value to me. It is literally the job description of journalists to sift through noise and filter out the signal so I don't have to do it myself.

(Having said that, it probably would have been sufficient to write a way shorter article that just points readers to the Reddit post instead of copy-pasting it. But I suppose that some corporate overlord won't let them do that.)

(And it would also have been more appropriate for this HN submission to link to the Reddit post directly.)

Seems like the widespread adoption of USB-C has been a bit of a mixed bag. The sheer combinatorial explosion of combined cables, accessories and devices has created a lot of surprises for manufacturers. I really like having one cable standard that can do so much, but I hope situations like this become rarer over time.

It doesn't really help.

Before you could tell what a cable and port could do by it's shape and connector. The wrong cable just wouldn't fit.

With USB-C it's a toss up as to whether this will work or release the magic smoke.

Bit of an exaggeration. I have yet to hit any device failures as an excessive user of USB-C. Worst I’ve had is devices that don’t support Thunderbolt or chargers/cables that don’t support USB-PD. The latter is usually a minor degradation.

I am aware that some faulty cables killed devices, but never ran across or even feared running across them.

adoption of usb is a problem. it is a complete nightmare of a standard and specification.

While I definitely can understand frustrations, as it definitely seems way too complicated, I'm not sure what the better alternative ever was. Being able to plug a USB device in and have it work almost every single time and usually automatically sure beats messing with serial ports.

With the benefit of hindsight, I'm certain a better standard could be made today, at least to some degree, but I think anything trying to be as 'universal' as USB is is probably doomed to a similar fate eventually.

Nintendo made a port based on the usb-c standard but changed it so it’s ever so slightly not usb-c. I think the problems stem from people trying to fit a 50-polygon into a circular hole, as it were (other way is a better analogy).

I’ve charged my Switch with third party chargers not designed for the Switch. Obviously I don’t get proprietary power delivery, but it does work in general. The issues with third-party docks don’t seem to apply to typical USB PD chargers.

for industrial applications, basically any other serial solution other than usb is a far better choice than usb. "almost" doesn't work for industrial applications. i personally don't think it works for consumer applications either, but we've all been programmed to deal with issues arising from poorly specified and implemented standards.

i am not for sure why i am downvoted so much, but i suppose it's because most here don't interact with hardware at all. usb, and the associated ecosystem, is a nightmare for anyone who has dealt with it on the job outside of charging their phone.

I agree that USB is terrible. RS-232 is good for some things, and MIDI is good for some things.

In nintendos defense, they say that is not a usb-c port and to use only nintendo accessories.

Then I'd say it's on Nintendo to make their connector proprietary so it is _obviously_ not interchangeable with a standard-shaped cable.

So these cheap docks are putting 9v on a pin that’s really only ever supposed to be 5v?

It doesn’t matter what the standard looks like that’s just bad.

It makes me wonder - did the manufacturers of these docks even really test them thoroughly?

It seems that given the explanation, the faults should be more reproducible though, so who knows.

The switch should include a simple series resistor of a few hundred ohm to protect this pin against over-voltage.

Nah it should have a voltage regulator.

IMO the PD chip should be internally protected to 20V, since accidental shorts between the power and CC pins are not implausible.

The plug shouldn't wiggle enough to cause a short when fully plugged in, and when you're inserting, the power pin is limited to 5V.

Perhaps, but as we've seen, USB-C power supplies vary in quality :)

In any case, protecting against or even 30V should cost essentially nothing. A few series diodes and a resistor should do the trick, no? Diodes are cheap on a CMOS chip.

After a few hundred plug cycles it’s probably going to deform enough that it will no matter how correctly you manufactured it.

How about it should have been an actual USB-C port instead of the horribly broken bodge than Nintendo designed?

USB-C PD 2.0 has quite good specifications that take into account the different failure modes and misbehavior that can occur.

It is an actual USB-C port on the Switch. The dock on the other hand has the non-standard part: a USB-C male connector that doesn't fit snug.

Basically, using a dock is equivalent to plugging a cable that resembles USB-C but isn't actually USB-C.

So just use the dock that comes with the Switch and you're fine? I don't see what the issue is...

The dock that comes with the switch is big, fragile, and unreliable.

There are DIY conversion kits like https://www.amazon.com/BASSTOP-Portable-Replacement-Nintendo... that use the original dock electronics in a different form factor

You forgot insanely expensive for what it is.

Eh, the expensive dock that works seems like a bargain compared to the cheap dock that destroys your docked device.

The obvious question is, if you can make a cheap dock that doesn't destroy the Switch, why make one that does instead?

And if you can't, how is that evidence that the working dock is "too expensive"?

> The obvious question is, if you can make a cheap dock that doesn't destroy the Switch, why make one that does instead?

Probably because it needs to go through a few iterations of failure and learning from failure.

The official dock is over 1/3 the price of a switch. If you figure a 15% chance of outright destruction of the Switch with an aftermarket dock, it may still be a good deal.

> how is that evidence that the working dock is "too expensive"?

Because it uses a standardised connector (USB-C) that has all the specs for power and video delivery well documented but a non-standard implementation of such features.

Nintendo chose to create their own flawed implementation, then charge a premium for their own dock.

Personally I prefer not to be over charged due to artificial constraints.

That's odd. I bought a Switch at launch and my dock's working fine. n=1 and all but it's worked great and I dock/undock my Switch nearly every day. I use it mostly like my cell phone - plug it in at night and grab it in the morning. I rarely actually play on the TV (pretty much only DooM) so I don't know how it's big, fragile, or unreliable.

The dock certainly appears to be fragile enough that I'm nervous about putting it in a suitcase. And, of the one dock that I know of that has seen moderately heavy use (again, n=1), the plug that connects to the Switch has gotten damaged in such a way that HDMI stops working if the Switch is pressed all the way into the dock. It works fine if I carefully insert the Switch very gently so it doesn't go quite all the way in.

For traveling, I would certainly prefer a small dongle just as a matter of size.

A blessing in disguise!

> there is a datasheet for Rohm Semiconductor's M92T30. The literal smoking gun here is the maximum voltage rating on the Configuration Channel pin

Here we see a curious attempt at using ‘literal’ to cancel out the figurative meaning... Aaaaaand it failed spectacularly.

Reply on news.ycombinator.com