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

Your DS18B20 temperature sensor is likely a fake, counterfeit, clone

github.com

Post reply on HN
Print view View original

111–120 of 239 posts

Re: Your DS18B20 temperature sensor is likely a fake, counterfeit, clone

#111

Post by pravda »

selectodude wrote: post #74
IgorPartola wrote: post #32

Earlier quoted context omitted.

This. I buy a decent amount of hobby electronics from eBay and AliExpress because yes it is cheaper. When I need quality, Mouser, DigiKey, and McMaster-Carr are there for me. But the minimum $7-9 shipping is the main reason is forego them. I’d gladly pay 2x for small electronic components (vs directly from China options) if shipping wasn’t a problem.

That's actually how much it costs to ship stuff when it's not being exceptionally heavily subsidized by the utterly broken Universal Postal Treaty.

Digikey would be charging more then $5, but competition forces them to not charge outlandish shipping rates!

It didn't used to be that way -- mail order suppliers used to charge high fake "Shipping & Handling" fees.

That's why Amazon introduced 'free' shipping.

Re: Your DS18B20 temperature sensor is likely a fake, counterfeit, clone

#112

Post by woah »

robomartin wrote: post #58
kregasaurusrex wrote: post #36

Trust in hardware supply chains when manufactring a PCB for a product can be quite fragile: when one component operates outside of spec, the entire device could be rendered useless. In the case of the DS18B20, the author states in the 'Warning' section that the primary way of determining counterfeit sensors is to check the ROM output compared to a known format. When counterfeit parts like this are added, it creates v…

> This anti-consumer behavior on behalf of companies I strongly disagree with this. I see no way to rationalize that a company should be responsible for ensuring that counterfeit devices work correctly by releasing drivers that are tolerant of them or do not stop them from functioning. FTDI's products are the combination of their hardware with their drivers. Both are required in order to delivery functionality and re…

Why are the two options you present either to a) fully engineer and qa the driver to work with every possible counterfeit, or b) to insert malicious code to brick counterfeits? Seems like kind of a false dilemma.

Re: Your DS18B20 temperature sensor is likely a fake, counterfeit, clone

#113

Post by fulafel »

robomartin wrote: post #104
LeifCarrotson wrote: post #76

Earlier quoted context omitted.

I agree that anyone who built a device and knowingly used a counterfeit FTDI chip deserved to be punished. However, the FTDI debacle didn't punish those people, they're not the consumers. It punished end users who have no idea what an FTDI chip is or does or that one exists in the products they buy. In your airliner microcontroller example, you have much more informed consumers. They could reasonably be expected to k…

@LeifCarrotson You are making the mistake of taking a hypothetical and arguing against it. These are contrived examples designed to communicate a concept rather than an absolute reality that one should argue against. I'll just say that it is very common to see comments relating to hardware issues on HN from people who obviously don't have experience building hardware at scale. And so, it is hard to discuss these thin…

You are talking about compatibility but the FTDI case was a case of the company intentionally shipping malicious code used to brick the clones, no? Quote from article: "the latest driver bricks the fake chips, rendering them inoperable with any computer"

Re: Your DS18B20 temperature sensor is likely a fake, counterfeit, clone

#114

Post by pravda »

metaphor wrote: post #68
mindslight wrote: post #33

Earlier quoted context omitted.

How else can you get anything from Maxim besides samples? (joke may be a decade out of date)

I understood that joke to mean, "Maxim components are so expensive that the grey market is about the only place the layman can afford them in more than 1-off prototype quantities." At least that was my experience 8 or so years ago when I made the mistake of purchasing a lot of 100 MAX7219 off eBay for $100 shipped from China...$1/ea was such a tempting deal to a younger, more naive self when these chips were selling…

>I made the mistake of purchasing a lot of 100 MAX7219 off eBay

Why was it a mistake? I've used the 'fake' ones. Did not have any problem with them.

Re: Your DS18B20 temperature sensor is likely a fake, counterfeit, clone

#115

Post by robomartin »

teruakohatu wrote: post #60
robomartin wrote: post #58

Earlier quoted context omitted.

> This anti-consumer behavior on behalf of companies I strongly disagree with this. I see no way to rationalize that a company should be responsible for ensuring that counterfeit devices work correctly by releasing drivers that are tolerant of them or do not stop them from functioning. FTDI's products are the combination of their hardware with their drivers. Both are required in order to delivery functionality and re…

> I strongly disagree with this. I see no way to rationalize that a company should be responsible for ensuring that counterfeit devices work correctly They destroyed devices that worked perfectly well, but maybe (or maybe not) had a fake FTDI branding on a chip inside the device. Even the manufacturer may have been a victim of commingled inventory. For this reason I stopped buying anything with FTDI in it, because I…

> I would hate to think what you would have Apple do to Hackintosh hardware.

You are looking at it precisely backwards. The key question here goes something like this:

Is Apple responsible for ensuring that fakes function correctly as it issues software updates for its own hardware?

In other words, just because someone decided to make a Hackintosh or a fake iPhone is Apple now instantly saddled with having to support this hardware for the lifetime of the fake products? And this is the case whether there's just one clone or 100 different variants?

As I have asked others, in what alternate reality does this make any sense?

My guess is that none of you have ever designed or manufactured hardware products at scale and don't fully comprehend the implications of what you so vehemently believe. No hardware manufacturer would ever take the side of having to ensure fakes work correctly; this would be sheer insanity.

Re: Your DS18B20 temperature sensor is likely a fake, counterfeit, clone

#116

Post by erdo »

robomartin wrote: post #58
kregasaurusrex wrote: post #36

Trust in hardware supply chains when manufactring a PCB for a product can be quite fragile: when one component operates outside of spec, the entire device could be rendered useless. In the case of the DS18B20, the author states in the 'Warning' section that the primary way of determining counterfeit sensors is to check the ROM output compared to a known format. When counterfeit parts like this are added, it creates v…

> This anti-consumer behavior on behalf of companies I strongly disagree with this. I see no way to rationalize that a company should be responsible for ensuring that counterfeit devices work correctly by releasing drivers that are tolerant of them or do not stop them from functioning. FTDI's products are the combination of their hardware with their drivers. Both are required in order to delivery functionality and re…

You're presenting this as a binary choice:

a) > company should be responsible for ensuring that counterfeit devices work correctly

or

b) > bricking [the counterfeit] device as soon as possible

The answer is almost always c) let it be, if it works fine, if not, not your problem.

Adding any kind of bricking code anywhere, unnecessarily introduces the possibility that it will be unintentionally (or maliciously) activated. Personally I don't have that kind of confidence in software or the people that write it (and I am one of those people)

Re: Your DS18B20 temperature sensor is likely a fake, counterfeit, clone

#117

Post by ajnin »

unnouinceput wrote: post #79

I don't get it. Why bother faking something that is very cheap to make genuine? I mean you pay peanuts for real deal, and you want to make something even cheaper then that?...why? Where is the profit?

First, when you make millions of a thing, then saving even fractions of a cent add up to a lot of money. Chinese manufacturers favor high volume, low margin products. Second, Maxim parts are not cheap, in fact they are among the most expensive in the IC industry. A single DS18B20+ goes for 3.47€ on Mouser right now. That's a juicy target for counterfeiters.

Re: Your DS18B20 temperature sensor is likely a fake, counterfeit, clone

#118

Post by robomartin »

woah wrote: post #112
robomartin wrote: post #58

Earlier quoted context omitted.

> This anti-consumer behavior on behalf of companies I strongly disagree with this. I see no way to rationalize that a company should be responsible for ensuring that counterfeit devices work correctly by releasing drivers that are tolerant of them or do not stop them from functioning. FTDI's products are the combination of their hardware with their drivers. Both are required in order to delivery functionality and re…

Why are the two options you present either to a) fully engineer and qa the driver to work with every possible counterfeit, or b) to insert malicious code to brick counterfeits? Seems like kind of a false dilemma.

There are a bunch of permutations, sure.

What you are calling "malicious code" could --and likely is-- quality control code.

The fact that fakes fail isn't the responsibility of the legitimate manufacturer.

I don't know how many hypothetical examples I have to concoct to drive the point home.

OK, here's another one. Fake Tesla. The real Tesla writes code that bricks the entire fake car. This is 100% pro-consumer. The fake Tesla could literally kill people.

What you guys are arguing for is that Tesla should ensure that fakes are able to function. And, at the extremes, that Tesla is responsible for their proper function and safety.

Think about that for a moment.

Post reply on HN