Author Topic: FTDIgate 2.0?  (Read 382295 times)

0 Members and 3 Guests are viewing this topic.

Offline suicidaleggroll

  • Super Contributor
  • ***
  • Posts: 1453
  • Country: us
Re: FTDIgate 2.0?
« Reply #275 on: February 02, 2016, 05:25:39 pm »
You said it yourself.

Sure you'd be better off not trying to trigger a problem

For fuck's sake, I'm trying to stop the hypotheticals, that's all the post was about.  Quit trying to dig an argument for or against FTDI's actions out of my words, there was none.  I can see both sides of the issue, and both sides have valid points.  The moral/ethical/legal legitimacy of FTDI's actions was not what I was discussing, I was talking about how ridiculous these hypothetical scenarios are, and suggesting that people put an end to them because they're pointlessly dragging down the entire discussion.
 

Offline Sal Ammoniac

  • Super Contributor
  • ***
  • Posts: 1662
  • Country: us
Re: FTDIgate 2.0?
« Reply #276 on: February 02, 2016, 05:32:41 pm »
As I wrote before: I have come across firmware doing safety critical tasks and it got upset from receiving data it didn't expect. There is nothing hypothetical about that! Also the assumption FTDI present and future detection algorithms will never be wrong is a false one. So even with a real chip there is a probability things can go wrong (Murphy's law).

This thread is getting absurd in the claims some people are making regarding safety critical products. Any competent designer of a product with a critical safety factor involved is going to take special care that the device cannot malfunction if it gets bad data from an FTDI chip or any other source. He will use error detecting algorithms to insure the integrity of the data stream and have safety interlocks to prevent any damage if something is not right. To not do so would be professional incompetence. For some reason, sloppy design seems to be tolerated more in the electronics/software industries than it is in other engineering professions.

Some have even argued that CRC algorithms are not perfect and errors can slip through. Sure, that's correct if you're talking about a few corrupted bits here and there, but if FTDI were truly sending random characters, then even the most basic checksum will detect that.
Complexity is the number-one enemy of high-quality code.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: FTDIgate 2.0?
« Reply #277 on: February 02, 2016, 05:37:56 pm »
As I wrote before: I have come across firmware doing safety critical tasks and it got upset from receiving data it didn't expect. There is nothing hypothetical about that! Also the assumption FTDI present and future detection algorithms will never be wrong is a false one. So even with a real chip there is a probability things can go wrong (Murphy's law).
This thread is getting absurd in the claims some people are making regarding safety critical products. Any competent designer of a product with a critical safety factor involved is going to take special care that the device cannot malfunction if it gets bad data from an FTDI chip or any other source. He will use error detecting algorithms to insure the integrity
You are being super naive here!!! You really don't want to know the shitty firmware I have come across and which still can pass safety regulation tests without problems.

It's always the naive people who say 'that shouldn't happen' which cause the problems. I put a reverse power protection diode or even a bridge rectifier in every DC powered design because even though people shouldn't swap the + and - they still do.
« Last Edit: February 02, 2016, 05:41:33 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Sal Ammoniac

  • Super Contributor
  • ***
  • Posts: 1662
  • Country: us
Re: FTDIgate 2.0?
« Reply #278 on: February 02, 2016, 05:50:34 pm »
You are being super naive here!!! You really don't want to know the shitty firmware I have come across and which still can pass safety regulation tests without problems.

Perhaps we live in different worlds then. In the world I live in engineers take pride in their work and engineer it properly. If there are any safety aspects to a design, they take care that even uncommon failure modes are taken into account and handled properly in a combination of hardware and firmware interlocks.

Just because some people and companies create shitty, dangerous products doesn't mean that everyone does. Tell me--before all this FTDI stuff started, did engineers designing safety critical products that relied on a serial data stream in a safety critical part of the product assume that this data stream was 100% reliable 100% of the time?
Complexity is the number-one enemy of high-quality code.
 

Offline AlxDroidDev

  • Frequent Contributor
  • **
  • Posts: 471
  • Country: br
    • Arduino Web Brasil
Re: FTDIgate 2.0?
« Reply #279 on: February 02, 2016, 06:04:40 pm »
Who the hell still needs FTDI?

They are a shitty company, with zero respect for the end-users (who may or most likely may not [which was my case] be aware that their chip is fake).

It's not like we don't have alternatives and we really need them. There are plenty of options out there, from complete software solutions (like V-USB) to other dedicated alternatives, like the CH340G chip, and other solutions from Prolific, Texas, Cypress, Silicon Labs, Microchip, and a few others.

The sooner people stop caring about FTDI and stop using their products, the sooner we'll be rid of the problem.

Regarding the driver, I NEVER let Windows Update automatically update my drivers, since I've had bad experience with that in a distant past. Most of my drivers are probably out-of-date, but they've been working, so I just let them be. Don't fix it unless it's broken. Neverthless, I was a victim of FTDI with my very first Arduino Nano 3.0, because I didn't know any better, and it was a fresh install.

On a side note, some PL-2303HX drivers will simply not work with fake PL2303 chips, but it won't even touch the chip in any way. Prolific even has a utility to detect fake chips.
"The nice thing about standards is that you have so many to choose from." (Andrew S. Tanenbaum)
 

Offline Monkeh

  • Super Contributor
  • ***
  • Posts: 7990
  • Country: gb
Re: FTDIgate 2.0?
« Reply #280 on: February 02, 2016, 06:06:07 pm »
On a side note, some PL-2303HX drivers will simply not work with fake PL2303 chips, but it won't even touch the chip in any way. Prolific even has a utility to detect fake chips.

They have a utility to detect old, legitimate, and perfectly functional chips, too; their driver. Which won't work with old chips.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: FTDIgate 2.0?
« Reply #281 on: February 02, 2016, 06:11:01 pm »
You are being super naive here!!! You really don't want to know the shitty firmware I have come across and which still can pass safety regulation tests without problems.
Perhaps we live in different worlds then. In the world I live in engineers take pride in their work and engineer it properly. If there are any safety aspects to a design, they take care that even uncommon failure modes are taken into account and handled properly in a combination of hardware and firmware interlocks.

Just because some people and companies create shitty, dangerous products doesn't mean that everyone does. Tell me--before all this FTDI stuff started, did engineers designing safety critical products that relied on a serial data stream in a safety critical part of the product assume that this data stream was 100% reliable 100% of the time?
Again: I have seen released-for-production safety critical firmware do weird stuff when/after receiving unexpected data. And I've seen much worse than that as well. So yes, competent engineers are very rare. Even with safety interlocks and so on poorly designed firmware can still cause lots of trouble.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Sal Ammoniac

  • Super Contributor
  • ***
  • Posts: 1662
  • Country: us
Re: FTDIgate 2.0?
« Reply #282 on: February 02, 2016, 06:26:46 pm »
Again: I have seen released-for-production safety critical firmware do weird stuff when/after receiving unexpected data. And I've seen much worse than that as well. So yes, competent engineers are very rare. Even with safety interlocks and so on poorly designed firmware can still cause lots of trouble.

These companies should be run out of business. I'm not a big fan of government regulation, but this is a case where it's needed. Poor engineering of products with safety critical aspects should not be tolerated.
Complexity is the number-one enemy of high-quality code.
 

Offline f4eru

  • Super Contributor
  • ***
  • Posts: 1086
  • Country: 00
    • Chargehanger
Re: FTDIgate 2.0?
« Reply #283 on: February 02, 2016, 06:31:41 pm »
If you search for 'woman falls in elevator shaft' you'll notice it -shockingly- happens very often! The case I was referring to happened in Germany.

"man falls in elevator shaft" >> 389 000  google results
"woman falls in elevator shaft" >> 116 000 google results
"FTDI accident" >> 71 000  google results
"man falls in elevator shaft due to FTDI malware" >> 7  google results

So men fall 3,5x more often into elevator shafts it seems, amd FTDI is responsible for 1/6 of elevator shaft accidents.

Offline f4eru

  • Super Contributor
  • ***
  • Posts: 1086
  • Country: 00
    • Chargehanger
Re: FTDIgate 2.0?
« Reply #284 on: February 02, 2016, 06:41:02 pm »
My favorite part is where you admit you're wrong and still keep going :-+
it's not wrong. typically, a properly done failure analysis for that specific looks like :
- problem : corrupt data on UART due to access conflict with other program/other Hardware
- probability : low
- severity : mid (after implementing CRC)
- risk level : low


Problem : FTDI just increased the probability, and therefore the risk to dangerous levels !

Offline f4eru

  • Super Contributor
  • ***
  • Posts: 1086
  • Country: 00
    • Chargehanger
Re: FTDIgate 2.0?
« Reply #285 on: February 02, 2016, 06:45:33 pm »
Quote
Any competent designer of a product with a critical safety factor involved is going to take special care that the device cannot malfunction if it gets bad data from an FTDI chip or any other source.

Wrong.
Any competent designer of a product with a critical safety factor involved is going to take special care to reduce the likelihood of  device malfunction down to an acceptable level.
There is no zero risk.
Lesson one in functionnal safety : You cannot eliminate risk. You only reduce it's likelihood or it's severity.

FTDI just raised that risk to an inacceptable level.

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 3575
  • Country: us
Re: FTDIgate 2.0?
« Reply #286 on: February 02, 2016, 07:08:38 pm »
The hypotheticals re: FTDI's shenanigans causing deaths may be overblown but the general principle and precident it sets- that is - companies reeking havoc on end user systems to combat clones - could easily eventually result in loss of life IMHO.

My end user experience with technology in the medical world has shown me that despite the best intentions of engineers - critical, unexpected faults can and do occur.

As things become more and more connected - there's been a rapid push to incorporate more technology into medical equipment and informatics - with efforts underway to allow more and more automatic connection and control - for example between implanted medical devices or bedside hospital medical equipment - and computerized medical systems accessible 24/7 to doctors and staff caring for patients.

The nature of the economics of the current system of hospital adoption of medical technology prevents the usual feedback mechanisms that force companies to compete based on the quality and reliability of their systems - so that salesmen and marketers are the focus - since once a hospital adopts a specific vendors technology they are pretty much stuck with it no matter how shitty it is.
 

Offline Karel

  • Super Contributor
  • ***
  • Posts: 2214
  • Country: 00
Re: FTDIgate 2.0?
« Reply #287 on: February 02, 2016, 07:14:26 pm »
Also the assumption FTDI present and future detection algorithms will never be wrong is a false one.

Can you provide a link to a documented event that shows that FDTI wrongly detected a non-genuine chip
while in reality it was genuine?
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 3781
  • Country: de
Re: FTDIgate 2.0?
« Reply #288 on: February 02, 2016, 07:32:01 pm »
These companies should be run out of business. I'm not a big fan of government regulation, but this is a case where it's needed. Poor engineering of products with safety critical aspects should not be tolerated.

Agreed. But that doesn't mean that we should tempt fate and try to blow up those poorly engineered products in the meantime.

You know, it is a bit like a kid poking nails into an outlet.  99.9% of time nothing happens - some outlets have shutters, kid is lucky and is poking in the "wrong" hole, etc. Then the chance strikes - and my former colleague woke up at 5AM to a blaring fire alarm and a smouldering wall because the junior managed to push a piece of naked wire into the outlet (that was in Switzerland where shuttered outlets aren't the norm). Pontificating about things being poorly engineered wouldn't have helped him any - he still had a major repair on his hands and was extremely lucky that nobody got hurt.
« Last Edit: February 02, 2016, 10:47:21 pm by janoc »
 

Offline Karel

  • Super Contributor
  • ***
  • Posts: 2214
  • Country: 00
Re: FTDIgate 2.0?
« Reply #289 on: February 02, 2016, 08:02:58 pm »
It's interesting to see how some people are clutching at straws and use all kinds of hypothetical cases to justify
their angriness against FTDI instead of pointing to the real culprits (the cloners)...
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: FTDIgate 2.0?
« Reply #290 on: February 02, 2016, 08:15:16 pm »
Also the assumption FTDI present and future detection algorithms will never be wrong is a false one.
Can you provide a link to a documented event that shows that FDTI wrongly detected a non-genuine chip
while in reality it was genuine?
I think I can make the driver do that with some effort (where can I send the bill?). I've seen enough instability issues with the FTDI USB-UART chip in question. Also it is hard to predict what the cloners will come up with next and how the algorithm in the driver will deal with those. Sooner or later the driver will be susceptible to false positives because only code which isn't there is 100% free of errors.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline diyaudio

  • Frequent Contributor
  • **
  • !
  • Posts: 683
  • Country: za
Re: FTDIgate 2.0?
« Reply #291 on: February 02, 2016, 08:55:27 pm »
Just glad my Xilinx Platform USB Download Cable Jtag Programmer arriving soon, dont have a FTDI bridge! 

cypress semiconductor all the way.
 

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 3575
  • Country: us
Re: FTDIgate 2.0?
« Reply #292 on: February 02, 2016, 09:21:44 pm »
It's interesting to see how some people are clutching at straws and use all kinds of hypothetical cases to justify
their angriness against FTDI instead of pointing to the real culprits (the cloners)...

The cloners loaded the gun.  FTDI pulled the trigger.  Who's the victim to blame?

And yes, there are real victims here not just hypothetical ones.
 

Offline Sal Ammoniac

  • Super Contributor
  • ***
  • Posts: 1662
  • Country: us
Re: FTDIgate 2.0?
« Reply #293 on: February 02, 2016, 09:27:02 pm »
Quote
Any competent designer of a product with a critical safety factor involved is going to take special care that the device cannot malfunction if it gets bad data from an FTDI chip or any other source.

Wrong.
Any competent designer of a product with a critical safety factor involved is going to take special care to reduce the likelihood of  device malfunction down to an acceptable level.
There is no zero risk.

In the specific case I mentioned (the device knows it has bad data from a data source) it is possible to ensure that nothing bad happens to a very high degree of certainty. I'll not split hairs with you, but it's that "nothing's perfect" mentality that leads to all of the crap designs on the market today. Most of the cheap consumer crap is that way because consumers are not willing to pay for quality. All they care about is price. That's why they buy poorly designed disposable crap at Wal*Mart (a U.S. chain of stores that caters to this crowd) that ends up in a landfill a few months later. Extending this to the realm of this forum, it's why many people here are so eager to buy cheap Arduino and Segger J-Link clones from China -- they don't care about quality or the fact that these products rip off the original designer of these items. They're the same people who'd buy counterfeit Rolex watches or Prada shoes.
Complexity is the number-one enemy of high-quality code.
 

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 3575
  • Country: us
Re: FTDIgate 2.0?
« Reply #294 on: February 02, 2016, 09:49:40 pm »
Extending this to the realm of this forum, it's why many people here are so eager to buy cheap Arduino and Segger J-Link clones from China -- they don't care about quality or the fact that these products rip off the original designer of these items.

As far as I'm aware, Arduino hardware is open source. Clones ok. Counterfeits (using Arduino logo) not ok.

Quote
They're the same people who'd buy counterfeit Rolex watches or Prada shoes.

What? you mean there are electronic hobbyists who are worried about status symbols and fashion and buy these things based on how they look and not function?

I think that is a poor analogy.

A better one might be someone who buys a cheap automobile to get themselves from point A to B, knowing full well that it is not the same quality and may not last as long as a quality product.  Still it works for their purpose.  For most the technical details of why it is lower quality are unimportant to them.  Patent or trademark infringements, "fake" versus real and underlying details of the technology they leave to the manufactures and legal system to sort out.

 

Offline Pentium100

  • Frequent Contributor
  • **
  • Posts: 258
  • Country: lt
Re: FTDIgate 2.0?
« Reply #295 on: February 02, 2016, 10:02:46 pm »
I have a car analogy:

Let's say the rubber diaphragm in the carburetor of my car failed, so I went and bought a new one, replaced the failed one and my car works again. Unknown to me, the new diaphragm is a counterfeit, made from cheaper materials and won't last as long, but should still last a while. However, while I was at a gas station one day, a man came by, opened he hood of my car, found out that the diaphragm is not genuine and cut it up making my car inoperable (FTDIgate 1) or fixed the throttle valve to full open (FTDIgate 2). I do not think I would be OK with that and my anger would be with the man who disabled my car. If and when the fake part failed prematurely by itself, only then my anger would be with the counterfeiter.

But then again, I can buy car parts made by whoever wants to, as long as they fit and work OK...
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 3781
  • Country: de
Re: FTDIgate 2.0?
« Reply #296 on: February 02, 2016, 10:58:45 pm »
I have a car analogy:

Let's say the rubber diaphragm in the carburetor of my car failed, so I went and bought a new one, replaced the failed one and my car works again. Unknown to me, the new diaphragm is a counterfeit, made from cheaper materials and won't last as long, but should still last a while. However, while I was at a gas station one day, a man came by, opened he hood of my car, found out that the diaphragm is not genuine and cut it up making my car inoperable (FTDIgate 1) or fixed the throttle valve to full open (FTDIgate 2). I do not think I would be OK with that and my anger would be with the man who disabled my car. If and when the fake part failed prematurely by itself, only then my anger would be with the counterfeiter.

But then again, I can buy car parts made by whoever wants to, as long as they fit and work OK...

Don't bother, people like Karel or suicidaleggroll will try to convince you that it is all your fault because you weren't supposed to be cheap and not do your due diligence by demanding a certificate of authenticity from whoever sold you the diaphragm and all their suppliers down to the grunt somewhere in Malaysia who actually made it. In fact, you should have stood behind him and watched him making it in order to be sure that he doesn't replace it with a counterfeit behind your back. See, you didn't do it, it is all your fault!

Then you will have some other folks who will tell you it is the car manufacturer's fault because the car shouldn't be so poorly designed as to let the nefarious guy open the hood and break your car.

And then you will have cops who are hopefully not engineers and thus will do the common sense thing instead of blaming the victim - put the saboteur behind bars.

You can't win with these trolls. Fortunately the people who actually are empowered to do something about it tend to know better - Microsoft pulled the driver last time and it is very likely they will do it again once a sufficiently big stink is raised.
« Last Edit: February 02, 2016, 11:01:30 pm by janoc »
 

Offline rch

  • Regular Contributor
  • *
  • Posts: 168
  • Country: wales
Re: FTDIgate 2.0?
« Reply #297 on: February 02, 2016, 11:02:09 pm »
But then again, I can buy car parts made by whoever wants to, as long as they fit and work OK...
That is only because some governments have decided, in the specific case of cars, but also in some other consumer goods, that the public good of there being cheap replacement parts available for cars of any age is a greater good than that of car manufacturers being able to extract maximum profit from their intellectual property.  It does remind us that the exploitation of intellectual property is not some sort of natural right,  but enabled by  political decision.
 

Offline Sal Ammoniac

  • Super Contributor
  • ***
  • Posts: 1662
  • Country: us
Re: FTDIgate 2.0?
« Reply #298 on: February 03, 2016, 12:08:42 am »
You can't win with these trolls.

Troll: (noun) Anyone who doesn't agree with my opinion.
Complexity is the number-one enemy of high-quality code.
 

Offline pickle9000

  • Super Contributor
  • ***
  • Posts: 2438
  • Country: ca
Re: FTDIgate 2.0?
« Reply #299 on: February 03, 2016, 01:11:52 am »
The dangerous part is all speculation.
Surprise that you said this.  Of course, it is speculation when you have no real body count to prove .   Once you have one, it becomes a crisis when something not suppose to happen happens, and people started to ask where are all the planning, thinking, and were the engineers sleeping?

Speculation is all I need to spec out another device. Unless there is a particular feature that no other chip has I just look at the next best choice.

FTDI could easily kill the clones by selling at a lower or equal cost. Either that or discontinue the chip and come out with a replacement that has a wiz bang feature. Easy no, but better than having the brand damaged.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf