FTDI driver kills fake FTDI FT232??

[Red Squirrel]

Avoid FTDI,  that is the right path

Yep, when I was in the market for a USB to serial chip I looked elsewhere after hearing about the FTDIgate and all that.  Found the Microchip MCP 2221, it's actually a pretty nice chip, it has other misc features like a few gpio pins and can also do HID and i2c.   I have not looked into any of those features yet, I'm still a newbie myself at electronics, but it's nice to know they're there.

[ebclr]

Silabs CP 2102 another option

[batteksystem]

Glad that I never use FTDI in the first place. Microchip USB MCU is another choice (like PIC18F4550)

[bookaboo]

This could be coincidence or unrelated but we have just had 3 Barth PLCs fail with a comms error, the PLC works fine with the previous program but when trying to update it just says "iThok-RS232: Error write data ti", the virtual COM port that was previously there dissapears. Barth are a low end PLC but are a German company distributed by RS Components. Inside the PLC I can see the FTDI chip. The first unit ran 2 weeks, the 2nd ran 2 days and the 3rd for 1 day.

If its the case that some fakes have gotten into the supply chain and causing this problem (from what I read here they dont fry them just ignore them now) what drivers can be used to read the fakes? (if indeed they are fake... no evidence for that yet).

[batteksystem]

Never use FTDI before, glad that I don't need to use them.

[bookaboo]

This could be coincidence or unrelated but we have just had 3 Barth PLCs fail with a comms error, the PLC works fine with the previous program but when trying to update it just says "iThok-RS232: Error write data ti", the virtual COM port that was previously there dissapears. Barth are a low end PLC but are a German company distributed by RS Components. Inside the PLC I can see the FTDI chip. The first unit ran 2 weeks, the 2nd ran 2 days and the 3rd for 1 day.

If its the case that some fakes have gotten into the supply chain and causing this problem (from what I read here they dont fry them just ignore them now) what drivers can be used to read the fakes? (if indeed they are fake... no evidence for that yet).

That's easy.  Replace Windows with Linux.  FTDI drivers are built into the Kernel, and there is none of this "clones vs. monopoly" bullshit, so it should "Just Work".

Friends don't let friends design anything industrial using any Windows O/S ...

If you can't change the O/S at this stage of the game, then get your supplier to fix it-- (it should be under warranty)-- replace the board(s) that have the clone chips in them with boards that have FTDI chips in them.

We got the issue sorted, nothing to do with FTDI chip in the end. Something weird deep in the .ini file causing an issue as we were pushing the little PLC a bit hard. We got it sorted with some good tech support from Barth and pro-sign.de so thumbs up for them.

Thumbs down as well for FTDI anyway.

[3roomlab]

just a general question, does FTDI still continue to brick CHINA FTDI products?
cos i just bought some FTDI equipped stuff from CHINA and the supplied driver is not working, i am contemplating to use the actual FTDI  driver  ...  or does anyone have a CHINA version driver like v2.12.14 i could "borrow" ... 

[crispytato]

I'm pretty sure that new drivers don't brick the chips, they just won't readout data correctly. http://hackaday.com/2016/02/01/ftdi-drivers-break-fake-chips-again/

[Unixon]

So, 10+ years later, has anyone returned to using FTDI chips in their products? 

[wraper]

So, 10+ years later, has anyone returned to using FTDI chips in their products? 

It was a good thing. Counterfeits have gone away almost entirely and cheap products started using CH340 instead of counterfeit junk. How much it impacted genuine FTDI sales? Who knows, what was seen widely used in inexpensive products were mostly counterfeits anyway.

[Unixon]

It was a good thing. Counterfeits have gone away almost entirely and cheap products started using CH340 instead of counterfeit junk. How much it impacted genuine FTDI sales? Who knows, what was seen widely used in inexpensive products were mostly counterfeits anyway.

We went to CP2102 even prior to FTDIgate just because of lower price and then CH340G has appeared and finally lowered the price bar to the absolute minimum. My only regret is that Prolific PL2303 has been left on the side for no obvious reason although I heard this bridge is less problematic one of FTDI/SiLabs/Prolific trio. WCH chips hadn't been around then yet.

[wraper]

My only regret is that Prolific PL2303 has been left on the side for no obvious reason although I heard this bridge is less problematic one of FTDI/SiLabs/Prolific trio.

Prolific is way worse than FTDI. They fucked over their real customers, not just counterfeit owners. Their way of fighting counterfeits was just quickly dropping driver support of chips that were not the latest version (+ removing older driver versions from the website). Like, your chip is 1 year old and we currently make one with a different suffix? Go fuck yourself. For example, PL2303HXA which was produced in late 2000s till 2012 did not have win 8 support (released in October 2012). AFAIK earliest FTDI chips are still supported in the latest drivers.
Also AFAIK FTDI are the least problematic and the most versatile. You could run native COM port bit-bang programmers on those (like PonyProg) which would fail on others.

[Karel]

So, 10+ years later, has anyone returned to using FTDI chips in their products? 

I never abandoned FTDI products, I never had any issues.

[free_electron]

So, 10+ years later, has anyone returned to using FTDI chips in their products? 

Never gone away. They just work. Screw those counterfeits.

[coromonadalix]

still using  lots of ftdi,  rs232 usb to rs232  rs485  etc ...  ft232h   etc .... 

once you know about drivers vesions who may brick fakes ftdi,  just a matter of blocking drivers updates, and have the ftdiprog software handy  to restore vid's and pid's ....

[Psi]

I'm still using CP210x, used to use FT232R back them.

At some point i'll upgrade to something USB3, or maybe just a MCU with USB3 built in.

[Kjelt]

My only regret is that Prolific PL2303 has been left on the side for no obvious reason although I heard this bridge is less problematic one of FTDI/SiLabs/Prolific trio.

Prolific is way worse than FTDI. They fucked over their real customers, not just counterfeit owners. Their way of fighting counterfeits was just quickly dropping driver support of chips that were not the latest version (+ removing older driver versions from the website). Like, your chip is 1 year old and we currently make one with a different suffix? Go fuck yourself. For example, PL2303HXA which was produced in late 2000s till 2012 did not have win 8 support (released in October 2012). AFAIK earliest FTDI chips are still supported in the latest drivers.
Also AFAIK FTDI are the least problematic and the most versatile. You could run native COM port bit-bang programmers on those (like PonyProg) which would fail on others.

Had that issue two weeks ago when I pulled an old project out of the closet.
But hey if you install the old 3.3.2.105 driver it still works even under win11.

[Unixon]

still using  lots of ftdi,  rs232 usb to rs232  rs485  etc ...  ft232h   etc .... 
once you know about drivers vesions who may brick fakes ftdi,  just a matter of blocking drivers updates, and have the ftdiprog software handy  to restore vid's and pid's ....

This is OK if you're in control of the software side, but giving/selling that hardware to people presumably using Windows is not a good idea.
Well, still using Windows is not a good idea nowadays... W10/W11 is not that good old XP anymore and not even W7.

[magic]

My only regret is that Prolific PL2303 has been left on the side for no obvious reason although I heard this bridge is less problematic one of FTDI/SiLabs/Prolific trio.

Prolific is way worse than FTDI. They fucked over their real customers, not just counterfeit owners. Their way of fighting counterfeits was just quickly dropping driver support of chips that were not the latest version (+ removing older driver versions from the website). Like, your chip is 1 year old and we currently make one with a different suffix? Go fuck yourself. For example, PL2303HXA which was produced in late 2000s till 2012 did not have win 8 support (released in October 2012). AFAIK earliest FTDI chips are still supported in the latest drivers.
Also AFAIK FTDI are the least problematic and the most versatile. You could run native COM port bit-bang programmers on those (like PonyProg) which would fail on others.

Had that issue two weeks ago when I pulled an old project out of the closet.
But hey if you install the old 3.3.2.105 driver it still works even under win11.

Not sure if this driver supports newer chip revisions, OTOH.
But who cares if PL2303, both genuine and clones, work fine in Linux

[coromonadalix]

i dont care about os'es   it's about knowing what you're dealing with  ...  sure  if  ex:  you buy cheap stuff like usb cables, duino's  or else, if you use them for serious stuff .. you have to know

the FTDI gate had stired the web, tons of endelss debates,    it is a known problem, solutions exists,  there was some debates here too

fan boys of nix  nux  windows etc ... it was a mess,  dont start that again


if not : buy CH340 CH341  based chipset  interfaces   loll   i have a few too,  once drivers installed   neved had any problems with them too


Prolific  is a mess, i reject anything based on them


once again they are not KILLED  they have their PID VID changed / voided, they are recoverable

[Kjelt]

Not sure if this driver supports newer chip revisions, OTOH.

Both drivers can be choosen, for newer hardware select the newer driver.
Yes it is one step more hassle but at least your old hardware is back in business.