6 ms·

Isn't there something like, just one company in the world that makes 95% of all USB to Serial chips?

by drvdevd 10y ago

kosma 10y ago
It used to be "just" FTDI... but then other manufacturers and endless copycats jumped in, which resulted in a XKCD #927 situation. As if we didn't already have a USB-CDC standard... (For an interesting side story, search for "ftdigate". At some point FTDI decided they're fed up with copycats and released a new driver pack (through automatic Windows Update) that bricked counterfeit chips. This led to a lot of angry people and number of amusing situations, including someone jokingly submitting a patch to Linux to do the same.)
- dfox 10y ago
  FTDI is not and never was especially big in the USB-to-Serial market, and their chips tended to implement CDC in variously broken ways not to mention that their windows drivers are major PITA. What they are big in is market for "just add USB to this custom device" because they supply mostly NDA-free documentation since forever and produce various ready-made modules.
  - kosma 10y ago
    Speaking out of curiosity, can you name some companies that provide NDA-laden USB-serial bridges? I'm painfully aware that it's hard to even know such products exist, especially when your production volume barely runs into hundreds or thousands and you're not established in the industry... and from my limited interactions with even the smaller players like SIMCom or SiRF, it takes an NDA just to talk to someone who can get you to sign a second NDA so you can see datasheets. Even if you have tons of money, giving it to such company in return for their product can be a challenge.
  - ajford 10y ago
    They also put the legwork into getting their drivers into nearly everything, which makes them great to use in a custom or small batch product that has users that aren't technically oriented or wont be installed by a knowledgeable installer. If you already have a client side install (like a software package) that can bundle the drivers for you, it's not a big deal. And with the emergence of USB-CDC, it's not nearly as necessary as it used to be, since most modern OSes support that now.
    - kosma 10y ago
      And then killed that convenience by literally bricking counterfeit chips. Most users have no way of knowing whether or not their chips are genuine - and, what's worse, most electronics manufacturers have no way of knowing either, because entire supply chains are rotten. These days we have our trusted supplier of FTDI chips but we still check every single shipment. We already had one supplier send us fake chips and then flat out exclaim they did it because "no one would pay premium to get the real thing". We would.. but since they already cheated, we had to switch suppliers.
      - cartoonfoxes 10y ago
        Can you share which suppliers have you been burned by? We've seen counterfeit Prolific parts, but I'm not aware if we've had similar problems with FTDI hardware.
        kosma 10y ago
        I don't remember, unfortunately, as I wasn't the one handling logistics, I just spec'd out parts we needed. If anything, here's a bit of advice: don't ever assume your supply chain is 100% clean - more than once we've been sold thousands of dumpster-grade components by a seemingly very reputable company. And don't ever assume suppliers won't try a bait-and-switch; we've been bitten by this more times than I can count.
        nitrogen 10y ago
        Can the bigger names like Digikey be relied on? Or do their suppliers sometimes give them bad parts, too?
        kosma 10y ago
        My point is that even if your supplier plays fair, the manufacturer may change the designs slightly and you either dont't get the PCN or fail to understand the implications of a change. We had a case once where a manufacturer added a minuscule hole on the case of a relay, presumably a pressure vent. Our devices are potted in resin in order to withstand rough conditions; in this particular case the resin penetrated the relay and blocked the contacts. We were halfway through a production run before anyone noticed. Needless to say, the run had to be scraped because there's no economical way to remove the resin once it's hardened. All for a saving of a few cents per relay. QA lesson: new batch of anything = tight quality control. When you go into manufacturing, it becomes a game of numbers. Manufacturing 10000 of whatever is statistically bound to generate some duds and failures. Speaking of prototype/hobbyist amounts: stick to the big suppliers and you should be safe, but if something feels fishy, don't be shy to order the part from a different supplier. Generally, identifying counterfeit parts in prototype amounts can be tricky without prior experience because your manual soldering skills will always be sub-par to a pick'n'placer; same goes for ESD precautions, accidental shocks applied to the circuit, passerby finger-pokers and your curious boss.
  - duskwuff 10y ago
    > FTDI is not and never was especially big in the USB-to-Serial market ... whaaa? They're probably the biggest company in that market, closely followed by Prolific (PL2303). > ...and their chips tended to implement CDC in variously broken ways No, their chips implement a proprietary protocol. Not CDC at all.
tyingq 10y ago
There are several. FTDI, Prolific, WCH (CH340)...others as well. It's a mess at the moment, because there are unauthorized clones of both FTDI and Prolific, and both companies release drivers that purposefully don't work (or worse...brick them) on the clones. But, there's not really a way for the end buyer to know for sure they are buying the real thing.
- cartoonfoxes 10y ago
  I can confirm the FTDI drivers are total shit. We had to obtain hardware info under NDA, and write our own driver in order to get a reliable solution build using their parts.
- Animats 10y ago
  The SiLabs CP2102N is useful for serial to parallel. It will talk to Linux with the standard Linux serial driver, although you need a free Windows program from SiLabs if you want to reconfigure it. (This is needed only for unusual applications.) I use them because they'll go down to 45 baud for antique Teletype machines. They're popular for Arduino applications, and there are lots of cheap breakout boards with 0.100 pins for Arduino interfacing.
  - i336_ 10y ago
    Interesting. I'd gotten the impression the FT232 was "rock solid" and filed them away as "they're good, use them", but that was before the bricking incident, and now I'm not really sure anymore. I guess on the surface the big thing I really like is device differentiation. Do CP2102Ns have unique serial numbers, or can that free utility burn in info I can use to differentiate? Going a bit deeper, can I bitbang with it?
    - Animats 10y ago
      You can set vendor ID, product ID, product string, serial string, release version, and max power requested. "Manufacturer string" is set to "Silicon Labs". After doing that, you can lock the device against further changes, if you want. This is all done via SiLabs "Simplicity Studio", which is a big IDE for their microcontrollers into which they wrapped up some of the device-specific tools for their simpler devices. Thus, you can force the host machine to demand a device-specific driver if you need to. By default, it appears to the OS as a USB to serial port device. Linux and Windows recognize it as such, without special drivers. Linux mounts it starting at /dev/usb0; Windows mounts it starting at COM3. No bit-banging, though; it doesn't have the hardware.

Isn't there something like, just one company in the world that makes 95% of all USB to Serial chips?

by drvdevd 10y ago

Isn't there something like, just one company in the world that makes 95% of all USB to Serial chips?

kosma 10y ago

It used to be "just" FTDI... but then other manufacturers and endless copycats jumped in, which resulted in a XKCD #927 situation. As if we didn't already have a USB-CDC standard... (For an interesting side story, search for "ftdigate". At some point FTDI decided they're fed up with copycats and released a new driver pack (through automatic Windows Update) that bricked counterfeit chips. This led to a lot of angry people and number of amusing situations, including someone jokingly submitting a patch to Linux to do the same.)

dfox 10y ago
FTDI is not and never was especially big in the USB-to-Serial market, and their chips tended to implement CDC in variously broken ways not to mention that their windows drivers are major PITA. What they are big in is market for "just add USB to this custom device" because they supply mostly NDA-free documentation since forever and produce various ready-made modules.
- kosma 10y ago
  Speaking out of curiosity, can you name some companies that provide NDA-laden USB-serial bridges? I'm painfully aware that it's hard to even know such products exist, especially when your production volume barely runs into hundreds or thousands and you're not established in the industry... and from my limited interactions with even the smaller players like SIMCom or SiRF, it takes an NDA just to talk to someone who can get you to sign a second NDA so you can see datasheets. Even if you have tons of money, giving it to such company in return for their product can be a challenge.
- ajford 10y ago
  They also put the legwork into getting their drivers into nearly everything, which makes them great to use in a custom or small batch product that has users that aren't technically oriented or wont be installed by a knowledgeable installer. If you already have a client side install (like a software package) that can bundle the drivers for you, it's not a big deal. And with the emergence of USB-CDC, it's not nearly as necessary as it used to be, since most modern OSes support that now.
  - kosma 10y ago
    And then killed that convenience by literally bricking counterfeit chips. Most users have no way of knowing whether or not their chips are genuine - and, what's worse, most electronics manufacturers have no way of knowing either, because entire supply chains are rotten. These days we have our trusted supplier of FTDI chips but we still check every single shipment. We already had one supplier send us fake chips and then flat out exclaim they did it because "no one would pay premium to get the real thing". We would.. but since they already cheated, we had to switch suppliers.
    - cartoonfoxes 10y ago
      Can you share which suppliers have you been burned by? We've seen counterfeit Prolific parts, but I'm not aware if we've had similar problems with FTDI hardware.
      - kosma 10y ago
        I don't remember, unfortunately, as I wasn't the one handling logistics, I just spec'd out parts we needed. If anything, here's a bit of advice: don't ever assume your supply chain is 100% clean - more than once we've been sold thousands of dumpster-grade components by a seemingly very reputable company. And don't ever assume suppliers won't try a bait-and-switch; we've been bitten by this more times than I can count.
        nitrogen 10y ago
        Can the bigger names like Digikey be relied on? Or do their suppliers sometimes give them bad parts, too?
        kosma 10y ago
        My point is that even if your supplier plays fair, the manufacturer may change the designs slightly and you either dont't get the PCN or fail to understand the implications of a change. We had a case once where a manufacturer added a minuscule hole on the case of a relay, presumably a pressure vent. Our devices are potted in resin in order to withstand rough conditions; in this particular case the resin penetrated the relay and blocked the contacts. We were halfway through a production run before anyone noticed. Needless to say, the run had to be scraped because there's no economical way to remove the resin once it's hardened. All for a saving of a few cents per relay. QA lesson: new batch of anything = tight quality control. When you go into manufacturing, it becomes a game of numbers. Manufacturing 10000 of whatever is statistically bound to generate some duds and failures. Speaking of prototype/hobbyist amounts: stick to the big suppliers and you should be safe, but if something feels fishy, don't be shy to order the part from a different supplier. Generally, identifying counterfeit parts in prototype amounts can be tricky without prior experience because your manual soldering skills will always be sub-par to a pick'n'placer; same goes for ESD precautions, accidental shocks applied to the circuit, passerby finger-pokers and your curious boss.
- duskwuff 10y ago
  > FTDI is not and never was especially big in the USB-to-Serial market ... whaaa? They're probably the biggest company in that market, closely followed by Prolific (PL2303). > ...and their chips tended to implement CDC in variously broken ways No, their chips implement a proprietary protocol. Not CDC at all.

tyingq 10y ago

There are several. FTDI, Prolific, WCH (CH340)...others as well. It's a mess at the moment, because there are unauthorized clones of both FTDI and Prolific, and both companies release drivers that purposefully don't work (or worse...brick them) on the clones. But, there's not really a way for the end buyer to know for sure they are buying the real thing.

cartoonfoxes 10y ago
I can confirm the FTDI drivers are total shit. We had to obtain hardware info under NDA, and write our own driver in order to get a reliable solution build using their parts.
Animats 10y ago
The SiLabs CP2102N is useful for serial to parallel. It will talk to Linux with the standard Linux serial driver, although you need a free Windows program from SiLabs if you want to reconfigure it. (This is needed only for unusual applications.) I use them because they'll go down to 45 baud for antique Teletype machines. They're popular for Arduino applications, and there are lots of cheap breakout boards with 0.100 pins for Arduino interfacing.
- i336_ 10y ago
  Interesting. I'd gotten the impression the FT232 was "rock solid" and filed them away as "they're good, use them", but that was before the bricking incident, and now I'm not really sure anymore. I guess on the surface the big thing I really like is device differentiation. Do CP2102Ns have unique serial numbers, or can that free utility burn in info I can use to differentiate? Going a bit deeper, can I bitbang with it?
  - Animats 10y ago
    You can set vendor ID, product ID, product string, serial string, release version, and max power requested. "Manufacturer string" is set to "Silicon Labs". After doing that, you can lock the device against further changes, if you want. This is all done via SiLabs "Simplicity Studio", which is a big IDE for their microcontrollers into which they wrapped up some of the device-specific tools for their simpler devices. Thus, you can force the host machine to demand a device-specific driver if you need to. By default, it appears to the OS as a USB to serial port device. Linux and Windows recognize it as such, without special drivers. Linux mounts it starting at /dev/usb0; Windows mounts it starting at COM3. No bit-banging, though; it doesn't have the hardware.