What is the history of FTDI? You’re the founder and CEO for 25 years, that is impressive! It would be interesting to go into some details around how FTDI started, what your motivation was for starting it, how it grew and what kind of challenges you had to overcome over its ~25 year history, etc.
Before FTDI, I ran a one-man consultancy designing PC motherboards and peripherals mainly for companies in Asia. Early designs were based on programmable logic ( PALS ) and discrete TTL, but then the focus changed to integrating much of this logic into “chipsets” which mopped up all the logic into a few ASIC devices. So, I took my existing skills and learned how to design in silicon as opposed to discrete logic. After designing a few chips for others, I decided to design and sell own branded chips and so FTDI was born.
In the early ( pre-USB ) days we designed and sold 286/386/486 PC chipsets but we were reliant on a customer base of just 2 or 3 customers including IBM at one point. This was very hard business for a small relatively unknown company of 5 people based in Scotland. When USB came along, I realised there would be a sea change in the peripheral market so I decided to move out of PC chipsets and into USB peripheral interface chips. After a false start ( USB keyboard, mice and joysticks ), we found our niche in designing legacy USB converter chips, like the USB UART bridge chips we are famous for today.
Our main challenge in the early days was lack of funding and consequentially lack of manpower resource which often resulted in time-to-market delays. The other challenge was building up a brand name as few people had heard about FTDI at that time. We were entirely self-funded and still are to this day.. It took 10 years before we finally achieved “critical mass” where we could afford to properly grow the company.
Can you describe the product lines at FTDI and goals for the company. What would you say your mission and cause is?
We are best known for our USB interface bridge solutions such as the popular FT232R USB-UART chips + drivers. These allow an engineer to interface technologies such as USB to a UART interface without having to understand the underlying technology, USB in this case and develop and support drivers for various platforms. This methodology works well – for instance a product developed 15 years ago using our FT232B chips can run on Windows 10 as can our more modern family – same drivers. More recently we’ve been looking at innovative TFT display interfacing with our EVE object orientated ( FT8xx ) family and are moving into mass production with our brand new FT9xx 32-bit MCU family which is capable of interfacing many different technologies together.
Our common mission in all these I would say is “Bridging Technologies” aimed at providing quality solutions for engineers (and Makers) to help bring their product development to market faster and with less ongoing technical support required.
How many people do you have? How many locations, how many products sold, etc?
Currently, we employ around 130 folks worldwide in R&D, technical support, sales, purchasing, logistics, finance and admin roles. We have 5 offices worldwide in Glasgow UK, Portland US, Taipei Taiwan, Shanghai China and Singapore. Singapore is not listed on our website as it’s an R&D centre only. We sell tens of millions of chips annually, mainly USB bridge products, though we hope to grow other areas as well.
To jump right to what everyone wants to know, what happened before and currently with the FTDI drivers from Windows update regarding counterfeit chips?
The problem first appeared some time back when we were sent some samples of a Chinese made “FTDI” USB-RS232 cable that seemed to be behaving in a way we couldn’t reproduce in our lab. On testing these cables, they seemed to work and install with our drivers, however the throughput was well below what we expected and usage was not very stable. We were puzzled at first, then suspicious. The markings on the”FT232RL” chip looked correct **but** on close scrutiny not exactly so and the date code did not match our records. Desoldering the chip and looking at the markings on the bottom of the chip confirmed our suspicions –it was marked as been Made In China! Contrary to rumours I’ve seen on the internet, FTDI neither fabricate nor assemble ANY of our chips in China. We had the chips de-capped and photographed and the die inside was totally different – seemed like a MCU programmed to emulate an FT232R, whilst the real device uses interlocked state machines – hard to design but gives superior real-time performance.
We seem to have caught it early, before it spread like a cancer so, an action plan was needed starting with containment and information gathering. The counterfeit chip was totally different design to the real FT232R and basically an imperfect copy, therefore it can be caught by our drivers in several ways which you’ll understand me keeping to myself. So, action number one was to detect counterfeit chips and stop them illegally using our drivers ( they steal our USB VID and PID in order to masquerade as an FTDI chip). This in no way affects the millions of genuine FTDI users and allowed us to analyse the situation a lot better.
Why do this instead of a notice on the web page or driver?
Genuine customers are not affected so it doesn’t help by causing a panic. We needed to analyse the situation and decide on a plan of action. Basically, what we discovered was that 90% of the problem were Arduino “bargain” copy/clone related, mainly sold on EBay, Alibaba, Amazon Marketplace by anonymous sellers. The rest was cheap “FTDI” USB RS232 cables sold likewise. I’m sure it occurs to most reasonable folks that a bargain bought in the internet version of a flea market may be cheap for a reason.
Do you see a lot of counterfeiting of your products?
Not really – it’s largely restricted to the FT232RL ( SSOP only ) and occasionally the older FT232BL.
Do you have an estimate of how many companies are using the FTDI trademark for counterfeits?
Just one or two only.
Do you have an estimate of how much business you’ve lost due to counterfeiting?
It’s really hard to put an estimate on this – I’d guess in the hundreds of thousands of dollars.
Is FTDI feeling more business/financial pressure due to counterfeiters or from competing companies?
We are ALWAYS under pressure from competitors who would love to “eat our breakfast” and we realise this. However, our reputation for providing robust USB bridge solutions help us maintain market share amongst genuine customers even though some of our competitors are large corporations. Financially, we are independent and debt-free, so we don’t worry too much on that front. Our main concern with counterfeit parts which are passed off as genuine FTDI chips is that their substandard performance and total disregard to quality will damage our hard-won reputation as people mistake these fakes for the genuine article. It’s our duty to protect ourselves and our much valued customers.
Do counterfeit chips ever make it into the ‘big disti’ network, e.g. arrow, digikey, mouser, element 14, RS etc. or is it purely gray market?
I can only speak for FTDI but in our case, the answer is definitely not. We’ve not had one single instance of a fake chip being sold by our listed sales distributor chains, most of whom we supply directly to avoid supply chain corruption. We don’t approve of the gray market as they sell fakes and real chips indiscriminately. A lot of these fakes are supplied through the infamous Shenzhen component market to the local Chinese manufacturers. Some are resold by anonymous sellers though EBay, Alibaba, BangGood etc. in small lots usually though China Post. You complain, send them cease to desist letters , but they disappear and re-appear shortly afterwards with a new identity.
How much support do you have to do for counterfeit products, do companies and customers call you with support problems due to counterfeits?
Yes, it causes a big headache for our support department – not from genuine FTDI customers of course. We’ve now made it easy to spot this by getting our latest drivers to flag any counterfeits by issuing a “Non Genuine Device” string which makes the issue obvious and saves time all round. The counterfeit device does not get harmed or re-programmed in any way – the drivers simply refuse to work with recognised counterfeit chips. Following our previous driver release, I’ve had many helpful suggestions on how the driver should respond when it finds a counterfeit chip. We can’t please everyone but the vast majority voted for this approach and that’s what we did.
What can chip companies do to stop counterfeiting? What are the issues with that?
Firstly, trademark your logo – not just the fancy one you put on your web site **but** the one you use on your chips too. We’ve worked very closely with the US customs teaching them how to identify counterfeit FTDI components and this has resulted in several shipments of fake ships to gray market re-sellers in the US being impounded and destroyed. Secondly, keep a tight control of your distribution chain and discourage them from supplying to gray market re-sellers.
Thirdly, should it happen, take action as soon as possible. Doing nothing makes the problem worse and harder to control.
Have you considered legal action against the counterfeiters?
Yes, we’ve appointed a legal firm in China to represent us there and we do our best. However, you should realise that we are dealing with a professional criminal gang here who counterfeit a lot more than just one or two chips. I’m pretty sure it’s the same gang that flooded the market with fake Prolific PL2303 chips a few years ago. They are very good at covering their tracks – in order to bring them to justice you would need an inside informer or an FBI style sting operation. Chinese law leaves a lot to be desired when it comes to counterfeiting.
The new FT231 series has the same essential functionality as the FT232, but is nearly half the cost – (the older chips costs more, how does that work?).
When you have a success like the venerable FT232R, don’t sit on your laurels and let your competiton eat away at your market share. Be your own competition! The FT232R uses a older large geometry 5V process – but the great thing is, it can drive out at full 5V levels. Our X series family of which the FT231X is just one, uses a much smaller geometry process, which combined with smaller packages reduces the cost of the X series substantially. The downside – it can only drive out at 3.3V max though it has 5V tolerant inputs. We recommend the X series for new designs, but despite this, many folks prefer to stick with the tried and trusted R series. Either way, I don’t mind – it’s good to offer customers the choice.
Anything you’d like to tell the maker community out there?
I started off as an electronic hobbyist, what you would term a maker today, as a teenager in the early 1970’s and it’s great to see this tradition revived by the Maker community today.
The best form of education is self-education where you lean by wanting to learn as opposed to being force-fed boring facts in a classroom. I would credit Arduino in particular as instrumental in reviving the interest in DIY electronics. They generously allows derivatives of the platform to be produced under an open-source hardware licence, though sadly a few rotten apples do not keep to the spirit of the agreement. When looking for a supplier, choose one that contributes to the community in the form of extensive tutorials, drivers and examples. Our good friends at Adafruit ( and Sparkfun too ) especially come to mind – buying from them helps reward their time and effort spent to improve our education. Lastly, if you’re tempted by an internet bargain from an anonymous supplier, well ok, but don’t be disappointed if it’s not what it seems.