RE: Cypress Semiconductor PSOC 4, The Engineers Magic Wand!

FredM — Fri, 05 Jul 2013 02:07:38 GMT

Hi Dab,

Yes - The PSoC devices are in a class of their own - When I first encountered them oh, 10 years ago (?) I was so taken by them that I designed them into many projects - even using multiple parts communicating with each other and eliminating most of the analogue in complex sensor / motor commutation designs - and became a certified CYPros in the process..

I loved (and still love) the PSoC 1 family - Most of the time in the applications I was building, the slow 8 bit processor wasnt too much of a bother..

Cypress IMO made one huge mistake - this was their "Express" venture - The wonderful PSoC Designer software became bloated in 2005 with this "Wot - no code?" express rubbish, and the compilers became almost unusable, and my clients left Cypress in droves.. And eventually I stopped promoting PSoC and just quietly used the parts in projects where I was the only engineer working on a project - it was just too embarrasing to get other engineers involved with Cypress product! - It took Cypress until about 2010 before they realised that the unusable development software was causing people like me to leave - and they withdrew Express and restored PSoC Designer to a functional state... Sadly, by this time some major players had given up - Hi-Tech, who produced the first good compiler to work with the PSoC was acquired by Microchip - I met Hi-Tech's director prior to this, and his frustration with Cypress was palpable.

I think Cypress is now back on track - they got their compiler sorted out, brought out their long awaited PSoC 3 and 5, and have now brought out their PSoC 4 which has a really competetive price tag.. All a little bit too late for me, alas - My main PSoC client went bust, and another switched to Atmel. I ended up using Atmel parts in several designs (In fact, I think the Q-Touch technology might be superior to Cap-Sense) - But PSoC 5 in particular has power in terms of user defined modules and analogue, which I think puts PSoC in the lead again..

But I really do fear that some STUPID IDIOT "boss guy" at Cypress will risk everything on some hobbyist idea like "Express" - And drag people like me into a pit again - I invested everything in PSoC, spent a fortune, built a large client base, promoted PSoC and Cypress - And they went and screwed it all up....... Oh, I spoke to Many Cypress employees and "insiders" at the time who were equally upset about the new direction, and a major project I was working on for / with Cypress (I was developing simulation models of the analogue blocks, and discussing incorporating these into a major EDA program to enable co-simulation of both the hardware and firmware) was abandoned in favour of Express....

As long as Cypress keeps their eye on the real market, and understands that its the "old school" engineers who they need to keep on-board, they should be ok. Its innovation that generates tomorrows products - There often arent any "ready made" UM's one can plug in to an innovative product, they need to be invented - Focus on allowing engineers to create and be engineers not Lego constructors and , even though slow, they will produce growth.... Appeal to the 'plug-and-play' "Wot?-No Code?" bunch, and you go fast.. But you go nowhere.

I hope Cypress has learned this lesson - I must say that it looks like they have!

Fred.