News for hackers. Stuff that matters.
Re:What Are the Tricky Parts of Embedded Development? by probablybanned on Tuesday February 21, 2017 @06:20PM (#13701159)
"Oh, so this feature is just totally broken, then? Cool."

Always read the errata, folks.

Re:What Are the Tricky Parts of Embedded Development? by ianhowson on Tuesday February 21, 2017 @07:54PM (#13701873)
"Also, we're not fixing the bugs, because customers have already built products that work around the bugs. Fixing the bugs would break their products."

Case in point: Microchip ENC28J60 (http://www.microchip.com/wwwproducts/en/en022889). Less than half of the advertised features actually work.

Re:What Are the Tricky Parts of Embedded Development? by bsder on Tuesday February 21, 2017 @11:01PM (#13702919)
Yeah, don't use their CAN interface IC's either. Exact same complaint.

In general, I now avoid Microchip like the plague. Which means I now avoid Atmel, Micrel, SMSC, etc. <cries>

Re:What Are the Tricky Parts of Embedded Development? by pryelluw on Wednesday February 22, 2017 @07:15AM (#13704909)
What chips do you use then?
Re:What Are the Tricky Parts of Embedded Development? by bsder on Wednesday February 22, 2017 @12:57PM (#13707502)
Generally I will go for a built-in CAN controller on a microcontroller. Those generally work fine.

Brands? A lot of TI and ST.

Re:What Are the Tricky Parts of Embedded Development? by vvanders on Tuesday February 21, 2017 @09:38PM (#13702459)
Or the cost of the tape-out outweighs the number of customers using the feature.
Re:What Are the Tricky Parts of Embedded Development? by ianhowson on Tuesday February 21, 2017 @09:42PM (#13702487)
Yup. The times when I've worked on small-quantity ASICs, tapeout cost and engineer availability have been the main reason that bugs weren't fixed.

The ENC28J60 did at least four more silicon revs after the errata were released, so... yay Microchip?