kicad-developers team mailing list archive

[Lorenzo Marcantonio <l.marcantonio@xxxxxxxxxxxx>]

On Mon, Jun 30, 2014 at 03:28:32PM -0700, Cirilo Bernardo wrote:
> Oh yes, the OMAP ones can be worse. A huge section of the chip errata have to do with USB issues, and I'm not happy with the length of TI's chip errata either. Some faults have persisted across 2 chip revisions. Sometimes I get the impression that TI has no interest in creating a working processor.

OMAP == consumer, don't be so surprised... another TI errata which is
absurdly fun is for an NFC controller/transceiver (don't remember the
part number). It can be used with parallel, 3-wire SPI or 4-wire SPI.
Errata #1: 3-wire SPI doesn't work. Workaround: Use 4-wire SPI :D:D

Not that other manufacturers are better; a dsPIC33 errata substantially
says: don't ever *hope* to get the part programmed over JTAG, we left
that part out (were they out of silicon area?!). And don't care to trap
DMA conflicts either since they trigger whenever they want (if they
trigger at all).

Seems that the whole industry is plagued by too short development
cycles...

-- 
Lorenzo Marcantonio
Logos Srl