Why STmicro not fixes errata in new revisions ?

[weird]

Posted on June 13, 2011 at 18:01

One really sad and annoying issue regarding otherwise very good ST MCUs, are ST behaviour/policy of not fixing errata in new chip revisions, inheritance of IP/peripheral blocks including erratic functions and NOT fixing it when new line with completely new docs and specs arrives.

I expected that the new STM32F2xx will be free of old bugs, peripheral issues fixed and when I look at fresh STM32F2xx errata and compare with STM32F1xx, most issues remains... I understand that new cycle of revision fix/test/mask set/validate is expensive and not economical for niche chips, but widely used and popular line like STM32 deserve for better policy, at least at generations boundary, when new products are launched. Competitive products from Texas Instruments and Atmel, go through many revisions through lifetime and most errata are fixed even within the same generation. Maintaining bugs cross generations is a total nonsense for my mind, when new features, specs, mask set, documentation etc. are defined and no bug-compatibility for large scale adopters is necessary. I'm asking all forum members and STM32 fans, please vote and push ST for change of this policy. #stm32-errata-policy

[infoinfo989]

Posted on June 13, 2011 at 22:09

What specific bugs are you talking about? I was reading the F2xx errata just last week, and it's a pretty short read.

[weird]

Posted on June 16, 2011 at 20:23

I'm probably oversensitive, but 28 page document, describing 31 issues in 2. gen product (better 3. gen, counting F105,7 as 2.gen), based mostly on same IP blocks, even after ES versions, is by my opinion undesirable burden which should be avoided and removed when suitable (at least in new product).

I agree that none of currently listed for F2xx are too serious, but widely spread necessity of product design by errata, is generally bad...