From: Henning Brauer (lists-openbsdbsws.de)
Date: Wed Jun 02 2004 - 04:27:12 CDT

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

* Damien Miller <djmmindrot.org> [2004-06-02 10:50]:
> J Moore wrote:
>
> > On Sun, May 30, 2004 at 11:22:54AM +0200, the unit calling itself Christian Gut wrote:
> >
> >>>Any possibility that Soekris might be added to the list of supported
> >>>architectures anytime in the near future?
> >>
> >>already is. look for i386.
> >
> > Your response implies that one can simply take the i386 distribution,
> > pack it on a CF card, and go to town. Is that what you mean?
>
> Of course! (though this isn't an approach that would be recommended,
> due to the limitited r/w cycles that flash supports)

actually this isn't really true any more.
while the flash cells itself don't survive THAT many cycles, the
vendors do a defect-sector mapping nowadays like in modern harddisk,
and have a lot of reserve cells in their CF cards. as every write is
verified on CF, the mapping is completely transparent.
recent tests have show that the number of reserve cells is that high
with recent cards that it is far more likely that you kill the
card's connector before you have a problem with that...

--
http://2suck.net/hhwl.html - http://www.bsws.de/
Unix is very simple, but it takes a genius to understand the simplicity.
(Dennis Ritchie)

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]