Author: David Bisset Date: To: balloon Subject: RE: [Balloon] CF slots on Balloon3
>Right, ballooners & potential Ballooners... >Compact Flash slot(s) are a bit contentious. I'm currently hoping (but not >guaranteeing) to fit 2 slots to Balloon3.
Interesting, where are you going to put the second one (roughly...)
I guess if we don't like where it is or can't fit our existing loon
cards round it we will be able to "nofit" it? (i.e. I assume it's not a
double slot, or that if it is there is a single slot fit option that is
one to one compatible).
>However, there is (almost) no
>chance of supporting 5V cards, only 3V.
Not an issue since the CF specification says that *all* CF cards must be
able to work at both 3v and 5v. And I've never seen otherwise.
>Since I posess exactly two CF
>cards, I don't know how much of a limitation the 3V-only will be. None.
>Has anyone with a microdrive any information on its requirements?
I've just checked an out-of-date spec sheet (circa 1999) and it says it
will work on 3v or 5v, so I hardly think they will have got worse in the
last 5 years. But then IBM....mmmm....
>Same goes
>for other CF cards - wireless, Bluetooth, Ethernet, etc. (Similarly,
>sketches of the bulges on the end of CF cards - antennas, connectors, etc) >that people would be hoping to use, would be interesting, to reduce the >likelyhood of mechanical clashes between CF cards (which plague my laptop >PCMCIA slots...)
I won't bore the list with this but I'll send you photo's (against a cm
grid) of my small collection of CF cards. (Mainly WLAN cards). The worst
I've found to date is the D-Link DCF-660W. The antenna bulge is massive,
and it incidentally has the worst WLAN performance on record according
to my tests.
(Indeed there is a rough *inverse* correlation between bulge size and
performance) (Best to date the Pretec card, still trying to find a
working driver combination for the symbol24 card which might be better)
>The second limitation is that I don't plan to support hot-swapping for >Compact Flash. I expect the CF slots to be filled at the factory, rather >than ever being exposed to end users.
Personally this is not a problem, since this was the case on 2.xx and
it's damn difficult to design enclosures to allow the user to get their
fingers round one; I see no need to complicate things.
>With PXA27x, USB should be able to
>support a huge range of random plugged-in toys, and allow them to be
>hot-swapped on a whim.
Agreed.
>The support to make CF (reliably) hot-swappable will cost, in pins,
>board-space, PCB complexity and design time. If anyone has a usage model >for Balloon which needs hot-plug on the CF slots, can they please make >their argument now, to try to convince me?
Sounds like a tough challenge :-)
>It's also of note that I have a fearsome amount of non-Balloon work lined >up for the next 6 months. (have to pay the mortgage somehow, and Balloon's >just infill work). The plan is still to relase the prototype design for >Balloon3 by the end of the year, but I would be very interested to hear if >anyone is relying on that date.
Well I'm trying not to rely on that date, since even once the first
design is delivered it'll be at least three months before we are likely
to see production runs. (On the assumption that there are bugs in the
first rev, and Nick will take a while (even if only a week or two) to
port over to the real board, and it'll take similar time to write the
CPLD, build manufacturing jigs, get the paste mask right, get hold of
chips in volume etc. etc.)
However if it's likely to be March next year before the first one rolls
out then I'd like to know as early as possible so I can think up some
really good excuses as to why I can't deliver a faster version of Hydra
by next summer....
(text/plain)