[Arm-netbook] EOMA68 A20 DDR3 SA15 (2GB addressing), finance records in bugtracker

[Luke Kenneth Casson Leighton]

http://bugs.rhombus-tech.net/index_by_target.html

i'm keeping a record of the amounts in and out.  thinkpenguin (my
sponsor) is handling the money... which if you've been following
(running) the assess_campaign.py program is pretty much gone /
accounted for.  there are a number of reasons why the amounts are
being published, not least is so that people can see exactly where the
money's going, but also to ensure that it is absolutely clear that
this is *NOT* being run as a "profit maximising business".  over the
past five years i've put at least $40,000 of my own personal funds
into this; over the past two years thinkpenguin has sponsored the
project with a further $60,000 approximately (so far).

one of the other reasons for making this clear statement on-list is
because now that there exists a wikipedia page, we're getting
well-meaning people try and contribute to it, and, so far, there's a
100% track record of the contributions being factually incorrect,
misleading and just plain wrong.  that's led to me having to step in
to correct them, which has led to a request that i make a (false)
declaration that there exists a "conflict" between the two roles
{wikipedia editor | guardian and editor} of the EOMA68 open Creative
Commons Licensed {wikipedia page | Standard} on the wikimedia site
{wikipedia.org | elinux.org}.  it's yet to be demonstrated that such a
"conflict" exists.  i absolutely flat-out refuse to make a publicly
auditable FALSE declaration that a conflict exists when it does not:
that brings both the EOMA68 standard *and* wikipedia into disrepute.

interesting situation - one that wikipedia's clearly never had to deal
with before.



http://rhombus-tech.net/allwinner_a10/news/

the DDR3 RAM address line that needed to be added has been done...
it's extremely tight and has been quite disruptive.  i'll need to do
some close inspection of what the auto-router's altered during the
many push-and-shove events required to make room for the extra address
line.  the problem is that if it wasn't planned for, the available
space is used up for VIAs and components that literally makes it
impossible to get the extra track into the available space without
considerable knock-on modifications.

at the time that the original A10 board was done, we actually didn't
*know* if it was possible to do 4x DDR3x8 2GB RAM layout: it was only
tom cubie's team who discovered the extra address line and used it
with the Cubietruck.

l.


---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68