[hpsdr] "FPGA Floor Planning"

[Scott Traurig]

How is it that the free version of Quartus is sufficient when just recently
both the Angelia and Orion firmware updates required multiple, rapid
iterations and still apparently either don't work at all, don't work for
everyone, or work but have to trade-off previous functionality for new
functionality?

Given the current complexity of the firmware (which, by the way, only
underscores the really tremendous amount of phenomenal work that Joe and
Phil have completed over the years), timing closure has had to become more
and more difficult with the free tool. This is even more of an issue with
the new firmware protocol where there is more data, moving faster, than
ever before.

Another factor that is driving the need to go to a more capable development
environment is the fact that there is no lab where a firmware build can be
fully validated on a range of Hermes, Angelia or Orion boards, over
temperature, before release. This makes achieving tight timing about the
nominal range even more important such that builds that work for some and
not others are less likely.

Here's another suggestion: the firmware source code remains open source, of
course. And those who wish to develop on the free version of Quartus can
still do so. But it would seem that Apache Labs might want to obtain the
standard version of Quartus and someone willing and able to use it to
polish up the timing of a given build such that it will reliably work on
all Apache Labs radios under most conditions.

73,

Scott/w-u-2-o
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20170210/74b2d6d9/attachment.htm>