Ah. I was going by the addressing conventions in the documentation I here: http://www.milkymist.org/socdoc/dmx.pdf (which seems to be the same as the documentation for the core on Github: https://github.com/milkymist/milkymist/blob/master/cores/dmx/doc/dmx.tex ).
I guess that's an older version of the documentation; it states that the 512 DMX addresses are mapped 0x000, 0x004, 0x008 ... 0x7fc. (Unless I'm misreading it, which is certainly a possibility.) Thansk, Chris --- Chris Dzombak http://chris.dzombak.name On Thu, Apr 21, 2011 at 5:50 PM, Sebastien Bourdeauducq <[email protected]> wrote: > Hi, > > On Thu, 2011-04-21 at 17:44 -0400, Chris Dzombak wrote: >> When I write any value to an address between 0x200 and 0x3fc in the >> DMX TX core (via the CSR bus), the core completely stops transmitting. >> Once the core goes into this state, I can still read and write >> addresses below 0x200, but whenever I read any address above 0x200, I >> get a 0, regardless of what I write. > > There are only 512 DMX channels (as per the DMX specification) which are > mapped 0 to 0x1ff. Address 0x200 is used to enable/disable "chained" > mode, that is, transparently forwarding what is received on the RX port. > In your case, with the RX port probably unconnected, enabling "chained" > mode effectively stops transmission. Addresses strictly above 0x200 are > invalid. > >> Additionally, even while the core seems to be transmitting properly, >> any reads from addresses 0x400 - 0x7fc in the core return 0, >> regardless of what I write. > > What would you expect it to return instead? > > S. > > > _______________________________________________ > http://lists.milkymist.org/listinfo.cgi/devel-milkymist.org > IRC: #milkymist@Freenode > Twitter: www.twitter.com/milkymistvj > Ideas? http://milkymist.uservoice.com > _______________________________________________ http://lists.milkymist.org/listinfo.cgi/devel-milkymist.org IRC: #milkymist@Freenode Twitter: www.twitter.com/milkymistvj Ideas? http://milkymist.uservoice.com
