| cladamw | wpwrak, did you try to let lables of differential signals as from something_[P:N] to [p:n]_something ? | 03:00 |
|---|---|---|
| cladamw | wpwrak, the ETH_INT_N, i just didn't carefully see it, now surely changed to nETH_INT | 03:01 |
| wpwrak | (foo_P/N -> P/N_foo) yes, i saw that. i think the P/N is different from the FOO_N vs. nFOO for negation. the nFOO is a characteristic of the information. the _P and _N is a way how that information is translated into signals. | 03:15 |
| wpwrak | e.g., you could conceivably have an nINT signal that passes through a conversion to differential and then becomes nINT_P/nINT_N .. and gets converted back to nINT at the end of the differential line | 03:16 |
| wpwrak | so i'd keep the foo_P/foo_N notation, even if it may seem somewhat inconsistent | 03:17 |
| wpwrak | (of course, also the differential example isn't bullet-proof) e.g., you could have an inverter that turns nINT into INT, and everybody would agree that the two signals should be called nINT and INT, not nINT and nINT_N or such. | 03:18 |
| wpwrak | so it's more a question of gut feeling and common choice. at least differential signals have only a few choices, basically P/N and +/-. | 03:19 |
| wpwrak | with inversion, we have a lot more. which also makes it easier to pick a relatively uncommon one ;-) (as long as it works nicely) | 03:20 |
| cladamw | so labels for differential signals expressed as foo_P/foo_N is more easier to understand its gut. | 03:58 |
| wpwrak | hehe ;-) | 04:08 |
| isa | :D | 05:25 |
| GitHub169 | [board-m1] adamwang pushed 1 new commit to master: https://github.com/milkymist/board-m1/commit/a9894324add88eb7bbb89243c4e4f4bf4583c1ce | 07:13 |
| GitHub169 | [board-m1/master] followed by: http://lists.milkymist.org/pipermail/devel-milkymist.org/2012-May/003032.html - Adam Wang | 07:13 |
| --- Fri May 18 2012 | 00:00 | |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!