Am 25.04.2011 um 21:08 schrieb Brian Lloyd: > There seems to be a lot of interest in the I2C interface on the Flexwire > bus. Certainly it is not difficult to interface to the I2C bus to transfer > data but I just don't see the advantage or desirability of using the I2C > interface over a USB interface on the computer.
Yes, based on today's features and functionality in the DICE-based custom firmware for FRS that is actually true. > My reasoning is as follows: > > 1. The source of all band and other control data is the computer, not the > RF hardware. You have an extra layer of indirection which provides yet > another place where bugs can creep in. For me, this works the other way round: Because you have to use the PC as the relay / sender of band data you have an extra layer which might be the source for trouble. Iff the band data were created and sent to the FlexWire port without any PC interaction this will reduce the nr o layers. > 2. There is no standard for data on the Flexwire bus. What comes out > depends on what Flex does with the firmware in the radio. If it makes the > radio work better, you can bet the messages on the flexwire bus will change. Which is fine. Because it's still a bit sequence being transferred via the I2C bus. And it requires software (firmware, in this case) to decode that bit sequence. So if FRS changes the semantics and/or the format, well, let's adapt the connected device's firmware. > 3. (I believe the following is true but I am not 100% certain so feel > free to correct me if you know.) Currently any control messages that come > out of the I2C interface are actually generated inside PowerSDR, not by the > firmware in the radio's hardware. When you change bands and PowerSDR sends > the bandswitching data to the RF hardware to switch filters and the like, > that does not produce a corresponding output from the Flexwire interface. > PowerSDR actually has to send a SECOND set of commands to output band data > on the Flexwire interface. Hm - That would be true, yes, if PowerSDR were programmed to do so. > Given the above, it seems clear to me that the simplest and most reliable > way to get control information for external hardware into and out of > PowerSDR is through CAT on a serial port, not through the Flexwire I2C bus. > If you are going to connect up a wire, does it matter to which box it > connects? Oh yes. It does. Because very expensive hardware depends on, well, not so good software (I am talking about the operating system). As another poster said it will be much more relieving to know that radio, antenna selector, amplifier, etc. are "on the same page" re band selection. Why not have a user definable bit sequence that is put out on the FlexWire bus on every band exchange? Why not even have different events definable (mode change, band change, power on, power down, ATU active, SWR too high, ...) ? All this in PowerSDR, all this definable even via CAT commands ? ZCAT PWRSDR DEFINE-EVENT BAND-CHANGE(80m, FLEXWIRE-SEND(2)) Meaning: Please, PowerSDR, create an event of category band change IN THE RADIO, for a change to 80m. And if that event occurs, then send the bit sequence for a value of 2 onto the FlexWire bus (that is, the radio does this, because this is defined in the Flex-xy00 radio, and not in PowerSDR). Bonus: All this works without PowerSDR even being running. Just turn on your Flex-xy00 and your good to go. Because that would programmed in an EEPROM and is available right at power on. It's software, after all. 73, Frank DG1SBG _______________________________________________ Flexedge mailing list [email protected] http://mail.flex-radio.biz/mailman/listinfo/flexedge_flex-radio.biz This is the FlexRadio Systems e-mail Reflector called FlexEdge. It is used for posting topics related to SDR software development and experimentalist who are using beta versions of the software.
