??? 02/25/06 00:45 Read: times |
#110739 - It's always a tradeoff ... Responding to: ???'s previous message |
Even when the volume is low, cost-effectiveness is a factor, though perhaps not for those who are spending someone else's money. Sure, there's a tradeoff between "getting 'er done" and spending a "fortune," which can mean different things to different people. If I'm building something for an underfunded research fellow at the university I may not have the liberty of spending hundreds of dollars on components, even though it might reduce the effort. When I donate my time, I still don't donate materials.
Consequently, I'm not willing to go out and buy the most generously equipped "evaluation" or "development" boards for every purpose. The main reason, however, is that, in providing "features," they often rob you of options. If they use the serial channel for their monitor console function, then I can't use the serial channel in my target application. That's the reason I don't like most "evaluation" boards. For example, the board that Maxim wants us to use for evaluation of their DS89C4x0 MCU's has no feature enabling use of their much-touted page modes. That's the reason I built my own. My board presumably allows use of all the features, peripherals, etc, in the device, though it's always possible I might have missed something. I'm using a monitor that uses the second serial port as the async link to the console. That way, if I need, I can use the first channel in my target app. If I use a different version of it, I can use an external SCC2691 and a 75155, the latter being a one-bit RS232-type receiver and transmitter. It works fine with +/- 5 volts, which, in this case, is readily available. That way I really don't ever have to choose between having serial comm with the console or other equipment or using the serial port pins for parallel port bits, for synchronous use, or for async. The SCC2691 has its own crystal, hence doesn't have any impact on what the MCU does. RE |