| ??? 02/14/12 15:00 Read: times |
#185936 - Where you sit determines what you see. Responding to: ???'s previous message |
Erik Malund said:
On the contrary, using a "classic" debug monitor - and getting to understand its requirements & limitations - would be a significant learning curve!
I'm not sure I can give you that last one, Andy. Now, I've used monitors for several decades, and have got to admit, I am comfortable with 'em You are arguing against a "learning curve" and, in doing so, recommend another "learning curve". That YOU are "comfortable with 'em" does not mean it is the right choice for a beginner. You're right, of course, but I maintain that it's just one learning curve versus two or more, and, in the case in which the learner is trying to learn the intimate details of the processor core, the one, which is easily handled with the monitor, is plenty. Now, if the goal is to approach all of the learner's programming problems from a high-level language, which has to go through a compiler that arranges low-level instructions in a way totally transparent to the user, well, it won't be as helpful. I would say that, if different, the learning curve for a monitor will be steeper than the learning curve for a "live" debugger, be it ICE, JTAG or whatever. I'd have to disagree. There aren't many monitor commands to learn, and finding one's way around the innards of the chip via JTAG is not so simple either. I will admit that one would, in any case, have to learn one's way around the innards of the chip, no matter how one gains access. The monitor, however, works with chips that provide no hardware debugging features, and, being "free", costs WAY less than an ICE. Now, Andy made it clear that this O/P said he was pretty familiar with the KEIL environment. How that mates up with his initial claim at being new to this area of endeavor puzzles me a bit. a "live debugger" does not affect the locations etc of your code and thus saves you the "learning curve" of how to make your thing work with a "monitor". Agreed, but it's little help with a chip that doesn't have debug support in hardware. So if you want to roam with the dinosaurs, go ahead, but please do not insist it is "easier to learn"
Erik That's where we'll have to agree to disagree. I find using a monitor very straightforward, probably because of my many years' experience with 'em. The result is that I know what to expect of 'em, and what to do in order to make 'em do what I want. You, on the other hand, have relied on hardware debugging capabilities for as long as I've followed your writings in this forum. Clearly, you know what it can do for you and how to exploit that. Naturally, you see things from the viewpoint of one who's been doing things this way for years. As I often say, "Where you sit determines what you see." I would point out, though, that neither approach provides much help with respect to the O/p's original problem, which was selecting and learning about a "small, modern" 805x chip available in through-hole packaging. The little ones lack enough code space to support development with the aid of a monitor, nor do they generally provide debug support of any sort. The newer SiLabs parts, however, do provide it. That might make the choice easier. RE |
