| ??? 09/26/09 14:38 Read: times |
#169165 - Investigation of "prior art" is what will solve the problem Responding to: ???'s previous message |
Andy Neil said:
Richard Erlacher said:
I'd not bet on the notion that 'C' fixes inability to do the job, Per. The advantage of doing it in 'C' (or similar) is that it lets you concentrate on the higher level of "what" you need to do, without getting bogged-down in the detail of "how" to send each individual character, etc... I wouldn't deny that there can be advantages to HLL. However, one still has to understand the underlying operations, in the case of 'C', normally performed by initialization code supplied with the compiler, and often written in ASM anyway. Usually the individual functions actually require as much typing in 'C' as in ASM. It's a bit like doing your high-level design in pseudo-code - except that you're actually using a real programming language, and you can actually run it. Exactly! It offers a level of abstraction ... but at a cost. First of all the resulting code is normally bigger. Secondly, it's slower. Now, in many cases it doesn't matter, since initialization, and calls to routines for I/O don't happen often, but it can be a problem. And, as Per says, the vast majority of it is entirely platform-independent: it's just stuff like building strings, sending strings, receiving strings, picking-out tokens from strings, etc - which can all be very easily written entirely generically in 'C' (or whatever) and tried and tested on any platform. Sadly, it's not entirely platform-independent, and that's why I'm pointing this out. One must understand what's going on. Far too many people believe that one needn't learn the individual MCU before proceeding. That's where it goes off the rails. It's only in the detail of actually transmitting & receiving the characters that it needs to be platform-specific at all.
So, on the contrary, I think I'd say that if you can't even manage to do it at the "high level" in 'C' (or whatever), then you'll have no hope if getting down to doing it in detail in assembler!
Again, only thing the modem cares about is that you send it valid AT commands; it neither knows nor cares what programming language - if any - was used! So, before you can do any programming in any language, you need to understand what AT Commands you need to send, and what responses to expect - and that is all entirely independent of any programming language or computing platform. I can't disagree on this point. The original question seems to be the format and content of the GSM modem's communication, i.e. the AT-command structure. I think GOOGLE would be a good place for the O/P to start. The ASM/HLL debate is completely separate. I submit that, if you don't know your MCU's assembly language, then you don't know the MCU you're using. Certainly when it comes to small systems like the 8051. What's really at the root of the question is the understanding of the task at hand. It matters not whether it's to be resolvedin HLL or ASM. In either case, the O/P is required to know what the format and content of the communication between the modem and the MCU will be. Investigation of "prior art" (via GOOGLE) is what will solve the problem. RE |
| Topic | Author | Date |
| Interface between GSM modem & 8051/52 | 01/01/70 00:00 | |
| Use the Search button... | 01/01/70 00:00 | |
| How about table-lookup? | 01/01/70 00:00 | |
| Cross-posting | 01/01/70 00:00 | |
| That borders on nonsense! | 01/01/70 00:00 | |
| Disagree (to a point) | 01/01/70 00:00 | |
| 'C' Vs 'Asm' "Please ! Not Again!!!" | 01/01/70 00:00 | |
| that will happen at every opportunity Richard sees | 01/01/70 00:00 | |
| I think Richard's point... | 01/01/70 00:00 | |
| It's not the language ... it's the approach | 01/01/70 00:00 | |
| The approach | 01/01/70 00:00 | |
That's bottom-up, not top-down. | 01/01/70 00:00 | |
| Investigation of "prior art" is what will solve the problem | 01/01/70 00:00 | |
| how .asm code will work to intrface with AT commands | 01/01/70 00:00 |