| ??? 06/19/10 04:45 Read: times |
#176762 - Sometimes the practical reality is of little consequence Responding to: ???'s previous message |
Justin Fontes said:
Erik Malund said:
I recently had a project where the limitations of an ARM with a RTOS in responding to multiple simultaneous events (all extermely time critical) led to 3 '51s instead of an ARM. Surely, one can use 3 ARMs just as well? The idea of of using multiple 8051/2s is not a new concept, and is largely put into practice everyday, however, a designer could potentially do the same with a 32-bit mcu and I have seen products that gloat about the sheer awesomeness of having multiple (3 in fact) 32 bit ARMs. The product is highly specialized in my field of practice. I guess depending on the complexity of the project a processor could be chosen? I do like the idea of being able to pick and choose the options by using multiple 8051/2s and increasing efficiency by using more than 1 chip. I'm not certain whether 3 32-bit ARM's could outperform 3 805x's. The answer hinges on how quickly the 32-bitter can isolate a specific bit or byte and act on its state. Again, however, it's not just the performance that determines the outcome. How easy is it to get 3 ARM's to play together. How easy is it to get 3 805x types to work together? It all requires thought ... thought which seldom goes into the MCU choice. Most of the time it's a hardware guy wanting to add the ARM to his CV or a software guy wanting to add the 805x to his ... I suppose you get the picture. Neither of these two care that much about whether the resulting design/firmware is optimal. The just care whether THEY got what THEY wanted. Of course, sometimes it's a successful outcome that they want, but that's the exception rather than the rule. RE |
