??? 03/26/05 16:56 Read: times Msg Score: +1 +1 Informative |
#90453 - I Would Like to Point Out . . . Responding to: ???'s previous message |
I would like to point out something that could be thought of just simply from common sense.
If all I/O pins of any microcontroller came out of power ON or RESET as outputs it would generally not be possible to connect any other logic or circuit that had its own "output" to that microcontroller. If you tried to do so there would be the chance for huge amounts of current flow to occur as these two outputs that are possibly fighting to establish a logic level on the I/O port pin. Now why would any self respecting microcontroller manufacturer design a chip that did not allow you to safely connect another legitimate periheral device to her/his part???? In all the years that I have worked with microcontrollers and PI/O type chips I have encountered NONE that, by design intentionally came up from power on or RESET with the I/O pins set as outputs. The only exception to this that I have ever seen was one part I worked with (sorry cannot remember its type) had the selection of the port directions set via a "fuse setting". I can tell you that this part was short lived too because with fixed port direction it was not possible to interface the pins to things like the bi-directional I/Os of tri-state devices. I would now like to venture to guess that one of the reasons that the 8051 style of microcontroller has had such a long lived existence is due to the versatility of its quasi-bidirectional ports. Ports of this design only require one flip-flop register bit per port pin and so are efficient from the silicon chip area used. This was of extreme importance back at the time of the first microcontrollers because it made the chip smaller and less expensive!! It also required less addresses in the SFR address space for special configuration ports. Contrast that to the other mainstream microcontrollers of earlier times such as those from Motorola which required a DDR (data direction register) for each port or the Microchip parts which had a TRIS (tristate control register) for each port that was configurable. Even one historical PI/O chip I recall had to have an extra set on on-board internal registers that controlled the port directions. And in the desire to save silicon in those early days the lousey chip was designed so that all of the bits of port A and B were either all in or all out at one time!! Thank goodness the port C of the part could be split as half in and half out. BUT even that goofey part powered on with all its port pins set as INPUTS. Michael Karas |
Topic | Author | Date |
format of hard disk | 01/01/70 00:00 | |
Why not? | 01/01/70 00:00 | |
its the 1st step... one more ques. | 01/01/70 00:00 | |
All High. | 01/01/70 00:00 | |
ports are output when you start 8051 | 01/01/70 00:00 | |
Your statement is wrong | 01/01/70 00:00 | |
Quite right! | 01/01/70 00:00 | |
Wrong!!!. 8051 ports. | 01/01/70 00:00 | |
Input/output | 01/01/70 00:00 | |
8051 ports. | 01/01/70 00:00 | |
Which Book? | 01/01/70 00:00 | |
8051 Ports on Startup | 01/01/70 00:00 | |
8051 Ports. | 01/01/70 00:00 | |
The End![]() | 01/01/70 00:00 | |
I Would Like to Point Out . . . | 01/01/70 00:00 | |
Why Hard disk. | 01/01/70 00:00 | |
thats the req. | 01/01/70 00:00 | |
CF Better in all aspects. | 01/01/70 00:00 |