| ??? 04/25/07 12:05 Read: times |
#137913 - not really Responding to: ???'s previous message |
that will take a lot of processor time... wouldn't it ? Is that ok in terms of performance ?
not really, and the majority of time spent is asynchronous It would, of course, kill you, if you polled, but doing the sensible thing and using interrupts, the overhead is very nominal. The beauty of the scheme is that the master is not forced to immediately respond to the data collected by the ISR and thus, while some time is, indeed, spent, there is no synchronization problems. For instance, I have a master running heavy tasks and also controlling a 115k M/S bus. Erik |
| Topic | Author | Date |
| multiple 8052 communication with PC | 01/01/70 00:00 | |
| RS485 | 01/01/70 00:00 | |
| More COM: Ports | 01/01/70 00:00 | |
| Further additional COM: ports | 01/01/70 00:00 | |
| multiple 8052 communication with PC | 01/01/70 00:00 | |
| Simple solution | 01/01/70 00:00 | |
| Recipe for a short circuit? | 01/01/70 00:00 | |
| Not on a standard 8051? | 01/01/70 00:00 | |
| whether that is a problem is for the OP to determi | 01/01/70 00:00 | |
| protocol ?? | 01/01/70 00:00 | |
| yes - no - see above | 01/01/70 00:00 | |
| master slave PC performance | 01/01/70 00:00 | |
| multi-GHz PC can get a lot done in a few ms! | 01/01/70 00:00 | |
| Slave comms controller? | 01/01/70 00:00 | |
| expanding on th above | 01/01/70 00:00 | |
| For just two devices | 01/01/70 00:00 | |
| been there, done that | 01/01/70 00:00 | |
In this case... | 01/01/70 00:00 | |
| Interrupt Request? | 01/01/70 00:00 | |
| not really | 01/01/70 00:00 | |
| Not on a PC | 01/01/70 00:00 | |
| Not that it can't be done | 01/01/70 00:00 | |
| Another recipe is a ring!!! | 01/01/70 00:00 | |
| sure, if you have all the time in the world | 01/01/70 00:00 |