| ??? 08/04/07 22:41 Read: times |
#142772 - table is not necessary Responding to: ???'s previous message |
Just a note, I've personally never met a case when in a '51 I would really need the table-driven CRC - the algorithmic method might be slightly slower but not significantly, and the tasks a '51 usually fulfills usually allow for a few extra microseconds. I see the nibble method for the first time now but I suspect on '51 it is slower than the usual algorithmic method (longer data word processors are different of course).
It of course does not really matter if you have kilobytes of flash spare, but starts to be interesting if you have to squeeze in the whole stuff within a tight space to be able to use the sub-$ micros. I am speaking of asm implementation, of course - this is so trivial that I see no point to fall back to C or anything similar. YMMV. JW |
| Topic | Author | Date |
| How to create a crc table? | 01/01/70 00:00 | |
| simple | 01/01/70 00:00 | |
| RE: simple | 01/01/70 00:00 | |
| I said it\'s simple | 01/01/70 00:00 | |
| Thank you | 01/01/70 00:00 | |
| Protocols | 01/01/70 00:00 | |
| examples | 01/01/70 00:00 | |
| Runtime Computed Table?? | 01/01/70 00:00 | |
| RE: Runtime...? | 01/01/70 00:00 | |
| table is not necessary | 01/01/70 00:00 | |
| Benchmark | 01/01/70 00:00 | |
| I did benchmarks as such on AVR | 01/01/70 00:00 | |
| Similar findings | 01/01/70 00:00 | |
| no beavers that I've tended to!! | 01/01/70 00:00 | |
| not bitwise.... | 01/01/70 00:00 | |
| Better is a strong word | 01/01/70 00:00 | |
| Also comment about SMBus PEC Code... | 01/01/70 00:00 | |
| The Table | 01/01/70 00:00 | |
| The Nibble Table for Poly 107 | 01/01/70 00:00 | |
| The formula | 01/01/70 00:00 | |
| Have you seen this? | 01/01/70 00:00 | |
| Now I have | 01/01/70 00:00 | |
Yes, that is a Key Document! | 01/01/70 00:00 |