| ??? 08/25/05 12:10 Read: times |
#100064 - why assembler Responding to: ???'s previous message |
Ghulam Mustafa said:
He told me that the hex file that is generated using C is quite larger than that of assembly programing. Andy replied: These days, memory is cheap - and you can easily get 8051 derivatives with 64K (and more!) of on-chip flash. So, even if the above were true, does it really matter? I used to write some assembly for the sake of space - no more, using a <64k chip and squezze the code to save a buck is counterproductive. What assembly I write these days is solely for the sake of speed and thus often take more space than the C. In my "world" all apps fit in the 64k code space (typically 30-50k) and one that does not is now split between 3 processors since running it on one would kill the speed. As you seem so intent on floating point, minimising code size can't be that much of an issue for you, can it...? </i Oh yes, everyone want to eat the cake and have it too. Erik |
| Topic | Author | Date |
| Mathematics of Fractional # in Intel ass | 01/01/70 00:00 | |
| Search | 01/01/70 00:00 | |
| Actual task | 01/01/70 00:00 | |
| Lookup table? | 01/01/70 00:00 | |
| Scale it | 01/01/70 00:00 | |
| echo........... | 01/01/70 00:00 | |
| sounds like fun, but it will take you ma | 01/01/70 00:00 | |
| Using C | 01/01/70 00:00 | |
| C | 01/01/70 00:00 | |
| Space occupied | 01/01/70 00:00 | |
| That old chestnut! | 01/01/70 00:00 | |
| Example | 01/01/70 00:00 | |
| Does it really matter, anyhow? | 01/01/70 00:00 | |
why assembler | 01/01/70 00:00 | |
| If actual fraction numbers | 01/01/70 00:00 | |
| And decimal number, | 01/01/70 00:00 |