??? 02/12/07 22:12 Read: times |
#132673 - My two cents Responding to: ???'s previous message |
This doesn't sound very useful in my book, since the count of instruction cycles for a given bit of code relates directly to execution time only for code that has no loops or branches. In the more interesting (and typical?) case where the code does contain loops and branches, you need either a simulator or a lot of patience to predict how long it will take to run. I'd say it's much easier and much less error-prone to just add a couple of blips on an unused output pin, load the code onto a physical target, and measure the timing directly with an oscilloscope.
-- Russ |
Topic | Author | Date |
Disassembler feature poll | 01/01/70 00:00 | |
My two cents | 01/01/70 00:00 | |
My one cent | 01/01/70 00:00 | |
Code writer? | 01/01/70 00:00 | |
my opinion | 01/01/70 00:00 | |
An opposing opinion | 01/01/70 00:00 | |
"my customers' " and no source ???? | 01/01/70 00:00 | |
C-Source | 01/01/70 00:00 | |
Two more cents | 01/01/70 00:00 | |
then, most likely you have stolen it | 01/01/70 00:00 | |
I strongly disagree | 01/01/70 00:00 | |
How about Windows, Erik? | 01/01/70 00:00 | |
waddayamean | 01/01/70 00:00 | |
see below ... | 01/01/70 00:00 | |
It can be useful | 01/01/70 00:00 | |
Purpose | 01/01/70 00:00 | |
I'd find that useful. It won't ever hurt. | 01/01/70 00:00 | |
no banning | 01/01/70 00:00 | |
go for it | 01/01/70 00:00 | |
Cool! | 01/01/70 00:00 | |
TU Clausthal Germany | 01/01/70 00:00 | |
this feature was independently requested here :) | 01/01/70 00:00 | |
I second that | 01/01/70 00:00 | |
New options | 01/01/70 00:00 | |
thanks for adding these options!) \"==\"![]() | 01/01/70 00:00 | |
One purpose. | 01/01/70 00:00 |