| ??? 04/19/05 02:01 Read: times Msg Score: +2 +2 Good Answer/Helpful |
#91880 - Catch 'em young ... Responding to: ???'s previous message |
... what I am saying is do not let a piece of code go with the thought "if something is wrong, it will come out in the test" To catch a bug before it bugs you, needs tremendrous experience and plenty of lateral thinking to analyse the myraid scenarios under which it will/can fail. And if you are so well endowed, the chances that you will write buggy code is also minimum. Its thus a vicious circle. When I read your comments, a recent incident came to mind: It was an endurance rig for a valve and one of the points in Quality Acceptance Procedure ( QAP ) was to run the rig for 1000 cycles non-stop. This meant a time frame of 8 hours roughly. Sure the rig was put to the test and passed. All happy. One month later, there was this complaint: after 9999 cycles, the next roll over was 00100000 on the eight digit counter!! While the problem was simple to solve, it still pricks your concsience that the client had to find it out. The problem was in modifying a six digit counter routine to eight digit. Sure it must have been checked by forcing roll over boundary values - which was not done. Testing complements good coding. Raghu |
| Topic | Author | Date |
| Bit Bang i2c | 01/01/70 00:00 | |
| You should | 01/01/70 00:00 | |
| Errr | 01/01/70 00:00 | |
| a slight problem | 01/01/70 00:00 | |
| I was | 01/01/70 00:00 | |
| Licence | 01/01/70 00:00 | |
| call it SMB | 01/01/70 00:00 | |
| maybe different | 01/01/70 00:00 | |
| Maybe! | 01/01/70 00:00 | |
| Correction! | 01/01/70 00:00 | |
| SETB SDA | 01/01/70 00:00 | |
| misunderstanding? | 01/01/70 00:00 | |
| re: | 01/01/70 00:00 | |
| Thanks to Atmel | 01/01/70 00:00 | |
| 'Stub' resistors missing? | 01/01/70 00:00 | |
| Pull ups ? | 01/01/70 00:00 | |
| Not those... | 01/01/70 00:00 | |
| Rp | 01/01/70 00:00 | |
| IIC is owned by philips why not go there | 01/01/70 00:00 | |
| You're right! | 01/01/70 00:00 | |
| Set port as input - NO SUCH THING! | 01/01/70 00:00 | |
| oh yes there is | 01/01/70 00:00 | |
| Yes, but not here? | 01/01/70 00:00 | |
| nowhere in the documentation | 01/01/70 00:00 | |
| one more reason not to rely on testing | 01/01/70 00:00 | |
| one more reason not to rely on testing | 01/01/70 00:00 | |
| Catch 'em young ... | 01/01/70 00:00 | |
Yes | 01/01/70 00:00 |