|
04.28.2008, 12:31 AM
I actually wanted to ask you guys for suggestions for making my testing better. It's pretty unscientific so far.
I like the fact that my tests are "real world" in an actual car on the ground, but that makes them hard to repeat pack to pack. I think I should do multiple runs per pack and test them within their first few cycles, then again like around 20 cycles like VintageMA suggested.
How long do you think I should run each test? I think that 10 minutes should be enough at 10 samples per second. How many times should I repeat each test per pack? An average of three sounds about right to me.
If Austin is willing to send me a pack for this, I definitely want to make the testing worthwhile. I'll probably take video of the testing as well so people can see the set-up and what-not.
Thoughts? Suggestions?
|