1 |
On Mon, 1 Dec 2008 15:40:11 -0500 |
2 |
"Joshua Murphy" <poisonbl@×××××.com> wrote: |
3 |
|
4 |
= |
5 |
> One thing that comes to mind... if one of those files is primarily |
6 |
> closer to the start of the disk than the other, and the drive tapers |
7 |
> down on speed as bad as some... well, that's one variable you can't |
8 |
> easily resolve. Whether that would influence enough to matter, |
9 |
> compared to seeks, is questionable though. |
10 |
> |
11 |
|
12 |
|
13 |
Yes. That has crossed my mind too, but I can't figure out if there's |
14 |
anything I alone can do about it. It would taint the results in a very bad |
15 |
way, because it is impossible to catch by increasing the number of |
16 |
repetitions. If my memory serves well its called "systematic error" in |
17 |
statistics. |
18 |
|
19 |
I'm open to suggestions. |
20 |
|
21 |
|
22 |
P.S. |
23 |
|
24 |
The way to eliminate the influence of this factor is to find |
25 |
many other people to make the test and share the results. |
26 |
|
27 |
|
28 |
|
29 |
-- |
30 |
Best regards, |
31 |
Daniel |