Michael wrote:
>Now something really strange. Page load time of ctk.ceskenoviny.cz when
>Arachne was launched entirely from RAM disk, is constantly 8 sec, while
>on drive C: it is 8 sec only before pages are cached - and then it is
>7 sec (!) (second reload from cache). On drive E:, it won't get under
>11 sec....
That doesn't make sense at all, weird.
>So what can be wrong with my drive E: ? Maybe the disk surface is too
>tired from storing my Arachne home directory for nearly 3 years ? ;-)
C: is at one "end" of the disk and E: is on the other "end" (I imagine).
Isn't a hard drive faster closer to the center point (of the physical disk)
then out on an end of the physical disk? (It's true for most CD-players).
Now I know that there are several physical disks in a HD but this can still
make a diffrence.
>And how it is possible, that disk C: is faster, than my RAM disk ? (?)
Perhaps page size vs. cluster size?
>Is size of clusters really so important ?
Perhaps if you use the same cluster size in FAT as the drive has physically
(which you set on older HDs when low formatting them - not something you
should do on newer HDs however) things will go faster? (More is read to the
HDs cache?)
>(I am running DR-DOS 7.01...)
You could always upgrade ;)
And from the other mail you recently sent:
>I am going to compare load times with RAMdisk only installation, but
>I am surprised how big the speed differences can be even on the same
>PC, same operating system and same hard drive.... maybe this may be the
>reason why some users are enthusiastic about speed, while others
>complain ?
Perhaps, but IMHO it's more likely because we have diffrent hardware. A new
computer is faster than an old after all - so people using a 386 will
complain why people like myself that have a newer computer will not.
//Bernie
http://bernie.arachne.cz/ DOS programs, Star Wars ...
