[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: HFS Utils?



sheldon@atlcom.net (Sheldon Simms) wrote:
>In article <5cb05d$1nkg@uni.library.ucla.edu>, pubpc1@library.ucla.edu
>wrote:
>
>> spec@vax2.concordia.ca (Mitchell Spector) wrote:
>> >
>> >    Yes, at present there are no utilities whatsoever for working
>> >with HFS on the Apple IIgs. This means you cannot repair or even
>> >optimize volumes...
>>  ^^^^^^^^
>>
>> There may be no utilities, but if you want to optimize, just copy
>> your files to another HFS volume, erase your original HFS one, and
>> copy the files back.  :)
>
>Has anyone who regularly uses an HFS volume on the IIgs ever
>optimized the volume (in whatever manner) and noticed an actual
>improvement in disk access speed?

Yes, optimizing seems to help after you used the HFS volume for
a long time.  The difference is very slight, but it is there.


>I ask because I dug out my "Inside Mac: Files" when this thread began
>and read up on the HFS format. It seems to me that optimization is
>unlikely to help very much because the way that the ProDOS File System
>and by inheritance, GS/OS, finds files on the disk is not well suited
>to the HFS format.

The access speed is hampered already, but optimization seems to
make the best of a bad situation.


>Using Mac file calls, a program needs only one traversal of the catalog
>B-Tree to find any file on the disk. However it seems likely to me
>that using HFS FST calls on a IIgs requires a traversal for every
>directory in the full pathname of a file. This may explain why an HFS
>volume seems so slow on even a accelerated IIgs while on even an very
>old Mac its pretty quick.

So that might be what's causing it.  I've read that it was because
the FST was programmed in a high level language rather than assembly,
with non-optimized code at that.

-Scott G.