Hans-Peter Diettrich <DrDiettrich1@[EMAIL PROTECTED]
> writes:
> Phil Carmody wrote:
>
> > Each of the above operations is atomic, so what the **** is a write
> > pointer? You're barely making sense, so it's hard to argue against
> > you.
>
> You have read about FSInfo.FSI_Nxt_Free?
>
> Isn't that a write pointer?
Given that one term has already been coined for it, there's
no need to coin another one. Economy of nomenclature.
> > The FAT32 system apparently (see posts by other posters upthread)
> > writes each new file at the lowest cluster it can.
>
> So you think that the system scans the FAT, each time a cluster has to
> be allocated to an file or directory?
My god, you really are as stupid as I first suspected!
No I do not think that at all.
What on earth made you think a scan would be required every time?
That's what maintaining such an index is for, so you don't have
to scan. Think about which operations could make that index change,
and whether the system knows what it would change to after
completion of those operations.
> > This is in agreement with all the prior FAT systems that I have been
> > familiar with.
>
> Then familiarize with the new features of FAT32, before posting nonsense
;-]
Care to provide citations for where the standards deviate in
this regard?
Damn, I'd forgotten that I said I wasn't going to waste time
spoon feeding someone who has painted himself from head to toe
in KlooRezist. Due to lack of compression content, I'm removing
comp.compression from followups so I don't have to suffer your
no-doubt inane and ill-conceived reply.
Phil
--
Dear aunt, let's set so double the killer delete select all.
-- Microsoft voice recognition live demonstration
|
