-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of
Todd Lyons
Sent: Wednesday, April 25, 2001 4:21 PM
To: [EMAIL PROTECTED]
Subject: Offlist: [expert] 8.0 final --brakes MANY applications
(Software Installeris first on that list)


"Jose M. Sanchez" wrote:


You are correct and he is correct.  Read on.

---

No, Just I'm correct... ;-)

---


> Win-Drives... yuk, yuk.
> Linux has no problem with WD drives as long as the hardware itself is not
to
> blame...
> I.E. bus chatter, IDE/PCI noise, etc... (as with the VIA chipsets...)

His assertion that activeX has anything to do with it is wrong, but he
is correct in that the drives are kind of Windrives.  The exact sequence
as I understand it is that with each block that gets sent to a hard
drive is accompanied by a CRC (checksum).  The hard drive puts that into
its cache and may or may not write the block to the physical hard drive
surface immediately.  When the controller requests a CRC to verify the
written data's CRC, the hard drive just sends back the CRC from its
cache...the same one that the controller sent it.  So it will ALWAYS
match properly.  If the cable toggeles one bit during the transfer, or
the heads toggle one bit during the actual write, the controller will
never know about it because the drive itself never checks.  The reason
he's partially accurate is because the Windows drivers force the drive
to flush the cache, then read it back, perform a CRC, then send the CRC
to the controller.  This is something that is done by default by the
hardware on the rest of the hard drives out there.  The drives which
remove functionality from the hardware and put it into the driver is
remarkably similar to modems which remove functionality from the
hardware and put it into the driver.

---

Oooh now this gets real interesting!

So this is NOT a driver removing a hardware function at all!
(ahem, which is why I'm correct... heh).

Rather this sounds more like a circuit flaw that updated drivers tend to
correct...

Sounds more like some engineer screwed up, not an OS dependency as
misstated...

Furthermore other WD models and more recent drives seem not to have this
flaw...
suggesting again a design problem or glitch.

The Windows drivers have merely been "adjusted" to overcome the problem...

Sounds like a small patch to the kernel source can do the same thing...

----


As I hope this all makes sense.  Not all WD drives are bad, just some.  I
fear that this trend will continue though.

---

Highly unlikely... it would mean:

1) WD would be shooting themselves in the back.
2) A completely NEW IDE "soft" drive standard married to Microsoft.
3) Problems galore with tens of thousands of second party devices and
applications.
4) There would have to be a preponderance of people wanting this type of
change...

Not I.

But anyway -THANKS- for the discussion... I'm going to have to keep a sharp
eye on the specs for this from now on to make sure the Windrive fallacy
never becomes true!

-JMS
[EMAIL PROTECTED]


Reply via email to