View Single Post
  #4  
Old July 30th 19, 06:26 PM posted to alt.comp.hardware.pc-homebuilt
Paul[_28_]
external usenet poster
 
Posts: 1,467
Default PCCHIPS M811 IDE UMDA

T. Ment wrote:
On Tue, 30 Jul 2019 07:10:14 -0400, Paul wrote:

You want 80 conductor wire for your ATA IDE cable.
This reduces crosstalk at higher speeds due to the interleaved
grounds on the cable.


Right. That's been true throughout this.


The signal integrity on IDE looks awful.


Having an 80 wire cable, with just one device on the end, should
give better looking signals than with two devices. As the "middle"
device is a stub capacitance and a reflection point.


All I know is, I get errors with a single hard drive (master), and they
go away when I add a cdrom slave to the middle of the cable.


it's really hard in this case to say why this is happening.


It's a known M811 motherboard flaw, others have reported it. When you
downgrade to ATA-66, it works fine. Windows XP downgrades it by default,
maybe they blacklisted the M811. But DOS can't downgrade or start with
ATA-66.

AFAIK, no one else tried loading the cable with two devices to fix the
problem, I just got lucky because I wanted to free up IRQ 15 and save a
cable.

Why extra cable load compensates for a motherboard flaw, is the mystery.
Maybe you just need the right stuff.


You'd need to compare two motherboards using the same
chipset, in the IDE connector area or next to the
Southbridge, to see what corners were cut to save
a few pennies. There is undoubtedly an explanation.

Some chipsets use a single "reference" resistor to set
the slew rate on pad drivers, and an incorrect value
there could affect the waveform.

Paul