LTR-48125W Secondary Master on Gigabyte Motherboard GA-7N400 pro 2 (PCB
rev 2.0)

Driver NVIDIA nForce 2 ATA Controller (v2.5) 5.10.2600.415 built by: WinDDK

Best nForce2 IDE Driver settings, NVIDIA multi word 2 using NVIDIA
drivers, NOT ultra-dma and NOT pio and NOT standard windows IDE drivers.
Tested on Gigabyte GA7n400pro2 rev2.0 (BIOS version FH), Liteon
LTR-48125W CDRW on secondary master IDE, Pioneer DVD-115 ROM on
Secondary slave IDE(UDMA 2 seems to work fine for it, must be a bug in
write modes or just Liteon/Plextor related bugs), 1024 meg 333mhz DDR
RAM (2x256 meg + 1x512 meg), Geforce 2 MX AGP card, 80GIG seagateIV HDD
on Primary IDE, SCSI controller in PCI (disabled),
Firewire/SATA/RAID/Serial/Parrallel/Gameport/MIDI disabled in BIOS.


All testing info follows...
Yes, I have been trying to work this stuff out for the last 3 days and
finally narrowed it down to driver settings and did some tests. Okay a
lot of tests...

CD burn "test writes" with buffer underrun enabled for 650meg data on
700meg cheap LASER "premium quality" 52x cdr which ONLY allows 32x,
(Smart Burn sensing the quality). Also a DVD-rom as Secondary Slave in
UMDA mode2. Primary Master is a Seagate Baracuda IV 80 gig hard drive at
Ultra DMA 5 - Ultra 100 (bios auto select on). Firewire, SerialATA,
Serial Port1&2, Parallel Port, Gameport, Midi, LAN Rom boot - ALL
DISABLED IN BIOS. Also have a SCSI card in a PCI slot which is disabled.

Changed the settings on the drive and rebooted for every test. UDMA2 has
been tested HEAPS in various configurations and has never worked on this
motherboard under any program or win98 or shipped and latest BIOS
version FH or gigabyte vendor drivers or NVIDIA drivers or windows drivers.

nVidia IDE drivers,
PIO barely works at ave 12x speed and heavy cpu use (a bit better than
windows driver)
Multi-word DMA 0 works, didn't check time
Multi-word DMA 1 works, 32x took 2:48
Multi-word DMA 2 works, 32x took 2:49
Ultra DMA 0 failed, 0% constant spinning getting faster and faster up to
speeds i've never even heard it reach before! argh she's gonna blow
capn'. In NERO it starts at about 7% and locks up the program so you
can't stop the drive, RecordNow also locks up like Nero. CD Burner XP at
least eventually will report a failure.
Ultra DMA 1 SAYS it works but it really does not. It takes 10 seconds
and thinks it worked. I do not know if it will work in real non-test
mode but I am assuming it will fail.
Ultra DMA 2 - Ultra 33 fails, 0% constant spinning getting faster and
faster up to speeds i've never even heard it reach before! argh she's
gonna blow capn'. In NERO it starts at about 7% and locks up the program
so you can't stop the drive, RecordNow also locks up like Nero. CD
Burner XP at least eventually will report a failure.

Then I uninstalled the NVIDIA nforce2 IDE drivers and rebooted then
chose Standard Dual Channel PCI IDE Controller driver when it asked for
drivers. After every detected hardware item it asked to reboot, which I
didn't do - there were about 6 devices so this would have really sucked
if I did reboot every time it asked... Hope this isn't wrong, funny that
it didn't ask to reboot after the final device was installed (secondary
IDE for the curios) . I checked the device settings before reboot
because I was curios and noticed the CD-ROMs were PIO, so I changed them
to "DMA if available" then rebooted finally.

Standard windows IDE drivers
"DMA if Available" failed with a similar error as usual but with more
info, IDE/SCSI reset. Failed at 0% and took a long time to stop and
report the error.
"PIO Only" Mode worked, 32x took 7:31 with a LOT of speed changes and
heavy CPU use which made the mouse pointer jerky and generally sucked
all round.

Then swapped to NVIDIA NForce2(tm) SPP/IGP ATA Controller drivers
version 5.10.2600.340 built by WinDDK
Ultra DMA 2 - Ultra33 failed at 0%, same as others just keeps spinning.
Multi-word DMA 2 worked, 32x took 2:48 and about 4% CPU
PIO mode barely worked, 32x took 5:50, MASSIVE cpu use.
assume UDMA1/UDMA0/DMA1/DMA0 same as other nvidia driver results

Best nForce2 IDE Driver settings, NVIDIA multi word 2 using NVIDIA
drivers, NOT ultra-dma and NOT pio and NOT standard windows IDE drivers.
Tested on Gigabyte GA7n400pro2 rev2.0 (BIOS version FH), Liteon
LTR-48125W CDRW on secondary master IDE, Pioneer DVD-115 ROM on
Secondary slave IDE, 1024 meg 333mhz DDR RAM (2x256 meg + 1x512 meg),
Geforce 2 MX AGP card, 80GIG seagateIV HDD on Primary IDE, SCSI
controller in PCI (disabled),
Firewire/SATA/RAID/Serial/Parrallel/Gameport/MIDI disabled in BIOS.


I also posted this here
http://www.cdburnerxp.se/forum/viewt...hp?p=3830#3830
someone said that the NVIDIA drivers could not be removed, but mine
seemed to remove fine and resintall fine. I don't know if the drivers
were lying or not but as far as I know these results are fair.