Problems accessing >137GB on HPT372 (4PEA+)

[LiQiCE]

Has anyone else had problems accessing drives bigger than 137GB when they're connected to the Highpoint 372 Controller on an EpoX 4PEA+?

I first was using the old 12/2002 BIOS and I was able to sucessfully copy all of my data from my old 120GB drive to a new 250GB drive except for the last partition (which happens to be the only partition above the 137GB mark on the new drive, partitions were 1GB FAT16, 20GB NTFS, 30GB NTFS, 60GB NTFS, and the remaining 120GB NTFS). Only the last 120GB NTFS had problems.

I then realized that the new BIOS claims to fix problems with copying large amounts of data, so I upgraded to that BIOS and had the same problem!

Let me also add that I used 3 different methods for trying to copy the data. I tried copying the data using Ghost 2003, I tried copying the data using Partition Magic 8, and I tried copying the data using plain old Windows.

Suspecting there was a problem with the Highpoint 372, I switched both my disks to the Intel IDE controller and I was able to successfully copy the data. I then switched the drive back to the Highpoint controller and I couldn't access the data anymore (it showed up, but when I tried to play an MP3 or tried to access a document they would show up all garbled or corrupted)! Upgraded from the v2.32 drivers (which I was told is bad to do since v2.32 is the BIOS revision that the 4PEA+ comes with), but I upgraded to v2.34 anyway and I was able to access the data. Tried copying again with Ghost and the drives hooked up to the HPT372 and the data was corrupted again under v2.34!

Finally, I read on AOAForums how I could use CBROM to place the latest HighPoint BIOS inside the latest EpoX BIOS. I upgraded to the v2.35.1 drivers and luckily my BIOS worked just fine (phew!). Re-ran the copies using Ghost 2003 and poof, I was able to access the data!

Has anyone else had these problems? I spent the entire weekend tracking this problem down, and I think it would be nice if EpoX would release an official BIOS with the latest HighPoint BIOS so those of us trying to use a >137GB drive with their HPT372 controller can do so without problems.

[danrok]

The HPT controller is a separate chip and BIOS to the main system BIOS. The two can be updated independently to each other.

If the two BIOS's were bundled together, flexibility would be reduced. Not everyone wants to update their SATA BIOS at the same time as their system BIOS.

Also, I'm not sure if switching a drive between two different types of controller is a good idea. The drive should be formatted on the HPT and not switched to the standard IDE.

Have you got things working OK now?

[LiQiCE]

Originally Posted by danrok

The HPT controller is a separate chip and BIOS to the main system BIOS. The two can be updated independently to each other.

If the two BIOS's were bundled together, flexibility would be reduced. Not everyone wants to update their SATA BIOS at the same time as their system BIOS.

Also, I'm not sure if switching a drive between two different types of controller is a good idea. The drive should be formatted on the HPT and not switched to the standard IDE.

Have you got things working OK now?

Actually the HPT controller BIOS is embedded into the EpoX Mainboard BIOS on the 4PEA+. Like you said, this reduces flexibility considerably because the latest EpoX motherboard BIOS is bundled with v2.32 of the HPT372 BIOS (a fairly old BIOS revision). Only by hacking up the EpoX Mainboard BIOS using CBROM was I able to embed a new version of the HPT372 BIOS into the EpoX Mainboard BIOS.

This is the HPT372 Controller built onto the mainboard that I'm talking about, not a PCI card which would offer the added flexibility that you refer to.

With the built-in HPT372, it is not possible to update the HighPoint BIOS without upgrading the System BIOS like you said. Reflashing a new BIOS for the EpoX 4PEA+ includes the HighPoint 372 BIOS. All of my previous boards which include an on-board RAID/IDE Controller have had this same limitation. Perhaps the newer motherboards don't do this? That would be nice.

While switching controllers for drives in a RAID configuration can be potentially dangerous, switching controllers for normal EIDE drives should not pose any sort of danger assuming the two controllers both support the size for that particular drive. For example, it would be a bad idea to put a 250GB drive onto an older external EIDE controller that does not support 48bit LBA, however moving a disk from one controller (HPT372) to another (Intel i845PE IDE controller) that both support 48bit LBA causes no negative problems. In my professional experience, I have moved EIDE disks from one computer to another in an attempt to fix a computer or recover data from the other drive and have never had any problems.

In truth, even a motherboard EIDE controller that does not support 48-bit LBA should have no problems with a 48-bit LBA drive so long as you are not trying to boot off of a partition which lies within the >137GB range. At least for Microsoft Windows XP (and Windows 2000 I believe), there is a patch that allows XP to access partitions beyond the >137GB barrier. This patch only works for partitions which are not needed during the boot process (The OS partition and any partitions where Page files live), and it only applies to the Motherboard's built in EIDE controller (ie. the Intel chipset EIDE controller or the equivalent).

I have gotten my disks working finally. It took me all weekend to track down the specific problem because it takes awhile to constantly clone the original partition back to the new drive. My solution was to hack up the latest EpoX mainboard BIOS for the 4PEA+ and embed the latest HighPoint 372 BIOS (available from www.highpoint-tech.com) using CBROM. My question mainly was if anyone else had this problem, and if so if they resolved it in the same manner that I did, and also if EpoX would consider bundling an official BIOS release with the latest HighPoint 372 BIOS in it.

[danrok]

I wasn't aware that it isn't possible to flash the two independantly. Now I understand the source of your frustration with this.

[LiQiCE]

Guess no one else has had this issue? I was surprised that I didn't find any other posts about this, but maybe I am the only one