• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

160GB and up drives are not compatible with Windows 98/Me, what about Serial drives?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

c627627

c(n*199780) Senior Member
Joined
Feb 18, 2002
160GB and up drives are not compatible with Windows 98/Me, what about Serial drives?

You need a PCI raid card to use 160GB and higher IDE PATA hard drives with Windows 98/Me because of the 137GB operating system limit, right?

What about SATA 160 GB and up drives?


Motherboards have an onboard RAID controller that accommodates the standard two IDE cable connectors PLUS two SATA connectors so:

If the original problem is taken care of through PCI cards, would it be taken care of by onboard RAID controllers?

Would they allow today's large capacity SATA drives to be connected without risking data corruption which would result if those drives were PATA and were accessed under Windows 98/Me.


 

Attachments

  • SATAonWin98.jpg
    SATAonWin98.jpg
    34.1 KB · Views: 231
c627627 said:
If the original problem is taken care of through PCI cards, would it be taken care of by onboard RAID controllers?
Even though running a 48-bit LBA capable BIOS and 48-bit LBA enabled drivers for the said controller, the use of a PCI controller card / onboard SATA controller doesn't actually resolve the "corrupted data" issue when the partition(s) is / are increased in size past the 137 / 128GB limitation.

Even though the specific controller manufacturer may have updated 48-bit LBA drivers available which work with Win98 / Me (like for instance the Intel Matrix Storage drivers), increasing the size of the volume(s) on a 48-bit LBA HDD (after the OS is installed, and from within the GUI) normally results in corrupted data.

48-bit LBA and Windows 98, 98 SE, Me
http://www.48bitlba.com/win98.htm

Partition Tools and 48-bit LBA
http://www.48bitlba.com/partitiontools.htm
 
redduc900 in the house :)

I've been using Windows Me and Silicon Image IDE Raid PCI Card SIL680RAID for years now to access a 200 GB drive filled to the brim.

I know what happens to the data if it is accessed when the 200 GB drive is connected directly to the motherboard with no onboard RAID controllers and without the use of the RAID card: .txt files turn to gibberish and other data is corrupted.

The use of above mentioned RAID card to this day did not result in any data corruption.


You bring up an excellent point however and I look forward to finding out when it is applicable and why do you think it wasn't to the RAID card I'm using. Thank you.


 
Also the use of the word "partition(s)" can be a source of much confusion as many people tend to think that partitioning the drive and having all partitions under 137 GB would get around this problem.

Of course it does not as the physical size of the drive being lower then 137 GB is the only thing that keeps you data from being corrupted if it is accessed directly from Windows 9x/Me, right?
 
Last edited:
"Drive size", not "Partition size" is correct... sorry for the confusion. I'll hazard an educated guess and say that the Sil controller drivers you're using are possibly more effective at communicating the controllers' 48-bit LBA capabilities / attributes to the OS than say for example comparative Intel or HPT drivers? I don't really think it makes a difference as to whether the drive(s), be it / them IDE or SATA are connected to an onboard or PCI controller, as long as the drivers provide the correct / needed information to the OS.
 
Just finished reading the article you linked to, redduc900.

They start off by saying not to ever use Windows 9x/Me system tools such as Scandisk.
...which can easily be disabled in Windows 9x/Me:
Start Menu > Run... > Msconfig > Advanced... button > CHECK: "Disable Scandisk after bad shutdown option" > OK > OK > Yes.

[Quick Off topic: I say easily because you may recall btw, this cannot be done permanently on Windows XP (despite dozens of threads on multiple forums to try to find out why.)]


They do say that "Use PCI ATA controller adapter to connect the hard drive to the system. A 48-bit LBA driver is provided with the controller."


so....


PCI ATA controller is a valid (full?)work around.


If yes, a side question would then be, the talk about the use of system tools they start off with, do they mean "even if you use PCI ATA controller"?


 
Also important: "...as long as the drivers provide the correct / needed information to the OS" implies they may not be. How is that determined specifically?
 
I've heard of the Scandisk / Defrag / Fdisk (System Tools) issue, and I'm not exactly sure if it's meant to address partitions and / or drives larger than 137 / 128GB, and whether or not that applies to drives on a 48-bit capable PCI controller card / onboard SATA controller. The whole 48-bit LBA issue with Win98 / Me is somewhat confusing, as there seems that there is no "concrete" answer as to whether Win98 / Me will work flawlessly when installed to a >137GB HDD or partition... or even when accessing data on a HDD connected to say an onboard RAID controller or PCI card. I think that more than anything, reliability (meaning no data corruption) depends most on the controller drivers used.
c627627 said:
Also important: "...as long as the drivers provide the correct / needed information to the OS" implies they may not be. How is that determined specifically?
Unfortunately I'm unsure of the exact specifics involved, but I'm sure the information could be gathered from the controller manufacturer's spec sheets specific to that controller, or from their developer's site... I would think so anyway. That's a good question regardless, but sorry to say I don't have an answer to it.
 
Thank you. I'll shoot an email over to the authors of that article when I have time then revisit the thread in the future. It would be nice to get them to revise the article they wrote to address the very questions it was written to answer...

redduc900 said:
I think that more than anything, reliability (meaning no data corruption) depends most on the controller drivers used.

Unfortunately I'm unsure of the exact specifics involved, but I'm sure the information could be gathered from the controller manufacturer's spec sheets specific to that controller, or from their developer's site...

Spec sheet info may be in the manual of the mobo with onboard RAID controller:
http://epox.com/USA/article.asp?ID=1202

...but what would I be looking for as far as which specs :(

I guess emailing tech support is also a way to go...
 
Email sent:

Thank you for writing a fine article 48-bit LBA and Windows 98, 98 SE, Me http://www.48bitlba.com/win98.htm

As we were discussing it on this forum:
http://www.ocforums.com/showthread.php?t=469331
we noticed that it could be clearer in addressing some important questions.

1. Without the use of any PCI ATA Controller cards, it is clear that data loss, data corruption may result if Windows 9x/Me is used to access hard drives over 137 GB.

2. Without the use of any PCI ATA Controller cards, it is also clear that use of System tools in Windows 98 and Me may cause data corruption.


However it is unclear whether the use of PCI ATA Controller cards FULLY solves the problem the article was written about and allows the use of high capacity drives under Windows 9x/Me in the same way as if a hard drive under 137 GB were connected directly to the motherboard.

Also, the linked forum question asks if motherboards have an onboard RAID controller that accommodates the standard two IDE cable connectors PLUS two SATA connectors,

and

if the original problem is taken care of with PCI ATA Controller cards, would it be taken care of by onboard RAID controllers as well?

Would they all allow today's large capacity SATA drives to be connected directly without risking data corruption which would result if those drives were PATA and were accessed directly under Windows 98/Me?



Please also note that the most common, widespread misconception about this problem is that if all the partitions on the drive were under 137 GB, that this would somehow solve the problem. This is of course not true, but in this case the obvious definitely merits mentioning.


 
Back