Archived from groups: alt.comp.periphs.mainboard.abit (
More info?)
Hi, Phr3d. The bus-clock-speed sensitivity issue is specific
to the particular IDE controller chip used. The HPT controller
chip is a PCI device which connects to the PCI bus completely
independently of the on-chip IDE controller in the BX chipset.
As a result, there can be completely different compatibility
results for the HPT controller and the BX on-chip controller.
Both individual controller chips' compatibility sensitivities can
respond to their own independent BIOS updates.
Eg: A motherboard BIOS update can solve an on-chip IDE
controller issue -- however if that BIOS update uses the
same HPT insert as the previous motherboard BIOS, the
compatibility issues with the HPT controller won't change
at all.
Conversely, an HPT BIOS insert into the motherboard BIOS
with no changes to the basic motherboard BIOS can fix
HPT compatibility issues without changing the rest of the
BIOS at all. This is what happened with the 72_2351.BIN
update.
Potentially, there is a big plus to removing the HPT 370 controller's
PCI bus-speed-sensitivity -- this update *may* allow the user to
select a synchronous memory/PCI clock set.
Many motherboards have problems maintaining an async
memory/PCI speed set when O/C above a certain threshold
memory clock speed. See the Abit forums on the Abit USA
website for more info on this problem. Also, some motherboards
simply run faster with synchronous memory/AGP/PCI clocking.
If (and I have not confirmed this -- which is why I put the caveat
in my original post) the PCI bus-speed-sensitivity problem has
been completely resolved, then you should be able to move the
PCI clock off the 33MHz mark with impunity.
However, please note that *some* hard disk firmware doesn't
like faster PCI bus-clock speeds -- this is *irrespective* of
whether or not the HPT chip likes higher clocks or not. Only
in-depth testing with a full offline backup in place will allow the
user to recover from the difference between "the documentation
says" and "this is what actually happens" disease.
Note: It is entirely possible for a high PCI bus-clock-speed to
induce the hard drive to completely corrupt its entire data
set. This is why it is of paramount importance to have a
complete offline backup in place -- and to have that backup
completely disconnected from the test machine during the
experimentation phase.
If you do decide to experiment, please let the newsgroup know
the results of your investigations. Also include the make/model
and firmware version of your Hard Disks in your results post, so
we can get an idea of which drives are compatible with
high-speed-PCI-bus operation and which are not.
Hope this helps your understanding.
Bill
Phr3d wrote:
> Hi Bill,
>
> Thanx! It might be worth my while to install, since I am oc. I
> presume the oc issue is with RAID only, as my wife's machine is 2
> 144/4 to the PCI and has had no noticable probs, but is only two
> standalone drives.
>
> knf
>
> "Bill Drake" <bdrake@telus.net> wrote in message
> news:gU3ic.3170$w63.1556@edtnps89...
>> The newest Highpoint BIOS supports the following features:
>>
>> 1. When reconnecting drives that were part of a failed RAID 1.0
>> array -- the BIOS will *recognize* the signature on the existing
>> drive at startup (in the BIOS screen itself) and offer to remove
>> the drive from the array so the working drive can operate as a
>> standalone item. (No, you don't need to go through a long and
>> involved rigmarole to fix this -- it's as simple as pressing
>> "y/n" at the "fix" prompt in the BIOS setup screen.)
>>
>> The above feature removes the need to rebuild the array with
>> a second drive -- simply to get the machine working properly
>> again.
>>
>> Note: You have the *option* to remove the drive from the array,
>> or provide a second drive to rebuild the array. The
>> main use for this function is to permit easy OS
>> upgrades using a single Hard Disk. You can disconnect
>> the secondary RAID 1 disk, reboot, allow the HPT BIOS
>> to discover the broken array, remove the primary disk
>> from the array (to bring the primary disk back to a
>> single-disk item) and then perform the OS upgrade on
>> the primary hard disk. You can perform the above task,
>> secure in the knowledge you have a complete backup of
>> the original OS on the secondary disk (which is
>> unplugged while you do the update). That secondary
>> disk can be used to painlessly return the machine to
>> its pre-update state should the OS update or OS install
>> fail to work properly for whatever reason.
>>
>> The above is particularly useful since the HPT 370
>> driver shipped with WXP does not reliably support RAID
>> arrays.
>>
>>
>> 2. There are a bunch of other fixes subsequent to 2.34 -- these are
>> listed in the readme file found inside the 2.34 driver archive.
>> One *important* bugfix is the clock-speed calculation -- which
>> the readme file *implies* should remove some of the overclocking
>> bus-speed sensitivity that causes the HPT chip to fall-over-dead
>> as soon as the PCI bus moves even slightly off the 33MHz mark.
>>
>> Note: I have not personally verified the PCI bus-speed-stability
>> of the HPT370/A chipset at higher-than-standard speeds.
>> This requires more time that I currently have available
>> for testing. Perhaps someone else who has both time
>> and full high-speed-backup capability (ghost to
>> firewire external drive) can experiment with this and
>> provide some real-world insight as to whether the
>> *implied* fix is actually working in reality.
>>
>>
>> Best I can do for now. <tm>
>>
>>
>> Bill
>>
>>
>>
>> Tom G wrote:
>>> the newer upgrade just incorporates the newest Highpoint bios
>>> (2.51 or something??) into the 72 mainboard bios...
>>>
>>>
>>> "Phr3d" <Gwaihir@Orthanc.com> wrote in message
>>> news:20040420210005.546$Q0@news.newsreader.com...
>>>> Hi Tom,
>>>>
>>>> Sorry to butt in here, I missed the HPT update. There is newer
>>>> than Z72? What benefits have been noticed? Thanx,
>>>>
>>>> knf
>>>>
>>>> "TomG" <tgeery-NOSPAM-@cox.net> wrote in message
>>>> news:hUghc.8854$55.4710@lakeread02...
>>>>> depending on the version of your board (there were four
>>>>> versions...), you may be able to benefit from updating the
>>>>> Highpoint bios. if the board was a version 1.2 or 2.0 with a
>>>>> Highpoint 370 controller, then you could benefit from running the
>>>>> Z72C or another custom bios that I have on my FTP site under the
>>>>> /beta folder, just to keep them separate from the regular released
>>>>> bios sets. look under BE6-II then under Custom. there is also a
>>>>> bios in that folder that updates the Highpoint bios all the way to
>>>>> the newest released version but the Z72C is a great update as
>>>>> well.
>>>>>
>>>>> --
>>>>>
>>>>> Thomas Geery
>>>>> Network+ certified
>>>>>
>>>>>
ftp://geerynet.d2g.com
>>>>>
ftp://68.98.180.8 Abit Mirror <----- Cable modem IP
>>>>> This IP is dynamic so it *could* change!...
>>>>> over 130,000 FTP users served!
>>>>> ^^^^^^^
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> "Jerry Manner" <goodminded@hotmail.com> wrote in message
>>>>> news:Nrchc.8884$Mg2.5171@amsnews05.chello.com...
>>>>>> Hi
>>>>>>
>>>>>> I have a ABIT BE6-II MB with W98 SE on it. I need to have new
>>>>>> drivers because I have a save conflict with my PCI Mass storage
>>>>>> controller that cannot be removed. Is it possible to get new
>>>>>> drivers for this old MB? Can I solve my problem with the new
>>>>>> drivers? or do I have to do something else?
>>>>>>
>>>>>> Kind Regards
Facebook
Twitter
Instagram