• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

Unable to boot from a drive connected through an Acard 6280M

superpantoufle

Well-known member
So it's like a normal behavior?

Ok, I can live with it. I'm just a (tiny little) bit disappointed, sine I thought that thanks to that card I'd be able to easily switch drives from different machines.

Thanks all!

 

equill

Well-known member
The nearest corroborations of the suggestion that drives should be formatted while they are attached to the ACARD that I can find in ACARD's PCI-to-IDE ATA-133 Controller for Macintosh User’s Manual (2001) is the third statement on p6:

■ Connect 2 hard drives in the same channel, check the Master/Slave jumper.

■ For the hard drive’s Jumper setting (Master / Slave), refer to hard drive’s Manual.

■ As hard drive contain old data which might cause the Mac OS installation problems. Before installing Mac OS, the hard drive must be processed the “initialize” first.

As I interpret the Taiwanglish, this may be no more than a reminder (as are the first two statements) of prudent drive management, but not a mandatory requirement.

Elsewhere in the manual the implicit assumption is that a blank drive will be used, but the preceding seems to allow use of a drive with data of some kind already on it. However, my own experience does not accord with the suggestion that drives used with the card must be formatted through the card. The G4 DA that I referred to above has two drives, both formatted from OS 9 with Silverlining 6.5.8—as I use Silverlining exclusively on more than 100 drives in my possession—before the ACARD was acquired and installed. Each drive already had an OS 9.2.2 and an OS 10.2.8 partition, and functioned perfectly with the ACARD.

Is it just possible that the formatting of drives is being confused with the updating of firmware on the ACARD? Most ATA card manufacturers insist that firmware should not be updated while attached drives are active, and preferably should be updated with drives disconnected from the card. If someone else's mileage has varied from this prescription, I cannot argue with that, but I find the suggestion that drive preparation needs attachment of the drive to the card, at the very least, odd.

Further, when my MDD DP was sulking with a blown PSU, I simply removed its drives to the DA pro tempore so that my work could continue. The drives not only have OS 10.4.11 and 9.2.2 on them, but also the OS ROM v10.2.1 needed for OS 9.2.2 to 'see' drives of more than 128GB. Again there was no problem in simply attaching them to the 6280M.

ACARD's manual, on p4, also assures us:

The ACARD AEC-6280M is the leading Mac add-on card, supports ANSI X3T9.2 CAM ATA4/ATA5/ATA6, data transfer rate up to 133 MB/sec and meets the demand of multimedia, real time video. It’s also backward compatible with traditional ATA modes.
The ACARD AEC-6280M supports all models of PowerMac which have a PCI slot, including PowerMac 7200, G3, G4. For PowerMac older than G4, ACARD AEC-6280M can really boost the disk performance up to 100%. For PowerMac G4, ACARD AEC-6280M allows Mac to add more drives and use software RAID.

It supports the true PnP function, in Mac 8.5 and later version, there are no software driver needed for installing an Ultra ATA133/100/66/33 HDD. With the advanced function, the adapter is really user friendly. It is coexist with the on board IDE controller. ACARD AEC-6280M with on board BIOS supports system to boot from ATA HDD. ACARD AEC-6280M also supports multitasking to improve the CPU performance.
de

 

trag

Well-known member
So it's like a normal behavior?
Yes. IDE drives initialized on an IDE/ATA card do not work on some of Apple's built-in IDE busses and vice versa.

It's been a long time since I experimented, but I do not think this is true in all cases, but it's on a by card and by Apple model basis. Certainly the older Apple models with IDE do something different with drives than any of the IDE cards do.

Once you are aware of this it is pretty easy to deal with. I am sorry I came so late to the thread. I could have saved you some headache.

I use an Acard 6280M in a Beige G3 and in a Umax S900 (8500/9500 clone). I boot both OSX and OS9.x on the Beige from drives attached to the Acard. The S900 doesn't have OSX installed on it. I boot from both the Acard and from SCSI drives on an Initio card and from optical drives on the Acard and on the built-in SCSI busses on the S900. That S900 has a lot of drives. :) Five hard drives and four optical drives.

The Acard 6880M (slightly different from your card) is nice because its built-in RAID capability is usable on both OSX and OS9. For example, if you create a RAID using the utiiltiies in OSX it will not be usable in OS9 because OS9 does not have support for OSX RAIDs. However, if you create a RAID on the 6880M, it will work in both OSX and OS9--at least that's my understanding.

Oh, and one other irrelevant thing, which might help someone in the future with a similar problem reading this thread. If you try to use a Western Digital drive as a single drive (only drive on cable) and set the jumper to Master, you'll get much the same symptoms as you had. At first when I read this thread, I guessed that you had the Western Digital problem, until you mentioned that you were using a Maxtor.

Western Digital drives have a 'Single" setting which must be used when the drive is all alone. Using Master when you should use Single will result in similar behavior.

 

Bunsen

Admin-Witchfinder-General
The Acard 6880M (slightly different from your card) is nice because its built-in RAID capability is usable on both OSX and OS9. / at least that's my understanding.
8-o

Before I stick that in ebay worldwide search, how do you come by this understanding? I gather you haven't actually had one in your hands to confirm?

 

trag

Well-known member
Before I stick that in ebay worldwide search, how do you come by this understanding? I gather you haven't actually had one in your hands to confirm?
I own one (6880M) and that's what the documentation claims. However, I have not yet set up a RAID to try it myself. I also assign it a slight doubt factor because it's been several months since I read the documentation, and I could be misremembering. I strongly remember that fact. However, I have no memory of the specific spot in the documentation where I read it. Hence the doubt factor. But their user's guides are downloadable, so it should be easy enough to look up.

Some time in the next month or (more realistically) two, I'll be selling some 6880Ms for about $50. I have a pile of PC style 6880s which I am going to convert to Ms. I'm still trying to find the top of my workbench though.

 

Bunsen

Admin-Witchfinder-General
user's guides are downloadable / I'll be selling some 6880Ms / I'm still trying to find the top of my workbench though.
Hah! I'm still trying to find my workbench!

I'll check out the user guide. Consider me in the queue for one when they're available. Ta.

 

superpantoufle

Well-known member
Ok, just to keep you helpful fellows updated, it's been almost two weeks, and the 9500 boots fine under OS 9 and OS X either from the card or the internal bus. It's pretty easy, once you know (and remember!) which disk has been formatted where!

Thanks to all!

 

trag

Well-known member
Resurrecting this old thread.

So after my confident pronouncements about formatting the drive with the Acard solving problems -- which it seems to have for the original poster -- I find that I have the same problem as the original poster.

I have an S900 logic board, which is essentially a PM8500 or PM9500.

It has an array of three striped SCSI drives with various boot volumes on an Initio card.

It has a 40 GB IDE drive on Channel 1 of the Acard 6280M. Channel 1 is shared with a DVD-ROM drive.

I recently added a pair of Seagate 160 GB ST3160815A drives to Channel 2 of the Acard.

Everything is formatted with SoftRAID 2.2 and my primary OS is 9.1, although some of those SCSI boot volumes go back as far as 7.6.1.

I created several Mirrored volumes on the pair of 160 GB drives. The volumes with OSs on them are all 1 GB in size and initialized as HFS Standard, not +.

And I cannot get the darned machine to boot from any of the volumes on the 160 GB drives. They're selectable in "Start-up Disk". I also tried de-mirroring one of them and booting from the singleton, thinking there was some problem with mirrored volumes. But that didn't make any difference.

I can boot from an OS9.1 volume on the 40 GB IDE drive (also connected to the Acard) and I can boot from my various SCSI (Initio) boot volumes. But the 160 GB drives refuse to be bootable.

Hunting around, it looks like some things for me to check are whether my Acard firmware is up to date -- although that update is only supposed to be needed for partitions larger than 120 GB, and these boot partitions are 1 GB even if the drives are bigger.

I should also check that the drive I jumpered as master is on the correct connector on the IDE cable and similarly for the Slave.

But as long as I was hunting the web for wisdom and Google dropped me off here, I figured I'd post my predicament.

Has anyone heard of any issues with this model of Seagate drive? I think that they were pulls from a Dell server (bought them on Ebay) and I've heard of server companies doing some unnatural things to drive firmware....

Thank you for any helpful or humorous suggestions.

 

trag

Well-known member
Almost a year later, I found the answer. The 6280M was installed in a Umax S900 logic board in one of the lower slots. I was overjoyed, years ago, to discover that the 6280M worked trouble-free in any of the lower four slots of the S900.

Apparently, that's not completely true. When SoftRAID is used to create Mirrored volumes of disks connected to the 6280M, the volumes are not bootable, unless the 6280M is in one of the two upper slots of the S900. Weird, but there it is.

Umax screwed the pooch more than a decade ago by using a PCI-PCI Bridge to implement the lower four PCI slots, and it causes all kinds of trouble in those machines.

When the 6280M is installed in an upper slot, mirrored drives boot fine.

 
Top