cjeffery@arizona.UUCP (07/13/88)
Hi people: for a year now I have been running an Atasi 3046 drive in my XT clone. Recently I purchased an AT clone kit, planning on using the Atasi as its hard drive. So far, the hard disk "fails initialization" and has resisted any attempts at low-level formatting in the AT. So my question: what differences are there between the XT (e.g. WD WX1) controller and the AT (e.g. WD WA2) controller? Don't most ST 506 drives work with either? Are there any little jumpers I need to pull? Any help would be greatly appreciated! -- | Clint Jeffery, University of Arizona Department of Computer Science | cjeffery@arizona.edu -or- {ihnp4 noao allegra cmcl2}!arizona!cjeffery --
davidsen@steinmetz.ge.com (William E. Davidsen Jr) (07/13/88)
In article <6201@megaron.arizona.edu> cjeffery@arizona.edu (Clinton Jeffery) writes: | Hi people: for a year now I have been running an Atasi 3046 drive in my | XT clone. Recently I purchased an AT clone kit, planning on using the | Atasi as its hard drive. So far, the hard disk "fails initialization" | and has resisted any attempts at low-level formatting in the AT. I believe that AT disks are *not* ST506, they are ST412. The difference in the two standards is that seek commands are buffered for ST412 and the wait comes at the end. BIOS commands to step five tracks: ST506 ST412 step step wait step step step wait step step step wait wait step wait step wait Looking at the standards as presented in a short article I found, this apperas to be the only difference, but it's enough to make the ROM BIOS on the disk controller not work. Warning: this is my understanding of the matter, but I don't sell or service disk drives. -- bill davidsen (wedu@ge-crd.arpa) {uunet | philabs | seismo}!steinmetz!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me
dale@wucs1.UUCP (Dale Frye) (07/13/88)
IBM modifies the cable so that all drives are to be configured as drive 2
(low = 1). Some of the clone kits followed IBM but most simply use straight
through cables. If the large cable does not have a twist in the middle of it
then it is a straight through cable. There is a jumper or a set of switches
on the drive that must be changed. These are usually located on the back
edge of the drive.
NOTE: This is not a difference between XT and AT but the difference between
IBM cables and clone cables. This difference occurs on both types of
machines.
REMEMBER:
If the cable has a twist then configure all drives as the second drive.
If the cable is straight through then configure each drive differently
starting with the first drive.
Dale Frye @ Washington University in St. Louisray@micomvax.UUCP (Ray Dunn) (07/15/88)
In article <11517@steinmetz.ge.com> davidsen@crdos1.UUCP (bill davidsen) writes: >In article <6201@megaron.arizona.edu> cjeffery@arizona.edu (Clinton Jeffery) writes: >| Hi people: for a year now I have been running an Atasi 3046 drive in my >| XT clone. Recently I purchased an AT clone kit, planning on using the >| Atasi as its hard drive. So far, the hard disk "fails initialization" >| and has resisted any attempts at low-level formatting in the AT. > > I believe that AT disks are *not* ST506, they are ..... Hmmm... A common misconception here. The "ST506" refers to the *low level* interface between the disk controller and the drive, and commonly is used for both XT's and AT's. The interface between hardware and software (i.e. to the BIOS, or to any BIOS ignoring badly behaved software) is an "XT interface" or an "AT interface" respectively. This software interface is independant of the low level interface, so that, for example, the higher capacity RLL drives still use the same software/hardware interface, but, if you like, the protocol going over the ribbon cable from controller to the disk drive is different. The XT and AT interfaces are very different, for example: The XT uses DMA ch 3, IRQ 5, I/O addresses 320-323 and an 8-bit data transfer. The AT does *not* use the DMA (fun eh?), uses IRQ 5, I/O addresses 1F0-1F7 and 3F7, and uses 16-bit data transfers. You can read about this in various places in the IBM manuals, which have changed over time, originally in the straight "Technical Reference Manual", but now, I think, in the "Options and Adapters" section. The situation is further confused by the fact that the BIOS interfaces to the disks are different in the XT and AT, and provide different services. Lower cost solutions are now quite common for clones, with the controller built into the drive (i.e. no separate controller card, with the disk connected directly to a special connector on the motherboard, or to a "dummy" board plugged into an option slot. This first happened with XT's, but is now available for AT drives as well. Now, why doesn't your XT drive work on the AT.... -- Ray Dunn. | UUCP: ..!{philabs, mnetor}!micomvax!ray Philips Electronics Ltd. | TEL : (514) 744-8200 Ext: 2347 600 Dr Frederik Philips Blvd | FAX : (514) 744-6455 St Laurent. Quebec. H4M 2S9 | TLX : 05-824090
jamesa@amadeus.TEK.COM (James Akiyama) (07/15/88)
In article <6201@megaron.arizona.edu>, cjeffery@arizona.edu (Clinton Jeffery) writes: > Hi people: for a year now I have been running an Atasi 3046 drive in my > XT clone. Recently I purchased an AT clone kit, planning on using the > Atasi as its hard drive. So far, the hard disk "fails initialization" > and has resisted any attempts at low-level formatting in the AT. > > So my question: what differences are there between the XT (e.g. WD WX1) > controller and the AT (e.g. WD WA2) controller? Don't most ST 506 drives > work with either? Are there any little jumpers I need to pull? > > Any help would be greatly appreciated! > -- > | Clint Jeffery, University of Arizona Department of Computer Science > | cjeffery@arizona.edu -or- {ihnp4 noao allegra cmcl2}!arizona!cjeffery > -- There are actually several differences between an XT (e.g. WD WX1) and AT (e.g. WD WA2) controller. This is a list of some: 1. XT controllers contain on board BIOS while the AT controller uses the AT BIOS. This is because the original PC did not support fixed disk and used an adaptation of the Xebec controller. 2. The XT uses DMA to transfer data between the fixed disk buffer and system memory while the AT uses the CPU. 3. The XT is a 8-bit controller while the AT is a sixteen bit controller. Since the WD1010 (or WD2010 in newer AT controllers) is only 8-bits, the controllers assemble two 8-bit values into one 16-bit word. This is part of the reason why most AT's can get by with a 3:1 (or 2:1) interleave. 4. Most AT controllers also contain the floppy disk controller. Either controller should work with a standard ST506 (ST412) drive. Note that the ST412 drives support "buffered seeks" while the original ST506 drives did not. All current drives support buffered seeks (at least those I'm aware of). Note that the standard AT controller, as shipped, requires the drive set to DRIVE #2 and requires that the 34-pin cable have four line twisted. Many third party controller have a jumper to make the controller "IBM compatible" or standard "ST412" compatible. Check with the OEM. I'm not sure whether this was true on the XT--part of the problem being that IBM had several variations to the XT controller (I believe). If you need more information, and have the WD1003-WA2 AT controller, let me know, since I have the OEM technical manual for that controller. One final note is that the WD1003-WA2 has two 20-pin connectors, one for each fixed disk. You must use the 20-pin connector closer to the 34-pin connector for the first drive. It is the one with the higher "J" number (I believe the two "J" numbers for the 20-pin connector are J30 and J40--you need to use J40). If you use the wrong connector, the controller will not be able to read any data from the fixed disk. Hope this helps. James E. Akiyama jamesa@amadeus.LA.TEK.COM UUCP: ....!tektronix!amadeus!jamesa ARPA: jamesa%amadeus.LA.TEK.COM@RELAY.CS.NET
mikes@mntgfx.mentor.com (Mike Stanbro) (07/20/88)
From article <6201@megaron.arizona.edu>, by cjeffery@arizona.edu (Clinton Jeffery): > Hi people: for a year now I have been running an Atasi 3046 drive in my > XT clone. Recently I purchased an AT clone kit, planning on using the > Atasi as its hard drive. So far, the hard disk "fails initialization" > and has resisted any attempts at low-level formatting in the AT. > > So my question: what differences are there between the XT (e.g. WD WX1) > controller and the AT (e.g. WD WA2) controller? Don't most ST 506 drives > work with either? Are there any little jumpers I need to pull? > > Any help would be greatly appreciated! > -- > | Clint Jeffery, University of Arizona Department of Computer Science > | cjeffery@arizona.edu -or- {ihnp4 noao allegra cmcl2}!arizona!cjeffery > -- The question more appropriately should be "What isn't different?". * The WX1 is an 8-bit peripheral, the WA2 is a 16-bit. * The configuration setup on the WX1 is jumper selectable, on the WA2 it is electrically programmed at boot-up using the disk-type data stored in the CMOS configuration RAM. * The WX1 is a hard-disk only controller, the WA2 is both a hard-disk controller and a floppy controller. I have installed a lot of hard disks in both XT's and AT's but never tried using a WX1 controller in an AT so I can't say that it is or isn't possible. Even if it is possible to use the 8-bit controller, I suspect there is a significant performance difference between the two due to 2X data width and less wait states for the 16-bit controller. You can get a WA2 for about $140 if you shop around. Mike Stanbro, Research Engineer (503) 626-1437 Mentor Graphics Corp., 8500 SW Creekside Place, Beaverton OR 97005 ...!{sequent,tessi,apollo}!mntgfx!mikes OR mikes@pdx.MENTOR.COM These are my opinions, & not necessarily those of Mentor Graphics. -- Mike Stanbro, Research Engineer (503) 626-1437 Mentor Graphics Corp., 8500 SW Creekside Place, Beaverton OR 97005 ...!{sequent,tessi,apollo}!mntgfx!mikes OR mikes@pdx.MENTOR.COM These are my opinions, & not necessarily those of Mentor Graphics.
dsd@hpsadla.HP (Donald St.Denis) (07/21/88)
>| Hi people: for a year now I have been running an Atasi 3046 drive in my >| XT clone. Recently I purchased an AT clone kit, planning on using the >| Atasi as its hard drive. So far, the hard disk "fails initialization" >| and has resisted any attempts at low-level formatting in the AT. > > I believe that AT disks are *not* ST506, they are ST412. The >... > bill davidsen (wedu@ge-crd.arpa) The drive interface is determined by the controller and drive, not the computer. I used an ST225 (ST506 Interface) taken from my XT to boot a new AT clone I'd put together. It worked. I had a bit more trouble getting a new ST225 formatted in the AT though. >IBM modifies the cable so that all drives are to be configured as drive 2 >... >Dale Frye @ Washington University in St. Louis I have not worked with many IBM PC's, but I thought the twisted cables were used for the floppy drives, not he HDs. Are you using a different controller? You could try what I did with my XT/AT, get the XT working with the drive, transfer over both controller and drive, and see if it boots. This may work. Good Luck. Donald St. Denis Signal Analysis Division
rps@homxc.UUCP (R.SHARPLES) (07/21/88)
In article <1988Jul19.100226.737@mntgfx.mentor.com>, mikes@mntgfx.mentor.com (Mike Stanbro) writes: > * The WX1 is an 8-bit peripheral, the WA2 is a 16-bit. > > * The configuration setup on the WX1 is jumper selectable, on the WA2 it is > electrically programmed at boot-up using the disk-type data stored in the > CMOS configuration RAM. > > * The WX1 is a hard-disk only controller, the WA2 is both a hard-disk > controller and a floppy controller. > > I have installed a lot of hard disks in both XT's and AT's but never > tried using a WX1 controller in an AT so I can't say that it is or > isn't possible. Even if it is possible to use the 8-bit controller, > I suspect there is a significant performance difference between the two > due to 2X data width and less wait states for the 16-bit controller. > > You can get a WA2 for about $140 if you shop around. We have several AT&T 6386s around here which have Western Digital 1006 WAH controllers in them. With these controllers they run with a 1:1 interleave. The CORE test program clocks the data transfer rate at 500kb/sec, about twice what my 28ms 70meg Core/WD1003-WA2 combination will do. These disks *without* the LIGHTNING disk caching program are as fast as mine with LIGHTNING! Needless to say, I would like to get a WD1006 WAH controller but I have never seen one advertised. Does anyone know about this controller, what it costs, who sells it? Russ Sharples homxc!rps NOTE: The above in NO WAY reflects the opinions of AT&T. These opinions are my own and the results of un-scientific and highly irregular analysis methods.