jim_d@cimcor.mn.org (Jim Dahlberg) (01/16/89)
I just installed a new Perstor hard disk controller in my Zenith 386 PC.
The controller is Perstor's new 16-bit advanced RLL model PS180-16F.
This controller gives me 90% more storage on my CDC Wren II model 51 hard
disk (originally on an MFM controller) resulting in 78 MBytes of storage.
The installation was fairly easy, and everything *does* work. The read
transfer rate went from about 160 KBytes/sec with the MFM controller to
399 KBytes/sec. I am very happy with this :-).
HOWEVER - There seems to be a problem with writing. My hard disk activity
light flashes on when I do a copy, for example, and then goes off for a few
seconds, then flashes on again, then stays off for a few seconds before I
get the DOS prompt. Copying a measily 178 KByte file takes about 6 seconds!
When copying multiple files to a floppy disk, the floppy disk activity
light goes out between files.
Anybody have any idea what might be happening?
Jim Dahlberg
Internet: jim_d@cimcor.mn.org
UUCP: uunet!rosevax!cimcor!jim_dsamc@hpccc.HP.COM (Samuel Chau) (01/21/89)
>HOWEVER - There seems to be a problem with writing. My hard disk activity >light flashes on when I do a copy, for example, and then goes off for a few >seconds, then flashes on again, then stays off for a few seconds before I >get the DOS prompt. Copying a measily 178 KByte file takes about 6 seconds! > >When copying multiple files to a floppy disk, the floppy disk activity >light goes out between files. >Anybody have any idea what might be happening? Sounds to me like a strange timeout problem. Perstor does have a new BIOS for the PS180-16F. According to their technical assistance, the new BIOS "traps reset calls", and just might fix your problem. I was warned that a low-level re-format may be necessary to accomodate the new ROM. Give Ted Buck a call at (602) 991-5451. Specifically mention your 386 motherboard hardware and hard drive. My copy of the new BIOS is expected to arrive today. It costs $25, less if you return your old ROM. I opted to keep the existing BIOS ROM in case the new one has other problems. I'll post again when I have the details. Sam Chau NSG/STD Cupertino 42LA samc@hpcup13 (408) 447-0238
samc@hpccc.HP.COM (Samuel Chau) (02/03/89)
Jim, I don't think I can reach you via e-mail, so I'm posting my response here. I too received BIOS 3.01 from Ted earlier this week. The "head banging" problem I experienced at boot time was gone, but was replaced by double slow seeks (drives making long "errrrrrrrrrr" sounds) just before the system boots up. Ted sent me two chips, a BIOS EPROM and a microcontroller with updated microcode. I'm pretty sure we have the same chips now, but I just want to verify this. I use the PS180-16F with a Mylex 386 motherboard at 20 MHz. I have a pair of Seagate ST-251-1 half-height drives (had to sort through some 5 drives to get a pair to work with the Perstor), and I use HP's DOS 3.3, which I believe is very close to Microsoft's MS-DOS 3.3. I also use SpeedStor from Storage Dimensions to gain access to all 78 Mbytes on each Seagate. It allows you to configure partitions of up to 512 Mbytes. It works with ST-506, ESDI and SCSI drives using MFM or RLL encoding. It is my favorite and I highly recommend it. I don't use the extended partitioning features of DOS 3.3, since I have no control over the cluster size, and it almost always sets the cluster size bigger than I would like to have. By the way, I believe if the Perstor controller is hooked up to only one drive, there is a 10-second "dead time" just before booting. It looks like the controller is trying to talk to the second drive, and doesn't give up and assume it's not there until 10 seconds later. It this what you're seeing? Ted reminded me that a low-level reformat may be necessary when the 3.01 BIOS is put in, but actually I didn't have to do that. You may also want to re-run the ATFSETUP routine to make sure the CMOS memory is properly configured for your drive(s). I hope this helps. I'll need to talk to Ted soon and I'll let you know if anything comes up. Sam Chau HP Cupertino samc@hpda #include <std_disclaimer.h>
jim_d@cimcor.mn.org (Jim Dahlberg) (02/05/89)
This is really to Samuel Chau, but I can't seem to be able to email directly
to him, so I'll post it here. It may be of benefit to someone else with a
similar problem now or in the future anyway.
I fixed the slow write problem with my Perstor PS180-16F HD
controller by getting a new board from the dealer. My original BIOS was
rev 2.05, and the new one is 3.01, which Ted @ Perstor tells me is the
latest.
Alas, I then had a different problem, which may relate to the new
BIOS "trapping reset calls" mentioned in Samuel Chau's previous
response. I then got an error when doing a warm boot (CTRL-ALT-DEL).
After the drive polling was complete, the computer went dead for a
minute or so, then a beep and an "ERROR FOUND. PRESS ESCAPE TO
CONTINUE" message was displayed. When I pressed ESC, the boot process
continued normally and everything was fine. Warm hardware resets did the
same thing, but power-off power-on "resets" worked OK.
I got in touch with Ted @ Perstor about this problem. He sent me a
new microcode ROM (not the BIOS ROM). The first time I tried it, the
computer booted into the Zenith ROM monitor program (definately NOT
normal, even for a Zenith machine), so I put the original ROM back in.
My local dealer advised me to change the drive controller cable to use
the untwisted connector end and change the drive unit number to 0. I
did this, and still had the warm boot problem. I then decided to try
the new microcode ROM one more time. This time it worked! I don't
understand why, or if changing the drive unit number really had anything
to do with it (logic tells me it shouldn't have), but everything now
functions properly.
I forgot to mention in my first posting that I am using the new
Zenith MS-DOS 3.3 Plus. This allows me to partition the entire 78
Mbytes the Perstor controller gives me, as one partition. It allows
partitions of up to 512 Mbytes.
Jim Dahlberg
Internet: jim_d@cimcor.mn.org
UUCP: uunet!rosevax!cimcor!jim_ddrew@geaclib.UUCP (Drew Sullivan) (02/08/89)
In article <5060047@hpccc.HP.COM> samc@hpccc.HP.COM (Samuel Chau) writes: >I too received BIOS 3.01 from Ted earlier this week. The "head banging" >problem I experienced at boot time was gone, but was replaced by double >slow seeks (drives making long "errrrrrrrrrr" sounds) just before the system >boots up. Ted sent me two chips, a BIOS EPROM and a microcontroller with >updated microcode. I'm pretty sure we have the same chips now, but I just >want to verify this. Be forwarned, if you are running xenix or novel with this controller you also have to have a set of wire-wrap changes to the board. There are some physical timing problems are not fixed by just the roms. I have the new wire-wrapped board running a maxtor(70Mb) and a miniscribe(80Mb) and am pleased to have 126 and 135Mb available (after formating) Lastly support was excellent, they returned my calls, shipped me new roms and latter a new board, and didn't charge me a cent. -- -- Drew Sullivan <drew@lethe.uucp> +1 416 225-1592 Toronto, Ontario, Canada
jim_d@cimcor.mn.org (Jim Dahlberg) (02/10/89)
I finally gave up on my Perstor controller, even tho I eventually got
all of the incompatibility problems fixed (slow writes & error on warm
boots). After they were fixed, I started to get data errors until the
computer warmed up. I had also read in a couple places (incl Feb Byte)
about what RLL and ARLL are, & that they require better drive
characteristics. I became worried that even if I got it working now, I
might start having problems later, after the drive started to age.
So thanks to an understanding local dealer, I was able to return the
card.
Jim Dahlberg
Internet: jim_d@cimcor.mn.org
UUCP: uunet!rosevax!cimcor!jim_d
UUCP: orbit!pnet51!jad