3 Feb
2023
3 Feb
'23
9:09 a.m.
Decided to spend some time working on my 11/73 with MFM drives.
Currently it has one of my RQDX3 boards (I have 3, 1 in attic), a 40mb
ST412 drive (the half height Seagate whatever) which works fine. No
issues there.
I'm trying to format an RD54 compatible drive and am running into major
issues. First, my two RWDX3's have different ROM dates, the old one is
1986 and the new one is 1990. This is important because I can't boot
RX33 disk images with my GoTek using the old card but I can using the
new one.
Question: I'm guessing the old ROMs only supported RX50 disks? Or is it
a secret jumper setting.
Anyway I do have both RX33 and RX50 versions of XXDP so not a big issue.
On to formatting.
The old controller (which I used for the 40mb Seagate) had pins 2-3
jumpered on W23. With that the RD54 was able to autoformat but then
would crash xxdp as soon as the initial format was done. Odd. So I used
the new controller with 1-2 and 3-4 jumpered. Same problem. Then I tried
having 1-2 jumpered and did a manual format (not autoformat, select
RD54, etc)
I noticed that on the old board it would ask me for the date when doing
this kind of format, on the new board it would just ask me for the
serial number. Odd.
Question: Is the ZRQCH0.BIN file calling different routines in the RQDX3
ROM?
Anyway after this the drive would format but then do endless seek errors
on the "read" portion of the disk check. Two drives did this, so it's
probably not the drives. Odd. Putting the drives on the Dave Gesswin MFM
reader showed all cylinders could be read.
Question: Can Dave G's board be used to low level format an RD54? Can it
test physical disk for errors (wasn't sure)
Now the drives only format for a minute or two before throwing errors.
Looks like something is very confused on XXDP. Not going to try any
other disks until I figure this out.
Thoughts? Different sites say different things about the RQDX3 jumpers,
some say to jumper 2-3 to allow more than 7 heads, some say to jumper
pins 1-2 and some say jumper pins 1-2 on "early ROM" and 1-2 3-4 on
"later ROM".
This is a serious pain, but just what settings should be done to allow
low level formatting, and did my previous attempts to low level wedge
these disks from the RQDX3 point of view? Can I do a low level wipe with
Dave Gesswin's board/software?
Thanks!
Chris
3 Feb
3 Feb
9:14 a.m.
On 2023-02-03 12:09, Chris Zach via cctalk wrote:
Decided to spend some time working on my 11/73 with
MFM drives.
Currently it has one of my RQDX3 boards (I have 3, 1 in attic), a 40mb
ST412 drive (the half height Seagate whatever) which works fine. No
issues there.
I'm trying to format an RD54 compatible drive and am running into
major issues. First, my two RWDX3's have different ROM dates, the old
one is 1986 and the new one is 1990. This is important because I can't
boot RX33 disk images with my GoTek using the old card but I can using
the new one.
Question: I'm guessing the old ROMs only supported RX50 disks? Or is
it a secret jumper setting.
Anyway I do have both RX33 and RX50 versions of XXDP so not a big
issue. On to formatting.
The old controller (which I used for the 40mb Seagate) had pins 2-3
jumpered on W23. With that the RD54 was able to autoformat but then
would crash xxdp as soon as the initial format was done. Odd. So I
used the new controller with 1-2 and 3-4 jumpered. Same problem. Then
I tried having 1-2 jumpered and did a manual format (not autoformat,
select RD54, etc)
I noticed that on the old board it would ask me for the date when
doing this kind of format, on the new board it would just ask me for
the serial number. Odd.
Question: Is the ZRQCH0.BIN file calling different routines in the
RQDX3 ROM?
Anyway after this the drive would format but then do endless seek
errors on the "read" portion of the disk check. Two drives did this,
so it's probably not the drives. Odd. Putting the drives on the Dave
Gesswin MFM reader showed all cylinders could be read.
Question: Can Dave G's board be used to low level format an RD54? Can
it test physical disk for errors (wasn't sure)
Now the drives only format for a minute or two before throwing errors.
Looks like something is very confused on XXDP. Not going to try any
other disks until I figure this out.
Thoughts? Different sites say different things about the RQDX3
jumpers, some say to jumper 2-3 to allow more than 7 heads, some say
to jumper pins 1-2 and some say jumper pins 1-2 on "early ROM" and 1-2
3-4 on "later ROM".
This is a serious pain, but just what settings should be done to allow
low level formatting, and did my previous attempts to low level wedge
these disks from the RQDX3 point of view? Can I do a low level wipe
with Dave Gesswin's board/software?
Thanks!
Chris
Have you tired manually telling ZRQC that it is an RD54?
Nigel
--
Nigel Johnson, MSc., MIEEE, MCSE VE3ID/G4AJQ/VA3MCU
Amateur Radio, the origin of the open-source concept!
Skype: TILBURY2591
9:15 a.m.
Yes, the no autoformat option allows me to specify RD54.
C
On 2/3/2023 12:14 PM, Nigel Johnson Ham via cctalk wrote:
On 2023-02-03 12:09, Chris Zach via cctalk wrote:
Decided to spend some time working on my 11/73
with MFM drives.
Currently it has one of my RQDX3 boards (I have 3, 1 in attic), a 40mb
ST412 drive (the half height Seagate whatever) which works fine. No
issues there.
I'm trying to format an RD54 compatible drive and am running into
major issues. First, my two RWDX3's have different ROM dates, the old
one is 1986 and the new one is 1990. This is important because I can't
boot RX33 disk images with my GoTek using the old card but I can using
the new one.
Question: I'm guessing the old ROMs only supported RX50 disks? Or is
it a secret jumper setting.
Anyway I do have both RX33 and RX50 versions of XXDP so not a big
issue. On to formatting.
The old controller (which I used for the 40mb Seagate) had pins 2-3
jumpered on W23. With that the RD54 was able to autoformat but then
would crash xxdp as soon as the initial format was done. Odd. So I
used the new controller with 1-2 and 3-4 jumpered. Same problem. Then
I tried having 1-2 jumpered and did a manual format (not autoformat,
select RD54, etc)
I noticed that on the old board it would ask me for the date when
doing this kind of format, on the new board it would just ask me for
the serial number. Odd.
Question: Is the ZRQCH0.BIN file calling different routines in the
RQDX3 ROM?
Anyway after this the drive would format but then do endless seek
errors on the "read" portion of the disk check. Two drives did this,
so it's probably not the drives. Odd. Putting the drives on the Dave
Gesswin MFM reader showed all cylinders could be read.
Question: Can Dave G's board be used to low level format an RD54? Can
it test physical disk for errors (wasn't sure)
Now the drives only format for a minute or two before throwing errors.
Looks like something is very confused on XXDP. Not going to try any
other disks until I figure this out.
Thoughts? Different sites say different things about the RQDX3
jumpers, some say to jumper 2-3 to allow more than 7 heads, some say
to jumper pins 1-2 and some say jumper pins 1-2 on "early ROM" and 1-2
3-4 on "later ROM".
This is a serious pain, but just what settings should be done to allow
low level formatting, and did my previous attempts to low level wedge
these disks from the RQDX3 point of view? Can I do a low level wipe
with Dave Gesswin's board/software?
Thanks!
Chris
Have you tired manually telling ZRQC that it is an RD54?
Nigel
10:20 a.m.
On Fri, Feb 03, 2023 at 12:09:27PM -0500, Chris Zach wrote:
Question: Can Dave G's board be used to low level format an RD54? Can it
test physical disk for errors (wasn't sure)
The read test you did should have told you if there were sectors with errors.
Using my tools to low level format is only sort of possible. DEC got overly
creative with their disk format so some sectors are formmatted differently for
controller use. I've haven't written code to be able to reproduce that.
If you have an image of a good disk mfm_write can write that to another drive.
Somewhat fiddly. Problem with that is it will not have the bad sectors on
the new drive properly mapped out so you will get read errors. If there is
a tool you can run to test the disk and mark new bad sectors that should fix
that problem.
I do have an RD54 image that can be used with mfm_write. Search in page for RD54
http://www.pdp8online.com/mfm/status.shtml
10:51 a.m.
I do have an RD54 image that can be used with
mfm_write. Search in page for RD54
http://www.pdp8online.com/mfm/status.shtml
Ok. Hm. So the command would be ./mfm_write --emulation RD54_A -d3?
This is possibly better than nothing: I could use this to put some sort
of format on the drive, then try letting xxdp reformat it. Right now I'm
getting this:
DR>START
CHANGE HW (L) ? Y
# UNITS (D) ? 1
UNIT 0
Enter controller IP Address (O) 172150 ?
What unit do you want to format [0-255] (D) 0 ?
Would you like to revector a single LBN only [Y/N] (L) N ?
Do you want to use the "AUTOFORMAT" Mode [Y/N] (L) Y ?
Enter unit serial number [1-32000] (D) 12345 ?
**** WARNING ****
ALL DATA ON SELECTED DRIVE WILL BE DESTROYED
Write protect all drives not being formatted.
Please verify that the selected drive is ON LINE
and NOT write protected.
If formatting RX33 media, insert media to be
formatted in the selected drive.
Do you wish to continue [Y/N] (L) Y ? Y
Autosizer found:
Unit Cylinders Drive Name
0 1225 RD54
1 RX33 Diskette (FORMATABLE)
MSCP Controller Model: 19
Microcode Version: 1
Formatting of Drive 0 Begun.
(Oddly enough the old controller doesn't display the progress. I really
think it's just talking to the RQDX3 and that is calling the shots as
the later ROMs do tell you how far it is and what it's doing which is nice)
ZRQC SYS FTL ERR 00007 ON UNIT 00 TST 001 SUB 000 PC: 105742
Controller has reported a fatal error in the FORMAT program.
Status: RCT write error
Drive 0 was not formatted successfully.
ZRQC EOP 1
1 TOTAL ERRS
11:39 a.m.
Wow, this is interesting Dave. Took one of the "formatted badly" disks
and did a write of your file.
root@beaglebone:~/mfm# ./mfm_write --emulation RD54_A -d3 -c1224 -h15 -s 17
Board revision C detected
Then a read of the disk:
root@beaglebone:~/mfm# ./mfm_read --analyze -e rd.dsk
Board revision C detected
Found drive at select 3
Returning to track 0
Drive RPM 3596.5
Matches count 34 for controller DEC_RQDX3
Header CRC: Polynomial 0x1021 length 16 initial value 0xffff
Sector length 512
Data CRC: Polynomial 0x1021 length 16 initial value 0xffff
Interleave mismatch previous entry 0, 10 was 2 now 0
Number of heads 15 number of sectors 17 first sector 0
Unable to determine interleave. Interleave value is not required
Drive supports buffered seeks (ST412)
Found cylinder 200 expected 1224
Disk has recalibrated to track 0
Stopping end of disk search due to recalibration
Number of cylinders 1224, 159.8 MB
Command line to read disk:
--format DEC_RQDX3 --sectors 17,0 --heads 15 --cylinders 1224
--header_crc 0xffff,0x1021,16,0 --data_crc 0xffff,0x1021,16,0
--sector_length 512 --retries 50,4 --drive 3
Retries failed cyl 0 head 0
Bad sectors on cylinder 0 head 0: 0 1 2
Retries failed cyl 0 head 3
Bad sectors on cylinder 0 head 3: 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 4
Bad sectors on cylinder 0 head 4: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 5
Bad sectors on cylinder 0 head 5: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 6
Bad sectors on cylinder 0 head 6: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 7
Bad sectors on cylinder 0 head 7: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 8
Bad sectors on cylinder 0 head 8: 0 1 2 3 4 5 6 7 8 9 10 11 12H 13 14 15 16
Retries failed cyl 0 head 9
Bad sectors on cylinder 0 head 9: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 10
Bad sectors on cylinder 0 head 10: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 11
Bad sectors on cylinder 0 head 11: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 12
Bad sectors on cylinder 0 head 12: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 13
Bad sectors on cylinder 0 head 13: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
So looks like the disk is either not taking the format or something is
very very wrong. Wonder what happened to all these disks....
C
11:48 a.m.
Wow, this is interesting Dave. Took one of the "formatted badly" disks
and did a write of your file.
root@beaglebone:~/mfm# ./mfm_write --emulation RD54_A -d3 -c1224 -h15 -s 17
Board revision C detected
Then a read of the disk:
root@beaglebone:~/mfm# ./mfm_read --analyze -e rd.dsk
Board revision C detected
Found drive at select 3
Returning to track 0
Drive RPM 3596.5
Matches count 34 for controller DEC_RQDX3
Header CRC: Polynomial 0x1021 length 16 initial value 0xffff
Sector length 512
Data CRC: Polynomial 0x1021 length 16 initial value 0xffff
Interleave mismatch previous entry 0, 10 was 2 now 0
Number of heads 15 number of sectors 17 first sector 0
Unable to determine interleave. Interleave value is not required
Drive supports buffered seeks (ST412)
Found cylinder 200 expected 1224
Disk has recalibrated to track 0
Stopping end of disk search due to recalibration
Number of cylinders 1224, 159.8 MB
Command line to read disk:
--format DEC_RQDX3 --sectors 17,0 --heads 15 --cylinders 1224
--header_crc 0xffff,0x1021,16,0 --data_crc 0xffff,0x1021,16,0
--sector_length 512 --retries 50,4 --drive 3
Retries failed cyl 0 head 0
Bad sectors on cylinder 0 head 0: 0 1 2
Retries failed cyl 0 head 3
Bad sectors on cylinder 0 head 3: 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 4
Bad sectors on cylinder 0 head 4: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 5
Bad sectors on cylinder 0 head 5: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 6
Bad sectors on cylinder 0 head 6: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 7
Bad sectors on cylinder 0 head 7: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 8
Bad sectors on cylinder 0 head 8: 0 1 2 3 4 5 6 7 8 9 10 11 12H 13 14 15 16
Retries failed cyl 0 head 9
Bad sectors on cylinder 0 head 9: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 10
Bad sectors on cylinder 0 head 10: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 11
Bad sectors on cylinder 0 head 11: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 12
Bad sectors on cylinder 0 head 12: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Retries failed cyl 0 head 13
Bad sectors on cylinder 0 head 13: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
So looks like the disk is either not taking the format or something is
very very wrong. Wonder what happened to all these disks....
C
12:55 p.m.
Running a non-reformatted disk through Dave G's mfm device got me the
following after running a mfm_write to the whole disk followed by an
mfm_read:
Expected 312120 sectors got 311766 good sectors, 282 bad header, 72 bad data
0 sectors marked bad or spare
49 sectors corrected with ECC. Max bits in burst corrected 2
Track read time in ms min 18.260750 max 2098.231833 avg 35.600018
So.... 354 bad sectors, and 49 more that were only readable with ECC.
That's not too great, I think I'll shelve the Maxtor drives and just go
with ESDI when needed. Those seem to be a *lot* more reliable than MFM.
C
7:12 p.m.
Well, formatting an RD53 seems to be working. And the rev 4 ROMs are
much nicer than the rev 3 in terms of talking to you.
Unit Cylinders Drive Name
0 1024 RD53
1 RX33 Diskette (FORMATABLE)
MSCP Controller Model: 19
Microcode Version: 4
Formatting of Drive 0 Begun.
------------ FORMAT PROGRESS REPORT -------------
1 minute into format ---- Formatting tracks, LBN # 28816
2 minutes into format ---- Formatting tracks, LBN # 57682
3 minutes into format ---- Formatting tracks, LBN # 86514
4 minutes into format ---- Formatting tracks, LBN # 115363
5 minutes into format ---- Reading defect list
6 minutes into format ---- Replacing defect # 3 on head # 0
7 minutes into format ---- Replacing defect # 7 on head # 0
8 minutes into format ---- Replacing defect # 1 on head # 1
9 minutes into format ---- Reading defect list
10 minutes into format ---- Replacing defect # 1 on head # 7
11 minutes into format ---- Third check pass, writing LBN # 31399
12 minutes into format ---- Third check pass, writing LBN # 62815
13 minutes into format ---- Third check pass, writing LBN # 93959
14 minutes into format ---- Third check pass, writing LBN # 125239
15 minutes into format ---- Third check pass, reading LBN # 17935
16 minutes into format ---- Third check pass, reading LBN # 49487
17 minutes into format ---- Third check pass, reading LBN # 80495
18 minutes into format ---- Third check pass, reading LBN # 104336
19 minutes into format ---- Third check pass, reading LBN # 126055
Format Completed.
00012 Rev LBNs
00000 Bad RBNs
00000 Bad DBNs
00000 Bad XBNs
00001 retired
FCT used successfully.
Drive 0 has been formatted successfully.
ZRQC EOP 1
0 TOTAL ERRS
On 2/3/2023 1:20 PM, David Gesswein via cctalk wrote:
On Fri, Feb 03, 2023 at 12:09:27PM -0500, Chris Zach
wrote:
Question: Can Dave G's board be used to low level format an RD54? Can it
test physical disk for errors (wasn't sure)
The read test you did should have told you if there were sectors with errors.
Using my tools to low level format is only sort of possible. DEC got overly
creative with their disk format so some sectors are formmatted differently for
controller use. I've haven't written code to be able to reproduce that.
If you have an image of a good disk mfm_write can write that to another drive.
Somewhat fiddly. Problem with that is it will not have the bad sectors on
the new drive properly mapped out so you will get read errors. If there is
a tool you can run to test the disk and mark new bad sectors that should fix
that problem.
I do have an RD54 image that can be used with mfm_write. Search in page for RD54
http://www.pdp8online.com/mfm/status.shtml
454
days inactive
454
days old
9 comments
5 participants
participants (5)
-
Chris Zach -
Chris Zach -
David Gesswein -
emanuel stiebler -
Nigel Johnson Ham