[cctalk] Re: Formatting and using RQDX3's and 2190's.
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
[cctalk] Re: Formatting and using RQDX3's and 2190's.
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 0x Sector length 512 Data CRC: Polynomial 0x1021 length 16 initial value 0x 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 0x,0x1021,16,0 --data_crc 0x,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
[cctalk] Re: Formatting and using RQDX3's and 2190's.
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
[cctalk] Re: Formatting and using RQDX3's and 2190's.
On 2023-02-03 12:09, Chris Zach via cctalk wrote: Decided to spend some time working on my 11/73 with MFM drives. You don't have a MicoVAX 2000 sitting around? Did you use the RD54 by any chance in a different system, and low level formatted it there?
[cctalk] Re: Formatting and using RQDX3's and 2190's.
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 0 Bad RBNs 0 Bad DBNs 0 Bad XBNs 1 retired FCT used successfully. Drive 0 has been formatted successfully. ZRQC EOP1 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
[cctalk] Re: Formatting and using RQDX3's and 2190's.
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 0x Sector length 512 Data CRC: Polynomial 0x1021 length 16 initial value 0x 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 0x,0x1021,16,0 --data_crc 0x,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
[cctalk] Re: Formatting and using RQDX3's and 2190's.
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 7 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 EOP1 1 TOTAL ERRS
[cctalk] Re: Formatting and using RQDX3's and 2190's.
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
[cctalk] Re: Formatting and using RQDX3's and 2190's.
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