This is Micropolis-1578-with-NeXT? in view mode; [Up]
Date: Sun 25-Nov-1991 14:22:33 From: meyergru@Informatik.TU-Muenchen.DE (Uwe Meyer-Gruhl) Subject: Micropolis 1578 with NeXT? Hi Nexters, me again with a new problem: I have two 330 Mbyte harddisks, a MICROPOLIS 1578 and a MICROPOLIS 1384. One is 5.25" full height, the other one is half height. Both worked perfectly well for years with an AT-compatible configured with an ADAPTEC 1542A controller. On my NeXTstation, however, they don't work. They are recognized, I can format them with 'sdform', I can even read data off them. What fails is writing to /dev/rsd1a. Interestingly, I am even able to 'write' with the 'disk' utility. The disk label and boot block are seemingly written with direct scsi commands. When I issue the 'init' command from within 'disk', all goes well up to the point where the 'newfs' is tried. Reading the disk is fine, 'dd if=/dev/rsd1a of=/dev/null bs=512 count=10' works. I tried also to copy the disk to itself, with: 'dd if=/dev/rsd1a of=/dev/rsd1a bs=512 count=10' which failed. I tried several blocksizes, to no avail. This command works with /dev/rsd0a (which is a FUJITSU M2624), but with none of the two MICROPOLIS drives. The message that is listed on the console states: 'Nov 24 18:23:00 mgsoft mach: sd1: Incomplete disk transfer; bytes moved = 0x3f2, resid = 0x200, retry 1' followed by 8 further retries, followed by a FATAL message. When I increase the block size to 1024, the bytes moved and residual are both '0x400'. The residual is always the block size given, the 'bytes moved' are always the greater of block size and 0x3f2. I have tried many jumper settings on the drive, but since I have no manual this has lead to no better result. So, my question is: Has anybody successfully installed a MICROPOLIS drive, does someone know the jumper settings for one or both of the 1578 and 1384(?), do I have to purchase a newer bios for the harddisks, is MICROPOLIS reachable via Internet or is it useless trying this at all? Some other things I've done are: Formatting with Rory Bolt's "formatter". Setting blocksize via MODE SELECT. All worked well. Maybe I just have to set the right MODE SELECT parameter? desparately needing more space, Uwe P.S.: I have also experienced problems using a FUJITSU and a MICROPOLIS drive connected to the same FUTURE DOMAIN controller. Strange, huh? P.P.S.: It seems, that the "c't" (a german computer magazine) is right when they state that the I/O part of the NeXT isn't programmed too well. While it might be asked too much to have a synchronous driver, I think it is not necessary to waste half of the time doing whatever is not I/O. Matter-of-fact these MICROPOLIS drives work at about 1 MByte/sec when connected to the PC. They rated at about 450 KByte/sec using 'disk' on the NeXT. And I think you can't compare faster drives, because the main PC bottleneck is the bus, which results in no greater speed than 1 MByte/sec at all, whereas I get 1.4 MByte/sec with my FUJITSU on the NeXT (values given are vaild for my configuration, maybe other controllers are faster, what I mean is that it should be possible to reach _at_least_ the PC figures). Maybe, though, the figures were distorted by some other influence factors, e.g. the drive not yet working right (see above). ----------------------------------------------------------------------------- ! Uwe Meyer-Gruhl ! ! ! Institut fuer Informatik ! I don't want to start any ! ! ! blasphemous rumours, but I think ! ! TU Muenchen ! that God's got a sick sense of humor ! ! Orleansstr. 34 ! and when I die, I expect to find him ! ! D-8000 Muenchen 80 ! laughing... ! ! FRG ! (Depeche Mode) ! ! ! ! voice: +49 89 48095-209 ! ! email: MeyerGru@gsradig5.Informatik.TU-Muenchen.DE ! -----------------------------------------------------------------------------
These are the contents of the former NiCE NeXT User Group NeXTSTEP/OpenStep software archive, currently hosted by Marcel Waldvogel and Netfuture.ch.