| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 549.1 | I'm not surprised | CANDID::steph | Constants aren't. Variables don't. | Tue Jul 25 1989 11:24 | 4 |
| I never got TURBODOS to work at all. I suppose that means I was lucky
to start with :^).
Steph
|
| 549.2 | DGC controller takes a kick or two | PRNSYS::LOMICKAJ | Jeff Lomicka | Tue Jul 25 1989 13:53 | 9 |
| TURBODOS will fail if it gets errors on it's first I/O to the hard
drive. In order to get TURBODOS to work reliably with the DGC disk
adaptor (thet's you,right?), I needed to have a program run BEFORE
Turbodos in the Auto folder that would try to open a file
(C:\DESKTOP.INF) twice. It fails the first try, but works on the
second try, and after that, TURBODOS loads and runs just fine.
I still use it when repeating a compile/edit/link cycle, but not for
backups or anything involving large data files.
|
| 549.3 | I'm sticking to FATSPEED. | CANDID::steph | Constants aren't. Variables don't. | Tue Jul 25 1989 20:19 | 11 |
| Oh. Since I tried TURBODOS, I wrote a program which does a RWABS of
block zero so that the next read (the read by the DESKTOP) would
succeed. Probably it would would ``work'' now.
However, I've felt unexperimental ever since I trashed a partition with
that fsck utility (didn't wreck anything important, but I decided that
I would be slow but a little safer).
Fatspeed gives a good speed-up by itself.
Steph
|
| 549.4 | | PRNSYS::LOMICKAJ | Jeff Lomicka | Wed Jul 26 1989 11:30 | 5 |
| What I suspect is going on is that some random I/O by the boot ROM is
putting the DGC adaptor into a funny state that it takes it a kick to
get out of. I think the code for this should go in the driver. One of
these days...
|
| 549.5 | This is what I found when I was building the interface. | CANDID::steph | Constants aren't. Variables don't. | Wed Jul 26 1989 18:11 | 18 |
| Actually, my considered opinion about this, at least on my system, is
that the first read after a reset or a power-fail fails (SCSI check
condition), with the extended status indicating that the drive was
reset. The driver is not currently written to handle this.
My drive is a native SCSI Quantum, though.
I suggested to Alan Pratt that the boot ROM try several times at each
SCSI address (it currently performs one read per address), before
failing. He got peeved enough about this that he responded to it on
the USENET, saying ``this is probably the right thing to do, but we're
not going to do anything about it now''.
So we can't boot from the hard disk. Personally, I could care less about this.
Anyhow, that's my understanding.
Steph
|
| 549.6 | Turbodos incompatible with HDX version 3 | CIM1NI::POWERS | I Dream Of Wires - G. Numan | Wed Aug 30 1989 12:22 | 15 |
|
>After using TURBODOS for two years with no noticable problems, this
>program has been removed from my AUTO folder!
I just removed it this past weekend, As I installed Atari's new Hard
disk software (HDX) version 3, and when I use turbodos, my machine forgets
it has a hard disk, and the partitions become unavailable.
By the way, the new HDX allows for partitions larger than 16 meg, and
you can I believe have more than four now.
Bill Powers
|