Scorpion HDD

I don’t understand what’s going on here. The partition does not want to be mounted programmatically. And when
trying to get the descriptor of the connected partition, I get an error
Tape loading error.

Examples:
...
ld c,35
ld a,%00001011
ld de,txt
rst 8
db #81
...

txt db "d:\tmp \Disk01"

After this, when I try to read the directory, I get the same disk that was
connected. There is no error in CY.

...
ld c,35
ld a,%00010011
ld hl,descr
rst 8
db #81
...

descr ds 25

This is where Tape loading error appears.

What am I doing wrong?

ZXNet: 500:8462/1 ~/ <~ http://www.zx.ru/zs/
Fido: 2:5057/21.93 /_ _> ftp://zx-museum.org.ru/zs/

One Sunday, May 27, 2001 at 16:20:40 Sergey Zotov wrote a letter to
All about Scorpion HDD:

SZ> Here I don’t understand what’s going on. The partition does not want to be mounted programmatically
SZ>.

[skip...]

SZ> what am I doing wrong?

You are doing everything correctly - the ProfPZU itself is not working correctly. This is a well-known glitch that I once had to solve.
In general, there is a special sector on the screw that stores information about
connected partitions. If you switch sections through the shadow manually, then all
information in this sector is updated normally. But if you use
the RST #8 subroutines, then the information will be updated only in the eighth bank, and with
any access to TR-DOS through 15619 and in several other cases, the information
will be restored:(
The only way out is to rewrite the information in the settings sector programmatically,
that is, through the screw ports. There are a few more subtleties, such as calculating
checksums, etc., but this is already little things...
In general, this topic was well covered in the 53rd or 54th ZXNews.
If you are interested in anything, please contact me.

I wish you good luck!
Sincerely, Fyodor Odegov.

FO> You are doing everything right - ProfPZU itself is not working correctly. This
FO> is a known glitch that I once had to solve.

Bring information to the public.

FO>> ProfPZU. This is a well-known glitch that I once
FO>> had to solve.
KF> Bring information to the public.

Everything is simple here: when switching partitions on the screw via TM, all information
about them is recorded in the emulation sector and remains in the eighth bank. That is,
problems do not arise here.
But if you switch sections through the RST #08 functions, the information will be updated
only in the eighth bank, and nothing will change in the emulation sector (this can be
noted even visually: the screw access indicator will not blink). If after
this you read from the connected subpartition using the RST8 or #3D13 functions, then
everything will be fine, but if you just turn to the screw on 15619, then reading
will occur from the previous connected partition.
Way out of the problem: rewrite the emulation sector manually. Since it is not accessible using RST8 functions, you will have to rewrite it using direct access to SMUC ports.
I will not dwell on this in detail: everything has already been written in ZXNews54 (there is a great article by Vlad Sotnikov about this). Those who need it will read it.

I wish you good luck!
Sincerely, Fyodor Odegov aka FION.