Atari 1050 Double Density

[Jack Tseng]

Atari 1050 is an enhanced density drive. ICD, Indus, Rana, and Trak can make
the double density drive but why can not Atari company make the double
density drive?

Thanks!
Jack

[Docrotcod]

The XF-551 was made by Atari and was double side/double density.

Post by Jack Tseng
Atari 1050 is an enhanced density drive. ICD, Indus, Rana, and Trak can
make the double density drive but why can not Atari company make the
double density drive?
Thanks!
Jack

[Larry]

Post by Jack Tseng
Atari 1050 is an enhanced density drive. ICD, Indus, Rana, and Trak can make
the double density drive but why can not Atari company make the double
density drive?
Thanks!
Jack

Atari was allegedly concerned about reliability in DD, and probably
with good reason. For that era of drives, DD was/is much more
difficult to make consistently reliable. Most any Atari drive in good
condition will read most any other Atari SD disk. That can't be said
of DD. My own experience with Rana, for instance, was many would not
even read their own disks in DD, let alone disks from on other drives.
So Atari "split the difference" electing to go with MFM, but 128 byte
sectors. As it turned out, the 1050 (and at least its Tandon
mechanism) was reliable enough for DD. Indus, too. If you want to
better understand the perils, pitfalls, and other problems with drives,
read "Beneath Apple Dos." Excellent and understandable, too.
-Larry

[TJ]

Post by Larry

Post by Jack Tseng
Atari 1050 is an enhanced density drive. ICD, Indus, Rana, and Trak can make
the double density drive but why can not Atari company make the double
density drive?
Thanks!
Jack

Atari was allegedly concerned about reliability in DD, and probably
with good reason. For that era of drives, DD was/is much more
difficult to make consistently reliable. Most any Atari drive in good
condition will read most any other Atari SD disk. That can't be said
of DD. My own experience with Rana, for instance, was many would not
even read their own disks in DD, let alone disks from on other drives.
So Atari "split the difference" electing to go with MFM, but 128 byte
sectors. As it turned out, the 1050 (and at least its Tandon
mechanism) was reliable enough for DD. Indus, too. If you want to
better understand the perils, pitfalls, and other problems with drives,
read "Beneath Apple Dos." Excellent and understandable, too.
-Larry

ICD sold the "US Doubler" 1050 upgrade that moved the drive up to full
double density. IIRC, it consisted of a replacement controller chip and
it seems like you had to check a jumper position. I had a drive with
that upgrade, as well as a couple of Indus drives and a couple of
XF551s. None had any trouble reading disks created with the others. (As
long as you stuck with single-sided in the XF, of course.) I used MyDOS
almost exclusively with my DD disks, if that makes any difference.

TJ

[Rick Cortese]

Post by Jack Tseng
Atari 1050 is an enhanced density drive. ICD, Indus, Rana, and Trak can make
the double density drive but why can not Atari company make the double
density drive?
Thanks!
Jack

As Larry said: Actually if I have it right, Atari had problems even
getting any drive to work. I'm not sure about which models but I am
pretty sure that at least one, like the 1050, was developed out of house
by an outfit like Tandon. At any rate, the 'not invented here' drive won
out when the in house drive failed. Compare the rock solid PC board,
shielding, etcetera, of the 1050 to something like the PC board made out
of saltine crackers in and flaky connectors in the XF551 to see what we
probably would have got from Atari.

I think I recall Bill Wilkinson saying the cost of the in house drives
like the 815 was so high that you could buy 2 1050s for the same price.
I imagine the 810 was an expensive drive to manufacture too and thus the
incentive to replace it.

The 1050 enhanced density is double density. Double density refers to
the bit density, ~bits per inch, laid down on the disk. There were all
kinds of single and double density formats back then.

Atari was limited in that they only had 256 bytes of RAM in a 6810 and a
PIA type chip. Can't exactly have a 6502 stack and still have 256 bytes
available for a write buffer. The XF gets away with it because the
different chip, 8052<?>, has a lot of general purpose registers that can
be used for storage of variables and what not.

I am of the opinion Atari would not have given us a drive with a lot of
ram in it even though it made sense. For one thing, you would only need
to add a single command to the ROM to get a drive that could defeat a
lot of copy protection schemes. Just something simple like load RAM with
code then JSR to it.

[Kryten]

I am pretty sure that at least one, like the 1050, was developed out of
house by an outfit like Tandon.

My 1050 had (c) Tandon on the EPROM sticker.

Atari was limited in that they only had 256 bytes of RAM in a 6810 and a
PIA type chip. Can't exactly have a 6502 stack and still have 256 bytes
available for a write buffer. The XF gets away with it because the
different chip, 8052<?>, has a lot of general purpose registers that can
be used for storage of variables and what not.

http://homepage.ntlworld.com/kryten_droid/Atari/800XL/1050/1050/1050_FDD.htm

There is RAM in the 6810 (128 bytes) and the 6532 (128 bytes RAM, I/O,
Timer) chips.

http://en.wikipedia.org/wiki/6532

I am of the opinion Atari would not have given us a drive with a lot of
ram in it even though it made sense. For one thing, you would only need to
add a single command to the ROM to get a drive that could defeat a lot of
copy protection schemes. Just something simple like load RAM with code
then JSR to it.

I copied a 'disk doubler' using a spare 6502, EPROM and a 2K RAM (6116).
It would be easy for me to fit more RAM than the Atari 800 had :-)

[Matthias Reichl]

Post by Rick Cortese
Atari was limited in that they only had 256 bytes of RAM in a 6810 and a
PIA type chip. Can't exactly have a 6502 stack and still have 256 bytes
available for a write buffer.

There exists an extension for the 1050 which does this trick: the
1050 Turbo module by Bernhard Engl (released in 1986). It's just a
replacement for the 1050 ROM and consists only of a ROM chip plus
some logic ICs to do bankswitching.

I've always wondered how this really worked (yes, you need a 256 byte
buffer if you want to use 256 byte sectors). Some time ago I talked
to Bernhard Engl and he explained the trick: If you can live with
a subroutine level of 1 you don't really need a stack. Instead he
just used the stack-pointer as kind of a function index.

The software in the 1050 (turbo) is quite straight forward and using
only one subroutine level is sufficient. Consider what happens eg
when reading a sector: first the 1050 receives the command frame,
then it has to read either 128 or 256 bytes from disk, then it
transmits the data frame (either 128 or 256 data bytes) to the
computer.

After command frame reception, the software knows if the sector
is either 128 or 256 bytes. So it sets up the stack pointer register
to, lets say, 1 for "send 128 bytes" and 2 for "send 256 bytes"
and then jumps to the read-disk-sector subroutine. At the end
of the routine, it checks the S register and then jumps to
the right "send bytes" routine.

The fine thing about the 1050 Turbo was that it was a lot cheaper
than all the other floppy-speeders (like happy/speedy/...). It cost
only 99 DEM and with 49 DEM more you could add a printer interface
(this interface was connected directly to the module inside the
1050). And, the 1050 Turbo came with one of the most sophisticated
backup tools. Very fine thing and it was also very popular here
in Europe (especially Germany and Austria).

so long,

Hias

[Jochen Schäfer]

Post by Matthias Reichl

Post by Rick Cortese
Atari was limited in that they only had 256 bytes of RAM in a 6810 and a
PIA type chip. Can't exactly have a 6502 stack and still have 256 bytes
available for a write buffer.

...

Post by Matthias Reichl
The fine thing about the 1050 Turbo was that it was a lot cheaper
than all the other floppy-speeders (like happy/speedy/...). It cost
only 99 DEM and with 49 DEM more you could add a printer interface
(this interface was connected directly to the module inside the
1050). And, the 1050 Turbo came with one of the most sophisticated
backup tools. Very fine thing and it was also very popular here
in Europe (especially Germany and Austria).

Fine thing. I had one with the printer cable. But it's dead.

Is there any chance that Bernhard Engl would release the layout and
software for this thing?


Bye Jochen

[Matthias Reichl]

Post by Jochen Schäfer

[... 1050 Turbo ...]

Fine thing. I had one with the printer cable. But it's dead.
Is there any chance that Bernhard Engl would release the layout and
software for this thing?

Good thing is: this already happened :-)

Bernhard Engl gave permission to ABBUC to release all the documentation
to his products and also to build new versions of the 1050 Turbo
and the Turbo Freezer. I built a highly improved, new version of
the TurboFreezer which is also available from ABBUC. The documentation
on my webpage isn't complete yet but I'm working on it.

You can find the documentation to the original products in the
ABBUC infothek (most docs are in german, but this shouldn't be
a too big problem for you, I guess :-)

http://www.strotmann.de/twiki/bin/view/Infothek/1050Turbo

so long,

Hias

[Jochen Schäfer]

Post by Matthias Reichl

Post by Jochen Schäfer

[... 1050 Turbo ...]

Fine thing. I had one with the printer cable. But it's dead.
Is there any chance that Bernhard Engl would release the layout and
software for this thing?

Good thing is: this already happened :-)
Bernhard Engl gave permission to ABBUC to release all the documentation
to his products and also to build new versions of the 1050 Turbo
and the Turbo Freezer. I built a highly improved, new version of
the TurboFreezer which is also available from ABBUC. The documentation
on my webpage isn't complete yet but I'm working on it.
You can find the documentation to the original products in the
ABBUC infothek (most docs are in german, but this shouldn't be
a too big problem for you, I guess :-)
http://www.strotmann.de/twiki/bin/view/Infothek/1050Turbo

Wow, this is great news. Now I know why I kept the dead board of my
1050Turbo.

Jochen