RX50 format
A.R. Duell
>In article <47147p$4...@mirv.unsw.edu.au>, sh...@solution.maths.unsw.edu.au () writes:
>> Just wondering what this RX50 format is, who invented it, who uses it and why.
>> Where did it come from and what is it compatible with?
>The RX50 format is 10 512-byte sectors per track, single-sided, 80 tracks,
>double-density. Digital used it in their PCs.
It was also used in some MicroPDP11's and MicroVaxen. Of course, you might
consider those to be PC's.
>As near as I can figure, it was invented as an attempt to save money on
>floppy drive hardware (not that you could tell based on how much DEC
I thought the main reason for the 2-drives-in-one-unit was that the earlier
DEC drives, the RX01 and RX02, has also been double drive units. The
reason it was built into 1 drive's worth of space was that it allowed the
2 floppies _and_ a winchester in the BA23 cabinet.
>charged for the drives...). The RX50 drive has two diskette drives in
>a full-height 5.25" drive bay. Both drives use the same spindle motor and
>head stepping motor; that is, if you move the heads on drive A, drive B
>moves as well. Mechanically, the drives contain about as much stuff as a
>normal double-sided 5.25" 96TPI drive.
>The drive mechanical stuff is located between the two diskette slots. I
>don't think my ASCII art skills are up to the job. Since the heads are
There's a wonderfully heath-robinsonish (Rube Goldberg?) drive belt that
drives the 2 hubs in opposite directions from a motor that's fitted
hoirzontally behind them.
>in between the drive slots, you stick the diskettes in upside down
>from each other; in order to move a diskette from drive A to drive B, you
>have to flip it over befor inserting it in drive B.
There are red lines on the front of the drive that you line up with red
marks on genuine DEC disks to help you do that.
>The driver is mostly plastic, so it's easy to knock out of alignment if
The RX50's I've worked on a mostly metal - metal chassis (which holds all
the bits that need to be in alignment and a very solid metal front
casting. The plastic bits are the side frames and the disk clamp arms.
>you're frogging around in the box. I have several drives that can't read
>each others' disks. The drives can also be damaged by diskettes with
>hub rings; the hub rings eventually ruin the plastic dealies that clamp
>the diskette to the spindle.
I've never had that - can you give me more info?
>Roger Ivie | "Once again we see that clowning and anarchy
--
-tony
ar...@eng.cam.ac.uk
The gates in my computer are AND,OR and NOT, not Bill
--
-tony
ar...@eng.cam.ac.uk
The gates in my computer are AND,OR and NOT, not Bill
David Evans
<sh...@solution.maths.unsw.edu.au> wrote:
>
>Just wondering what this RX50 format is, who invented it, who uses it and why.
>Where did it come from and what is it compatible with?
>
Sounds like the DEC floppies used on the Rainbow, the Professional series
("Two Professionals: You and the Computer". Indeed.), and I think PDPs and
some VAXen as well.
Sorry I can't help much more than that.
--
David Evans (NeXTMail OK) dfe...@bbcr.uwaterloo.ca
Computer/Synth Junkie http://bbcr.uwaterloo.ca/~dfevans/
University of Waterloo "Default is the value selected by the composer
Ontario, Canada overridden by your command." - Roland TR-707 Manual
Roger Ivie
> Just wondering what this RX50 format is, who invented it, who uses it and why.
> Where did it come from and what is it compatible with?
The RX50 format is 10 512-byte sectors per track, single-sided, 80 tracks,
double-density. Digital used it in their PCs.
As near as I can figure, it was invented as an attempt to save money on
floppy drive hardware (not that you could tell based on how much DEC
charged for the drives...). The RX50 drive has two diskette drives in
a full-height 5.25" drive bay. Both drives use the same spindle motor and
head stepping motor; that is, if you move the heads on drive A, drive B
moves as well. Mechanically, the drives contain about as much stuff as a
normal double-sided 5.25" 96TPI drive.
The drive mechanical stuff is located between the two diskette slots. I
don't think my ASCII art skills are up to the job. Since the heads are
in between the drive slots, you stick the diskettes in upside down
from each other; in order to move a diskette from drive A to drive B, you
have to flip it over befor inserting it in drive B.
The driver is mostly plastic, so it's easy to knock out of alignment if
you're frogging around in the box. I have several drives that can't read
each others' disks. The drives can also be damaged by diskettes with
hub rings; the hub rings eventually ruin the plastic dealies that clamp
the diskette to the spindle.
The speed tolerance for the drive is not tight enough to allow the drive
to format diskettes. If I recall correctly, you need to have a drive
within 1% of 300 RPM to format the drive, but the RX50 is specced at
1.5% speed tolerance. That's not a problem reading or writing diskettes
normally as the data separator will track the drive's speed variations,
but that doesn't work for formatting; since you're not reading data off
the drive while formatting, there's no way for the data seperator to lock
to the speed of the drive.
When Digital finally released formatting softwre for the Rainbow, they
would replace the drives of people that complained they could not format
diskettes. There is a pot on the side of the drive that controls the speed
and there used to be software that would display the speed of the drive so
you could tweak it in. Those of us with DECmates and Pros were still out of
luck, though, as there is a microcontroller sitting between the processor
and the diskette controller that does not include formatting firmware.
By the way, the Otrona Attache also used 10 512-byte sectors per track.
Diskettes formatted on a 96TPI Otrona could be used by RX50 drives.
--
-------------------------+---------------------------------------------
Roger Ivie | "Once again we see that clowning and anarchy
iv...@cc.usu.edu | don't mix." -- The Tick
http://cc.usu.edu/~ivie/ |
Tim Shoppa
The format used on RX50 floppies :-). More seriously, 80 tracks, 10 sectors,
single sided, on a double-density 5.25" floppy. Note that while the
track spacing on a RX50 is the same as on a high-density 5.25" floppy,
the data density is the same as on a double-density 5.25" floppy.
Because of the narrower track spacing, it is essential to get good
registration on the center hub of the floppy; many double density
floppies have a hub which isn't accurate enough to put this narrower
track spacing down and reliably read it again later.
> who invented it,
DEC
> who uses it
People with DEC machines that have RX50 drives. i.e. DEC Rainbows,
DEC Professionals, and PDP's and uVaxen that have a RQDXn/RX50.
> and why.
Because it's the only format I know of that you can read on a
RX50 drive.
>Where did it come from
DEC
> and what is it compatible with?
Nothing. Well, there are software tools for reading (and writing)
RX50's on a High-density PC drive on a PC.
Tim.
Roger Ivie
> iv...@cc.usu.edu (Roger Ivie) writes:
>
>>The RX50 format is 10 512-byte sectors per track, single-sided, 80 tracks,
>>double-density. Digital used it in their PCs.
>
> consider those to be PC's.
For that matter, I have an 8200 with RX50s on it. My particular machine is
a VAXstation 8000 less the graphics hardware; it's essentially an 8200 in
a nice, small, roll-around box with a card that does MFM hard disks and
floppies. The floppies are connected to the disk controller rather than the
CPU, as is normal for 8200s (the RX50 is the console medium for the 8200);
i.e., my floppies appear as DUA23: and DUA24: (when I'm running MicroVMS
4.5; I forget the names 5.x gives them) instead of CSA0: and CSA1: as they
do on normal 8200s.
> There's a wonderfully heath-robinsonish (Rube Goldberg?) drive belt that
> drives the 2 hubs in opposite directions from a motor that's fitted
> hoirzontally behind them.
Hmm. Maybe I'll have to find that one drive we have lying around that doesn't
have the metal case on it...
>>you're frogging around in the box. I have several drives that can't read
>>each others' disks. The drives can also be damaged by diskettes with
>>hub rings; the hub rings eventually ruin the plastic dealies that clamp
>>the diskette to the spindle.
>
> I've never had that - can you give me more info?
The hub ring problem is a rumor that's been floating around the DECmate
community as far back as I can recall. Personally, I've never worried
about it; all of my machines with RX50s also have hard disks, so I don't
usually use floppies a lot (the chief exception being one rather large
embedded firmware project that I assembled on the DECmate using M80; the
assembly required more directory entries than CP/M made available, so I
had to shuffle six floppies to assemble the code. My submit file would
switch between drive A: and drive B: for the modules, so I could swap
floppies when it was done with one module and had moved on to the other
drive).
Charles Lasner should be able to give more information on the hub ring
problem; ask over in alt.sys.pdp8.
John Hanson
>
>
>Just wondering what this RX50 format is, who invented it, who uses it
and why.
>Where did it come from and what is it compatible with?
Used on PDP-11, and VAX machines.
>
>Shawn
>
Tim Shoppa
A.R. Duell <ar...@eng.cam.ac.uk> wrote:
>iv...@cc.usu.edu (Roger Ivie) writes:
>
>>you're frogging around in the box. I have several drives that can't read
>>each others' disks. The drives can also be damaged by diskettes with
>>hub rings; the hub rings eventually ruin the plastic dealies that clamp
>>the diskette to the spindle.
>
>I've never had that - can you give me more info?
I've also heard that you shouldn't use disks which have hub rings, but
for a different reason: the hub rings tend to make the floppy disk
less likely to be aligned in the drive properly if they are not
exactly on center. And even if they are on center, I've heard stories
about them drifting around because they weren't glued on securely.
Roger Ivie
> I had a problem to install 10*512-format in my multi-format bios. Some
> controllers like the 1797 in my mashine seem not to like this format.
> 5*1024 (Wave-Mate) was no problem.
Can't imagine why not. I was on the design team for Digital's SCSI floppy
controller. The original model used the 2793, essentially a fancy 1797. The
gaps get quite short for 10*512, which may be what your BIOS is complaining
about. Most modern drives have speed control good enough to format the
disks; for every day use the data separator should track the drive's speed,
allowign you to use the disks even on crappy drives like the RX50.
When we were working on the formatting routine for the SCSI floppy controller
(yes, you can format RX50s with Digital's SCSI floppy controller), we did
come across some problems with the 9224 controller used in the RQDX3 and
the 765 used in the AT. In the case of the 9224, there were some gap values
that just didn't work; the 9224 took too long to switch from finding IDs to
reading data, IIRC. I don't recall what the problem with the 765 was.
Newer models of Digital's SCSI floppy controller had to a 765-type controller
so that we could deal with 2.88 MB floppies. The 2793 is a much nicer
controller, but you can't argue with market share...