PET index - disks

• CBM-Hackers summary
• Inside Commodore DOS
• CBM-Hackers discussion

In June 2015 Spiro summarized the discussion neatly on the CBM-hackers list, so I put it here as a tl;dr summary ;-)

For brevity reasons, I will use the term 1540 for the 2031, 1540 and the
1541 with -01 and -02 ROM, and I will use the term 1541 for the 1541
with -03 ROM and later.

The 4040 used 8 bytes. Anyway, it used 8 bytes before conversion to GCR,
which was done in hardware. Thus, the gap was 8 * 10 bit = 80 bit long.

The 1540 just reused the value 8. Unfortunately, here, it is 8 *real*
bytes, resulting in a gap of 8 * 8 bit = 64 bit.

This could result in problems: After the last byte of the header, the
drives start writing a SYNC mark after this gap. On the 1540, the gap is
64 bit, on the 4040, it is 80 bit.

Now, if the 4040 writes on a 1540 formatted disk, it is already in the
SYNC when it starts writing the SYNC. It has to change from READ to
WRITE, which could result in "unclear" bits that could be read as "0".
If this happens, the reading drive has already read 80 - 64 = 16 bit of "1"
(resulting in SYNC), then a 0 bit - thus, for the reading drive, the
SYNC has just ended. Unfortunately, no data follows. Instead, there is a
SYNC. Thus, the drive has a read error.

The other way around, it is not the problem. If there is the 80 bit GAP
of the 4040, and the 1540 writes on this disk, it just starts writing
its SYNC too early - no big deal.

Why did the change from 8 to 9 bytes in the 1541 change the rules?
Because now, the 1541 GAP is 72 bit long. Thus, the 1541 only writes
80 - 72 = 8 bit too late on a 4040 disk. As 8 bit do NOT form a SYNC yet
on reading, there is no risk for an erroneous SYNC mark.

Why didn't Commodore use 10 byte, which would result in a GAP of 80 bit,
like in the 4040? Most probably, it was for compatibility reasons with
the 1540. As disks with 8 byte GAP already existed, the same problem
would result if the 1541 writes on 1540 disk. With the 9 byte GAP, the
problem does not exist.

9.12  Write Incompatibility with 4040

Programs or data stored on a diskette formatted on a 1541 disk drive can be
READ using a 2040 or 4040 disk drive.  Conversely, a 1541 disk drive can
READ a diskette formatted on either a 2040 or 4040 disk drive.  However,
these drives are not completely write compatible.

This write-incompatibility problem appears to be caused by two things:

1.  Differences in the header gap length.
2.  Alignment problems (particularly with the 1541).

Let's consider the differences in the header gap length first.

Differences in Header Gap Length
                                                                        :209:
The 2040 and 4040 drives use a header gap that is nine GCR bytes long while
the 1541 uses a header gap that is only eight non-GCR bytes long.  On this
basis we would expect the header gaps to be 90 and 64 bits long respectively.
However, when we use a bit-grabber to view the gap we find that the actual
header gaps as recorded on the disk are 100 bits for the 4040 and 92 bits for
the 1541.  In read mode, this makes no difference.  After reading the header
bytes to check that this is the correct sector, all the drives simply wait
for the next sync mark.  The number of bytes in the header gap does not
matter.  Once the sync mark is over, the first character in the data block is
read.  This is the data block ID character.  If it is not a $07, the DOS
reports a 22 READ ERROR (data block not found).

In write mode, however, the length of the header gap is important.  After
reading the header bytes to check that this is the correct sector, all the
drives count off the bytes that make up the header gap.  Once the correct
number of bytes have been read, the drive flips to write mode and begins
writing out the data block sync characte.  Since this is reputed to be an
important aspect of the write incompatibility problem, let's examine what
happens in some detail.

The last part of the header gap and the start of the data block sync mark in
a sector of a diskette that has just been formatted on a 1541 disk drive look
something like this:

               Sync mark
1541 xxxxxxxxxx111111111111111111111111111111111111 -> 92 bits

The last part of the header gap and the start of the data block sync mark in
a sector of a diskette that has just been formatted on a 4040 lokos something
like this:

                                         Sync mark
4040 xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx111111111111 -> 100 bits

When a sector of a diskette that was ORIGINALLY FORMATTED ON A 4040/2040 disk
drive is REWRITTEN ON A 1541, the result is as follows:

Original                                 Sync mark
4040 xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx111111111111 ->

Rewrite         Sync mark
1541 xxxxxxxxxx-1111111111111111111111111111111111111 ->

Result          Sync mark
     xxxxxxxxxx-1111111111111111111111111111111111111 ->


NOTE:  The "-" marks when the drive switches into write mode.  A transient
current appears to flow through the record/play head during this time
interval.

The original sync mark on the diskette has been completely overwritten by the
new one.  This sector can be read cleanly on any drive.  It appears that a
1541 drive should be able to write data onto a dikette that was originally
formatted on a 4040 without causing any problems.

When a sector of a diskette that was originally formatted on a 1541 disk
drive is rewritten on a 4040/2040, the result is as follows:
                                                                        :210:
Original       Sync mark
1541 xxxxxxxxxx11111111111111111111111111111111111111 ->

Rewrite                Sync mark
4040 xxxxxxxxxxxxxxxxx-111111111111111111111111111111 ->

Result         Pseudo-sync   Sync mark
     xxxxxxxxxx1111111-111111111111111111111111111111 ->


NOTE:  The "-" marks when the drive switches into write mode.  A transient
current appears to flow through the record/play head during this time
interval.

In this case, the original sync mark on the diskette has NOT been completely
overwritten by the new one.  The start of the old sync mark is still there.
What actualy gets recorded at the start of the "new" sync mark depends on the
speed of the drives, the polarity of the magnetic field used to record the
original "1" at that spot on the diskette, and any transients that flow
through the head as it switches into write mode.

Before you read this next section, be sure that you understand Section 9.7 on
the Recording Process.

Let's take a look at an "exploded" view of that spot just before the new sync
character is written.  Remember, a "1" is not recorded as magnetization in a
particular direction.  It is simply a change in the direction.  Now that
you've got that straight, here is what that spot might look like.

          +-----+-----+-----+-----+-----+-----+-----+-----+-----+
Original  |N   S|S   N|N   S|S   N|N   S|S   N|N   S|S   N|N   S|
          +-----+-----+-----+-----+-----+-----+-----+-----+-----+
          1     1     1     1     1     1     1     1     1

Everything appears normal.  Now let's write that sync mark.

          +-----+-----+-----+-----+-----+-----+-----+-----+-----+
Original  |N   S|S   N|N   S|S   N|N   S|S   N|N   S|S   N|N   S|
by a 1541 +-----+-----+-----+-----+-----+-----+-----+-----+-----+

                                  +-----+-----+-----+-----+-----+
Replacement sync mark           ??|N   S|S   N|N   S|S   N|N   S|
written by a 4040                 +-----+-----+-----+-----+-----+
                                ?? = effects of transient currents

          +-----+-----+-----+-----+-----+-----+-----+-----+-----+
Result    |N   S|S   N|N   S|S  ??|N   S|S   N|N   S|S   N|N   S|
          +-----+-----+-----+-----+-----+-----+-----+-----+-----+
          1     1     1     1     1     1     1     1     1

Everything worked out just fine.  We have a clean sync mark and the sector
can be read cleanly by either drive.  However, suppose our 74LS74 flip-flop
(UF6) had been in the opposite state or the speed of this drive did not
exactly match this new one.  What would happen?  Take a look.
                                                                        :211:
          +-----+-----+-----+-----+-----+-----+-----+-----+-----+
Original  |N   S|S   N|N   S|S   N|N   S|S   N|N   S|S   N|N   S|
by a 1541 +-----+-----+-----+-----+-----+-----+-----+-----+-----+

                                  +-----+-----+-----+-----+-----+
Replacement sync mark           ??|S   N|N   S|S   N|N   S|S   N|
written by a 4040                 +-----+-----+-----+-----+-----+
                                ?? = effects of transient currents
          +-----+-----+-----+-----------+-----+-----+-----+-----+
Result    |N   S|S   N|N   S|S   ??    N|N   S|S   N|N   S|S   N|
          +-----+-----+-----+-----------+-----+-----+-----+-----+
          1     1     1     1    ?0     1     1     1     1

Argh!  Potential problems.  Because the magnetic polarity of the new "1"
happened to match the polarity of the existing zone, we appear to have just
created a double-length magnetic zone.  If we have, this will be interpreted
as a "0" bit.  From a study of the bits actually recorded on disk, this
appears to happen every time!  If there are more than 10 preceeding "1" bits,
this single "0" will be interpreted as the end of the sync mark and the drive
will interpret the rest of the sync bits as data.  Since this will definitely
NOT be decoded as a $07 byte, the drive errs out with a 22 READ ERROR.

Since th header gaps only differ in length by 8 bits, we should always have
only seven 1's in the pseudo-sync.  An examination of the bits recorded on
the disk seems to support this conclusion.  As a further test we did some
testing using recently aligned drives.  We found surprisingly few errors when
we use a 4040 disk drive to rewrite all non-directory sectors on a 1541
formatted disk.  On a freshly formatted diskette, we found no errors at all
after rewriting over 2400 sectors.  If the sectors of the 1541 diskette has
been rewritten several times using a 1541 before they were rewritten on a
4040, we did start to find a few errors.  However, the error count was low.
Usually less than two errors when rewriting all 640 sectors and these tended
to occur in two specific areas:  on tracks 25 or 26 or on tracks 31 or 32.
These findings lead us to conclude that the differences in header gap length
is NOT the cause of write compatibility problems between the 1541 and 4040
disk drives.

If for some reason you want to reduce the difference in header gap further
when writing onto a 1541 formatted diskette using a 4040 drive, enter the
following magic incarnation in either program or immediate mode.

OPEN 15,8,15
PRINT#15,"M-W"CHR$(157)CHR$(16)CHR$(1)CHR$(8)
CLOSE 15

This will change the header gap length of the 4040 drive from 9 to 8 GCR
bytes (actual length = 90 bits).  You can now write to the 1541 diskette with
little fear of damage.  However, you must remember to reset your 4040 drive
(turn it off or issue a UJ command) before you insert one of your 4040
formatted diskettes.  Otherwise, a magnetic plague will develop among your
4040 formatted diskettes.  Don't say you weren't warned!

Head Positioning Problems
Since we encountered so few errors using properly aligned drives, we feel
that most of the reported problems of incompatibilities are the result of
head positioning errors.
                                                                        :212:
If a sector is rewritten on a different drive and the position of the read/
write head is different, the new data will not completely replace the old as
indicated below.

           +---------------+-------------------------------+
Original   | N           S | S                           N |
on one     | N     1     S | S     1               0     N |
drive      | N           S | S                           N |
           +---------------+-------------------------------+

           +-------------------------------+---------------+
Rewritten  | S                           N | N           S |
on another | S     1               0     N | N     1     S |
drive      | S                           N | N           S |
           +-------------------------------+---------------+

           +---------------+-------------------------------+
Original   | N     1     S | S     1               0     N |
           +---------------+---------------+---------------+
Rewritten  | S                           N | N           S |
by another | S     1               0     N | N     1     S |
drive      | S                           N | N           S |
           +-------------------------------+---------------+

When this sector is read on the original drive, the head will pick up both
the new signal and the old signal.  The relative strengths of these two
signals depend on the amount of the original signal remaining.  If the two
drives are sufficiently different, the read signal will be garbled and
produce a abundance of 22 and 23 READ ERROR's.

Summary

In conclusion, although there is a difference in header gap size between the
1541 and the 4040 drives, this does NOT appear to be the cause of write
incompatibility problems.  Most complaints about the write incompatibilities
of various disk drives are probably due to problems in head positioning.
Further evidence for this is the fact that some schools are experiencing
similar difficulties when students use several different 1541 drives for
saving programs on a single diskette.

Hello,

* On Fri, May 06, 2005 at 12:35:55AM +0200 Patrycjusz R. ?ogiewa wrote:

> >OTOH, the 4040 used a gap between the data block header and the data
> >block itself of 10 bytes. The 2031 and 1540 (and 1541 up to $E000-ROM
> >version -02) used 8 bytes, while the 1541 (-03, -05 and later) and
> >157x use 9 byte. I believe 8 bytes are critical if a disk is to be
> >interchanged between a 154x and a 4040, as there might be "double
> >SYNC marks", making the data block unreadable. With 9 byte, it is
> >much more unlikely that this happens than with 8 byte.
>
> But...  "... note that a header block is never rewritten after
> formatting is complete. The data block of a sector, including the sync
> character, is completely rewritten every time data is written to that
> sector. The [...] header gap is counted off by the DOS to determine
> where to start writing the data block."

Yes, this is wrong. But think about the following layout of a 4040 disk,
which shows the part between the data block header and the data block
itself:

byte:    -2 -1  0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17
content: xx yy 55 55 55 55 55 55 55 55 55 55 FF FF FF FF FF aa bb cc ...

Here, xx yy are the last bytes of the data block header, aa bb cc are the
first bytes of the data block. As the 4040 has written this part, there
are 10 $55, followed by 5 $FF byte. This is immediately after formatting
the disk.

Now, think about what happens if the 2031, 1540 (or 1541-01, -02; I will
call this the the 1540 case) write to this disk:

byte:    -2 -1  0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17
content: xx yy 55 55 55 55 55 55 55 55 FF FF FF FF FF aa bb cc ...

As these disks use only 8 bytes for the gap, the $55 part is just
getting smaller. Everything else is correct.

Now, what if a 4040 writes to this very same disk again? Have a look
here:


byte:    -2 -1  0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22
content: xx yy 55 55 55 55 55 55 55 55 FF FF FF FF FF aa FF FF FF FF FF aa bb cc ...
gap count:      0  1  2  3  4  5  6  7  8  -  -  -  -  9

I added the byte count, that is, how the write mechanism sees this disk
(I assume the byte count mechanics are the same as with the 2031).

The 4040 recognizes the $55 bytes 0-7. The 8th byte, a $FF, is
recognized as a regular byte (a SYNC mark has to be 10 bits of 1, but we
only have 9 up to the first $FF). Anyway, as we are now IN the SYNC
mark, no more S.O. (byte marks are generated until the SYNC mark is
over. Thus, the first data block byte (aa, byte 13) is recognized as 9th
(and last) GAP byte, thus, the 4040 does not start before byte 14, thus,
the 4040 writes its own SYNC mark, starting at byte 14. Of course, as
the byte 13 (aa) is not a $FF byte, we have a double SYNC mark which
makes this disk unreadable afterwards. Furthermore, this block is 6 byte
longer than before (1541 format), or 4 byte longer than before (4040
format), which could potentially overwrite parts of the SYNC mark of the
next sector.


Ok, why doesn't this happen with the 1541-03, -05 (which I will call
1541 in the following) and newer, which write a 9 byte gap?

Have a look again at the disk as it has been written with the 1541:

byte:    -2 -1  0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17
content: xx yy 55 55 55 55 55 55 55 55 55 FF FF FF FF FF aa bb cc ...

You see, there is one more $55 than in the 1540 case.

Now, the 4040 wants to write to this disk:

byte:    -2 -1  0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22
content: xx yy 55 55 55 55 55 55 55 55 55 FF FF FF FF FF FF aa bb cc ...
gap count:      0  1  2  3  4  5  6  7  8  9

Here, the 4040 correctly counts the bytes 0-8, and it still recognizes
byte 9 as a data byte (as no 10 "1"-bits are recognized until then).
Thus, it starts writing its SYNC mark at byte 10, which is exactly where
it would have started its SYNC mark if no 1541 would have been used
in-between. The only difference here is that the SYNC mark is 6 bytes
long now, as byte 9 is the rest of the 1541 SYNC mark.

Thus, this 9 byte gap of the 1541 is much more uncritical than the 8
byte gap of the 1540. Anyway, there is still a chance for problems. The
problem is that going into write mode takes some time. Thus, it could be
that the 4040 emits a 0 bit before it starts writing the SYNC mark. If
this does not occur with the first bit, but with the 2nd bit or later,
than the drive would again have a double SYNC mark:

1. case: first bit is 0. Then we have
   55 FF 7F FF FF FF ...
   This is not critical, as the last bit of $55 and the 8 bit of the
   first $FF only give 9 "1" bits, thus, no SYNC mark is detected.
2. case: first bit is 1, but some subsequent one is "0". Then, we have
   55 FF BF FF FF FF ...
   or
   55 FF DF FF FF FF ...
   or
   55 FF EF FF FF FF ...
   or so on.
   Here, the "1" from the last bit of the $55, the 8 "1" bits in $FF and
   the first "1" of the next byte result in 10 "1" bits, thus, giving a
   SYNC mark. The next "0" then *clears* this sync mark, but there is a
   new SYNC mark immediately after (after 1 $FF byte!). Thus, we drive
   reads a $FF as first data byte, which makes this disk unusable.

I do not think that this is very likely, but it can happen. Anyway,
chances for problems are MUCH smaller than with the 8 byte gap.

> >Thus, Commodore might have had some problem and tried to solve this
> >by reducing the number of written bytes.
>
> At the off-bytes then? Note also that there is enough buffer at the
> end of the track. The tail gap of the last sector is usually much
> longer than any other (up to 100+ bytes),

I have not checked any disk, but are you sure about 100+ bytes? This
sounds like VERY much for me. My own probing formatter just has a
variance of less than the number of sectors per this track (that is,
less than 21 bytes for tracks 1-17).

Furthermore, it is not only the tail gap of the last sector; it's the
other way around: If the last sector's tail gap is that high, then
the blocks are not written very balanced all over the track, thus,
increasing the probability about one drive writing over the beginning of
the next block if the formatting was done with a very slow drive, but
the new writing drive is very fast.

> OTOH as we all more or less speculate on this then who really knows?

Yes, this is true. Anyway, it IS remarkable that Commodore reduced this
gap by 2 between the 4040 and the 2031, and increased it again by 1
between the 1541-02 and 1541-03. I'm sure there was some reason for it.

BTW: You all know the "wrong" empty sectors of the 1541 disks, don't
you? That is, instead of being all $00, the sectors are $00, $01, $01,
...  on track 1, and $4B, $01, $01, $01, ... on all subsequent tracks?
It is interesting that this "bug" was introduced between the 1540 and
the 1541. Thus, the 1540 had the "correct" empty pattern. Thus, to me,
it seems to be a kind of "version information" from Commodore to use
that blank pattern.

Ok, again, this is pure speculation.

Regards,
   Spiro.

Hello Glenn,

* On Fri, May 06, 2005 at 04:44:20AM -0500 Glenn Holmer wrote:

> Immers and Neufeld discuss this in chapter 9.12 of "Inside Commodore
> DOS."

Thank you, I did not know this one.

What is interesting:

1. My analyzes seems to be correct in most ways. ;-)

2. The gap written by the 4040 is 9, not 10 as I had read elsewhere.
   Thus, the 1541-03 ROM and later use the SAME gap as the 4040, it is
   only the 2031, 1540 and 1541-01 and -02 ROMs which use a shorter GAP
   (and thus, give a potential problem).

Regards,
   Spiro.

Hello,

* On Fri, May 06, 2005 at 07:59:46PM +0200 Patrycjusz R. ?ogiewa wrote:

> I read through this explanations and they seem really cogent. The
> question remains if you have really verified this?

Well, I cannot verify this as I do not own any IEEE drive.

> I believe that we would have to take down the bit images of the 4040
> and 1541 (according to your terminology) tracks freshly formatted and
> then cross-rewritten and compare it this way. This would be something
> that I am tending to believe the "incomdos" authors did and they have
> somewhat similar, yet different explanations, starting from p. 208.

Yes, thank, I now had a look at it. I believe I can explain their 100
bit vs. 90 bit gap problem for the 4040:

According to your explanation, the 4040 writes 9 gap (GCR-)bytes. This
gives 9 * 10 = 90 bits.

Anyway, when the drive executes the last BVC *, the "write shifter"
already latched the byte to be written. Thus, even if we change to write
mode immediately, the last byte will be written as 10th byte, resulting
in a gap length of 10 * 10 = 100 bits.

One remark: This is not true if the disk is to be formatted, as the
formatter does not have to switch into write mode, and thus, it
generates a 90 bit long gap.

What is happening with the 1541? Here, 8 * 8 = 64 bits are expected, but
the authors see a gap of 92 bits. Here, I have the following idea:

On a freshly formatted disk, newer 1541 disks write a 9 bit non-GCR gap
of $55. Anyway, the last 2 bytes of the data block are $0F $0F, which
translates into $x5 $55 $55. Thus, there are chances the authors of
Inside Commodore Dos counted these last two bytes as GAP bytes,
resulting in the count of (9+2)0 * 8 = 88 bit.

Now, the 1541 write mechanism has the same "problem" as the 4040 write
mechanism when writing a block: If it wants to go into write mode after
counting the last byte, the shifter already latched the last byte, which
is exactly what it has read before ($55) since the PA of the VIA is
still in read mode. Thus, it writes another $55 byte on the disk,
resulting in a GAP of (9+2+1)*8 = 92 bit which can be observed.

Of course, this all is pure speculation which would have been to be
proven. Furthermore, if the 4040 ROM has that $0F $0F filler for the
data block header, it would not explain why the authors count
differently for the 4040 and the 1541. This would only make sense if the
4040 ROM does not write the $0F $0F filler as part of the data block,
but as part of the intra-sector GAP. This might be the case as:

1. $0F $0F translates into $x5 $55 $55 in GCR
2. The 4040 can write "half" GCR bytes, while the 2031/15xx cannot


Of course, the authors tell that tests could not show this problem to
happen very often.

But remember: The authors examined 1541-03 drives (I conclude this from
the section "late news" on page 215, which compares -03 and -05 ROMs).
Which older drives (-02 and before), the other explanation from my mail
some days ago holds true, which is a totally different beast!


> Yet it seems to me unlikely (speculation) that people at CBM have cut
> off last BVC from the sector write routine (the thing we started with
> ;-) to improve compatibility with 4040.

Yes, I remember. ;-) In fact, I did not know that the 4040 counted in
GCR bytes, thus, I thought that Commodore reduced 10 (non-GCR) bytes to
8 (non-GCR) bytes, and then went back to 9 (non-GCR) bytes.

Anyway, it now seems to me that Commodore used 8 (GCR-) bytes, took over
that value for the 2031/1540 (which use non-GCR) and wondered about the
problems. Then, they increased the number to 9 (non-GCR) bytes on the
1541-03 ROMs to minimized the problem.


> >I have not checked any disk, but are you sure about 100+ bytes?
>
> Again - to be sure I would have to take the bit image, which I didn't
> but I quoted the same source, which I thought to be reliable.

Well, "sometime", I will do this test. If this is true, the probing
routine of the drive is not very good, despite the fact that it is very
slow.

> >Furthermore, it is not only the tail gap of the last sector; it's the
> >other way around: If the last sector's tail gap is that high, then
> >the blocks are not written very balanced all over the track,
>
> Yes. There is supposedly a bigger gap at the end of the track.

Well, "bigger" gap does not mean that one has to be 100+ bytes long.


> Well, such situations _will_ happen if the drive(s) are misadjusted
> but I think they have put enough buffer (as you calculated yourself)
> to cover the regular, acceptable deviations in drives' speed anyway.
> If that doesn't prove to be enough (since the drive exceeds the
> acceptable limits) then the drive needs servicing anyway.

Well, the arguments I showed in another mail where for ONE drive. These
arguments were not meant for different drives.

Let's calculate the length of a track in bytes. As I am lazy, I'm just
quoting from a mail of WoMo here. I came to the same formula
independently from him, so it is just for the sake of not writing too
much that I quote him, not that I did not follow the formula: ;-)


  f     :=  base clock frequency, which is 4MHz
            (this is not completeley true, the base clock is 16Mhz, but after
            beeing divided by the speed zone divisor, it is divided by 4
            again,
            so the overall virtual base frequency results to 4Mhz again)
  omega :=  rotation frequency of the main disk motor
            (300Upm +-3% on most drives, aka. 5/s)
  dvsr  :=  speed zone divisor of 13, 14, 15 or 16


                                            f
   Tracklength (Bytes)  :=  TlB  =  ------------------
                                     omega * dvsr * 8



Now, take speed zone 3 (track 1-17, dvsr = 13). Then, we get a tracklength with 300 rpm of

                      4 MHz
  TlB(300 rpm) = ---------------  = 7692,3 Byte
                   5 * 13 * 8

Take the same for 303 rpm:

                     4 MHz
  TlB(303 rpm) = --------------- = 7616,1 Byte
                  5,05 * 13 * 8


And the same for 297 rpm:

                     4 MHz
  TlB(297 rpm) = --------------- = 7770,0 Byte
                  4,95 * 13 * 8


You see, the track length differs by (7770 - 7616 byte) = 154 byte
between the drive with 303 rpm and the drive with 297 rpm! That is, PER
sector, the difference is 154 byte / 21 = approx. 7,3 byte!

With this, you see why a probing formatting routine is very important if
you want to share disks between different drives.


> >Yes, this is true. Anyway, it IS remarkable that Commodore reduced this
> >gap by 2 between the 4040 and the 2031, and increased it again by 1
> >between the 1541-02 and 1541-03. I'm sure there was some reason for it.
>
> AFAIR it was 9 GCR bytes vs. 8 non-GCR, with the actual bit patterns
> taking even more for some reason - interesting for what reason though.
> It would have to be verified if that was indeed changed _twice_ as you
> suggested. Don't have resources at hand to verify it.

Well, here, a simple byte compare suffices. Take the 1541-e000-02 and
1541-e000-03 ROMs from funet an perform a byte compare. You will see
that both images differ in exactly three positions:

 l_f586: jsr     l_f78f
         jsr     l_f510
-        ldx     #$08
+        ldx     #$09
 l_f58e: bvc     l_f58e

and

         bne     l_fcc2
-        ldx     #$08
+        ldx     #$09
 l_fcd1: bvc     l_fcd1a

and

-l_fee6: .byte   $10
+l_fee6: .byte   $0E

The latter is just the ROM checksum. The other two differences are the
counter for the intra-sector gap of the write routne ($F58D) and the
counter for the intra-sector gap of the format routine ($FCD0).

I have checked every 2031, 1540, 1541 and 157x ROM I found on funet:
There are only these two variants, anything before 1541-03 has 8 byte,
1541-03 and the later (including 157x) have a 9 byte gap.


> >BTW: You all know the "wrong" empty sectors of the 1541 disks, don't
> >you? That is, instead of being all $00, the sectors are $00, $01, $01,
> >...  on track 1, and $4B, $01, $01, $01, ... on all subsequent tracks?
>
> Yes. There is one INX too much at $FC86, which both increments the
> value passed to A by one (hence $01) and skips the first byte when
> writing to the buffer later on (hence the first byte different). Some
> ages ago, I recall wondering what did they have in mind to make it this
> way... ;-)

Yes. Interesting, the 1540 and 2031 ROM have a NOP ($EA) there, while
even the 1541-01 ROM has the INX.  Interesting: INX ($E8) and the NOP
differ in only one bit.

> At least the 1541 leaves its clearly distinguishable mark on the disk
> ;-)

Yes, I believe this was the intention of this change.

Regards,
   Spiro.