An improvement on the MFM encoding technique used in earlier hard disks and used on all
floppies is run length limited or RLL. This is a more sophisticated coding
technique, or more correctly stated, "family" of techniques. I say that RLL is a
family of techniques because there are two primary parameters that define how RLL works,
and therefore, there are several different variations. (Of course, you don't really need
to know which one your disk is using, since this is all internal to the drive anyway).
FM encoding has a simple one-to-one correspondence between the bit to be encoded and
the flux reversal pattern. You only need to know the value of the current bit. MFM
improves encoding efficiency over FM by more intelligently controlling where clock
transitions are added into the data stream; this is enabled by considering not just the
current bit but also the one before it. That's why there are is a different flux reversal
pattern for a 0 preceded by another 0, and for a 0 preceded by a 1. This "looking
backwards" allows improved efficiency by letting the controller consider more data in
deciding when to add clock reversals.
RLL takes this technique one step further. It considers groups of several bits instead
of encoding one bit at a time. The idea is to mix clock and data flux reversals to allow
for even denser packing of encoded data, to improve efficiency. The two parameters that
define RLL are the run length and the run limit (and hence the name). The
word "run" here refers to a sequence of spaces in the output data stream without
flux reversals. The run length is the minimum spacing between flux reversals, and
the run limit is the maximum spacing between them. As mentioned before, the
amount of time between reversals cannot be too large or the read head can get out of sync
and lose track of which bit is where.
The particular variety of RLL used on a drive is expressed as "RLL (X,Y)" or
"X,Y RLL" where X is the run length and Y is the run limit. The most commonly
used types of RLL in hard drives are "RLL (1,7)", also seen as "1,7
RLL"; and "RLL (2,7)" ("2,7 RLL"). Alright, now consider the
spacing of potential flux reversals in the encoded magnetic stream. In the case of
"2,7", this means that the the smallest number of "spaces" between
flux reversals is 2, and the largest number is 7. To create this encoding, a set of
patterns is used to represent various bit sequences, as shown in the table below
("R" is a reversal, "N" no reversal, just as with the other data
encoding examples):
Bit Pattern |
Encoding Pattern |
Flux Reversals Per
Bit |
Bit Pattern
Commonality In Random Bit Stream |
11 |
RNNN |
1/2 |
25% |
10 |
NRNN |
1/2 |
25% |
011 |
NNRNNN |
1/3 |
12.5% |
010 |
RNNRNN |
2/3 |
12.5% |
000 |
NNNRNN |
1/3 |
12.5% |
0010 |
NNRNNRNN |
2/4 |
6.25% |
0011 |
NNNNRNNN |
1/4 |
6.25% |
Weighted
Average |
0.4635 |
100% |
The controller these patterns by parsing the bit stream to be encoded, and matching the
stream based on the bit patterns it encounters. If we were writing the byte
"10001111" (8Fh), this would be matched as "10-0011-11" and encoded as
"NRNN-NNNNRNNN-RNNN". Note that the since every pattern above ends in
"NN", the minimum distance between reversals is indeed two. The maximum distance
would be achieved with consecutive "0011" patterns, resulting in
"NNNNRNNN-NNNNRNNN" or seven non-reversals between reversals. Thus, RLL (2,7).
Comparing the table above to the ones for FM and MFM, a few things become apparent. The
most obvious is the increased complexity: seven different patterns are used, and up to
four bits are considered a time for encoding. The average number of flux reversals per bit
on a random bit stream pattern is 0.4635, or about 0.50. This is about a third of the
requirement for FM (and about two thirds that of MFM). So relative to FM, data can be
packed into one third the space. (For the example byte "10001111" we have
been using, RLL requires 3 "R"s; MFM would require 7, and FM would need 13.)
|
2,7 RLL, FM and MFM encoding write waveform for the
byte "10001111".
RLL improves further on MFM by reducing the amount of space required
for the same data bits to one third that required for regular FM encoding. |
Due to its greater efficiency, RLL encoding has replaced MFM everywhere but on floppy
disks, where MFM continues to be used for historical compatibility reasons.
Next: Partial Response, Maximum Likelihood (PRML)
