FDC Gap Length?

[Allison]

>
>Subject: FDC Gap Length?
>   From: Dave Dunfield <dave04a at dunfield.com>
>   Date: Sat, 16 Jul 2005 13:55:05 -0400
>     To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at classiccmp.org>
>
>Hi Guys,
>
>I've been doing a bit more work on my replacement for TeleDisk, and I have
>it working quite nicely, however I do have one area that I could some help
>with...
>
>The 765 FDC requires two "Gap Length" values, one for formatting (total gap
>between sectors), and one for writing... I have been unable to find any
>information about calculating these values from arbitrary sectors/track,
>encoding and transfer rate. All of the documents I have found simply give a
>table of suggested values for common disk formats, but give no hint as to
>how to calculate them for uncommon or arbritary formats. (for example, the
>fairly common 9x512 format is not included in the table).
>
>Currently, I am using the values from the table, and some "guesses" for
>values for some other formats which I have tried, however I do not know
>how to derive the correct value for these items. A couple hours on Google
>turned up numerous articals, ALL of which either:
> a) use a single fixed value
> b) use the NEC table
> c) call then "magic" numbers
>  or
> d) say they can't provide more information on how to select GPL due to
>    lack of information.
>
>With the 765, I cannot determine the gap length used on the original
>disk, so the best I can do is to try and determine a suitable gap length
>when formatting based on the # sectors, sector size, encoding method,
>transfer rate and drive type....
>
>Can anyone provide any clues?
>
>Regards,
>Dave

The total number of bytes, gaps plus data, must not exceed the total 
number of possible byts on the media at nominal rotation speed.

When formatting the last gap is longer than the track so that you write 
gap until you see index to fill out the track.

For reading there is a minimum number of bytes for the gap and if 
memory serves 16 (decimal) is it for DD and 10 for SD. These are not 
optimum for writing however.  For writing you need to insure there are
enough gap bytes in a gap to assure the minimum needed for PLL data 
seperators to aquire and track after the splice point.  The splice 
point is the location on media when after reading sector-n and 
finding that sector n is next then switching to write.  The result
is always a discontinuity in the mid gap area from switching in the
erase head and beginning the first gap byte written. 

Is there a calculation, NO. Unless your formatting your own, then 
it's use biggest gaps that allow sector data and a resonable end gap
on off speed drives.   Most of the time you have to know how 
that disk was formatted originally to arrive at the correct values.
Fortunatly there is a fair amount of wiggle room if you do not run
too close (too close is below) to minimums. A good example of that
is the RX50 (10sectors 512bytes on 5,25 floppy) as the 765A can 
read it but barely (not really) formats that correctly, to format
that with non WD you need upD7265.  


hope that helps some.

Allison