Re: How to make .nib disk images?

[russotto@grace.speakeasy.net (Matthew Russotto)]

In article <3D67A803.3E1E5995@swbell.net>,
Rubywand  <rubywand@swbell.net> wrote:
>
>     Yes; that's very likely. However, at least on the .nib images I
>checked (in sets from 1999 and 2001), the formatting was messed up, too.

I managed to pull a nearly-good Trinity from at least one set of
images.  Good enough that by comparing bits from a Trinity of a
different version, I was able to make an accurate Z-code file (with proper
checksum). 

>     It turns out that Side 1 of the original disk is formatted for
>17 sectors/track. (A scan of the side for the address header bytes used
>gave 595 hits. 595sectors/35tracks = 17 sec/trk.)

I think most likely you're seeing duplicates caused by "over-reading".
That is, you've kept reading the track even after the disk has already
turned 360 degrees.

>     The big problem is Side 2. The original seems to be formatted for
>35 track-size "sectors". Both of the Side 2 .nib images checked were
>formatted for 16sec/trk. The original Side 2 can hold a lot more data;
>and, it seems to be nearly full.

This is correct; that's a format sometimes called "18 sector".  Each of
those big "sectors" consists of 18 sector data fields in a row, which is
where the name comes from.
-- 
Matthew T. Russotto                                mrussotto@speakeasy.net
===== 
Every time you buy a CD, a programmer is kicked in the teeth.
Every time you buy or rent a DVD, a programmer is kicked where it counts.
Every time they kick a programmer, 1000 users are kicked too, and harder.
A proposed US law called the CBDTPA would ban the PC as we know it. 
This is not a joke, not an exaggeration.  This is real.
http://www.cryptome.org/broadbandits.htm