Re: Commodore 15 Second disk format routine disassembled

Am 19. Juni 2026 16:54:18 schrieb gsteemso <48bitsorbust_at_gmail.com>:

> Hi,
>
>> On Jun 19, 2026, at 6:34 AM, André Fachat <afachat_at_gmx.de> wrote:
>>
>> Am 19. Juni 2026 15:10:17 schrieb gsteemso <48bitsorbust_at_gmail.com>:
>>
>>> Hi,
>>>
>>>> On Jun 19, 2026, at 5:59 AM, groepaz <groepaz_at_gmx.net> wrote:
>>>>
>>>> After having developed stuff for the catweasel (which involved hundreds of
>>>> different physical formats) i dare to say i know a little bit about the
>>>> physics involved :)
>>>>
>>>> (There have been commercial copy protections doing this, even on tracks > 35)
>>>
>>> OK, I'll happily believe you if I can see it in action...  That said, I 
>>> will have to remain skeptical until I see it survive (a) drives that vary 
>>> slightly amongst themselves in spindle speed and (b) mediocre media quality.
>>
>> Commodore drives are very good at compensating spindle speed. They 
>> synchronize the bit counter on every 1-bit. Not like other systems
>
> All good news!
>
> If it is as straightforward as that, I'm trying to understand how all this 
> was not taken advantage of by computer makers like Commodore "back in the 
> day".  Did they just assume that some fraction of the user base was still 
> using the very earliest drive mechanisms that _couldn't_ keep up with this, 
> even though they themselves controlled what mechanisms were in use?  We 
> know that was the case to some degree, it is why the double-density disk 
> formats all ignore tracks 36 and higher, but why they would overlook the 
> rest of it is confusing to me.

The original Shugart disk drives (SA390 for Apple II and Commodore as well, 
SA400 for PCs) could only use 35 tracks due to physical limitation.
Commodore probably wanted to keep compatibility with the older DOS.

They could have gone beyond that with the QD drives (100tpi drives), but as 
I figured they were already above the floppy disk spec (see link posted 
earlier) they probably decided to not overdo it with the bit density.

André
>
> G.