Someone on sci.electronics.design mentioned the following article:
This is a Floppy Disk Data Separator Design Guide for DP8473 which is a floppy disk controller like the infamous NEC 765 controller. It’s a little heavy on the math end, but overall it discusses many of the issues that I first fought with when first coming up with how to design the whole PLL clock and data recovery mechanism.
I really want to revisit the way I read data to make things more reliable. There are a certain class of disks which are unreadable by my project, but work fine in a normal Amiga. I don’t know how or why or which ones. Not yet. Sometimes mass-produced floppies don’t work. I’ve had problems with HD disks, although using a TEAC drive helped.
I think it’s related to the read speed or the write speed.
This article covers “performance measures” and while it uses a bunch of fairly proprietary and expensive/complicated test equipment, I’d like to come up with some type of metric to say, “I’m working at 95% of best case, or 90% of best case” or whatever.
My end goal is to have something that works BETTER than the Amiga…….. and I think it’s reachable…..plus add in my error correction stuff and I’m sure I’ll get past it. But I’m not there yet, and so it’s time to figure out why not and how to fix it.