> I believe the idea was constant linear velocity recording was "better" than constant angular velocity, by virtue of right-sizing the recording density across

> I believe the idea was constant linear velocity recording was "better" than constant angular velocity, by virtue of right-sizing the recording density across the entire surface. Yeah. If you think about it, the outer sectors on a disk rotating with constant speed are flying under the head much faster than the inner ones. So the slow inner sectors basically decide what magnetic density your disk has to have, as the bits will end up really close together there! But the disk will have the same magnetic density on the outer sectors, over a much longer distance. This will go wasted, if your drive reads the bits at the same constant speed (because the gaps between the bits are much larger, meaning you don’t get to use that high density). Drivers like the C64’s 1541 employed workarounds like still keeping a constant rotational speed (300RPM for the 1541), but divide the tracks into 4 different “speed zones”, which determine at what speed the bits are read off the disk. So, on the outer tracks the bits are read off faster, and there are actually more sectors than on the inner tracks.