1311 vs 2311 disk capacity 2476
There was a whole art to calculating optimized block sizes for various S-360 tapes and disks. As you said, the maximum was theortical only. Not only did you have to worry about core memory limits for the necessary buffers, but sometimes very high blocking simply didn't write out very well. I remember trying very high blocks on our 2415 tape drives but it ran horribly. I think I ended up with a 5,000 character block size (512 byte records by a 10 factor).
1311 vs 2311 disk capacity 2477
Okay, let's try this again; a power glitch caused my machine to crash, but apparently not before it spat out fragments of...
I remember Deltak training had various formulas for calculating block size, part of the determination was that large blocks didn't gain very much so their was a cutoff point for each particular device.
Today we say BLKSIZE=0 and let MVS (Z-OS?) calculate it for us appropriate to the device. Today we (application people) don't even know what device we're writing too; we use a generic UNIT= value.
It always amazes me how incredibly large today's disks are. The mainframe is in terrabytes, and PCs are in gigabytes. I thought we were in the big leagues with our double density 3330s with 200 meg on a pack.
I can't even conceive how they can get the recording density so tiny yet be so accurate.