ryandesign
commented
11 months ago I added logging to show the head position:
--- a/Storage/Disk/Drive.cpp
+++ b/Storage/Disk/Drive.cpp
@@ -103,6 +103,8 @@ void Drive::step(HeadPosition offset) {
track_ = nullptr;
}
+ fprintf(stderr, "head position: %5.2f %*s|\n", head_position_.as_quarter() / 4.0f, head_position_.as_quarter(), "");
+
// Allow a subclass to react, if desired.
did_step(head_position_);
}With enhanced IIe, INIT fails immediately on track 0 every time due to #1211 which does now look like it's a separate problem.
With unenhanced IIe, which track it fails on is random. The amount of time it takes to initialize each track is also somewhat random. I'm surprised by the randomness; I would expect the same result each time. Once in a blue moon, the initialization "succeeds" but the disk that was created is not actually initialized correctly. Compared to a disk initialized by Virtual ][, several 256-byte sectors on the disk initialized by Clock Signal contain only nulls.
TomHarte
Boot an Apple II or II+ or IIe with the 1983-01-01 Apple DOS 3.3 System Master disk, then eject it by inserting a blank disk that was created with this command:
At the BASIC prompt, type
INIT HELLOand press Return.In Open Emulator 1.1.1-202203110628 and Virtual ][ 11.4 this succeeds, returning you to the BASIC prompt and the disk has been initialized and can be used to boot the machine.
In Clock Signal 23.10.29,
I/O ERRORoccurs and only parts of the disk have been written and trying to boot it either hangs or crashes into the monitor. Curiously, what gets written seems to vary (by Apple II model? or randomly?).This might be the same bug as #1211 but this reproduction recipe works on any Apple II model, not just the enhanced IIe, so it is not limited to or caused by the 65C02 CPU.