Exit on UEF baud rate chunk (&117)

[Diminished0]

Elkulator exits when encountering UEF baud chunk &117.

In one sense, this behaviour is reasonable, as the Electron does not support 300 baud, so no baud change is possible. However, encountering this chunk type is fatal even when the chunk indicates 1200 baud, which the Electron of course does support.

My UEF tools (including the tape overhaul patch for B-Em) tend to emit chunk &117 as a matter of routine, making such UEFs incompatible with Elkulator. I would opine that 1200-baud chunks &117 should be permitted. I will also point out that you can play a 300-baud tape into a real Elk if you want to, but the Elk won't hear much, so there is a case for permitting 300-baud &117 as well. Chunk &117 describes what is on the tape, not how the serial decoder should be programmed.

I do not know what Elkulator's policy on unknown UEF chunk types currently is; it is possible that UEF files containing more obscure chunk types may begin to surface in the near future as my work on tape infrastructure continues. Opinions on this may differ but I would suggest that unknown chunk types might perhaps be ignored.

I enclose a couple of failing UEFs; one with 300 baud chunk &117s, and the other with 1200 baud chunk &117s.
uefs.zip

[dboddie]

I think that reading chunk &117 and recording the current baud rate would be useful, but it would require changing the way that other chunks are handled.
If the current baud rate is 1200 baud, data chunks would be read as normal.
If the current baud rate is 300 baud, data chunks would be skipped. This wouldn't give completely accurate behaviour because, as you say, the Elk may actually hear something, but I'm not personally looking to implement some level of signal processing model.