mirror of
https://github.com/TomHarte/CLK.git
synced 2024-11-26 23:52:26 +00:00
Actually, I think this is the correct conversion from received speed to clock rate. It'll become obvious if it's not when I get back to working on the 1541 itself.
This commit is contained in:
parent
19ee430d4a
commit
66895d3ac7
@ -120,7 +120,7 @@ std::shared_ptr<Track> G64::get_track_at_position(unsigned int position)
|
||||
|
||||
PCMSegment segment;
|
||||
segment.duration.length = number_of_bytes * 8;
|
||||
segment.duration.clock_rate = current_speed;
|
||||
segment.duration.clock_rate = 4000000 / (13 + current_speed); // the speed zone divides a 4Mhz clock by 13, 14, 15 or 16; TODO: is this the right way around? Is zone 3 the fastest or the slowest?
|
||||
segment.data.reset(new uint8_t[number_of_bytes]);
|
||||
memcpy(segment.data.get(), &track_contents.get()[start_byte_in_current_speed], number_of_bytes);
|
||||
segments.push_back(std::move(segment));
|
||||
@ -142,6 +142,9 @@ std::shared_ptr<Track> G64::get_track_at_position(unsigned int position)
|
||||
resulting_track.reset(new PCMTrack(std::move(segment)));
|
||||
}
|
||||
|
||||
// TODO: find out whether it's possible for a G64 to supply only a partial track. I don't think it is, which would make the
|
||||
// above correct but supposing I'm wrong, the above would produce some incorrectly clocked tracks
|
||||
|
||||
return resulting_track;
|
||||
}
|
||||
|
||||
|
Loading…
Reference in New Issue
Block a user