Posted By
Fraser
on 2022-07-25
09:26:58
|  endianness of tape file bits
I have seen the page here on TAP files but I can't find out the endianness of bytes recorded by the ROM saver/loader. I think it is big endian (bit 7 to bit 0).
|
|
Posted By
Fraser
on 2022-07-25
10:10:24
| Re: endianness of tape file bits
I have found a couple of sources saying its using the little-endian system.
|
|
Posted By
Csabo
on 2022-07-25
16:26:35
| Re: endianness of tape file bits
Welcome back after 15 years? 
I'm not very well versed in this topic, but I figured your question might be easily answered by saving some files (using the KERNAL save) and taking a look a the TAP file... But just by looking at that saved data - I have no idea how that corresponds to the actual bits saved.
Someone might know, perhaps Gaia?
|
|
Posted By
gerliczer
on 2022-07-26
01:51:11
| Re: endianness of tape file bits
The definitive answer is in the KERNAL source. A few years ago it leaked to the Internet, IIRC. Fraser should find them and look up the answer.
|
|
Posted By
Fraser
on 2022-07-26
04:31:03
| Re: endianness of tape file bits
Both of the sources say the first byte is $89 which I have found with CSW Viewer and it is little-endian. I have not had much interest in CSW Viewer from the Commodore community. It is not obvious that a TAP file can be looked at in the same way as a CSW file.
|
|
Posted By
Krill
on 2022-07-26
05:28:25
| Re: endianness of tape file bits
In files and memory, bytes are almost always stored with the bit order 7..0.
Endianness almost always refers to the order of multi-byte integers, not the bits in individual bytes
The .TAP file data size field (4 bytes) is stored in little-endian order, with the bits in each byte in the usual 7..0 order.
It is extremely likely that the file data pulse-length bytes are stored in the 7..0 order, too.
KERNAL stores bytes in LSB->MSB order, bit 0 appears first on tape.
|
|
|
|
|
|
Posted By
Krill
on 2022-07-26
07:38:59
| Re: endianness of tape file bits
How many more do you need?
There's also $FA04 and $FC0C in the C-64's KERNAL.
|
|
Posted By
Fraser
on 2022-07-26
07:51:58
| Re: endianness of tape file bits
It is confirmed that the bits on tape in the standard formats are little-endian. Sorry to have confused you with my mistake in the original post.
|
|
Posted By
gerliczer
on 2022-07-26
09:55:38
| Re: endianness of tape file bits
If you asked whether written bit order on tape is MSB first or LSB first people would have understood your question better. Endianness is a whole different thing as Krill wrote.
Asking the right question is always the fastest route to get the information. How I wish I would ever be able to get my questions right from the start.
|
|
Posted By
Csabo
on 2022-07-26
10:20:33
| Re: endianness of tape file bits
Actually I got his question right away (and replied that I didn't know which order the bits were saved), I dunno how the TAP format works. Nevertheless we got to the bottom of it, yay! 
|
|
