I uploaded a sample song with one of that special loops in this thread
https://www.psx-place.com/threads/managunz-ps3-backup-manager-by-zar.1198/page-49#post-223049
Take a look at it cypher, and play it in the PS3 to see how the loop has been synced
The bad news is the files generated by the "loops v4" tool made by spirax doesnt works in PS3

Im not sure why it happens... but the files generated by it can be used to see how the times are controlled in the loop
So... what you should do is to create the loop with GWAT3, that tool is relliable, after that we need to edit 1 single value (4 bytes length) in a hexeditor
Following my example before...
Code:
Offset(h) 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
00000000 52 49 46 46 14 FA 07 00 57 41 56 45 66 6D 74 20 RIFF.ú..WAVEfmt
00000010 20 00 00 00 70 02 02 00 44 AC 00 00 99 40 00 00 ...p...D¬..™@..
00000020 80 01 00 00 0E 00 01 00 00 10 00 00 00 00 00 00 €...............
00000030 01 00 00 00 66 61 63 74 08 00 00 00 00 28 15 00 ....fact.....(..
00000040 00 08 00 00 73 6D 70 6C 3C 00 00 00 00 00 00 00 ....smpl<.......
00000050 00 00 00 00 94 58 00 00 3C 00 00 00 00 00 00 00 ...."X..<.......
00000060 00 00 00 00 00 00 00 00 01 00 00 00 18 00 00 00 ................
00000070 00 00 00 00 00 00 00 00 00 08 00 00 FF FF 14 00 ............ÿÿ..
00000080 00 00 00 00 00 00 00 00 64 61 74 61 ........data
The value 0xFFFF1400 is the total number of samples of the audio track... this is dependant of the total time lenght, but also to the "samples per second" of the original wav, i was using 441000... anyway you can consider this value is the 100% of the time
And btw... all this values needs to be converted to big endian... so the 0xFFFF1400 is 0x14FFFF
The 4 bytes before (0x00080000) indicates the loop start... GWAT always uses the value 0x00080000 for it, so always generates a loop from the beginning of track
For the song i uploaded in managunz thread i used value 0x117E00 (as you can see comparing them is a big amount of the total 0x14FFFF... this is because my loop is near the end)
So the result was this:
Code:
Offset(h) 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
00000000 52 49 46 46 14 FA 07 00 57 41 56 45 66 6D 74 20 RIFF.ú..WAVEfmt
00000010 20 00 00 00 70 02 02 00 44 AC 00 00 99 40 00 00 ...p...D¬..™@..
00000020 80 01 00 00 0E 00 01 00 00 10 00 00 00 00 00 00 €...............
00000030 01 00 00 00 66 61 63 74 08 00 00 00 00 28 15 00 ....fact.....(..
00000040 00 08 00 00 73 6D 70 6C 3C 00 00 00 00 00 00 00 ....smpl<.......
00000050 00 00 00 00 94 58 00 00 3C 00 00 00 00 00 00 00 ...."X..<.......
00000060 00 00 00 00 00 00 00 00 01 00 00 00 18 00 00 00 ................
00000070 00 00 00 00 00 00 00 00 00 7E 11 00 FF FF 14 00 .........~..ÿÿ..
00000080 00 00 00 00 00 00 00 00 64 61 74 61 ........data
I had to do 4 or 5 tests with the tool made by spirax to realize it was not working... then examining his files to get the confirmation of where the loop info was located... then 4 or 5 more tests to find the exact value when i was trying to do the loop, that was tricky, but in audacity you can get an idea at the value by changing the counters to "samples"
Is a bit tricky but it works pretty well... and btw, when the XMB is playing the song and does the jump to "restart" loop, in that transition it does it very smoothly... much better than the PSP, i remember to be doing this for the PSP and the results was not so good, but in PS3 is perfect... you are going to have problems trying to imagine where i did the "cut" for the loop :P
The PS3 supports ATRACT3 132kbps STEREO
The 66kbps was the max for PSP, but it seems in PS3 we can use 132kbps
Not tested by myself, but is mentioned in the reddit tutorial
At this point this is tested, when i made the conversion from .wav to .at3 i used the max quality that is 132kbps
Souns quality is good, btw, i made that (weird ?) step mentioned in the reddit tutorial with goldwave to change the "playback rate" from 44100mhz to 40200mhz, not sure if this was really needed or a good idea because i dont even understand why is needed this step... but anyway the results was good