Messages

When you say it 'does not play as intended', do you mean it sounds normal, just the wrong speed, or that it's really messed up?

The sound messed up.

Can you please try playing this file and tell me if it works?

That one appears to play correct. Well done! Let's do that same treatment to the other 8 channels modules and we should be set. Does the format description need to be updated as well?

Hmm.  Yes.  I do not like that at all.

There's nothing wrong with the format specification - there appears to be a bug in Milkytracker (or whatever you're using).  It's obviously a very common mistake - I ran into the exact same problem when testing at this end, until I found an older and less generic player which handled them correctly.

The difference between the two is in the patterns.  StarTrekker (as detailed in the text from Exolon here) uses an unusual layout, in that the patterns are 'doubled up' - that is, each number in the sequence is multiplied by two, and the odd-numbered patterns contain the data for channels 5 to 8, which is the format I converted them to.

The test file I just posted here does not adhere to this format - rather, all eight notes are kept together in the same pattern, with one pattern being twice as long as usual.

So I could convert them to this specification, but it would be wrong to do so.

Thank s for the upload, excellent job converting the OMF files to StarTrekker modules. However I can tell that all converted 8 channel modules (FLT8) do not play as intended, hence there is some flaw in the conversion process. I hope you can solve that. Maybe Asle can assist you here?

When you say it 'does not play as intended', do you mean it sounds normal, just the wrong speed, or that it's really messed up?

Can you please try playing this file and tell me if it works?

OK, sent.

OK - I put them in uploads/omf_conversions.  The format specification's there as well.

How should I do that?  Can I just upload them from my FTP application, and if so, is there a certain place where they should be put?  (Maybe a subdirectory in 'Onyx Music Files'?)

Here they are!  (I could send the converter source code as well, but as it's written in BBC BASIC for Acorn machines, it probably wouldn't mean anything to you...)

I chose to use StarTrekker because it was the only 8-channel MOD derivative I knew about (and could play), but I'm aware that one or two things aren't quite correct ('Biologii', for example, uses CIA tempo commands).

Hope you enjoy them!

Hey deus-ex,

Is MODLAND down?  I just tried to download some OMFs to test my new converter, but I just got 'connection failed' - and the same message on any further attempts to access the FTP.

Just to let you know - I'm looking into this.  I've converted quite a few esoteric formats in my time, and this one isn't proving too difficult (helped massively by the ProTracker conversion Larry found, of course!).  I should have something ready by tomorrow.

Points of interest: the notes seem to be stored backwards in every pattern (first time I've seen that one!), there is (as far as I can tell, completely useless) three bytes padding before every pattern, and the sample data is unsigned.

I'm not sure I understand your problem with this song. What do you call "length-of-sequence", and what means " too large" ?. I've loaded up the song in Milkytracker and there's no "empty" space in any sample.
And what player are you using to replay this ? (DP is fine for this).

ModInfo says:

Pattern order: 1 ,9 ,2 ,3 ,5 ,4 ,6 ,7 ,8 ,0, 10 ,11 ,7 ,8 ,12 ,4 ,6 ,3 ,5 ,13, 14 ,15 ,16 ,17 ,18 ,0 ,0 ,0 ,0 ,0, 0 ,0 ,0 ,0

I'm pretty sure those nine zeroes at the end aren't meant to be there.  (It's more obvious if you listen to the tune.)

There are two possible reasons:

One, the author put them there as padding when composing the tune and forgot to remove them afterwards (I'm not familiar with ProTracker, so I don't know how easy it is to do this), or:

Two, the tune was ripped from something and the ripper misinterpreted the byte giving the length of the song (or the byte was corrupted at some stage), resulting in it thinking the sequence data (or 'pattern order', above) was longer than it actually was.  This is far easier to do in ProTracker than in other formats, because the length of sequence data is constant - 128 bytes.

Regarding Ghost and Goblin's, well, you can get it here : http://www.exotica.org.uk/wiki/Ghosts_%27n_Goblins
It's no tracker music (as far as we know) and, as such, not on AMP.

Oops - sorry, betraying my ignorance of Amiga music history there!

Take the following Cutcreator tune:

http://www.amp.dascene.net/analyzer2.php?idx=12277

This exhibits a problem I've seen on a few other tunes too, such as some of Nuke's 'Universe' compilations.  The length-of-sequence byte appears to be too large.  This results in the zero padding beyond the end of the sequence being counted as part of it, and interpreted by the player as an order to play pattern 0 nine times over.  Any idea what causes this problem?  I assume it's the result of a faulty ripping program somewhere along the line...

Another, unrelated, quick question: What was the original Ghosts and Goblins tune?  Searching for 'ghost' turns up two pages of remixes!

They've been converted to an Acorn specific format, unfortunately, so I have no idea what their original format was.  However, the samples used (monobass, synbrass and all that lot) and the fact that none of them go over sample 15 suggests Soundtracker.

I've converted the first one to ProTracker and attached it here - let me know if you have any thoughts.

OK, I'm wondering about the author of four ProTracker files I have here: 'thomsong1', 'thomsong2', 'thomsong3' and 'thomsong4'.  They certainly sound as if they were written by the same person.  The filenames are, respectively, 'germ', 'claves', 'popcorn' and 'stealth'.  The nature of 'popcorn' should be obvious!

I haven't managed to find any of them on this website.  None of them contain any indication of their authors, so it may have flouted that rule if somebody tried to upload them, but otherwise does anybody know anything about these?

I don't have Windows of any variety, so DOSbox is not an option.

hum .. DOSBox is _not_ only for Windows ... see their download page.

Yeah, not for RISC OS, though.

Thanks Deus-Ex

Thank you indeed!

'AdLib' - right.  As far as I can tell from Google results, that's some sort of chip sample generator, right?  I'm guessing the instrument definitions only hold parameters for that, then.

It was a look at the binary data of the file that led me to think it was truncated.  There was obviously no sample data, but I couldn't see much in the way of patterns either - so even if I could play it, it looks as if it would have been a very short song!

Thanks for all your help, anyway.  I don't have Windows of any variety, so DOSbox is not an option.

Completely by accident, AMP has just led me to discover a hugely talented artist of whom I was previously completely unaware: Falcon.

Having downloaded a few tracks at random (not having any other way to choose), I'm seriously impressed.  Many thanks for making this stuff available.

Unfortunately, I've just discovered one of the uploads is faulty.  The track is named 'Voices and visions vol.1' (http://amp.dascene.net/downmod.php?index=84109), and is listed as being 1K in size.  The reason for this is that most of the file is missing.

Please can this be fixed?