FMK song import accuracy

[GoogleCodeExporter]

http://kulor.arnoldascher.com/TFM/
FMK import is rather broken, portamentos in particular.

Original issue reported on code.google.com by pissmasterplus on 19 Apr 2014 at 9:55

[GoogleCodeExporter]

Hmm, that sure was of help, though :-) Try now...

Original comment by subz3ro....@gmail.com on 3 Jun 2014 at 6:54

Attachments:

• ADTRACK2.EXE

[GoogleCodeExporter]

Ooops, I misplaced some text/links.

https://dl.dropboxusercontent.com/u/16644842/AT2/MENUONE_pooppegio.rar

Original comment by pissmasterplus on 3 Jun 2014 at 7:16

[GoogleCodeExporter]

https://dl.dropboxusercontent.com/u/16644842/AT2/SPASE_newfix.rar
It appears the fix has fixed that issue and broken other things.

Original comment by pissmasterplus on 3 Jun 2014 at 7:44

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Yeah... so back to the old 'Force instrument volume' concept. I've adapted it a 
little bit for AM-connections and it sounds ok now.
As the popping arpeggio - don't know exactly which one of samples was AdT2, but 
as I performed some tests with state-change machine, I found some bug in effect 
post-processing, which is now fixed. Please try again if something has changed 
- if not, then this must be really some bug (or custom feature :) of arpeggio 
effect in FMK :-)

Original comment by subz3ro....@gmail.com on 4 Jun 2014 at 7:47

Attachments:

• ADTRACK2.EXE

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

The first halves of the recordings should be AT2, the second halves FMK.
I am changing that now, so you get individual recordings per wav with tracker 
and patch number(corresponding to post) in filename.
https://dl.dropboxusercontent.com/u/16644842/AT2/MENUONE_arp.rar

The arp speed appears identical now after the fix.
You will see that AT2 arps start with note 0, while FMK ones with C, for 
0C0/JC0.
That tends to sound a bit different with short notes.
And the weird FMK popping thing.

Oh, and stereo is still reversed.

Original comment by pissmasterplus on 4 Jun 2014 at 11:39

[GoogleCodeExporter]

PissMaster+, are you even reading what I've been writing lately? :-) Stereo is 
reversed because of replay routine bug in FMK! Check the Flip L/R option in FMK 
and it will sound ok. If I reverse it for instruments, it will be crippled when 
there is S0x command used somewhere in the module. If I reverse it both, it 
*can* be crippled for somebody who has used Flip L/R option. This is really 
LOSE-LOSE situation what we would do with touching it anyhow. I will keep it as 
it is, sorry...

As the arpeggio stuff -> the only one bug that was there (and appeared under 
some circumstances) is now fixed. Arpeggio effect *MUST* start with base note, 
just then adding 1st note and after that second. This is *NORMAL* processing, 
so if FMK starts with 1st additional note immediately, this is another bug in 
it's replay routine. I'm sorry, I really much appreciate effort you have given 
to this chase for FMK conversion improvement, but I will not intentionally 
cripple my replay routine with bugs just to have few FMK modules sounding like 
their are sounding in it's native tracker. Only option here would be to 
introduce special effect command for conversion, but I won't do that, sorry. 
This will stay as it is, together with this weired 'popping' thing which 
(whatever it is) seems to be another bug in the replay routine. :-)

What about the sound volume of this lead instruments in SPASE? Is it ok now?

Original comment by subz3ro....@gmail.com on 4 Jun 2014 at 12:58

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

One last idea regarding the panning situation -> what would be the *only* 
correct solution is to have an option in config file, where this reverse could 
be optionally switched off by user. I will do that for next version.

Original comment by subz3ro....@gmail.com on 4 Jun 2014 at 1:01

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Oh wait, Flip L/R is tracker option, not song option :<
My bad, here. Config file option sounds good.

I'll go check volume/timbre with SPASE.

Original comment by pissmasterplus on 4 Jun 2014 at 1:44

[GoogleCodeExporter]

Yes, they are ok.

I found another thing in pattern 0A, chan 15 that has more to do with effect 
interpretation. Seems as if modulator volume is not changing.
https://dl.dropboxusercontent.com/u/16644842/AT2/SPASE_pat0A_chn15.rar

Also, the arp in channel 17, after this section is panned to center in FMK(chn 
18 is panned to side). This is probably an FMK "feature".

Original comment by pissmasterplus on 4 Jun 2014 at 3:44

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Yes, good finding. Txx/Uxx effect commands must be in second effect column, so 
they are processed after setting base instrumemnt volume ;) Ok, here it is 
fixed. Also option for disabling stereo flip is introduced (refer to option 
"fmk_flip_stereo" in the config file and set it to "off" if you like) ;)

Original comment by subz3ro....@gmail.com on 5 Jun 2014 at 8:37

Attachments:

• ADTRACK2.ZIP

[GoogleCodeExporter]

Wooo wooo \o/

Original comment by pissmasterplus on 5 Jun 2014 at 12:12

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Welp, I spotted a playback change that started with patch #57 from this thread, 
outside FMK loading, in an unrelated A2M song.
/Modules/o2star/O2LOVEHR.A2M
pat 02 03
chn 08
row 0C

Compare to patch #51

Original comment by pissmasterplus on 5 Jun 2014 at 9:00

[GoogleCodeExporter]

Indeed. You have good ears ;)

Original comment by subz3ro....@gmail.com on 9 Jun 2014 at 7:34

Attachments:

• ADTRACK2.EXE

[GoogleCodeExporter]

Txx/Uxx thing fixed in patch #68 seems to have been reverted in patch #72.

Original comment by pissmasterplus on 9 Jun 2014 at 11:38

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Yes, I forgot to put last source change to my DropBox backup, so didn't have it 
here at work (they would be merged in the evening when I'm home). As you seem 
to test it thoroughly every time, I put the change brand new for you to have it 
now all together :-))

Original comment by subz3ro....@gmail.com on 9 Jun 2014 at 11:54

Attachments:

• ADTRACK2.EXE

[GoogleCodeExporter]

All that is reasonable seems well.
I'm still curious about the portamento strength and frame counting bugs you 
mentioned, though.

Original comment by pissmasterplus on 9 Jun 2014 at 4:00

[GoogleCodeExporter]

If you find any reasonable constant or conversion table for portamento strength 
adaptation, which accurately works for all FMK modules, it is no problem to 
have it applied during conversion phase. However, I'm sceptic about frame 
counting, as here is very small range of effect command values for some decent 
treatment or manipulation.

Original comment by subz3ro....@gmail.com on 9 Jun 2014 at 8:28

[GoogleCodeExporter]

...or try to ask Sami Wilenius for sourcecode of replay routine for 
FM-Kingtracker :P

Original comment by subz3ro....@gmail.com on 9 Jun 2014 at 8:31

[GoogleCodeExporter]

Fingers crossed :P

Original comment by pissmasterplus on 10 Jun 2014 at 10:20

[GoogleCodeExporter]

You asked him for it, or something? :-) I dunno, but for me the tone 
portamentos sound quite the same in Cruisin' Mah '95 Pimpmobile song, where it 
is heavily used...
Btw, just out of the curiosity I found the root cause for this 'popping' sound 
with the arpeggio in Menuone song -- it is (as expected) a bug in replay 
routine (no range treatment for notes within arpeggio effect processing), so we 
can't keep this one officially, but -- just to show you there really *are* bugs 
in the FM-Kingtracker's replay routine -- here it is (not that *loud* like in 
FM-Kingtracker as some part of the frequency overflow is probably catched by 
other range check elsewhere in my replay routine :-))

Original comment by subz3ro....@gmail.com on 10 Jun 2014 at 11:26

Attachments:

• ADTRACK2.EXE

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Yes, I emailed him.

https://dl.dropboxusercontent.com/u/16644842/AT2/SPASE_lead_porta_AT2_FMK.wav
https://dl.dropboxusercontent.com/u/16644842/AT2/SPASE_lead_porta_AT2_FMK_02.wav
Actually, I am not even sure it's JUST portamento strength anymore :/

Original comment by pissmasterplus on 10 Jun 2014 at 4:28

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

From what I hear, the vibrato is still a little bit deeper with this command 
setup. As the portamento, I can't even notice it within this arrangement. With 
vibrato can't be done more - this is the best generic solution possible. As the 
portamento, for me it's not something really significant, but as mentioned - 
it's no problem to adapt it, if you find good transform constant or conversion 
table.

Original comment by subz3ro....@gmail.com on 11 Jun 2014 at 6:39

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Fixed in version 2.3.48 / 2.4.15

Original comment by subz3ro....@gmail.com on 12 Jun 2014 at 11:56