couple more file formats?

[caffine77]

I was just bragging to my friend's fiance the other day how cool winamp was 'cause it can play WAY more music formats than WiMP (I sent him an .mtm it wouldn't play). But then I found one today that it wouldn't! Eek! Install another media player besides WinAmp? Here are the two formats I've found so far: .far and .amf. I guess they're both along the same lines as .mods, so I was kinda surprised they weren't already supported since you have other similar formats in there (.669, .mtm, .mod, etc.) Here's a player that does support them, but crashes a lot:
http://www.castlex.com/modplug/player.php3 Hopefully that helps. Don't tell him there are formats WinAmp doesn't yet support, then I'll never get him to switch! thanks and ttyl,

mich

[X-Fixer]

Winamp can't play really lots of modules formats (to say the truth, it can play only few of them). And the mod plug-in develompent is almost stopped. If you really want to play them you will need another player (the best one for Windows I know is DeliPlayer).

[caffine77]

oh yeah, I forgot about DeliPlayer, had that one a while ago. So it's pretty much to be assumed that WinAmp will *never* support these formats?? :( I'll take this opportunity to post something completely ignorant and stupid, but how hard would it be for me to try to add this myself (write it)? I've never written any kinda media player type of stuff, but I'd WAY rather have WinAmp play all of my media than have two media players... especially for music since I'd like to add the songs together in the same playlist. put my mp3s, .wavs, .oggs, .mtms, and .amfs all in the same place. Such a fun player... I think I have 1.90 burnt on a CD somewhere from a long time ago. :p Anyways, so what can I do to help WA and how hard is it? Are there any good links to start with? :winamp:! ttyl,

mich

[X-Fixer]

Well, I can do this, say, a bit later. But you should provide links to files of all formats you need.

[caffine77]

here's a link to a .far song:

http://debian.attica.net.nz/kosmic.org/1993/bdda-klf.zip

here's a couple .amfs:

http://debian.attica.net.nz/kosmic.org/1993/way-klf.zip

http://debian.attica.net.nz/kosmic.org/1993/sund-klf.zip

http://debian.attica.net.nz/kosmic.org/1993/relg-klf.zip

hope these help! :winamp: ttyl,

mich

[caffine77]

hey wow... that mod player that always crashes plays Unreal Tournament .umx files... that'd be pretty cool, huh? :D Obviously I can't post one, but I bet someone around there has one.

[peter]

umx are just s3m/it/xm containers, they all do work in Winamp if you only extract them with proper software.

[X-Fixer]

OK. I'll look what I can do here.

btw, if you have files in any other unusual mod format do not hesitate to give me a link ;)

[X-Fixer]

So here it goes. Changes:

- FAR loader
- info box close on esc
- remember position of info box (relative to main window)
- fixed lame bug in STM loader (detection)

available from here:

http://x-fixer.narod*****download/in_mod.zip

I really want to test it more thoroughly. Any help is appreciated. Cause the most interesting point is effects, I can make a special version that will blow up a message on any unimplemented effect it finds (so you can send the module to me).
Formats under testing are

- FAR
- 669
- AMF (comming soon, hopefully)

If someone is interested, post here ;)

peter: please, move this to your forum. I think there are better chances to find volunteers there.

[X-Fixer]

Next version (beta9). Changes:

- 699
- added tempo effect
- fixed displaying of sample names
- fixed reloading bug (hit play on a module, open info box, play another one, play this again)
- some fixes (now sounds more like in MPT)
- more info on samples (in info box)
- AMF loader

Also under testing:
- MED

caffine77: the work is done. where is my beer? ;)

peter: send me pm if you read this

[caffine77]

Originally posted by X-Fixer


caffine77: the work is done. where is my beer? ;)

peter: send me pm if you read this

:up: Two beers! cheers X-Fixer :D Can't wait to get home and try this out... on vacation for the weekend, drat! :p excellent (and quick!) work. Bravo. :) :winamp: ttyl,

mich

[caffine77]

Originally posted by peter
umx are just s3m/it/xm containers, they all do work in Winamp if you only extract them with proper software.

How does this work/what does this mean? sorry, I'm kinda more of a networking/systems kinda guy... don't do much with the media but listen to it while I'm working :) So I need to copy my .umx files somewhere and convert them, huh?

[X-Fixer]

UMX files are archives (like TARs). You should extract files from .umx before you'll be able to play them in Winamp. Try this (http://extractor.far*****cgi-bin/bsdcc.pl?utpt16beta3), or anything else you can find.

[X-Fixer]

Update (beta10)

- IT
- recognises "end of song" in orders list
- displays format version instead of created with
- 669
- tempo setting reverted back (almost)
- fixed samples looping
- FAR
- fixed panning
- AMF
- more effects
- load 1.4 files now
- fixed panning
- MTM
- fixed comment reading
- fixed lame internal bug
- corrected loop info on samples (in info box)
- OKT loader (MPT sucks!)


It will be really cool, if someone will get a blow-up build and test new loaders on a large amount of modules. This is the only way to add support of all/most effects in loaders.

caffine77: if you have more FAR files, send them to me.

[RIV@NVX]

I will integrate it into 4PSetup.

[X-Fixer]

Originally posted by RIV@NVX
I will integrate it into 4PSetup.

Great, but help with testing is a bit more appreciated ;)

[bigboo]

hi

I found the original zip archive of Composer 669 v1.2: it contains the software, documentation and ten 669 demo files


I also found some modules in various formats and i don't know if there are played correctly with in_mod:

Name: modules.zip
Size: 188Kb
Decompressed Size: 316Kb
2 files in archive:

------------------------------------------------------------------------------------------------------
Name | Size | Packed | Modified | Path
------------------------------------------------------------------------------------------------------
BEERBOTT.669 154 637 82 939 01/01/80 00:00
DARKRIDE.MTM 168 753 109 329 09/01/94 19:06




Name: stm.zip
Size: 409Kb
Decompressed Size: 724Kb
6 files in archive:

------------------------------------------------------------------------------------------------------
Name | Size | Packed | Modified | Path
------------------------------------------------------------------------------------------------------
AGONY.STM 102 773 68 000 23/10/92 20:20
CREAM.STM 167 436 88 116 23/10/92 20:20
DEATHSTA.STM 139 132 90 035 23/10/92 20:20
ICE_T.STM 42 017 21 474 23/10/92 20:20
MASQRADE.STM 94 880 46 899 23/10/92 20:20
TRANSF.STM 194 836 103 869 23/10/92 20:20


if you are interested i can send you the files

[X-Fixer]

Cool. Upload composer and modules.zip somewhere (do not send by email!), I wanna look at them.

MTMs and STMs seem to be played all right.

[RIV@NVX]

I have only mods, so I can't test it.

[bigboo]

x-fixer i send a pm to you with the address to download the files

and if you want i can test mods i have with your special in_mod version just tell me where i can download it

[Bedeox]

This (and also Peter's) in_mod incorrectly plays
xm instrument envelopes.
I've made two identical modules in two formats
to show it.

They should sound indistinguishable.

They're attached and zipped.

[X-Fixer]

Update (beta11)

- OKT
- more effects
- some changes in effects handling
- one small change, but I won't tell you :-p
- MOD
- added 2 more protracker-clones
- Old soundtrekker
- fixed rare bug
- MTM
- fixed rare crash
- 669
- made pitch slides linear, again
- XM
- fixed lame bug in panning envelopes
- fixed arpeggio bug in player
- fixed info box position saving


Blow up build (shows messages on unimplemented effects in extended669, FAR, AMF and OKT) here:
http://x-fixer.narod*****download/in_mod_bu.zip

It would be great if someone will test 669 with original composer and say how much it differs, especially songs with vibrato effect, like culture.669. (I can't run composer, because it won't run with emm, but my soundcards' legacy drivers require emm).

Bedeox: "my" and "peter's" plug-ins is basically the same one, based on MikMod (and most of my changes are to MikMod, not to plug-in). "Not correctly". LOL. It was absolutely ignored. Thanks for good report. And there was no need in sending me pm ;)

[bigboo]

arf this is a mod that was working correctly before but is broken in beta11:

[bigboo]

669 files are played differently (and I think better but I don't know) in modplug
for example with in_mod:
Beerbott.669 there are strange "click" in the sound (not in modplug)
Fonetag.669 tempo is much faster than in modplug


and for composer 669 maybe you can try DOS BOX, it's an emulator of old 386 with real mode and sounblaster 1.5 wich run on windows and linux:
http://sourceforge.net/project/showfiles.php?group_id=52551


voila that's all everything else is ok for me ;-)

[X-Fixer]

>arf this is a mod that was working correctly before but is broken in beta11

fixed. will be available with next beta.

>669 files are played differently

clicks are known. I'm trying to get rid of them with no success.
tempo is incorrect in MPT (also known).

DOS BOX won't help, because composer needs extended mode.

[bigboo]

hehe you're right dosbox doesn't work

but maybe this emulator would work:
http://bochs.sourceforge.net/

it seems very complete! but I have not tried it yet

[caffine77]

not sure if this is something that's supposed to be disabled or not, but we used to be able to bring up info about the song for the .mtms (I used the alt-3, but I think it's also in the right click menu).

also, found an amf that says it's not supported (I think you're trying to track those down?) here's the link:
http://debian.attica.net.nz/kosmic.org/1993/klfjpach.zip
but yay! I tried a bunch of amfs, and that was the *only* one that it didn't like :D

Sorry I've been MIA the last few days! I can't seem to find more .far files... grr! I searched both my main computers, and am looking around the net, but not having a lotta luck... and those RAR people that made a different .far format... yikes! However, while doing that, I ran across this page... http://fileformat.virtualave.net/ind_wave.htm it appears to have a bunch of specs, including .far. Didn't make much sense to me, but prolly would to you. Hope it's helpful :) I'm still lookin' for those .fars.

Hey, I haven't installed it here yet, but I noticed that Winamp3 plays .umx files natively... think that might sneak its way into this? ;)

this is soo cool :) :up: ttyl,

mich

[caffine77]

well, here are two .fars that I found, but neither seems to work (in modplug either), so I don't know if they're corrupt, invalid, or what. But just in case they're of any use to you, here is the link:

http://www.artificiallives.com/mods/modarch/far/

ttyl,

mich

[X-Fixer]

>not sure if this is something that's supposed to be disabled or not

for me info box works well on MTMs. What's wrong?

>also, found an amf that says it's not supported

now it's supported :)

>I tried a bunch of amfs

Please, look what versions of AMFs these were (look in info box).

>Hey, I haven't installed it here yet, but I noticed that

have you read peter's and my posts about umx (above)? if wa3 can play them this just means that it supports extraction from umx (like from zip/gzip/rar/etc that currently exists in wa2).

>well, here are two .fars that I found

thanx. they're both OK and play in MPT for me. and now in winamp, too ;)

bigboo: can you give me file from the modules2.zip (starttrekker tune with title "song3") again, but WITHOUT others?

And may be it's good time to say. Guyes, I've made file detection proggy that detects over 150 modules formats. I know most of specs resources. Giving me link to one or sending me spec is absolutelly useless, unless you have something really rare (like AMF spec).

[bigboo]

ok this is song3.mod alone in his little zip

[bigboo]

I found 2 mods not playing correctly (okt and xm) (you can download it where I posted the old zips)
modsoktxm.zip (soory its still a zip file i'll download winrar one day maybe ;-)

the xm one sounds better in modplug(in the begining of the file), the okt is maybe broken because it doesn't play correctly in deliplayer

I know that formats are not supported but tell me if you are interested i'll send them to you:
1 polytracker
2 digibooster
3 futurecomposer
11 med files

voila that's all the strange mods I have on my HD so now be happy I'll stop posting all the time here ! hehe

[bigboo]

in the in-mod preferences/mixer tab

1. size of the white boxes (for output format / max voices / mixing rate) may be reduced

2. there are 2 typo in the text under the interpolation checkbox:
improvement
efficiency


and no I am not boring LoL

[caffine77]

Originally posted by X-Fixer
>not sure if this is something that's supposed to be disabled or not

for me info box works well on MTMs. What's wrong?

Here's my setup, I just use the playlist editor, fyi... but when I'm trying to look at the info for an .mtm, I press the alt-3, and the playlist editor loses focus, but nothing else happens. It's like it knows it's supposed to do something, but doesn't. I just dled your in_mod.dll fixes a second ago too. :( I can get the info boxes if I put the original in_mod.dll back in, but not with the new ones. I'm on Win98SE, if that means anything to anything.

>also, found an amf that says it's not supported

now it's supported :)

sure is... yay! :D


>I tried a bunch of amfs

Please, look what versions of AMFs these were (look in info box).

Can't. info box isn't working for those either :( They all work, tho. But you wanna see how many versions we do know work, right? When I can get the info box working, I'll post :)


>Hey, I haven't installed it here yet, but I noticed that

have you read peter's and my posts about umx (above)? if wa3 can play them this just means that it supports extraction from umx (like from zip/gzip/rar/etc that currently exists in wa2).

Yeah, I saw the postings about extracting them myself, but I was hoping that eventually there wouldn't be a need to do that - that the files would be able to played in their own place. Fundamentally, that's the whole reason for playlists, right? So you don't have to copy and rearrange your music files around your hard drive all the time? (and I spose so you can dictate the order in which the songs are played).

But anyways, never mind... Winamp3 doesn't actually support them, so I have no idea why it's even in the media preferences. I tried playing some in there, but it doesn't pay any attention to them, but seemed more than happy to (uselessly) associate them with itself. Enigmatic. :igor:


>well, here are two .fars that I found

thanx. they're both OK and play in MPT for me. and now in winamp, too ;)

I can't get 'em to play in winamp with your new in_mod.dll :( Am I still doing something wrong? Between this and the missing mod info boxes, I must be screwing something up... I'm just extracting the new in_mod.dll from the archive and putting it in my Winamp\Plugins directory. What else should I be doing? Would there be any reason to think that if I'm keeping all three versions (original, bu, and fixed) in Winamp\Plugings\old that this would cause a problem? In the Input Plug-ins box in Winamp, it says I'm using Nullsoft Module Decoder v 2.2.10 beta11. That's the latest and greatest, right?


And may be it's good time to say. Guyes, I've made file detection proggy that detects over 150 modules formats. I know most of specs resources. Giving me link to one or sending me spec is absolutelly useless, unless you have something really rare (like AMF spec).

okey dokey. sorry 'bout that. wasn't sure if it'd help or not, but figured in case it did, I'd do what I can to help :)

[X-Fixer]

about last fixes (amf & far) - i was referencing to the version, that I have not uploaded yet ('cause I'm not going to upload new version every hour ;) ).

about info box - try to open winamp.ini and delete info_x=, info_y= lines from it.

about wa3 and UMX read here (http://forums.winamp.com/showthread.php?threadid=97325).

[caffine77]

Originally posted by X-Fixer
about last fixes (amf & far) - i was referencing to the version, that I have not uploaded yet ('cause I'm not going to upload new version every hour ;) ).

about info box - try to open winamp.ini and delete info_x=, info_y= lines from it.

about wa3 and UMX read here (http://forums.winamp.com/showthread.php?threadid=97325).

ah, well, then no prob about the .fars, I understand. I noticed that the date and time on the in_mods were pretty recent, so I thought they were the latest thing. I don't like to upload with every single little change too :)

Took out the in info lines in the winamp.ini, and still no avail. The playlist editor still loses focus, but nothing comes up. The info_x and y lines were both set to 0. There was also a stray "=0" without a var name, but removing that didn't help either. :( I can get an info box for mp3s, wavs, etc if that helps you at all. Just not the mtms, amfs, mods,... You don't think a whole new install would help anything do you? All I've done is change the in_mod.dll... I haven't molested it in any other way.

That's fine that UMX doesn't work in wa3, that's kinda outside of what we're working on here. I should have actually tried to play one in wa3 before I said something, but I've had such good luck with WA, I thought if it said it would work, then I didn't need to :( Still seems goofy to me that it'd be in the media preferences if it's not actually supported (ie. separate dl needed :weird: ). Maybe when wa4 comes out ... :p I only mentioned wa3 thinking that if they'd done it there, maybe it would be easy to bring into here. But since the former doesn't exist, the latter doesn't matter. ttyl,

mich

[peter]

X-Fixer: nice job, i'm only afraid that i will have some sort of trouble porting the new changes to Winamp3 (perhaps i'll redo entire wa3 port), i dont think i can give the sourcecode away, this time for real.
i'd love to move this to my forum, but i am afraid that vbb doesn't allow me to do that (i can put a sticky with link to this thread).
if you want weird mod files - sure, i have a bunch of them (been hacking Jazz Jackrabbit 2 .j2b format, managed to get through the compression but i still can't play what i get in output), i'll upload them somewhere later.

[X-Fixer]

Update (beta12)

- Old soundtrekker
- refixed patterns counting
- AMF
- changes in effects handling
- FAR
- more effects
- fixed bug with high positions
- lame bug fixed
- MOD
- more protracker-like formats
- IT & XM
- fixed retrigger effect
- OKT
- reverted some things back
- fixed "fadeout with no envelope" bug in player
- clipping of info box position


bigboo: okt and xm, that you've sent, should play ok now, but, please, check it. and no, I do not need any files now.

caffine77: this should help with your info box probs. (btw, what winamp version do you use?)

[peter]

or better: any ideas WTF are .AM files ? (thats apparently what i've got)
they've got RIFF headers and stuff, i think they're possibly converted from impulsetracker (but i'm not sure)

[X-Fixer]

when asking about RIFF files it's good idea to give RIFF magic field. If it's "AMM2" I can give you specs. You can try my filetype prog - if I have any idea about a format it's very likely to be detected by it.

[peter]

the main riff type is 'AM ' (with 2 spaces, damn vbb). i've found a proggie to convert other mod formats to AM, i'm working on my own hack; since only one game (Jazz Jackrabbit 2) appears to be using it, i'll put an AM->IT converter into my new game music ripper.

[X-Fixer]

seems I've never heard of it. does this proggy say the name of format?

[peter]

seems that i need a little bit of clarification here.
1. Jazz Jackrabbit 2 game uses their own .J2B format.
2. i've reverseengineered J2B itself, it's a mod container, it appears to contain .AM files compressed with regular zlib. thats where i got my AM files from.
3. the official mod2j2b.exe proggie (available on their official pages) says that it supports "AM" format so i was starting to think that this format is used more widely, but i can't find any more info on it.
4. the "converter" i've found is mod2j2b.exe; it converts everything to .AM, then packs into J2B. that's all. since i know how to unpack J2B, i have whatever->AM converter, so i can continue reverseengineering stuff (jazz2 music has been a royal pain in my ass for years).
5. i could write J2B loader later, just for the hell of it (and to piss epicgames guys off for using weird formats).
6. i believe that all (or most of) .AM files got converted from impulsetracker anyway; making an AM->IT converter should be relatively simple now.

[X-Fixer]

OK, I see. If you'll make AM->IT converter, it'll help to write a mikmod loader. though IT loader is wicked and I prefer to do thing in another way.

[bigboo]

ok okt and xm are ok now
I found some Ultimate Soundtracker files that are playing wrong, but maybe the are just bad files

(rename .zip to .7z because I can't attach .7z files only .zip)

[bigboo]

-arf and your fix to xm modules seems to have broken something else these xm files were working properly before
-the .mod one is another wrong playing Ultimate Soundtracker

(it's a 7z too)

[caffine77]

Originally posted by X-Fixer
Update (beta12)

caffine77: this should help with your info box probs. (btw, what winamp version do you use?)

I'm using Winamp 2.80 (needed the ogg support :) )

My .amfs are reporting to be DSMI AMF 1.0 and 1.1. The message boxes are back! :D Yay for beta 12 :)

Those two .far tunes that I posted still don't play. It says it's corrupt or unsupported. I'll try to dl it again and see if maybe both of the dls got messed up, but doesn't seem like it would have... Good job so far! (no pun intended) ttyl,

mich

(Hey, we're loading weird j2b files, and umx gets pushed aside? now I gotta find that Jazz2 game and uninstall Unreal Tourney and Deus Ex :p ha ha )

[peter]

about .umx - my old game music ripper (http://www.blorp.com/~peter/old/wr.zip) does them all. making umx loader wouldn't be extremely difficult, i did one for wa3, but its kinda hacky (it does dumb header scans, i still haven't actually reverseengineered umx format, i can only rip stuff from umx by scanning for actual mod file header).

about AM->IT converter - i plan to convert to IT, because i believe that at least 90% of existing AM files got originally converted from IT, so IT has all features used by AM. i know that IT loader is nuts, i spent many hours on fighting it (#@! sample compression and other shit).

[bigboo]

I found a web page with all pc trackers avaible to download:
with composer669 unis669 farandole etc...

http://www.hornet.org/music/programs/trackers/

maybe it will help you


Peter : about .am format
it seems it supported by deliplayer
http://www.deliplayer.com/

here the plug-in info:
A.M.Composer 1.2 1989 by Marc Hawlitzeck,adapted by Mr.Larmer/Wanted Team

it is an old amiga format maybe its not the same as the one in your game

[caffine77]

Originally posted by peter
about .umx - my old game music ripper (http://www.blorp.com/~peter/old/wr.zip) does them all. making umx loader wouldn't be extremely difficult, i did one for wa3, but its kinda hacky (it does dumb header scans, i still haven't actually reverseengineered umx format, i can only rip stuff from umx by scanning for actual mod file header).



That works really well. :) It there some kinda plugin or something that would take a file (say, .umx) if unrecognized, try to run it thru the ripper, then queue/play the file in winamp if it returns true? Also wondering... I'm not seeing it anywhere (so assuming that it's not supported) that the ripper doesn't do multiple songs at a time?

[peter]

you should be able to scan entire directory with the old ripper (i haven't used it for a while, but typing "c:\unreal\music\*.umx" should work). new ripper beta will come out within a few days, probably as soon as i'm done with .AM format.

[caffine77]

okey dokey. I was just trying to select multiple files in the dialog box, or even just a folder. I tried running it in DOS (with an argument) to see if I could batch 'em thru that way too :p Tried it and it works like a charm... bing bang boom! thanks :) ttyl,

mich

[Bedeox]

your xm fix works, but...:(

it breaks lots of XM volume envelopes
and loops. (This probably has something
to do with "fadeout" envelope)

[X-Fixer]

bigboo: Do not send me files, that plays "wrong" unless you now player (preferably DeliPlayer or MPT), that can play them "right". Most of your mods are broken. And I have fixed XM, I guess. And I'm 99% sure that your guess on AM is wrong.

caffine77: how many AMF 1.0 files did you play? also that's bad, that there is no 1.2 and 1.3 files (code assumes that 1.2 are like 1.1, and 1.3 are like 1.4 ('cause I have one), but it will be great to check it).

About UMX. caffine77 - I gave you a link to unpacker, do you remember. peter - I can give you spec, if you can't find one.

Bedeox: see above ;) I'll upload update as soon as I'll kill obscure bug in seeking.

[caffine77]

Originally posted by X-Fixer
caffine77: how many AMF 1.0 files did you play? also that's bad, that there is no 1.2 and 1.3 files (code assumes that 1.2 are like 1.1, and 1.3 are like 1.4 ('cause I have one), but it will be great to check it).

4 files were DSMI AMF 1.0 files. 10 1.1. Sorry I don't have more to check with :(

About UMX. caffine77 - I gave you a link to unpacker, do you remember. peter - I can give you spec, if you can't find one.

I know there was the link for that unpacker, but I was hoping for something that would play the files without needing to make a second copy of them on the computer. Peter's extractor does that well, as far as i know... I was just looking to see if there was a way to automate it more. Maybe it still copies files around in the temp dir or something, but at least it's transparent to me :)

[X-Fixer]

beta13

- S3M
- recognises "end of song" in orders list
- undid fadeout fixes
- refixed retirg effect
- fixed strange seeking bug
- improved comment reading


That's all for now.

It would be really great, if someone who knows how to use trackers will test all effects with zeros and half-zeros (I mean when one nibble is zero) params, especially when it comes as the first effect of that type (yes, I mean when "same as previous" is used without previous - I think that should use some default settings, but mikmod tends to ignore such effects).
Bedeox, would you try?

[bigboo]

ok beta 13 is working great!
no more bugs, everything is working correctly, at least for me.

nice job!

[bigboo]

ahah ok I found 2 website with .amf not playing at all with winamp but working in modplug:

http://www.thetaband.com/npdsmi/nptest.html
http://www.doofus.org/audio/midi.html

there are 10 .amf in total but I don't have the time to download them all sorry
gslinger.amf amf1.4 seems ok
testamf.amf doesn't load modplug ok
nilesong.amf doesn't load modplug ok


I have also a little suggestion about 669 title:
it seems that you use the comment section of 669 files to display the title but in some 669 comment is long or strange (with many space) and it doesn't look good at all in the winamp main window.
maybe you should display this comment in a comment tab in the in_mod info box (like modplug) and use only the filename as title (like deliplayer).
but it's just my point of vue of course ;)

[Bedeox]

That works great!

You're the man :up: :up: :up: :D

I'll see what I can do.

[bigboo]

here there is info about the am format (the good one this time I think)

http://pipin.tmd.ns.ac.yu/extra/fileformat/modules/index.htm

http://pipin.tmd.ns.ac.yu/extra/fileformat/modules/am/am.txt

[Bedeox]

There is a more generic problem with plugin.

Oscilloscopes in 24bit mode crash Winamp.
Obviously, non-working scope crashes vis plugins
as well.

Peter had fixed this in earlier version,
but his hack resized scope by half.

edit:
I have done some testing, and it seems,
that IT v2.16 and FT2 skip continue-without-value
effects.

edit 2:
Got my hands on Scream Tracker v3.20
and it ignores them too.

[Bedeox]

To easily create such testing mods,
I use ModPlug Tracker (http://www.modplug.com/modplug/download.php3?session=&download=Tracker) and then play
them with original tracker(FT2, ST3, IT)
I play MODs in ModPlug Player (http://www.modplug.com/modplug/download.php3?session=&download=Player) or Winamp,
so I can't help much
in this case. (I don't have ProTracker!)

[caffine77]

Originally posted by bigboo
ahah ok I found 2 website with .amf not playing at all with winamp but working in modplug:

http://www.thetaband.com/npdsmi/nptest.html
http://www.doofus.org/audio/midi.html

there are 10 .amf in total but I don't have the time to download them all sorry
gslinger.amf amf1.4 seems ok
testamf.amf doesn't load modplug ok
nilesong.amf doesn't load modplug ok


All the songs on those pages:
k_shalam, gslinger, panic, and testems, all report being DSMI 1.4, and seem to play just fine (and the red dwarf mod too, but it's not as good as the real theme).

goodbye, bluesky, testamf, anybody, ovrdrive, and nilesong don't work in Winamp, but do in modplug.

I'm using Winamp 2.80 with in_mod beta 12 (hadn't picked up 13 yet, but if you didn't change anything with the amf stuff, then it wouldn't matter). ttyl,

mich

[Bedeox]

caffeine, there is UMX reader at
http://www.umx-decoder.de.vu.

But it does only extract files
and send them to in_mod, so you can
read only one part of Deus Ex UMX.

[Bedeox]

Deus Ex UMX are multipart modules,
which are of this format:

Pattern 1 (...) Jump to pattern 1start
Pattern 2 (...) Jump to pattern 2start
.
. (max 5 parts )
. (order: Ambient, Action, Conversation, Death, Exit)
.
pattern 1start
(...)
pattern n (...) Jump to Pattern 1start
(...) something here or not
pattern 2start
(...)
pattern n2 (...) Jump to Pattern 2start

[Bedeox]

Link to the UMX format draft:
http://www.home.mastersites.com/umxdecoder/files/UT_Package_File_Format.pdf

[caffine77]

Originally posted by Bedeox
caffeine, there is UMX reader at
http://www.umx-decoder.de.vu.

But it does only extract files
and send them to in_mod, so you can
read only one part of Deus Ex UMX.

huh? I didn't say something right, it appears :( Until winamp reads umx files on its own, I'm diggin' peter's ripper. I think I'm a happy camper in the UMX respect unless you've found it doesn't work right... is that what you're saying about that multipart stuff? ttyl,

mich

[Bedeox]

both WinRipper and this UMX reader
and UnrealED
fail on Deus Ex files,
because they're non-standard
and must be broken down into parts,
because Winamp in_mod ends playback
on first jump to already played pattern.

UMX reader is only for automation.

[Bedeox]

x-Fixer: Could you add SSRC resampler
to the plugin? My ears can distinguish
between FIR resampler used in In_Mod
and SSRC. SSRC is just better and it could
be implemented as an configurable VeryHQ option.

[X-Fixer]

bigboo: I've fixed amfs (MPT sucks again, try testamf when i'll upload beta14). About 669 - I've thought about this, but I think that file name is somewhat boring/tends to be changed by evil bastards, who destribute files over the net. Song message doesn't look good in playlist, but looks ok in main window imho, where it scrolls. I can't use 2 different versions of title for main window/playlist and I do not want to stupidly truncate name to 16 chars, like MPT.

bedeox: are you sure/what effects have you tried? at least I know, that multi-retrig has default value of 1 for speed. (i have protracker, but hope that MPT and, say, IT or FT2 should be enough). as for plug-in, as I've said, I'm not responsible for winamp front-end, it's done by air, justin and peter and I don't feel like touching it (24-bit output is on peter's responsibility, afaik). also, I do not feel (at least now), like changing resampling - I can't even feel big difference between FIR/no-FIR (because of DFX, I guess). I'd first fix resonant filters, because they do not work at all and make much bigger difference.

[Bedeox]

sorry, i havent't tremor and tremolo, but most effects
(portamento, slides, vibratos, arpeggio, panbrello, offset)
work as I said

retrig, glissando, controls(Sxx) have NO continue,
so they must have default values.
Speed set to zero gives speed=1.

[X-Fixer]

why do you think so? at least in MPT and FT2 multi-retirg seems to have separate continue on both nibles (I mean "continue speed" and "continue slide"), with default values 1 for speed and 0 for slide (though my version of mikmod works another way).

[Bedeox]

default=0 equals "do nothing" in IT

(don't know about FT though)

[X-Fixer]

huh? what?

I've noticed that IT and FT work different. FT has continue on both nibles of Rxy (as I've said), but IT's Qxy has continue on x and no continue on y (y=0 means y=1). Damn. This also means that in IT there is no way to make retrig without slide (unless you will use x=8 - that is unused).

[X-Fixer]

beta14

- AMF
- fixed lame bug (oh, that 1-based indexes!)
- loads 1.3 files
- fixed tracks mapping
- new effects
- IT & XM
- final fixes to retirg


hopefully, retrig works correct now, in both IT and XM files. actually, continue is used fairly rare (and retrig, too), so this will not affect most tunes.

[Bedeox]

xfixer: great work, maybe Winamp
will become most accurate module player EVER!

:up: :up: :up: :D

[Lion King]

http://www.modarchive.com/cgi-bin/download.cgi/K/KnullaKuk.mod

one sample is played like a beep but it's not supposed to play so long
the error is there for some time now in mikamp 2.x
you may not be able to hear the error if you seek throug the song
it's a Protracker mod

[X-Fixer]

Lion King! (and to all others, who want to report "incorrect" playback)

1. Do not send me files, that plays "wrong" unless you know player (preferably DeliPlayer or MPT), that can play them "right".

2. If this files was not created specially for testing and/or the bug is not in the few first seconds - provide exact position in seconds AND orders where I can hear it.

That's not too much, I think and will save me some time to actually fix the bug.

Bedeox: thanks, but...

[Lion King]

bass.dll, mikit and mikmod 1.x don't play this beep
mikmod 1.3 shows 7.xx minutes playing time, latest version 14.xx
that's why no song position and trust me, you will be able to hear the beep (it's very long)
I'll check with different players if you want, do you?

[X-Fixer]

eh? why no position?

about the difference in time
1) that's why I ask position in orders as well (it should not be different, whatever happens).
2) go to preferences/player and set loop count to 0. you'll get 7:58 again.

the beep is on ~4:30 (position 36 and 37). if you had reported this yourself, it would have saved me TIME, that I value some much ;)

Anyway, I have fixed it.

[Lion King]

o.k. sorry and thanks for fixing!

I didn't know that loop count thing, I thought maybe something else was wrong also
why is it on by default?

[X-Fixer]

Originally posted by Lion King
o.k. sorry and thanks for fixing!

I didn't know that loop count thing, I thought maybe something else was wrong also
why is it on by default?
have no idea. I'll make it 0 by default, but this will not affect everyone, who already uses plug-in :(

[X-Fixer]

beta15

- fixed/optimized notedelay effect
- fixed vis on 24-bit playback
- some minor changes


minor update for Lion King and Bedeox specially ;)

[Bedeox]

Thanks XFixer, that bug was just pissing me off.
You see, my friend has 24bit soundcard.

Anyway, thank you very much. :D :D :D

[bigboo]

hi

sorry I was not here the last few days but don't worry now I'm back with many new bugs to fix (tested with beta15) because I know you like it ;)

ok here we go:

1. about the 669 clicking problem I found this in the unis669 1.1 documentation FAQ section but maybe you know it already:
->I've found the files created works well with tc669, p669gu0 and of
->course unis669. But all the same results if I play them on other 669
->compatible players like Inertia Module PLayer, DMP, and PMP. It
->doesn't play correctly and also produces lots of static. And this is
->with loading cryname.669 (that came with unis669) and saving it as
->under a new filename. (update: I tried cryname itself on the other
->players and got the same problems)

>Note that many of the multi format trackers do not play 669's correctly.
>I'm not trying to bash them, but they either (i) doesn't seem to play samples
>that have C-1 as notes, or (ii) doesn't process the Break location properly.
>The static is probably due Ultraclicking.


2. the sample/comment window in the info box is not large enough for this farandole module :
http://xmp.sourceforge.net/mod/Farandole/FAR10._1darkdr.zip

(in modplug sample/comment window is ok whith this one and can be resized if needed)


3. in the in_mod preference/player there are 2 options "enable smart seeking" and "play unlisted patterns" and there are always greyed. how to enable them? and if they can't be enbled why not remove them completely?


4. in the in_mod preference/player why not remove the option "use resonant lowpass filters" because (I think) you can disable them already in the loader window "disable resonant filter effects". It seems that there is 2 different options for the same thing, but maybe I'm wrong.


5. there are problems in okt playback (compared with deliplayer):

OKT.Necronom-9.okt (at the begining)

OKT.Grundlagen_2.okt (the channel repartition between left and right is different in winamp compared to deliplayer it's the case with many okt modules but I don't know which one is accurate)

OKT.Gumba_Vahim_Enim.okt (not working at all in winamp and deliplayer but play fine in modplug)

OKT.Freestyle.okt (not working at all in winamp and deliplayer but play (a little) in modplug

>oktunk.7z


if you want to test more okt files go there: http://exotica.fix.no/tunes/pages/OKT-Oktalyzer.html
or maybe you could make a new Blow up build of in_mod if you want me to test them (I already downloaded them)

[bigboo]

other small things:

I think you should remove the fadeout option in preference/player because pp's output plug-ins now have complete fading options so there is no need to keep fadeout in input plug-ins


I found a bug crashing in_mod only with this module (under win98se winamp2.80a in_mod beta15):
1. double-click on "Yes-PartII.okt" to play the file
2. open module info box (alt-3) and keep it open
3. click the play button in winamp to restart the song
>>winamp crash:

WINAMP a causé une défaillance de page dans
le module IN_MOD.DLL à 017f:10011417.
Registres :
EAX=0057b65c CS=017f EIP=10011417 EFLGS=00010206
EBX=000004a0 SS=0187 ESP=006ed0cc EBP=00000000
ECX=00000128 DS=0187 ESI=30313020 FS=2a9f
EDX=0057be08 ES=0187 EDI=0057b904 GS=0000
Octets à CS : EIP :
f3 a5 8b cb 83 e1 03 f3 a4 8b b2 10 02 00 00 85
État de la pile :
0000000e 0057b4c0 006ed1f8 00000001 0057bdac 10003cc3 006e0000 00008294 006ed254 005e0032 00000200 006ed1d8 006ed118 bfb8e234 00000764 00000200

[bigboo]

there is a game (crusader: no remorse) that use .amf for musics.
there are not "DSMI AMF" but "ASYLUM Music Format V1.0" (maybe the original format) and are not working in winamp (modplug play them).
this is one of them:

[X-Fixer]

bigboo:

1. this FAQ is useless blah-blah-blah-ing.

2. this FAR file does not have a comment field at all. if you mean that width of samplename field is too small - that's it, it's not for writting comments.

3, 4 etc. I won't remove obsolete features for now, there is some more urgent work.

5.

OKT.Necronom-9 (and some others) - there is a problem in sample loops (hopefully fixed), blame stupid software used to save this OKTs.

OKT.Grundlagen_2 (and others) - this is due to panning. changed.

OKT.Gumba_Vahim_Enim & OKT.Freestyle - this files are broken (do not contain all listed samples, so they are so small) and do not play fine in MPT (since MPT does not play OKTs fine at all, it's just too lazy to show error message).

6. crash is known. I asked peter to fix it, but since he haven't I'm gonna do this myself.

7. this file is not AMF (you can clearly see this with FileType).

and blow-up build is still available.

btw, and have I told you that your guess on AM is wrong? :p

[X-Fixer]

beta16

- OKT
- changed panning
- fixed sample loops problem
- fixed crash in infobox
- some file-handling-related fixes


this build contains experimental code for infobox (should not crash, if will work at all ;) ) and file-handling (especially urls).

[bigboo]

ok crash bug fixed

but okt playback seems still wrong for me ;)

OKT.Necronom-9 at the begining it doesn't sound good at all (listen with deliplayer it is "smooth")

OKT.Grundlagen_2 seems a mono file now (there is a strong stereo separation in deliplayer)

[X-Fixer]

1. DeliPlayer is A BIT smoother (due to it's general smoothness, I think). Anyway, I'm not sure I can make it any better.

2. OKT is mono by design. Any "stereosness" in it is artificial. Next beta will have configurable separation.

[X-Fixer]

beta17

- 669
- improved title/comment logic
- more effects
- OKT
- fixed panning!
- finally fixed sample loops!
- fixed song loops
- AMF
- corrected 1.4 handling
- channel remapping for 1.0
- changes in effects
- FAR
- more effects
- STM
- some corrections/optimizations
- MOD
- more notes
- skip spaces at the beginning of song titles
- support of national characters in titles
- merged file types visible in Open dialog
- fixes in default panning config
- refixed infobox crash
- fixed bug in mixer settings (surround is back!)
- lots of small things


bigboo: about OKTs. I should really trust my ears more than documentation ;) And upload your ptm somewhere, I'll make a loader (seems to be pretty easy).

caffine77: if you're still here, please retest all your AMFs (blow-up build now blows up more often).

Also it will be great, if someone will test STMs, because I only have few and they do not use wicked note-off things.

And other things should be tested, to check what I have broken this
time :rolleyes:

[bigboo]

Hi

- about blow up:
does "blow up" info works with all supported formats or just amf/okt?
does it detect unsupported effects during file loading or you have to play the module entirely to the end to be sure?

- I think the new 669 improved title/comment logic is the more elegant solution good work!

- okt playback is very good now! yay you should use your ears more often ;)

- my 6 STMs are playing fine (but i think they were playing fine before already hehe)


- I haven't find any bug this time (except this horrible 669 clicking bug ahhhhhh)
and 2 amf (both 1.4) unsupported effects in blow up build:
testems.amf

gslinger.amf
(i have the original protracker version of this one and this amf isn't playing fine both in modplug an winamp it's horrible (the flute "echo effect" on the right channel is too much lood) but maybe it's just a bad protracker>amf conversion

download here http://www.thetaband.com/npdsmi/nptest.html



- about polytracker there are 2 ptm here:
http://xmp.sourceforge.net/mod/PolyTracker/

and you can download a third ptm (perfect.7z) on my site (i can pm you the URL if you forgot it)

[bigboo]

...and 2 okt with unknow effects:

[X-Fixer]

ok, thanks for files and testing.

Blow-up works in 669, amf, far, stm and okt. But in 669, okt and stm it's very unlikely to blow anymore.
You do not have to play all file, loading is enough, because all data is converted to the internal format before playing.

[bigboo]

ok ok so in fact you can test many files very quickly using the playlist window it will blow up directly.

would it be difficult to extend this blow up system to others supported file formats so I (and millions of in_mod testers) can quickly test all my mod/s3m/xm/it to be 100% sure that all effects are supported in them?


oh and maybe you could make an "official" in_mod test thread, maybe on Peter's forum, with short infos and links to download your 2 versions (normal and blow up) because I think the current thread is a little messy and it may be difficult to newbie to find the links.

[X-Fixer]

There is nothing interesting in MOD loader. I can add blowing in XM/IT loaders later.

I would make a better thread somewhere, but I'm not sure that someone, except me and you, bothers to test this stuff, so I don't think that it makes any sense for now.

And rename your file from perfect.7z to, say, perfect.zip, because your stupid Netscape HTTP server treats .7z as text files and only sends me first 6 bytes.

Oh, and send me gslinger.mod, I want to compare it with amf version. (the convertion is done automatically - it's pretty obvious - with the m2amf proggy; and there is no "echo effect", it's just another track duplicated and shifted).

[bigboo]

ok renamed to perfect.zip

I found an amf not working and i put it with gslinger.mod in "amf.zip" (a 7z file too)

[bigboo]

I found a site with modules in .mt2 and .psy format (maybe some new modules formats).
there are not recognized by FileType 1.25.
but modplug plays mt2 files correctly as "MadTracker 2 modules".
I haven't found a player supporting psy files, so if you want to add a loader for them... :)

if you want them go here
http://www.tump.net/

at the bottom of the page check "file type" and type "mt2" or "psy" in the search bar to dowload them (oh yeah sorry maybe you know how to use a web page already ;)
other mods here are standard mod/xm/it files

[bigboo]

dope.mod Startrekker
there is one more channel (8 seconds after start) in deliplayer and modplug than winamp

[bigboo]

another unsupported okt:
http://www.artificiallives.com/mods/modarch/oktalyzer/pianoses.okt

oh and when playing mtm files in the module info box in the type field it displays "MTM" but it should be "Multitracker"

ok that's all for today ;)

[Bedeox]

these old formats are indeed useful,
as I have a friend with Amiga 1200 computer.

:D somebody has interest with this plugin :D

edit: checked all his mods play correctly
(mostly ProTracker 1)

[bigboo]

ok a new day has just begun so I post here again!


I just download dmp 4.0 and I quickly read the revision history and this is my theory about amf format (yeah I have nothing to do at the moment :))

dmp 2.43 to 4.0 > amf 1.4
dmp 2.42 > amf 1.3
dmp 1.33 > amf 1.2
dmp 1.31 > amf 1.1
dmp 1.3 > amf 1.0

oh and I played a little with M2amf.exe 4.0 and I found interesting things:
download on my site amf2-7z.zip and convert included mods with m2amf then play them with winamp ;)

you should find the remaining missing effects (and also some crash) easily with that!
I'll convert some more mods with m2amf to test but i'll wait for your next beta of in_mod!

[bigboo]

ahah a quick search on the web and i found the following dmp versions:

dmp141.zip (m2amf 1.02)
dmp153.zip (m2amf 1.02)
dmp255.zip (m2amf 1.50)
dmp300.zip (m2amf 2.70)
dmp301.zip (m2amf 2.70)
dmp32295.zip (m2amf 2.70)
dmp400.zip (m2amf 4.00)


amf made with m2amf 1.02 are not played at all in winamp
amf made with m2amf 1.50 are played as amf 1.1
amf made with m2amf 2.70 are played as amf 1.4
amf made with m2amf 4.00 are played as amf 1.4

they are all in this 7z

[X-Fixer]

beta19

- 669
- fixed clicks (somebody shoot me!)
- AMF
- corrected speed and tempo effects
- fixed "much too loud" problem (yay!)
- fixed low octaves problem
- new commands
- loads 0.8 files


AMF. Loader seems to be mostly complete (though I'd like to test it with 1.2 files). Everything should be retested, cause I've changed most things. bigboo: aladd.pc tune you've sent plays ok for me and I have not tested your files in amf2-7z.zip (please, do it yourself).

OKT.

pianoses.OKT effect=filter control time=0:00 status=ignored
demosong1.OKT effect=? time=1:10+ status=not changed
hurricane.OKT effect=? time=0:00 status=not changed

yes, I have not done much, but I just can't. strange effects, used in tunes, are undocummented (except filter control, that is ignored by all PC software in all modules types) and seem to be ignored by all players I have. if you can hear any difference and explain me what these effects do (they are used once per tune in the position I've written above with parameter 0), I will support them.

bigboo: MadTracker2 is successor of FT2. MT2s are recognised by filetype if you have the lastest patch. And I have no idea what PSY files are. And I, obviously, won't be able to make a loader, if I will not know how they should be played. And I can't here any difference/find anything strange in dope.mod.

Bedeox: great! may be he can donate me his workbench? ;)

[bigboo]

ok here is some fast and wonderful testing:

1. 669 clicks: ok

2. OKT : i can't hear a difference between deliplayer and in_mod for demosong1.OKT and hurricane.OKT so it's ok.
but hey you are not a perfectionist too bad you'll never be able to say "in_mod has accurate support for okt files" :p ok I'm joking muahaha

3. Level.amf/aladd.pc now works but it only plays 2.13 minutes only the first 12 patterns but there are 122 patterns in the file! (modplug only displays 2.13 minutes but it can play after that try it!
(in fact this file is taken from the game aladdin and it contains all the musics in one file)

4. amf2-7z.zip > they are still not playing/crashing winamp you should really look at them! (after amf conversion of course)

5. dope.mod > did you try with deliplayer 1.30 (or modplug too)? I can clearly ear one more channel on the left from 8 seconds to the end (it's not very loud but in winamp it is not played at all)

6. are you planning to add MadTracker2 loader?

7. 16h intro.mod
there is one more channel/melody in deliplayer 1.30 (but not in modplug and winamp)

[bigboo]

8. other not working amf modules (when converted with m2amf 4.0):

"intro-music" amf version is working but is not played correctly at 14 second and +
you'll hear the problem easily

[bigboo]

maybe you don't need them but just in case 2 new m2amf:
m2amf 1.6 > amf 1.1
m2amf 2.5 > amf 1.3

[bigboo]

maybe you have it already but just in case I found a very good doc about polytracker format (it's a real zip file) :)

[X-Fixer]

beta20

- AMF
- changed end of track detection
- changed volume setting
- removed obsolete limitation
- fixed crash on freaky files
- small changes
- STM
- last pattern was not played


2. in_mod IS one of the best OKT players out there. it's much better than MPT, and may be even better than deli now (deli seems to ignore arpeggio).

3. it's because file consists of several songs, each ending with jump to the beginning. when first jump is encountered it's treated as start of song loop (and playing is terminated, if you have loops count set to 0). I'll make something with it later.

4. no they seem to play, but I have not tested carefully.

5. deli plays drums on the left a bit louder. MPT plays absolutely same. that's all I can hear.

6. not now.

7. deli is *the only* player I have to play this mystic channel. if you can find another player (or, better, tracker) that plays it - send it to me. otherwise, I can't do anything with it.

8. seems to be fixed now.

and do you remember what have I said about format specs? :mad:

[bigboo]

1. about amf : ok you have fixed almost everything now!
I tested over 500 mod/mtm/669/s3m converted to amf and only 2 of them have missing effects (in the included 7z)

I think you should change the description about amf in the preference/loader window now that it's fully supported!


2. dope.mod > don't listen with crappy speakers you should try with headphones! the problem is even more obvious with headphones!
8 second after start there is one more channel on the left (hehe ok you know that)
in fact in_mod play that too but not correctly
it plays a short note "do" "re" "mi"
modplug and deliplayer play "dooooooo"reeeeeeee"miiiiiiii"

sorry it's difficult to describe it would be easier for me in french but hey



3. about 16th-intro mystic channel
sorry but it's the same for me deliplayer seems to be the only one with this feature. it's even in the doc of the player:

- subsong support in all major tracker formats
*discover previously hidden melodies*

[Lion King]

psy files: http://www.pastnotecut.org/psycle
winamp2 plugin is available there
sources here: http://sourceforge.net/projects/psycle/

there is a madtracker2 plugin for winamp2 at http://www.madtracker.org/

I have no idea if/how those plugins work though

x-fixer: Could you add a checkbox or something like that to not register compressed modules (itz,mdz,stz,s3z,xmz) with in_mod.dll?
I don't use them, but my filetype list in winamp2 grews longer and longer and slows down the 'open file(s)' dialog.

[X-Fixer]

beta21

- AMF
- new effect
- MOD
- volsilde fix


bigboo:

1. fixed. and the description already shows the lastest updates ;)

2. fixed, but this can break some other MODs.

3. subsongs support is another thing (subsong play sequantally, not simmulationusly). and it's not so useful at all.

Lion King:

thanks for info. and, if you have not noticed, I have merged all file types, visible in infobox, into one, so it should not abuse the poor thing. blame peter for in_wave and in_mid.

[bigboo]

arf I have nothing to say this time! good job!
maybe AOL will pay you for all this hard work (and me too for hard testing!) muahaha
everything seems to work fine now (even your small fix for mod)


just some suggestions to improve sound quality:

1. I don't like FIR interpolation it's too aggressive with old chip tunes and they don't sound "chip" anymore because all high frequencies are completely removed.
even modern modules sound "muffled" (I don't know of it's the good word)

It would be cool if there was an intermediate option between basic interpolation and FIR interpolation (like cubic interpolation in modplug it's a good alternative I think).
Of course SSRC resampling would be the best but I think it's GPL and maybe you can't use it in in_mod.


2. another cool thing would be an "automatic control gain" like modplug player or even better a replaygain/clipping prevention system but I don't know if it's possible to calculate replaygain with modules.


voila c'est tout!

[Lion King]

Originally posted by X-Fixer
I have merged all file types, visible in infobox, into onecool, an even better solution!

btw. where's the changelog for beta18?

[bigboo]

hey what a crappy module player with crappy amf support!
it doesn't even work with amf made with my m2amf proggy ! ;) :p

maybe this file will help you!
yay amf 1.2!

[bigboo]

far with unsupported effect:
ftp://ftp.dti.ad.jp/pub/demos/music/contests/mc_winners/__symph1.far

ftp with many .stm to download (only 3MB in total):
ftp://ftp.kiarchive*****pub/.1/misc/sounds/stm/

(sorry I don't have the time to test them now)

[X-Fixer]

bigboo: thanks for AMF 1.2. i had to change 1 byte in sources to support it. another mistery about AMF is finally revealed. now the only version untested in 0.9 (and, I guess, it won't load - if it exists at all). about FAR - i now all unsupported effects, but I can't do anything for now, because they do not have corresponding mikmod effects. I know, MPT might play them, but it plays them wrong!

Lion King: I guess I've skipped beta18 ;)

Bedeox: could you test how different trackers treat MODs 8xx panning range? DMP and MPT says it's 00-80, but FT2 docs says it's 00-FF (may be it's used that way only in XMs?). IT tries to play it 00-80 (but fails due to bug in oldeffects mode). and also, could you test how volslide in MODs is played: some trackers use (up_speed - down_speed), but some always slide up, if up_speed is not 0.

[bigboo]

about stm:
default panning (in preference/loader) is too high! (compare with deli/modplug with headphones)
I think it should be the same as s3m (slider n°13)

there is also a different panning (maybe inversed) between in_mod and deli in some stm
listen to this one (with headphones again of course ;) )
ftp://ftp.kiarchive*****pub/.1/misc/sounds/stm/2.zip
but I think there are other stm with the same problem on this ftp

[Bedeox]

sorry, but i can't test it no more :(

my brand new shiny Windows XP has no
SB16 emulation for SBLive and i got rid of Win98 :(
Impulse Tracker doesn't work too even on DSound
(it uses vxd to work and WinNT/2000/XP refuses
to load it :()

(but it's :cool: that it is soooo stable)

For MODs, there were many trackers
and every one used other max for this value.
There should be some detection in in_mod
eg. when there is anything over 80
use higher range (00-FF), below use (00-80) range.
(this will require some parsing)

[Rz_Ten1]

Hi, I'd like to report a bug in in_mod. I'm currently using beta21.
The error occurs in whiteshapesmall.it and causes WinAmp to totally freeze at 1:15.

The Player info dialog is displays this at that point:
BPM: 165
Song Spd: 6
Position: 13 of 53 (13)
Row: 50 of 64
Voices: 13 (25)

If you tell WinAmp to stop playing, you'll eventually get a "error asking thread to die!" warning. I'd attach the song, but it's 4.6MB :p
It can be downloaded from http://www.modarchive.com/top10.shtml, just search for "whiteshapesmall.it". It's under the "The 35th edition of Top10 - April 2, 2002" listing. I'd link directly, but I'm fairly sure that modarchive is blocking direct links.

Thanks for your help.

[Bedeox]

bigboo: Peter could add SSRC because it's
not GPL but LGPL, so you can use it even in commercial
programs, but you must give a link to the source of the library.
It's up to Peter to implement this or not.

xfix:
You can throw in a checkbox: MOD panning compatibility mode.

MODArchive is very slow. (bleh)

/edit:
X-Fixer: Is that X for eXtreme or rather eXtra? ;)

/edit2:
HORROR! Winamp 2 just crashes, but Winamp3 not only hangs,
but starts eating 100% CPU! Closing it took me 10 min.

[caffine77]

hi x-fixer, you've been working so hard and I just vanished... I'm sorry! It was easier to check in when I was getting notifications about postings. don't know why they stopped :( Anyways, I've run the blowup version against all my AMFs, and nothing popped up and everything sounded fine. Unfortunately, I don't have any 1.2 AMFs like you want to see tested :( Here's how many of what kind I played:

4 - 1.0
10 - 1.1
5 - 1.3
5 - 1.4

But they all sound and seem fine to me. Makes it kinda tricky when the composers put in crackly sounds and record scratches :p anyways, good job. :)

[X-Fixer]

bigboo: MPT uses "closer" pannnigs for all modules (if you have tracker, you can see: it uses 64 and 192, where in_mod uses 0 and 255). if you do not like it - just move the slider. that's what this slider is done for. ;)

Bedeox: too bad. may be you can do something about it? who have decided to abandon DOS support, I wonder? :mad: (may be you have onboard sound chip. shitty AC'97 legacy support seems to be harder to get rid of ;) )
And yes, I've thought about auto-detection, but this will require total rewritting of MOD patterns loading, and since 8xx is a rare effect, I do not think it worths it now.
If adding checkbox, should checked state be for 00-80 or for 00-FF? :p
X is for everything (like in VxD or ActiveX, you know) ;)
eh, what?

Rz_Ten1: thanks for info, but if the file is so large I don't feel like downloading it. May be you can pack it and upload somewhere? Also, does it freeze without infobox open?

caffine77: it's ok. the only untested version for now is 0.9, but I'm not sure that it exists at all.

[Rz_Ten1]

I'm uploading it to:
http://www.sysreset.com/whiteshapesmall.rar
and
http://www.sysreset.com/whiteshapesmall.zip

Pick the RAR if possible it's 2.5MB (vs 3.2MB)
My server isn't very fast, it might be better to get it from modarchvie. :P

And yes, it freezes regardless of the state of the infobox.

[X-Fixer]

ok, thanks. and believe me, I have archivers, that you haven't even heard about ;)

and your server is ok, faster that my dial-up at least ;)

[X-Fixer]

beta hmm.. 22?

- AMF
- loads 1.2 files
- small changes
- PTM loader
- "continue play after loop" mode
- fixed look up in mixer


check out new experimental mode - "continue play after loop". enable it in config (disabled by default). it's useful for games' music. should work for DeusEx and aladd.pc (bigboo's).

PTM support is not complete yet. It does not support few effects, I'm not sure about 16-bit samples and volume scale seems to be wrong (like in most players ;) ).

I've fixed lock up on whiteshape.it. It was caused by using low frequencies (10 Hz). I've simply added bounds checking in mixer, so it might slow down things a bit. But I just have no idea, what I can do instead (limit freqs to 20 Hz?). Anyway, report any bug ;)

btw, although whiteshape.it is played now, it does not sound good: 1) FIR eats percussion 2) resonant filters does not work (and moreover, they are disabled, if FIR is on).

[Bedeox]

It is time to go bug Peter about
SSRC resampling in in_mod ;)

I have SBLive and it supports SB16
only by emulation :(

(everything(X)-Fixer)
You can fix everything?
Go and add SB16 emulation to kX drivers
(or any WinXP SBLive drivers)!
ROTFL :p

/edit:
as to mod compatibility:
00-80 was used mostly by Amiga
and maybe you could add an option to
'Enforce Amiga limits'

[Bedeox]

additional:
When will Peter update Winamp3 Module component
to this?

And will he throw in compression component into wa3update?

[Rz_Ten1]

I've fixed lock up on whiteshape.it. It was caused by using low frequencies (10 Hz). I've simply added bounds checking in mixer, so it might slow down things a bit.
Thanks for fixing it so that it will at least play. Now I won't have to panic when it comes up in the playlist to skip tracks before it freezes the system. :) Is there any way to mix those frequencys in while it's playing? I don't know too much about audio reproduction, but I think thats technically below hearing range. I think that cutting it might have a bad effect on bass reproduction though. I dunno. *shrugs*

I know the mik libs can play it, I don't know what they do different though.

[X-Fixer]

Bedeox: how about using creative's drivers for emulation? or using onboard audio (if you have a modern mainboard you surely have it)?
(about adding support to drivers. unfortunatelly, I do not have expierence of writting drivers. but I didn't have expierence in writting modules players, before I've started to dig in_mod ;) )
about panning. Amiga? 8xx is not an official protracker effect, it was introduced by DMP on PC. And it seems that the only tracker to use 00-FF is FT2. But it is the tool for *creating* MODs! Damn!
And any question to peter should go directly to peter - you can use posts or pms just as good as I can :p

Rz_Ten1: in_mod plays these low freqs ok. the problem was in integer overflow in calculating the size of mixing block (speaking common language). the thing I do not like in my fix is bounds checking in time-critical code. possibly, I will be able to move it somewhere in less critical part.

[bigboo]

X-Fixer: about polytracker: have you tried perfect.ptm that I sent to you?
it says "unsupported effect" when I want to play it
but maybe you know that already and it's one of the few effects not included in this version

and what about replaygain support and clipping prevention?
maybe it's not technically possible with modules

Bedeox (or X-fixer, or anybody ;) ) : I don't know much about interpolation/resampling
why FIR interpolation sounds so bad with most modules? it seems that it's one of the highest quality technique
will SSRC sound really better?

is resampling done before or after channel mixing in module player?
resampling 32 channels independently with SSRC would be a rather cpu intensive task!

[X-Fixer]

perfect.ptm works for me. I'll test again it when I'll get home.

if you want replaygain - move to winamp3, peter've done a generic RG component for it.

I have not looked on FIR code. may be peter made a bug or just choosen a bad FIR table. And interpolation is done before mixing, of course, because all samples could have different sample rates.

[X-Fixer]

beta23

- PTM
- added note slide effects
- added "reverse sample" effect (yay!)
- small changes
- improved "continue play after loop" shiz


PTM. stupid VC overwritten my settings, so my internal build did not blow up! damn! thanks for info, bigboo. thought, I'm not really sure about note slide effects (scale). if someone can give me a link to polytracker, it would be really great.

and has someone tested "continue play after loop"? bigboo? Bedeox? caffine77? is it needed, or not?

[bigboo]

now perfect.ptm is perfect (hum hum)

"continue play after loop" works for me and yes of course it's needed (at least by me ;) )

about polytracker
I read somewhere that it works only with GUS
I'll do some search but it seems that it was never released to the public (only for ViC / AcmE and The REW / Nostalgia)
but hey I already found m2amf 1.8 (the only one which make amf 1.2 (yay!) so if pt is somewhere on the net I'll get it! LoL

[bigboo]

ok I found a page with Vic's ptm modules:
http://www.mpoli.fi/files/music/VIC/

this one not work in winamp:
Fm-algor.zip


another site with vic's ptms:
ftp://ftp.scene.org/pub/music/artists/vic/

[bigboo]

ok don't shoot me! but this is another ptm format description!(because you said volume slide was not correct in your loader)
it's an updated version including complete Poly Tracker Volume Table (at the end of the doc)


ah and another non-working ptm
ftp://ftp.scene.org/pub/music/artists/vic/logicwar.zip

[Bedeox]

creative's XP/NT/2000 drivers do not contain
SB16 emulation (it's a pity... :()
only Win9x/ME drivers have the function

And i have SiS735 Mainboard with builtin
sound, but it is not SB16 compatible
(but has it's own drivers)

[bigboo]

this s3m crash winamp after playing for 15 seconds.
i'm pretty sure that this bug wasn't there before
Starshine - PM

STARSHIN-7z.zip (on my site)

[bigboo]

another non playing ptm:
ftp://ftp.scene.org/pub/music/artists/vic/improvem.zip

and this ptm has "unsupported effect"
ftp://ftp.scene.org/pub/music/artists/vic/bugfixed.zip

all other ptm are loading

[X-Fixer]

ok, loading and crash are fixed. effect is not fixed yet (it's not documented). I'll upload new version later.

Bedeox: look in BIOS, if you have AWARD one, there'll possibly be a separate switch for onboard SB emulation. also my win2k SBLive drivers had legacy support, too.

[kode54]

Originally posted by X-Fixer
I have not looked on FIR code. may be peter made a bug or just choosen a bad FIR table. And interpolation is done before mixing, of course, because all samples could have different sample rates. [/B]

I helped port the FIR code and tables from ZSNES. The original method was based on pure C++ code from libmodplug, which is a part of the Modplug XMMS project (http://modplug-xmms.sourceforge.net). The original code dynamically generates the tables, but for some reason, use of that code was pushing .data or .bss in a funny way, causing strange glitches to manifest in the newgfx engine. It is still unknown if those bugs have always been there, just waiting for something to push the alignment around, or what.

Anyway, the tables I used were generated with the original code, which called for a frequency cut-off of 0.90, which is a scaler value. The value that seeds the sine generator is scaled by that, then all of the values in the window are scaled so they total 1.0.

I'm not really a mathematical genius or whatnot, and don't really have a perfect understanding of the techniques, so I can't really describe it all here. You're free to grab their source code and have a look at it, though. Peter might also still have firtable.cpp, which is just the table generator with a main function that writes the table to a C include. Cutoff won't require any changes to the interpolation function, but changing the window size will require new shifts and masks in the MMX code.

I still have no idea if this FIR code, being from libmodplug, is the same as the "quality" resampling featured in the binary-only player and tracker.


Oh, before I forget. I'm the one responsible for those lame placeholders that were in the 669 info, as well as infinite loop (I couldn't *)!@# find how to get the text box to accept negative sign without setting it to alphanumeric), and seek/display by patterns, since I wanted pattern exact seeking for Deus Ex and In Pursuit of Greed modules. Some of the changes were rather rushed, and I guess I didn't expect the changed code to go out verbatim.

I have a very limited collection of files to test with, unfortunately. I'll see if anything goes wrong with what I do have, though. One of the things that kind of peeved me was the way 8 bit samples were handled in the interpolation code. It seems that the mixer interpolated to 8 bit scale, then just scaled the result by an extra 8 bits. I tried to scale then interpolate in the FIR code, so there'd be less noise on lower pitched samples, but it seemed to break a few things. I'll have to check again with the new code, in case it was something else.

Whew. I think that's enough.

EDIT. Fixed link, adding... I've been using a plug-in I made based on libmodplug, but it's rather cheap. No configuration or options, and does that loader format list flooding like in_wav and in_mid. Oh, well. You can also find their tiny UMX loader, unlike in_umx, it parses the UMX header then passes the contained data to the respective loader, not sure if this is a worthwhile loader for the Winamp2 plug-in, as opposed to calling out to an archive reader. Are there any other WA2 UMX readers that don't require an input plug-in? Yes, I know about WinRipper and rip directly to Winamp2 w/ partial://.

[kode54]

Using the latest at this time, beta23.

Okay, the single AMF I have plays fine, which is to be expected after all the testing with weird files that you've done ;). The two 669 modules I have play fine, but one seems go too light with pitch slides, but I'll have to get the Composer 669 and see how it plays wit that first.

I have one MTM module that plays okay, except for some weird behavior with loops, possibly related to sample offset effect. Attached.

Also, in one module from Deus Ex, Area51_Music.umx, an 8 bit cymbal sample seems to be a bit too loud, and turning off FIR doesn't seem to have an effect. I am not attaching it, for obvious reasons.

[X-Fixer]

kode54: thanks for bothering and testing. I'm somewhat busy for a while. if you have time, I'll appreciate your help in hacking in_mod later (peter seems to be too busy... and he've sold his soul to the devil :( )

I'll give you a more detailed answer later, just remind me what you've asked ;)

ps. attached where?

[kode54]

Blast, extension wasn't in the list, and its slightly too big.


Here it is for you (http://home.earthlink.net/~kode54/here-klf.rar). I'll leave it there until I really need the extra 115KB.

[X-Fixer]

ok, I've got it. CWYL

[kode54]

Some of the multi song S3Ms from In Pursuit of Greed have blank orders between the songs, and do not have the first orders of each song conveniently located together at the start of the order list.


Oh, and another odd format request, support for older UNIMOD format modules, as there are a few games out there that used libmikmod and included converted modules. Attached is an archive containing the converter and loader for v0.3 of the format. I once tried to make a loader, but obviously more work was needed to convert everything to the newer internal format.

I'll also put up an archive of the songs from one such game if anyone is interested. Unfortunately, the game is no longer available from its creator's site because its name "Acid Tetris" infringes on the "Tetris" copyright. Hmm... considering copyright and such, maybe that isn't such a good idea after all. Bleh.

[X-Fixer]

> Some of the multi song S3Ms from In Pursuit of Greed...

do you mean "continue after loop" does not work, thought it should? so, could you send me such a song?

about UNIMOD. currently in_mod does not support unimod at all (because someone - who? - have fucked out the loader).
I might do 0.3 loader very-very later. I do not think the format is any popular.

btw, yes I have FIR table generator, but I did not looked into it and into FIR code, because I have no idea about how FIR works (I used to work with graphic). May be you could review the code and try to improve table? I (and other here, I guess) can point out what "should not be" (for example, FIR eats percussion in some songs - like whiteshapesmall.it, posted above).

and to get umx loader we should moan on peter. I guess, read_file and unpack.wac share same code, as usuall.

[kode54]

My hearing must suck, because I can't really hear any less percussion with FIR interpolation enabled. Have you tried again with a table built with WFIR_CUTOFF defined to 1.0f instead of the default of 0.90f?

Oh, and those two links are 404, but I did find it at the MOD archive, which I am now raiding.

[X-Fixer]

no, I have not played with table yet.

whiteshape.it: I've choosen the wrong word. I mean the sound on 0:02-0:05 (pattern 1, channel 10, instruments 13 and 14) is almost absent with FIR. with FIR you ony get buzz.

also, I've listened to your here.mtm and compared with MPT and 1) have not found serious difference 2) have not found any looped sample or song loops 3) there are lots of 9xx, you're right. they are used to get looping-like effect. so, if you're sure that something is wrong, please give more details.

btw, I really have fixed 9xx behaviour with looped samples recently (not in beta23 yet) :)

ps. you're not a junior member any more, so you can change the picture ;)

[kode54]

Rebuild the table with a higher cutoff value.

[X-Fixer]

[damn, have not noticed, that you've replied already...]

ok, I'll try. btw, are there any other values to play with except cutoff and window size? (seems that there should be much more freedom in generating so large table...)

[kode54]

You can change the quant bits, but that doesn't affect the table size and requires a change in how the final sample is down shifted. (Quant size 14, down shift result by 14, etc)

Fracbits controls the length of the table, but you'll need to change the code that masks the fixed point sample offset into a physical sample offset and FIR table offset.

LOG2WIDTH controls the size of the sample window. Changing that requires even more changes to the mixing code.

CUTOFFBITS doesn't seem to be used at all in the table generator.

TYPE controls the method of generating the coefficients, which I don't really understand too well.


By the way, you shouldn't have too much trouble adapting it so that it uses that constructor to generate the table at startup, and possibly allow users to tweak the non-table-size affecting parameters, such as the cutoff and type, in some sort of advanced options box, with an apply button to allow semi-real-time auditioning.

Just an idea. The generation at least would free up considerable space from the binary file, and generation isn't really slow or anything.

[X-Fixer]

beta24

- PTM
- fixed loading of songs with first sample empty
- fixed reverse effect
- fixed pitch and note slides
- fixed volume table
- S3M
- fixed rare effects mess up
- OKT
- fixed pitch slides
- blow up shows only one error message
- some obsolete internal stuff removed


PTM loader seems to be complete. it plays all effects, supported by PolyTracker (thanks to JAL, who have checked polytracker's source code to ensure this). bigboo, please check all PTMs again ;) btw, the "unsupported effect" in bugfixed.ptm is not supported by the tracker (!), so it's simply ignored now.

kode54: have you read about here.mtm above? and could you send me a wierd song from In Pursuit of Greed?

[bigboo]

Hi

mmmm everything seems ok, all ptm loading.
I listened to most of them and they were playing fine.
great work, as usual ;)

hunting new bugs in in_mod is now very difficult hehe :)

[kode54]

Sorry I missed that last part. Strange, even in beta 24, the offsets have an annoying buzzing sound. It all goes away if I change the UNI_OFFSET case in mplayer.c to set a->shared.start like the other offset handlers instead of setpos. Does this change break anything? I don't really have any other modules that use offset. Not sure if this is how the latest does it because I have some really old libmikmod code. (v2.2.9)


Uploaded song0.s3z (http://home.earthlink.net/~kode54/song0.s3z), let me know if the filename causes download problems. RAR compressed.

[X-Fixer]

> Strange, even in beta 24, the offsets have an annoying
> buzzing sound

really? I have never noticed that. does anyone else here hears it? and UNI_OFFSET sets setpos var instead of start because it can change sample offset while playing (not only when specified with the note, like others).

[kode54]

Hmm. Yeah, I forgot about that. Maybe I need to make a pair of comparison clips? Would it be possible for the loaders to specify a flag when the effect is on the same row as a note on, and then the command sets start instead of setpos? It sounds like when its using setpos on the same row as the note, it retriggers repeatedly for each frame in the row. I have no idea why. Maybe the way the note on handles setpos? ....

[kode54]

Okay, this is really weird. I inserted debug messages into the code, and only one UNI_OFFSET is being triggered, but for each one of those, six Voice_SetPosition() calls are occurring. I tried setting .setpos back to SF_START_CURRENT after that call, to no avail. Something is causing the position to repeat itself. I'll see if I can find a solution on my end.

[X-Fixer]

yes, position was really set on every tick. it's strange that I have not heard this myself. I have fixed this using brute force approach. (I think that this should be rewritten in future).

[kode54]

Does the S3M loader support surround mode?

if (!is_it && inf>0x80)
{
if (inf == 0xA4) // Surround mode
{
effdat.param.s = PAN_SURROUND;
effdat.effect = UNI_PANNING;
utrk_write_local(ut, &effdat, STMEM_SEFFECT);
}
break;
}

[X-Fixer]

no, S3M loader supports surround only by using S9x. where have you taken this code from? does this method is *actually* used in S3Ms (I thought it's only in MODs)? if so, I'll add this to loader.

also, the song you've sent does not contain any strange looping commands (the looping scheme in it is very straightforward). the reason that only the first part is played, is that they are separated with FF (in orders), which is recognized as end-of-song. I'll fix this soon. For now you can just change FF to FE, which is ignored.

[kode54]

Sorry if I wasn't clear on that.

Right now, I'm working on libmodplug's PSM loader. It sure made a lot of incorrect assumptions about the PSM format, several of which could have been corrected with the skeleton PSM format draft Joshua Jensen posted to some demo/sound groups 9 years ago. The rest have been made based on trial and error, such as portamento and volume slide sensitivity.

The code also converts to single effect per row structures, which would not work if any of the PSM files I'm working with used its full capability. They all seem to be converted directly from S3M.

I was also unaware of that S9x command, because I was originally working with legacy S3M. I am not sure if there are any S3M files floating around that really use XA4 for surround mode.


EDIT: A side note on the PSM loader, the convert utility that comes in epictest.zip cannot be used reliably because it doesn't convert effect scales properly. Where volume slides seem to be twice as sensitive as S3M and portamento up/down are 4x as sensitive, the converter translates them literally, with a nasty result even from the included players.

Here is the skeleton PSM format, more than enough for the modules included with several Epic games, but I don't know what the story is with the MASI revision used by Sinaria, nor do I know if there are any other games that use MASI.

[X-Fixer]

I have added your surround code to s3m_it.c (it does not break anything anyway).

I have never seen PSM files, so I can't make the loader right now. But I have file that names itselft "Protracker Studio Module Format Specification v1.00", and according to it, it does not seem too hard to write a good loader (it's very s3m/ptm like), though there are some "dark points" (like finetune and deltas in samples, 255-channels patterns format and effects, of course). It also says there should be efx.doc file somewhere, and everything->PSM converter.

[X-Fixer]

beta25

- S3M/STM
- added surround panning
- S3M/IT
- undid "end of song" recongision
- fixed sample offset effect
- removed some more internal stuff


kode54: please, test other files with "strange" looping. also, I want to fix resonant filters. any comments/suggestion?

[kode54]

http://groups.google.com/groups?selm=1993Nov16.140329.3970%40cc.usu.edu&oe=UTF-8&output=gplain


That's the spec that Joshua Jensen posted to usenet 9 years ago.

Most of the PSM files actually floating around fit within the limits of S3M.

You'll find the converter and players in a file called epictest.zip, mirrored on various demo and/or programming ftp archives. The player follows the sensitivity scale they used in the games, but the converter doesn't translate at all. As I said, the PSMs from Epic's games seem to use a volume slide scale twice as fine as S3M, while the converter copies values from S3M without doubling. Portamento, 4x, same story.

No mention of where it stores initial pan positions or channel modes (surround, pan surround, middle) either, but I guess you could reverse engineer the player, or less likely, get ahold of Joshua Jensen.

I'll test the beta later, when I have Windows working again. I was using Windows .NET RC1, with an evaluation that would expire some time next August, but then I removed my two gamepads, then accidentally dropped the machine two inches, causing it to lock up. When I restarted, WPA told me my copy has expired and to go buy the retail version. Might also have something to do with the really lame crack I was trying out for the equally lame SoundFX 2000, which I think was what set my system date to 1921.

I can still play around with libmodplug, though. Their code is more limited than unimod, as it converts to a structure that only holds one effect per channel per row. I haven't actually tried to write a PSM loader.


For some example production PSM files, look to the shareware versions of a number of Epic games, such as Epic Pinball and Jazz Jackrabbit. There is also One Must Fall: 2097, which has modules that don't use special pan modes, but use more effects than the other two. All from OMF convert to S3M with no problem, since S3M was the original source format.

There's also Sinaria, but it seems to use an older OPLH or SONG header format. I'm not sure about the patterns, though. The notes seem to be the wrong scale or something... and I think the people or company that developed the game seem to have dissolved.

[Bedeox]

Yay, now I can play music from my old Epic games :d
without resorting to ModPlug :p

You guys are doing the great job!
Xfixer: Throw a link to the plugin here,
because I'm lazy ;) and don't want to look for it
through the whole thread

[RIV@NVX]

Originally posted by Bedeox
Yay, now I can play music from my old Epic games :d
without resorting to ModPlug :p

You guys are doing the great job!
Xfixer: Throw a link to the plugin here,
because I'm lazy ;) and don't want to look for it
through the whole thread

Download link for in_mod (http://x-fixer.narod*****download/in_mod.zip)

I use it too for 4PSetup, so here it is.

[X-Fixer]

I've finally made a link on wy web page :)

/edit:
RIV@NVX: where do you host 4P now?

and may be some one can give me a link to jazz jackrabbit 1 game? As I can remember, it was 4MB unpacked.

[RIV@NVX]

Originally posted by X-Fixer
I've finally made a link on wy web page :)

That's nice to hear, I just visited your web. Well, maybe a change of font?


/edit:
RIV@NVX: where do you host 4P now?


Nowhere. Peter doesn't want it. But I make it for me and my personal usage... contact me if you want to see my current progress.
ICQ: 60692470
e-mail: riva_station@yahoo.com

I must say that I really do not understand Peter. He said on his site:

3rd August 2002

Link to 4P page got removed - sorry guys, i don't see you updating things, and even if someone actually has time to maintain that crap, it will always have old versions due to daily-ish update cycle. Not to mention the fact that 4P was obvious bloatware, made mostly for "i want to have all the plugins" kiddies.

So, these WA2 plugz are generally final now and very rarely updated.
Why he still doesn't want 4P only he knows.

BTW, only Thor wasn't updating. And I couldn't publish my builds.

Take a look at the screenshot to see what have I done to original package by Thor.

[kode54]

But you weren't keeping up with his daily out_ds and out_wav updates, not to mention the occasional in/out_vorbis or wtf. Not that I keep up with these either.

If you really care about following that crap, you could have a simple program that checks his zips directory for newer dates than whatever date you record, then tell you when he has new crap. Automated regeneration of the install script to reflect the new version numbers, as well as automated uploading and possibly reflecting some server file so an automated PHP download page can show all the version numbers, etc. My, already more of a thorough system than Peter has set up, but definately not a lazy man's job. ;o

[Bedeox]

I only use FlashGet (http://www.flashget.com) daily
and its option to "check file for updates" and I have
always newest version of plugins
(updating ones go to the Downloading tab
and download themself again):D

[RIV@NVX]

My updating.

[Bedeox]

Throw there this (http://www.inf.ufpr.br/~rja00/files/lameDLL.zip) Lame DLL, it's modified to have alt-presets
and is newer (3.93a)

[X-Fixer]

RIV@NVX: what's wrong with the font?

Bedeox: have you tried "continue aftre loop" with DeusEx modules?

[Bedeox]

Works great! 200% better than Modplug!!!
:winamp: :up: :up: :D

(but you could add native support for the extension
umx and break the files to multiple parts
that show up in playlist like that:

Area 51 - Part 1
Area 51 - Part 2
etc.

And maybe that "part" word customizable in configuration.)


I think you should clear all notes
on "continuing after loop" after every loop.

[Lion King]

RIV@NVX: latest in_tfmx is 1.25 (the readme says 1.24 though)
vf3.ape/orbz.ape: http://www.blorp.com/~peter/old/powerfrac.zip

but afaik vf2/vf3/orbz apes need gf4/powerfrac intalled to run, publishing them separately is pointless

[RIV@NVX]

Originally posted by X-Fixer
RIV@NVX: what's wrong with the font?

Bedeox: have you tried "continue aftre loop" with DeusEx modules?

I think that it should be Verdana or Tahoma, it would look nicer, but it is your choice.

Thanks to everyone for info. Many things Thor didn't tell me in the begining.

I have decided not to put alpha versions of LAME, but I will think about it.

This 3.92 (1.30 DLL) I took from CDex (19/7), so I guess it has Alt presents.
And that 3.93a has 1.28 DLL and it is older (18/4).

But I found 3.93a2 20/9. Should I put one with SSE optimizations or without?

[RIV@NVX]

Originally posted by Lion King
RIV@NVX: latest in_tfmx is 1.25 (the readme says 1.24 though)
vf3.ape/orbz.ape: http://www.blorp.com/~peter/old/powerfrac.zip

but afaik vf2/vf3/orbz apes need gf4/powerfrac intalled to run, publishing them separately is pointless

I don't know how to put it. Any help?
I have actually never touched VisFracs part of script.

I think that GF4 is here:

SubSection "Visualization Plug-ins"

Section "VisFrac AVS Render APE v1.3" SecVF1
SectionIn 1
Call QueryWinampVisPath
SetOutPath $1\avs
File "vf.ape"
SetOutPath $1\avs\VisFrac
File "VisFrac1.txt"
File "XenexFrac (vf1 sample).avs"
File "SYNFrac (vf1 sample).avs"
File "SYNFrac (vf1 sample) - Clear Frame On.avs"
SectionEnd

Section "VisFrac2 AVS Render APE v2.0" SecVF2
SectionIn 1
Call QueryWinampVisPath
ReadRegStr $2 HKLM "SOFTWARE\PP\GF4" "engine_dll"
StrCmp $2 "" vf check2
check2:
StrCmp $2 $1\avs\VisFrac\engine.dll vf novf
novf:
MessageBox MB_YESNO "A blah GF4 blah, skip blah ?" IDYES exitsection
vf:
SetOutPath $1\avs
File "vf2.ape"
SetOutPath $1\avs\VisFrac
File "engine.dll"
WriteRegStr HKLM "SOFTWARE\PP\GF4" "engine_dll" $1\avs\VisFrac\engine.dll
File "ABSolute (vf2 sample).avs"
File "Athlonology (vf2 sample).avs"
SetOutPath $1\avs\VisFrac\help
File /r `VisFracHelp\*.*`
SetOutPath $1\avs\VisFrac\Presets
File "cyclic madness.pr4"
file "OldSk00l FS.pr4"
exitsection:
SectionEnd

SubSectionEnd

Update of 4P:

Build 20/9/2002
* Added:
- PowerFrac v1.0 Beta 5+ (VisFrac3 and Orbits AVS Render APEs)
* Updated:
- TFMX Player Plug-in v1.25
- LAME 3.92 Alpha 2 MMX (DLL 1.32) (20 September 2002)
* Installer updated to lastest NSIS 2.0b0 (CVS version). That NSIS is
getting better and better daily. Amazing.
* Thanks to Lion King from Winamp forums who informed me about fractals
and newer TFMX.

Lame is without SSE (400 kb smaller DLL), but if you think otherwise, tell me.

[Bedeox]

Can you throw in a link to that DLL?
(with SSE, it is faster even on non-PIII cpus)

(DLL version = version of an interface)
(LAME version = encoding version :p)


Could you compile LAME from CVS yourself?

I don't have VC+ and getting Borland CBuilder
to work on its source is a pain in the ass.
I'll try, but my compiled LAME won't be a speed
daemon :blah:, but you could include it as an option.
Now I'm going to download newest source from CVS.

[RIV@NVX]

Originally posted by Bedeox
Can you throw in a link to that DLL?
(with SSE, it is faster even on non-PIII cpus)

(DLL version = version of an interface)
(LAME version = encoding version :p)


Could you compile LAME from CVS yourself?

I don't have VC+ and getting Borland CBuilder
to work on its source is a pain in the ass.
I'll try, but my compiled LAME won't be a speed
daemon :blah:, but you could include it as an option.
Now I'm going to download newest source from CVS.


http://home.attbi.com/~jerrybaker/lame/lame-3.93_2002092006_Intel.zip

I do not have VC+ too.

[Lion King]

RIV@NVX: the nsis script registers gf4's engine.dll so that vf2.ape probably runs without gf4, the same should be possible with powerfrac

please make a new thread for 4p instead of invading this thread

[Bedeox]

:D downloaded it and CVS source
maybe I'll try with MINGW.

[X-Fixer]

Originally posted by Lion King
please make a new thread for 4p instead of invading this thread

yes, I like this idea, too ;)

[RIV@NVX]

Originally posted by X-Fixer


yes, I like this idea, too ;)

That's no point. Nobody wants 4P (except me), so I think talk about 4P is over. Thanks to everyone for suggestions, I really appreciate that.

[kode54]

Originally posted by Bedeox
(but you could add native support for the extension
umx and break the files to multiple parts
that show up in playlist like that:

Area 51 - Part 1
Area 51 - Part 2
etc.

And maybe that "part" word customizable in configuration.)



I thought this was done to death, but apparently nobody gets it yet. Check in the configuration for an option to seek by orders. You'll notice in that mode that the trackbar skips an area near the beginning of the song. Deus Ex keeps all of the starting orders for the multi songs at the beginning of the order list. Also, they organize their own looping songs to deal with multiple notes.

Unfortunately, the way I designed the order seeking does not clear hanging notes. It shouldn't be much of a problem with seeking to the first orders in those Deus Ex modules, though, as they are mostly designed to clear all active channels so the game doesn't have to between switches.

Oh yes. The time seeking system also uses two distinct copies of the module at once. The one that plays actually renders. The second minimal load occurs when you have the info dialog open in the normal time mode. It serves to display render time offset pattern and row information, based the render position returned by the output plugin. So in that mode, you see the pattern/row that you are actually hearing.

I never bothered to construct that mechanism for seek by orders, either, so the info dialog shows the row/pattern currently playing inside the decoder thread.

Also, the time method uses yet another module loading function to briefly load the module and seek through it start to finish, including any pattern jumps. (Unless you configured it to ignore jumps) Along the way, it collects the pattern/row info for every second, which it uses for seeking. As opposed to seek by orders, which throws the whole order list onto the seek bar.

End of current ramble post.


EDIT: Okay, for those that seem to have missed it somewhere, UMX is just an archive format, like all of the other Unreal objects. In the case of UMX, it contains only a single MOD/S3M/XM/IT, or in the case of Deus Ex, exclusively IT, module file.

At the beginning of the module's order list is a range of single orders that start different variations of the file/level's theme. First is usually the ambient theme, with an action theme, death, talking, and leaving the level. ParisCathedral_Music.it also contains the music that plays on the Paris streets level, as well as the apartment building and sewers that precede it.

Hopefully the final end to this long message, and hopefully it isn't so long that nobody decides to read it completely before asking the questions it tries to answer.

[Bedeox]

But why can't you implement it in more
"user friendly" way?
(Yes, I could move to the next pattern
manually :blah:, but it's unnatural
and I want to listen to some parts
of this music in background)

[kode54]

Yes, I suppose a method of specifying the start pattern would be nice, for playlists.


If you set it to loop always, and not to ignore loops, you can use seek by order anyway to go to any spot in the order list, and loops will be natural.

[Bedeox]

Maybe there is a way, Peter's old WinRipper
used some kind of sending position to Peter's plugins.
Example:
partial://000000A2-00126FD0:C:\UT\Music\Botmca9.umx\Botmca9_0063.IT

But to work with module, you need the header
which is on the beggining :(, so you can't
just skip patterns with it.

[kode54]

And that module only contains a single song anyway.



Playlisting the modules from those games wouldn't really serve a point for someone like me who will listen to music so rarely; rendered files get the one shot now and then.

But when it comes to modules, I'll load the sucker up and often leave it running for hours at a time. I once left OceanLab_Music.it running on the ambient track before I went to sleep (at 9:00am no less) and my family became annoyed after a couple of hours of this, even though my sub only has a 5 1/4" speaker and the levels weren't too high. It is sort of repetetive, and most of the background sounds are just looping sounds started in the first pattern.

I've also been known to leave MIDI or SPC files looping for hours as well. Whatever.

[Bedeox]

But I have different preferences
and I like to listen to different
Action parts from DeusEx modules.
And I can't send them to background,
because it skips to the Ambient track.

:( :( :( CAN SOMEBODY DO SOMETHING ABOUT IT???

[kode54]

Skips on song change or skips while playing?

Turn on "Unconditional Looping" as some files need it (of UT and Deus Ex, I don't know, but I always leave this on) and turn off "Continue Playing After Loop" otherwise it'll skip past the song's own pattern jumps. You can still seek around the starting orders with this option turned on if you have "Seek by order" turned on, and the different song modes will play properly.


Now, a specification for start order in the playlist, coupled with finite looping or even timed looping (loop for X seconds, possibly configurable in playlist the same way as the start order) and either fade on complete, or crossfade in the output plug-in.

I was just stating my preference above, not for the effect of the plug-in development. Both features would quite useful, I'm sure.

[X-Fixer]

Can someone send me one deus ex module? I think it's not too large.

If I've got you right, in_mod (with "continue after loop") should play the first part ok, but with all others it should play the first pattern at the end. It's not too bad, imho, but I can fix even this.

[Bedeox]

When changing tracks
(it's not skipping!)

I'd only like to see an option for playlist
to skip to pattern #n,
because when I want to listen to Action DeusEx tracks,
and playing starts from Ambient track after track change.

(DeusEx file is too big, I'll PM you with link
to my own FTP server 100kb/s (~2MB file)
Will be up till 2100 GMT, then will go down,
and up again from 2200 to 2400)


(I didn't mean it,
but you can fix that as well)

Could you add note clear after
"continuing after loop"?



Crossfading DirectSound output is there (http://www.blorp.com/~peter/wa2.htm)!
(C) Peter Pawlowski (famous ;) Winamp plugin developer)
(I've seen him called this way on some webpage :cool :)

Don't forget to look into options! (mmmmh... Fades :D)

[Bedeox]

Anyone can port these changes to Winamp3?

It would be great!

[RIV@NVX]

Originally posted by Bedeox
Anyone can port these changes to Winamp3?

It would be great!

Yeah. X-Fixer, do you know how to use #488 SDK?

[Bedeox]

I think that this is Peter's responsibility :D

[edit: typo]

[RIV@NVX]

X-Fixer, what's going on?
There were not updates for plenty of time.

[X-Fixer]

actually, right now I'm waiting for JAL to send me a bug report on PTMs playing. anyway, expect update on next week.

[X-Fixer]

beta26

- PTM
- removed mess in sample names
- loader makes some validations on modules
- improved "continue after loop"
- code reorganization
- support for url-styled parameters
- resonant filters are temporary off


Bedeox: try continue after loop again. and now you can play only the part you need :) btw, do you still need a DOS driver for soundcard?

[Bedeox]

Nah, I don't need it anymore,
I used dos support with modified dos disk and dos drivers from Liveware I, but now I'm not able to do so anymore, since I have WinXP on NTFS.

I'm downloading new in_mod now...

<update>
COOL!!! X-FIXER YOU ARE GREAT!!!
:up: :up: :up: :D

It works even in Winamp3! (except for title selection: always shows URL)
</update>

[X-Fixer]

I ment: do you need dos support in winxp?

[Bedeox]

It would be useful. Can this dos support read NTFS partitions?

[X-Fixer]

I'm writting a native SB support for win2k's dos box (not a dos emulation). so it does not have problems with partitions or something. but it's in very alpha stage now.

[Bedeox]

If it'll work in XP then count me in!
I'm VERY interested.

[kode54]

Using your latest and greatest.

Jump commands in the mod:// URL seem to disable looping completely. Song stops when it hits a jump command in the song itself.

[X-Fixer]

define "seem to disable looping completely". what loops count do you use? does specifying loops count in URL works? does conditional/unconditional looping works? is correct time predicted (or it equals to the time played)?

[kode54]

I use a loop count of -1. (Infinite, a feature I coded into the plug-in myself. Maybe you could make something better of the config input field, which I had to make into a text field instead of numeric because it wouldn't accept the minus sign.)

Time prediction should not be used at all in this case.

Unconditional looping is not necessary since the module has jump commands, since it contains multiple songs. Adding the setting to the URL made no difference.

Setting the loop count might have an effect if I set it to something above zero, since the problem appears to be mishandling of that crappy infinite case I coded in there.

Setting continue after loop seemed to make it jump to the order immediately after the one I specified in the URL when it hit the end of the sub-song.

I will supply you an URL to the module in a PM.

[X-Fixer]

sorry to say, but I've found that infinite loops are totally not functioning with or without jumps or any of my stuff at all (just checked with beta7, which does not contain any of my hacks).

since you seem to be the only user interested in it, I ask you: I want to reimplement it, but with the following changes: 1) no time prediction 2) no seeking 3) less exact info in info_box (2 and 3 are caused by 1). I think they are useless in "forever" mode, and this will speed up loading.

[kode54]

See also: Seek by order, whereby:

1) You could also simply show time for one pass, and jump back when the loop repeats. In seek by orders mode, length is the number of orders in the list.

2) Again, assume one iteration for prediction and support seeking within the available order set. Irrelevant for seek by order mode, which always displays the full order set. Might be an optional feature to mute all voices when pattern jumping.

3) Already dealt with for the seek by order mode, which I originally implemented so I could do the pattern jumping in songs with multiple sub-songs. As I said with 1, why not predict length based on no-loop, then update the time position relative to the actual play positon? I suppose the same output lag correction applies as well. Maybe this isn't as easy as I'm making it sound...

Oh, and you said its broken in beta7.... hmmm.. Well, crud, this code I have claims to be 2.2.9, with no "beta" branding. The code is based on a snapshot of Peter's in_mod source directory, ca. October 12, 2001.



EDIT: Last I checked in my version of the code, infinite looping is only handled by three cases of code, which, upon further checking, I don't now if it would cause infinite looping on 0 as well....


if(config_loopcount > 0)
Player_SetLoopStatus(mp, config_playflag & CPLAYFLG_LOOPALL, config_loopcount);

at the three places in main.c and one spot in info.c.

Oh. Bleh, I just noticed. My original seek by orders code sets PF_LOOP flag for Player_InitSong in main.c, for the song that's used for actual playback. This forces enless loop unless you actually specify a loop count with Player_SetLoopStatus(). I suppose you could force loop 0 and/or create a song without PF_LOOP for the prediction, as I think the info.c already does.

[X-Fixer]

yes, I like the idea, I'll do it like that (though, I'm not sure how this will work in wa3). btw,

if(config_loopcount > 0)

was fixed before me (by peter?). but all "handling code" seems to be eliminated. anyway, it's hard to tell now, because I've rewritten main.c ;)

[kode54]

Okay, damnit, I can't seem to find where infinite looping is/was implemented, even in my own code. Bleh.

The Player_BuildQuickLookups code should be amended if infinite looping is added, as it will need to behave the same as length prediction, presumably turning off looping entirely for its preps MPLAYER object.

[B3d30x]

I've found a bug in the URL code:

Title adding code works incorrectly

when you use title+ code, then try
first a module with a long name,
then one with short and you'll see.
A screen is attached ;)

[kode54]

Seems to lose the last channel from this module:


b3wm (http://home.earthlink.net/~kode54/sblazer.itz)

[X-Fixer]

both things fixed (hopefully)

[B3d30x]

EUREKA, IT WORKS!!!!:)

[B3d30x]

'title fix' breaks UMX decoder's title display.
It just erases song title, when I play UMX.
What can you do about it?

UMX decoder (a cheap hack): http://www.umx-decoder.de.vu

I will extract these mods anyway :p

[X-Fixer]

beta27

- fixes to "continue after loop"
- changes to URL handling
- fixed validation of compressed samples
- updated title formatting library
- fixed infinite looping
- fixed small seeking bug
- redone "seek by orders"
- added "track song" feature to infobox
- some interface changes


Bedeox: do you mean it started working before I've uploaded new version? ;)

btw, could someone test how in_mod works over HTTP/FTP? there is a code to save stream to a file and play

it, but I have never tested it.

also, I think about removing fadeout stuff from plug-in, because it's kinda obsolete with out_ds2. does

anybody use it? any comments/ideas?

[Lion King]

http://forums.winamp.com/showthread.php?threadid=111536

[B3d30x]

Tried it, it didn't :confused:
Sorry for useless posts, like that one :blah:

narod.ru is veeeeery sloooow, could you put your files
somewhere else, like wz.cz (fast, many features for webmaster, advertisements (but not popups) :) )

Maybe Peter will include it on his page...

It does something weird to configuration,
as I got "[syntax error in ...]" as a title of song,
till I've set tagz to default,
and title+ code STILL doesn't work...
weird... some messup in config writing code?

<edit>
As for that thread...
AFAIK, Kode worked on that code,
maybe he messed it up? :p
Or maybe peter, but I can't believe he did
such a nasty mistake :)
</edit>

[kode54]

I worked on the code a year ago. Anything since then is just going by any possible similarity between the ~v2.9 code I still have and the latest.

[X-Fixer]

RTFM (completely) and define "still does not work"

and that lame code looks like it's written by Air, so I'm kinda surprised that Lion King says 2.2.8 works :weird:

btw, have anybody tried "track song" feature?

and as for narod.ru. it's fast enough for me. I thought about moving from it though, when they've made yahoo-like ads in the right-top corner and even more, when they've added a stoopid disclaimer appearing when you try to download a file.
if you have nothing to do, go ahead and create neat in_mod homepage ;)

[kode54]

Thanks, works great.


Odd behavior I noticed, it has a buffer underrun at each repeat. No noticeable glitches are caused by this, though. Is it a side effect of forever looping?

[X-Fixer]

Originally posted by kode54
Odd behavior I noticed, it has a buffer underrun at each repeat. No noticeable glitches are caused by this, though. Is it a side effect of forever looping?
surely it is. increase 'buffer-ahead' value in out_ds options (it's 0 by default) - I have it 500ms and have no underruns. you can tell peter about that, but I think he'll flame on you ;)

Bedeox: could you check if in_mod handles NNAs (with DCT/DCA) correctly? especially handling of S7x effects.

[kode54]

You are quite correct.

And Peter will probably flame me just for saying that here in the first place. ;)

[X-Fixer]

actually, I've updated the code to work fine even when 'buffer-ahead' is 0.

[X-Fixer]

another thread I will use
http://forums.winamp.com/showthread.php?threadid=56408&perpage=40&pagenumber=2

[Rz_Ten1]

Hey again, got a new one for ya. :D

The file scifi.xm seems to have a timing problem... it drifts around thoughout the song, but is most noticable in the snare/drum tracks.

I've found a copy online here:
http://ftp.stenstad.net/pub/modules/unsorted/r-s/scifi.xm
(stupid 100k forum limit)

Edited: Forgot to say, using beta 27 blow up. I'm not getting any messages during playback though.

Also, I've noticed that when changing tracks with the lastest version, it will very quickly flash [SYNTAX ERROR IN FORMATTING STRING] in the tracklist. Not really a bug, just a tiny bit annoying. ;)

[kode54]

And was already said the last time for the SYNTAX ERROR problem, RTFM. The manual states in no uncertain terms that the plug-in contains a new version of the title formatting library. Unless you had some special custom title string, click the Default button. Otherwise, either try adding trailing percent signs to all of the variables, or just follow the new syntax and recreate your format string.

[X-Fixer]

> Forgot to say, using beta 27 blow up. I'm not getting any messages
> during playback though.

blow-up actually does nothing in XMs.

> Also, I've noticed that when changing tracks with the lastest version, it will
> very quickly flash [SYNTAX ERROR IN FORMATTING STRING] in the tracklist.

do you mean that this nice and informative error message gets replaced with something else? that's very interesting. post your settings (title formatting string, winamp title displaying options etc). also this "default" button was introduced by me (to make life more easy ;) ), but the default string is far from ideal (it was updated in 27.5 to handle songs with no title, but I was unaware of peter's restriction on using brackets, so the %part% tag is actually never displayed; I've removed the restriction recently).

also, Rz_Ten1, could you give more details on time drifting?

[B3d30x]

Your default formatting string doesn't display PART.
Probably round brackets do not work in formatting strings.
:weird:

Non working:
$if2(%title%,%filename%)[ (%part%)] (doesn't display PART)
$if2(%title%,%filename%) (%part%) (complains about [SYNTAX...])

Working:
$if2(%title%,%filename%) %part% (Displays question mark if no PART present)
$if2(%title%,%filename%)[ - %part%] (Works great!)

<edit>
Song tracking works great!!! :up:
</edit>

[kode54]

What is the "track song" feature, anyway? I couldn't see any difference, unless I wasn't looking in the right place.

Try using single quotes ' around the rounded brackets. See the readme or formatting help dialog for what they do.

[B3d30x]

Check infobox...
There is a checkmark called "Track song".
This feature automatically updates
infobox contents to currently played module.

<edit>
Yes, quoted brackets work as expected.
X-Fixer, change default formatting string
to something that's working, e.g.

$if2(%title%,%filename%)[ '('%part%')']
</edit>

[kode54]

Which info is it supposed to be tracking? I opened the info box, and it was unchecked. After checking it, no effect. After unchecking it, no effect. Does it only take effect if I restart the module, or restart Winamp? Either way, the Player Info box appears to track the decoder as I hear the sound, which of course only works in normal time seeking mode.

[B3d30x]

It changes info on track change :)
It is all it does.

<edit>
Sorry, XFiXer, but I won't be able to do any tracking
because of free time shortage :(
</edit>

[X-Fixer]

Originally posted by me
but I was unaware of peter's restriction on using brackets, so the %part% tag is actually never displayed; I've removed the restriction recently
anything is still not clear?

Originally posted by B3d30x
It changes info on track change
It is all it does.
indeed

Originally posted by B3d30x
Sorry, XFiXer, but I won't be able to do any tracking
because of free time shortage
this should not take much time ;) how do I manage to do various crap having no free time at all, I wonder :rolleyes:

[Rz_Ten1]

The problem with the song is occuring in track 4. It looks like it's one long drum sample loop, and that it's playing slightly out of sync with the rest of the song. (Timing or playback speed error, I think) I'll post some audio examples:

The 'good' clip is out of good 'ol DOS based Cubic Player. :D
The 'bad' clip is, obviously, out of winamp.

Edit: the [SYNTAX] message only seems to be appearing during the slight delay between tracks (while nothing is playing). It's title string is: %title%[ (%part%)] (the default). In additon, if the title string is bad, I've noticed the real song title will flash inbetween song changes (where the error only flashes when it's 'correct') :weird:

[B3d30x]

quote:
--------------------------------------------------------------------------------
Originally posted by B3d30x
Sorry, XFiXer, but I won't be able to do any tracking
because of free time shortage
--------------------------------------------------------------------------------


this should not take much time how do I manage to do various crap having no free time at all, I wonder


Quoing quotes :p :
I'm not so much into trackers...

[kode54]

Oops. Thanks for clearing that up. I never noticed because I rarely queue up more than one file at a time. Guess I just don't mind listening to the same song for 20 minutes at a time, or the usual, Dead Silence. I usually only queue more than one file when scanning through a set of badly named files for a particular song.

Example of badly named: All of Datschge's SPC rips, which are named using an abbreviation of the game title, a hyphen, and a number.

So, Secret of Evermore:

soe-#.spc, where # is between 01 and 67. I still haven't bothered to grab the SPC renamer from Alpha-ii.com or whatever.

[Rz_Ten1]

Hey, Dead Silence is one of my favorate groups, right next to 60 Cycle Hum. :D

And I think I found the reason for the syntax flashing... it looks like I was actually seeing the playlist title update itself after I had accidently loaded an entire playlist with a title error. What actually appeared in the playlist were around 500 [Syntax Error] entries and winamp was just fixing it on load. :o

[X-Fixer]

I've tried to fix this (XMs "Amiga periods" were seriously messed up). check if it works now. or, at least, if is better or worse than it used to be.

[cgfiend]

What program is X-Fixer working on? Out of this whole thread I haven't been able to decipher what player he's fixing bugs in.

[kode54]

in_mod.dll, the Nullsoft Module Decoder. See the first page of the topic for the download link. He always reuploads new versions over the same file, so you just have to watch for changes. (ie. newer date than local file, if your downloader supports timestamping files to match the remote server)

[Rz_Ten1]

27.6 sounds a little better, but the drifting is still pretty bad.

[X-Fixer]

after half an hour of hacking with modplug and cooledit I've finally figured out that this was a typical of-by-one error! mikamp decreased note number by one for internal purposes and then just used formula and table from FT2 doc (that are quite correct, but for the original note value, of course)!

check out beta27.7!

since I suspect that this bug is still present in other period modes (there are 4 period modes: "Amiga", "linear", "XM Amiga" and "XM linear" and only "XM Agima" - used in scifi.xm - is tested), I will appreciate if someone test this carefully.

Rz_Ten1: may be you can do some testing?

cgfiend: I don't believe that you've read the whole thread anyway :p

[cgfiend]

I skimmed through it looking for the link. I must've scrolled past your post on the first page. :igor:

[X-Fixer]

my post is the second one, but it does not show my intention to work on module decoder plug-in ;)

[Rz_Ten1]

scifi.xm looks good! :D

However, looks like there are now some other problems. :p It sounds like either the pitch is way off or a note is playing that does not exist in certain songs.

I'll attach one of them. You'll notice some very, very high pitched beeping in it that isn't supposed to be there. They come in at 22 seconds and last for most of the song.

I've noticed some other oddess in another song with track volume, but that might be just because I've been listening to songs with the volume too high for 3 hours or so. ;)

[X-Fixer]

that's a lame bug, that I've fixed just after uploading ;)
(index in table sometimes became negative. after all, the initial code was not as brain-damaged as I it looked like - it gave errors of about 1/4 tone)

check it out now

Rz_Ten1, could you do some testing, I've asked about previously?

[Rz_Ten1]

I suppose I can, what did you have in mind?

[X-Fixer]

I want someone to

1) test in_mod over http/ftp
2) test NNAs carefully (with S7x effects), but with older version of plug-in (before 27.x)

btw, have you tried 27.77?