drastikterror wrote on 2020-08-30, 10:48:

Hello!
I was thinking about the possibility to implement a generic midi mpu-401 driver to make Ft2 work with external devices, in a way like Impulse tracker did with the command "it /s19 /a330"
Thank you!

As I have written in this tread before: I don't really know how to write low-level MIDI stuff, and I don't want to touch the low-level drivers used in FT2 without the proper knowledge... Sorry.

OK, maybe someone in this forum can help you, even without sharing the precious source code of FT2!

i could do that, but i'd need the full source code. disassembling would be not an easy task, time consuming and tedious...

drastikterror wrote on 2020-09-09, 10:55:

OK, maybe someone in this forum can help you, even without sharing the precious source code of FT2!

Well, are you fluent in Pascal and i386 assembler? There's also a lot of inline Pascal asm with hardcoded hex instructions because of limitations in Pascal for 32-bit asm instructions (FT2 actually runs in 16-bit protected mode). If you think you are up to the challenge and promise to not share the source code with other people, I can probably hand it over.

😀 😀 😀

i had quite good experience with both i could recall the stuff because of my great interest and involvement. as i said before, i use FT2 as my main sequencer/workstation, that's why i want it to be really useful and powerful even by contemporary demands. as for the source code, of course i promise not to share it in any case. i want to look into to find the possibility to add important features and i hope other users support my position and intention. let me know what do you think, and thanks in advance!

8bitbubsy wrote on 2020-09-11, 10:56:

Well, are you fluent in Pascal and i386 assembler? There's also a lot of inline Pascal asm with hardcoded hex instructions because of limitations in Pascal for 32-bit asm instructions (FT2 actually runs in 16-bit protected mode). If you think you are up to the challenge and promise to not share the source code with other people, I can probably hand it over.

I sent you the source code and a readme.txt file to explain how to use it.

This is an historic moment!
Thank you guys for your effort!
Let's keep updated!

thank you!!! i'll look into it tomorrow and start to analyze possibilities. and of course will keep you updated!

8bitbubsy wrote on 2020-09-12, 10:33:

I sent you the source code and a readme.txt file to explain how to use it.

Hello guys, I have a problem with the reproduction of a module, there is an instrument that does not sound, at minute 1:03, easy to realize, with other players it works, it is possible to see if it will not be something from my configuration or is something from the player, greetings

The party 1993 invitation , Trip To Heaven - WOTW

http://artscene.textfiles.com/music/mods/MODS … to%20heaven.mod

I hear nothing wrong at 1:03 (pattern 07) in FT2... Sounds the same in ProTracker.

It is strange, it can reproduce the error, Inertia Player fails too, Pattern 18 and above, FT2 win64 makes me the same fault, pro tracker ( win ) reproduces it well, the Dual Module Player ( DOS ) reproduces it well ,I would have to make a video, it shows like in a moment the note is cut off, i think Vibrato+Sample offset its the problem , because that's where you see that it cuts,Another error that I saw is that at the end of the module when it loops the module slows down, but the version that appears in the invitation to the 1993 party does not do that slow, but it is somewhat less, it is only the version of the module or some bad rip by modland

Aha, that's at 2:22, not 1:03. was too lazy to listen to the whole song, as I didn't know exactly what to look for. 😀

Anyway, that's related to a difference in how the 9xx (Set Sample Offset) effect is handled in the replayer. Remember that replayers are different across different players/trackers! Most .mod files were created with ProTracker (or NoiseTracker), with an exception of some coming from older or newer trackers.

ProTracker behavior (used for PT2 clone and OpenMPT in .mod mode, and probably XMPlay in .mod mode):

• Non-looping samples: If 9xx is after the end of the sample, cut off note
• Looping samples: If 9xx is after the end of the sample, start the sample from the loop start point (this is what should happen in the .mod in pattern 24)

FT2 behavior (also used for a lot of other replayers):

• Non-looping samples: If 9xx is after the end of the sample, cut off note
• Looping samples: If 9xx is after the end of the sample, cut off note (this is what happens in FT2 in the .mod in pattern 24)

8bitbubsy wrote on 2020-09-14, 08:34:

Aha, that's at 2:22, not 1:03. was too lazy to listen to the whole song, as I didn't know exactly what to look for. :) […]
Show full quote

Aha, that's at 2:22, not 1:03. was too lazy to listen to the whole song, as I didn't know exactly what to look for. 😀

Anyway, that's related to a difference in how the 9xx (Set Sample Offset) effect is handled in the replayer. Remember that replayers are different across different players/trackers! Most .mod files were created with ProTracker (or NoiseTracker), with an exception of some coming from older or newer trackers.

ProTracker behavior (used for PT2 clone and OpenMPT in .mod mode, and probably XMPlay in .mod mode):

• Non-looping samples: If 9xx is after the end of the sample, cut off note
• Looping samples: If 9xx is after the end of the sample, start the sample from the loop start point (this is what should happen in the .mod in pattern 24)

FT2 behavior (also used for a lot of other replayers):

• Non-looping samples: If 9xx is after the end of the sample, cut off note
• Looping samples: If 9xx is after the end of the sample, cut off note (this is what happens in FT2 in the .mod in pattern 24)

It is good to know, because very few reproduce well, will it be something that can be fixed or does it have to remain as part of the player?

It has to remain like this. The reason is that when FT2 loads MOD/S3M/STM files, it converts them to an XM module internally. FT2 has only one replayer, and if you bugfix it for better MOD-accuracy, it will suddenly be less accurate for XM (I care very much about extreme FT2 accuracy, so I don't touch the replayer).

traffkin wrote on 2020-09-12, 16:34:

thank you!!! i'll look into it tomorrow and start to analyze possibilities. and of course will keep you updated!

8bitbubsy wrote on 2020-09-12, 10:33:

I sent you the source code and a readme.txt file to explain how to use it.

Any updates on this?

Fast Tracker 2 does not work with Dual Covox Plugs or any other Sort of Dual D/A Converters.
It also makes crackling sounds when playing MODs over single Covox Plug.

working on this, trying to put filters into parallel buffer prior to mixing channels, not an easy task at all... some midi functionality will be added too

8bitbubsy wrote on 2020-09-30, 21:25:

traffkin wrote on 2020-09-12, 16:34:

thank you!!! i'll look into it tomorrow and start to analyze possibilities. and of course will keep you updated!

8bitbubsy wrote on 2020-09-12, 10:33:

I sent you the source code and a readme.txt file to explain how to use it.

Any updates on this?

Great!
Please, give attention to midi part! I think it's an important link between yesterday and today!!!

Yeah, MPU401 support would be cool! Also traffkin, please be sure to send the code back to me if you manage to complete these implementations/fixes, so that I can continue with bugfixing in the future. I'm sure that there are more of the old bugs in there.