ModPlug Central

OpenMPT => Development Corner => Topic started by: Saga Musix on May 02, 2012, 13:02:31

Title: OpenMPT 1.20 releases
Post by: Saga Musix on May 02, 2012, 13:02:31
Hi all,
it has been far too long since the last major update of OpenMPT, so I plan to not delay the release of OpenMPT 1.20 any longer. So here's the current test version available for some last tests. If everything goes well, no further changes will be made and this will be the release version of OpenMPT 1.20.

Download: http://openmpt.org/openmpt-1-20-01-00-released
Changelog: http://openmpt.org/release_notes/History.txt
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Rakib on May 02, 2012, 19:46:18
Crazy big changelog. Thanks and will be tested as soon as possible.
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Saga Musix on May 02, 2012, 20:14:37
Quote from: Rakib on May 02, 2012, 19:46:18
Crazy big changelog.
Not necessarily a good thing - if I still was to implement everything I wanted to have in OpenMPT 1.20, I'd have to delay it even more, but the list of changes is already too damn long. The following releases will have smaller changelogs again. :)
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Rakib on May 02, 2012, 22:07:53
I have been testing and compiling new version almost as fast as they, I am currently on version 1.20.00.77 and have not encountered any problems. But I only use mptm-format and with vst's.
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Harbinger on May 03, 2012, 12:51:04
 :o  WOW! That's alot!!

Fortunately, i have some time right now as i've finished a couple of projects, so i can review the build and the changes, and complete the OHM for this version, which i haven't worked on a for quite a few months. This should inspire me to finish the CHM file as well...

I'll post questions later if i run into difficulty.

Thanks, dev(s)! 8)
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Saga Musix on May 03, 2012, 13:47:14
FYI, most (if not all) changes in OpenMPT 1.20 have of course been written up in the wiki (http://wiki.openmpt.org/Main_Page) already, ready to be compiled into a PDF manual.
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Saga Musix on May 05, 2012, 15:39:56
OpenMPT 1.20.01.00 is now officially available (http://openmpt.org/openmpt-1-20-01-00-released).
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Harbinger on May 06, 2012, 16:46:24
Notes as i'm wandering:

The MIDI Macro help is better than nothing, but somewhere there should be more centralized help on HOW to set these macros up. I saw all the possibilities, but i had no idea what you could do or how to set it up. I could definitely use more knowledge on this subject for the CHM.

The tooltips and other changes in the Res and Cutoff sliders in the Instruments page is GENIUS.

The draggable mouse across the keys of the Sample Map is also a welcome change, but i would like to see the drag across the black keys separate from the white keys (possibly by separating by halves). It would also help if you had a "Reset" button to cancel out changes but allow the user to keep editing. (I'll make a FR.)

I also like the option of selecting a channel by double-clicking a field in the channel. Now to just click and drag to select more columns...

I do have a few questions:

Quote
[Imp]  When trying to paste but no pattern data was found in the system clipboard, previously copied pattern data is now regained from an internal clipboard.
I don't understand. If the clipboard does not have valid pattern data and a paste is attempted into the PE, is Modplug storing a backup clipboard of PREVIOUSLY copied pattern data? That means MPT has 3 clipboards -- the OS clipboard, this backup clipboard, and Modplug's "Quick-Paste" clipboard?

Quote
[New]  Added "mapped" MIDI channels (like in Impulse Tracker). If a plugin's MIDI Channel is set to "mapped" in the instrument editor, note data is transmitted on a triggered note's pattern channel modulo 16.
Could not find how to set a MIDI channel to "mapped" on the Instrument page or the plugin's GUI window. How do you activate this?

UPDATE: Never mind, i found it. Now my question is, is only MIDI data being transmitted for any instrument set to Mapped (any associated samples are ignored/not played)? I'm guessing that the channel in the pattern editor where the instruiment is called defines which MIDI channel to send; "modulo 16" means the PE channels are divided into groups of 16, so that a MIDI-Mapped instrument whose note is played in PE channel 16 will transmit on MIDI Ch. 16, and for PE channel 17, it's MIDI Ch. 1, and so on? Can you fill us in on how we can best use this feature, and what you initially intended with it? Is it designed mostly for calling external MIDI hardware?

Quote
[New]  Experimental feature: A Sample Tuner which tunes the sample to a specified note.
If you read my description of the different possible frequencies in the OHM (Reference-Audio Processing:Pitch:Frequency Calculation), i describe (admittedly with some verbal trudgery) of 3 frequency settings of a given sample in ModPlug -- the "apparent" frequency (the actual pitch of the sound when played back at the same speed it was recorded), the "declared" frequency (the value that appears in the sample's header, which doesn't have to do anything with the actual pitch that was recorded), and the "playback" frequency (which is the transposition that takes place according to the note that you're playing the sample at). I'm wondering which frequency the new Sample Tuner is modifying and its relationship to the Freq field in the Samples page. Does the Sample Tuner change the internal 'recorded' frequency (the actual waveform pitch) or is it changing the value that was in the sample header, or some other program variable that is neither?

Quote
[Imp]  OpenMPT also checks for the new VST DLL entry point ("VSTPluginMain") when loading plugins now, so plugins that only export "VSTPluginMain" and not "main" will now also load (if there are any).
Not sure what this means for the average user. Can you clarify this new capability?...

Quote
[Mod]  Knob mode is set to linear (instead of circular) mouse actions for VST plugins that support it.
Does this means that round knob elements in plugin GUIs can now be changed with a click-and-drag that moves forward or backward, instead of trying to emulate a knob "turn" while dragging the mouse?

Quote
[Fix]  When finishing a sustain loop that is behind a normal sample loop, sample playback is not simply stopped anymore.
Not sure what this means. The Sustain loop of a sample or an Instrument? "Behind" a normal sample loop; what do you mean by "behind"? Can you give an example of what triggered the playback to stop?

Quote
[Fix]  Multiple fine pattern delays on the same row are now added up.
Wait a minute. If you put pattern delay commands on the same row but in different channels, Modplug now ADDs them all together so that you can delay playback more than the number of ticks in a row?!?  :o

Quote
[Imp]  Saving: Samples that are not associated to any instrument are now tried to be written to XM files anyway.
This means that normally ModPlug would not normally save a sample that wasn't called in any pattern? You say it "tries;" what does that mean if it fails? The save is not completed, or the XM file has a missing sample slot?

Quote
[Imp]  PixPlay panning heuristics only apply to modules made with versions of ScreamTracker older than 3.20.
Never heard of PixPlay. Does this have something to do with more recent S3M tracks made in ScreamTracker? Some background and "overknowledge" is necessary to understand what this means for ModPlug users who track, edit, or play in S3M format.

Quote
[Imp]  UMX Loader: Instead of "scanning" Unreal files for modules, they are actually parsed correctly now. As a nice side effect, it is now also possible to load UAX files (collections of samples) from older Unreal Engine games as modules.
Excellent! As you may or may not know or remember, i got into MPT because i wanted to compose UMX tracks. Now i can unlock the UAX files too?! Incidentally i tried this out, and i couldn't Import any UAX sounds in the Samples page, but UAX files do open in the Folder Tree, and it will show all the sounds in the package. However, the preview playback (double-clicking on the sample icon in the Folder Tree) leaves something to be desired, and the only way i could hear the sounds correctly was to drag it into the sample window and play it back. Perhaps a contextual menu with different playback rates for previewing so you don't have to touch the module...

Quote[New]  Template modules can now be loaded and saved through the "File" menu.
Please provide a link on how to set up and use templates, for newcomers and the lazy...

Quote[Mod]  Compatibility export hints are now hideable.
Can't seen to find the option or toggle. Is it an INI setting?

Seems like one thing you forgot to mention in the changelog was access to the Help file...
I've been waiting with anticipation to see how you'd implement the Help manual call, and it seems that Modplug looks for the properly named PDF in the root directory of the application. You sure that's the best way?
For those who don't know, stick the latest PDF in the same folder as the Modplug EXE and make sure it's named "OpenMPT manual.pdf" -- just like that (it's apparently hard-coded).
I would also think we should have two Help menu items, one for an offline Help manual, and one for the wiki manual.

I also noticed that our skinny little application is now nearly twice as bulky. What did you insert that nearly doubled its size? I'm going to guess this is partly refactoring, partly backward compatibility retention, and partly MIDI enhancements, no?...

Most all of these changes will affect the text of the Reference section in the OHM, but not so much the upcoming Windows Help File for MPT.

All in all, an excellent set of fixes and new capabilities. Now hopefully with all these bug features, you can invest some serious time in requested features. (Yeah, i know... you just found time to catch your breath and we're already asking for 10 more laps!!  ;D I think we'll understand if you take some time off to pursue other projects.  8) )
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Saga Musix on May 06, 2012, 17:57:07
Quote from: Harbinger on May 06, 2012, 16:46:24
Quote
[Imp]  When trying to paste but no pattern data was found in the system clipboard, previously copied pattern data is now regained from an internal clipboard.
I don't understand. If the clipboard does not have valid pattern data and a paste is attempted into the PE, is Modplug storing a backup clipboard of PREVIOUSLY copied pattern data? That means MPT has 3 clipboards -- the OS clipboard, this backup clipboard, and Modplug's "Quick-Paste" clipboard?
Exactly. The backup clipboard is a foreshadowing of an upcoming feature: Multiple patterns clipboards, or a "clipboard history", if you will.

Quote from: Harbinger on May 06, 2012, 16:46:24
Quote
[New]  Added "mapped" MIDI channels (like in Impulse Tracker). If a plugin's MIDI Channel is set to "mapped" in the instrument editor, note data is transmitted on a triggered note's pattern channel modulo 16.
[...]Never mind, i found it. Now my question is, is only MIDI data being transmitted for any instrument set to Mapped (any associated samples are ignored/not played)? I'm guessing that the channel in the pattern editor where the instruiment is called defines which MIDI channel to send; "modulo 16" means the PE channels are divided into groups of 16, so that a MIDI-Mapped instrument whose note is played in PE channel 16 will transmit on MIDI Ch. 16, and for PE channel 17, it's MIDI Ch. 1, and so on? Can you fill us in on how we can best use this feature, and what you initially intended with it? Is it designed mostly for calling external MIDI hardware?
To be honest I have no idea what it could be useful for, I just implemented this because it's also there in Impulse Tracker (and thus the IT format). I guess you could use it for controlling a multitimbral plugin using a single mod instrument.

Quote from: Harbinger on May 06, 2012, 16:46:24
Quote
[New]  Experimental feature: A Sample Tuner which tunes the sample to a specified note.
Does the Sample Tuner change the internal 'recorded' frequency (the actual waveform pitch) or is it changing the value that was in the sample header, or some other program variable that is neither?
The sample tuner calculates a new C-5 frequency so that the sample sounds like the desired note. It does not apply pitch-shifting or whatever. So if you for example have a sample recorded at 44100 Hz playing a C#, and you tell the sample tuner to make it sound like a D#, it will set the C-5 frequency to 49492 Hz (theoretically; practically autocorrelation is not that exact; it will set the C-5 frequency to something close to that value).


Quote from: Harbinger on May 06, 2012, 16:46:24
Quote
[Imp]  OpenMPT also checks for the new VST DLL entry point ("VSTPluginMain") when loading plugins now, so plugins that only export "VSTPluginMain" and not "main" will now also load (if there are any).
Not sure what this means for the average user. Can you clarify this new capability?...
It means nothing. A program (or plugin in this case) needs an "entry point" that the operating system (or the plugin host) uses to start the program / plugin, and by convention this entry point is normally called "main". However, Steinberg decided at one point that the entry point for VST plugins should be called "VSTPluginMain" instead, so they recommend plugin authors to use that entry point instead of "main". However, practically speaking you can of course not simply remove the old entry point because many old hosts still rely on it, so it is a safe bet that all plugins that are still being published have both entry points enabled.

Quote from: Harbinger on May 06, 2012, 16:46:24
Quote
[Mod]  Knob mode is set to linear (instead of circular) mouse actions for VST plugins that support it.
Does this means that round knob elements in plugin GUIs can now be changed with a click-and-drag that moves forward or backward, instead of trying to emulate a knob "turn" while dragging the mouse?
Exactly. Keep in mind that many plugins don't support this opcode, though, so they will still use circular movement.

Quote from: Harbinger on May 06, 2012, 16:46:24
Quote
[Fix]  When finishing a sustain loop that is behind a normal sample loop, sample playback is not simply stopped anymore.
Not sure what this means. The Sustain loop of a sample or an Instrument? "Behind" a normal sample loop; what do you mean by "behind"? Can you give an example of what triggered the playback to stop?
First off, don't mix instrument loops with sample loops, they don't have anything to do with each other (I think trying to find a link between them has confused you already in the past, so simply see them as two completely different things). Of course I'm talking about sample loops only here. And I literally mean sustain loops that are "behind" sample loops. Imagine a sample with a sustain loop from 10000 to 20000 and a normal loop from 5000 to 8000. In old versions, if a note-off (==) was triggered, the sample was stopped after reaching the end of the sustain loop. Now the sample continues playing in the normal loop (so the sample cursor jumps backwards if you will).

Quote from: Harbinger on May 06, 2012, 16:46:24
Quote
[Fix]  Multiple fine pattern delays on the same row are now added up.
Wait a minute. If you put pattern delay commands on the same row but in different channels, Modplug now ADDs them all together so that you can delay playback more than the number of ticks in a row?!?  :o
Well, that's exactly what was done before. Previously the line was just repeated as often as the last pattern delay command in a row dictated, so if there was a S64 command and right next to it a S62 command, the row was extended by 2 ticks - now it extended by 4 + 2 = 6 ticks.

Quote from: Harbinger on May 06, 2012, 16:46:24
Quote
[Imp]  Saving: Samples that are not associated to any instrument are now tried to be written to XM files anyway.
This means that normally ModPlug would not normally save a sample that wasn't called in any pattern? You say it "tries;" what does that mean if it fails? The save is not completed, or the XM file has a missing sample slot?
No, this does not have anything to do with samples being referenced in patterns or not. This fix prevents samples that are linked to an instrument in an XM file, but not found in that instrument's sample map to be lost when re-saving the XM file. For this, OpenMPT tries to associate samples that are not associated with any instrument yet to some instrument. If this fails (because the instrument already has 16 or 32 sample associations), the sample is simply not stored in the file. As said, this is only important for samples that were linked to an instrument but not referenced by its sample map, so if this mechanism fails to associate such samples, the module will not sound different or anything. This mechanism is mostly there to preserve sample texts in existing XM files. See this thread (http://forum.openmpt.org/index.php?topic=4688.0) for details.

Quote from: Harbinger on May 06, 2012, 16:46:24
Quote
[Imp]  PixPlay panning heuristics only apply to modules made with versions of ScreamTracker older than 3.20.
Never heard of PixPlay. Does this have something to do with more recent S3M tracks made in ScreamTracker? Some background and "overknowledge" is necessary to understand what this means for ModPlug users who track, edit, or play in S3M format.
PixPlay was apparently a rarely used S3M player back in the days. It used commands Z00 to Z0F as panning commands. So OpenMPT looks if a S3M module contains those commands (it won't do anything if higher Zxx commands are found, as then they are clearly not indended to be PixPlay panning commands) and if thinks they are panning commands, it converts them to real panning commands (S80 to S8F) instead. As I know exactly one module which makes use of "PixPlay panning" and this module was composed with an old version of ScreamTracker, I have changed the heuristic to ignore S3M files made with ST3.20 or newer ST3.20 has its own panning commands). For OpenMPT users, this means nothing because OpenMPT does not claim to be ST3.00. So this fix only changed the heuristic to generate less false positives.

Quote from: Harbinger on May 06, 2012, 16:46:24
Quote
[Imp]  UMX Loader: Instead of "scanning" Unreal files for modules, they are actually parsed correctly now. As a nice side effect, it is now also possible to load UAX files (collections of samples) from older Unreal Engine games as modules.
Excellent! As you may or may not know or remember, i got into MPT because i wanted to compose UMX tracks. Now i can unlock the UAX files too?! Incidentally i tried this out, and i couldn't Import any UAX sounds in the Samples page, but UAX files do open in the Folder Tree, and it will show all the sounds in the package.
That's why it says that you can load them as modules, not samples. How would you imagine OpenMPT to pick one sample out of hundreds if you load the UAX file into a sample slot?

Quote from: Harbinger on May 06, 2012, 16:46:24However, the preview playback (double-clicking on the sample icon in the Folder Tree) leaves something to be desired, and the only way i could hear the sounds correctly was to drag it into the sample window and play it back. Perhaps a contextual menu with different playback rates for previewing so you don't have to touch the module...
You can use the pattern/sample/instrument editor note keys in the treeview to trigger samples, I think that's much better than a context menu...

Quote from: Harbinger on May 06, 2012, 16:46:24
Quote[New]  Template modules can now be loaded and saved through the "File" menu.
Please provide a link on how to set up and use templates, for newcomers and the lazy...
There's not much to say because templates are simply saved modules. They cannot be "set up". You simply save an existing module as a template and then re-use it later.

Quote from: Harbinger on May 06, 2012, 16:46:24
Quote[Mod]  Compatibility export hints are now hideable.
Can't seen to find the option or toggle. Is it an INI setting?
This is not a hidden setting, it simply adds a "Don't show this message again" checkbox to the compatibility export hint that is shown when trying to compatibility-export a file.

Quote from: Harbinger on May 06, 2012, 16:46:24I've been waiting with anticipation to see how you'd implement the Help manual call, and it seems that Modplug looks for the properly named PDF in the root directory of the application. You sure that's the best way?
Why should it not be a good way?

Quote from: Harbinger on May 06, 2012, 16:46:24For those who don't know, stick the latest PDF in the same folder as the Modplug EXE and make sure it's named "OpenMPT manual.pdf" -- just like that (it's apparently hard-coded).
There is no need to replace the manual or anything - I do not plan to do "in-between" releases of the manual. The latest version will always be included with the appropriate version of OpenMPT.

Quote from: Harbinger on May 06, 2012, 16:46:24I would also think we should have two Help menu items, one for an offline Help manual, and one for the wiki manual.
The online help is available through the "Web Resources" menu item, but it might indeed be a good idea to add it as a separate link again.

Quote from: Harbinger on May 06, 2012, 16:46:24I also noticed that our skinny little application is now nearly twice as bulky. What did you insert that nearly doubled its size? I'm going to guess this is partly refactoring, partly backward compatibility retention, and partly MIDI enhancements, no?...
mptrack.exe OpenMPT 1.19.04.00: 2.38 MB
mptrack.exe OpenMPT 1.20.01.00: 2.72 MB
That's not really twice the size. I guess you mean the installer package, which has indeed been doubled in size - have you already forgotten that OpenMPT now comes with a 2 MB big manual? :P
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Relabsoluness on May 06, 2012, 21:18:13
The release looks very good with help manuals and all and I don't think that the whopping 3.53 MB is too big an installer for an application like this ;)

Great work.
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Rakib on May 06, 2012, 21:35:05
Absolutely,  Jojo has really outdone himself. OpenMPT is really moving fast forward with all these new features.
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Diamond on May 07, 2012, 08:35:57
Good work and thanks for all the updates.  This is pretty minor so it's probably not worth having it's own thread,but I have F1 mapped to the "General" tab and in the latest version it does not work since it seems to have been hard coded to open the help file.
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Diamond on May 07, 2012, 08:45:15
Actually, after a bit more testing it seems that the action is performed, but I first receive an error about not being able to find the help file.  I downloaded the ZIP package and did not copy over the help file which explains the error so it seems the problem is more that the check to see if the help file exists is still performed even if the key is assigned to something else.
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Saga Musix on May 07, 2012, 12:12:35
I will have to check if the F1 key is really hardcoded - If that's the case, it's MFC's fault and I will have to change some stuff. Of course this behaviour is not intended.

Edit: OK, tracked down what caused F1 to trigger the help. This will be fixed in the next release.
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Diamond on May 07, 2012, 19:43:56
Quote from: Jojo on May 07, 2012, 12:12:35
OK, tracked down what caused F1 to trigger the help. This will be fixed in the next release.

Thanks.
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: .^o on May 22, 2012, 16:17:33
suppression of pattern multiple selection ?
I found nothing about this in releases notes ...
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Saga Musix on May 23, 2012, 13:23:13
Quote from: .^o on May 22, 2012, 16:17:33
suppression of pattern multiple selection ?
I found nothing about this in releases notes ...
No? Please elaborate on the problem you're having, I don't see any trouble with selecting multiple patterns here.
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: Saga Musix on May 23, 2012, 14:06:03
I think that I changed the hard-coded Shift shortcut to the normal "Selection" shortcut that you can find in the keyboard settings, though. So if your select shortcut (which is also used for pattern selections) is *not* shift, you will have to change this shortcut or get used to using that shortcut for order selections as well.
Title: Re: OpenMPT 1.20 release candidate available for testing
Post by: .^o on May 23, 2012, 14:34:14
ah ....^^
ok.
thx.
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: .^o on May 28, 2012, 11:42:35
impossible to check "pitch / tempo lock" in the "instrument" tab.

... ?
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: Saga Musix on May 28, 2012, 12:21:51
Use the MPTM format for OpenMPT-exclusive features. They are not available in other formats.
I hope this FAQ entry (http://wiki.openmpt.org/Manual:_Frequently_Asked_Questions#Why_is_previously_usable_Feature_X_unavailable_in_the_latest_version_of_OpenMPT.3F) answers this kind of question once and forever.
Title: More Questions for 1.20
Post by: Harbinger on May 30, 2012, 14:03:51
Quote[New]  It is now possible to record incoming MIDI aftertouch messages to patterns using volume commands or MIDI Macros.
I don't have a controller so i can't test this, but which commands are entered? Note FX commands (third-column vxx), Channel FX commands (Vxx), or does it depend on the module type? When entered as MIDI macros, is simply Zxx entered and the user is reponsible for setting up the correct SFx asssignment?
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: Saga Musix on May 30, 2012, 14:16:03
When recording aftertouch as volume, vxx is used whenenver possible, otherwise Cxx is used. Vxx is global volume, so of course it is not used. When recording aftertouch as a MIDI macro, \xx is used wheneve rpossible, otherwise Zxx is used. There is no automatic SFx assignment done, because it is not possible to associate aftertouch commands with plugin parameters or anything. It works exactly like recording MIDI CCs.
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: Harbinger on May 30, 2012, 14:29:44
Thx for the Nfo!
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: LPChip on May 30, 2012, 19:32:42
Ah, yeah... the aftertouch. I helped Jojo test this one using my Continuum Fingerboard. :)

Its great that I can now use my CF as a midi device in OpenMPT, multi-channel even! :)
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: Saga Musix on May 30, 2012, 22:21:24
Yeah, damn, totally makes me want to get a MIDI device with poly aftertouch *lol* - most of them are pretty damn expensive, though, and there aren't many of them either.
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: .^o on June 08, 2012, 13:02:45
crash when "remove unused samples / instruments."
not in 1.20.00.45
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: Saga Musix on June 08, 2012, 23:25:37
I use that function a lot and it certainly doesn't happen here. Do you have an example module with which the crash appears?
OpenMPT rescues modified tracks to the temporary folder when a crash is encountered; maybe one of those would be suitable.
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: .^o on June 09, 2012, 06:48:27


I'll try to explain it : 
to start my projects, I used large modules with lots of samples (xxx Mo), in which I after cut new modules, then  save and clean. it happened several times on the first large modules, not on the small others.
1.20.01.00 version only, no problems with previous versions.
it looks like "Windows" crashes. no crash report or file,  it crashed before.
i use only portable versions of OpenMPT (on several drives) without autosave (i always work on copies).
[in this case: intel dual core, 2 GO ram, win vista 1]



édit:
this crash-notify.dmp file will help ?
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: Saga Musix on June 09, 2012, 10:54:04
Try if the latest test version at http://sagagames.de/stuff/mptrack.exe helps. Your problem could be realted to this (http://bugs.openmpt.org/view.php?id=232) bug, which has been fixed in the meantime.
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: .^o on June 09, 2012, 11:14:17
i can't test with the same modules because i saved them yesterday but it's ok with 1.6 with a large similar module.

thx Jojo.
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: Saga Musix on June 09, 2012, 14:42:25
The size doesn't matter. The problem depends entirely for which samples an undo point has already been created.
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: BooT-SectoR-ViruZ on July 07, 2012, 10:07:59
sry for not reading the entire thread to check if it has already been pointed out, but i'm not too happy that the modern tempo mode isn't available for standard .it files anymore.....
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: Saga Musix on July 07, 2012, 13:05:58
Just use the MPTM format. The only difference between the MPTM and IT format is that the MPTM format has more features. There are no other differences or drawbacks. There is a hidden setting (http://wiki.openmpt.org/Manual:_Hidden_Settings) to set up a different default module format than IT, if you are concerned about your workflow.
Title: OpenMPT 1.20.02.00: Final version available
Post by: Saga Musix on July 07, 2012, 15:42:11
Hi all,
OpenMPT 1.20 has been updated with some new features and bugfixes. I've made this release available immediately, i.e. this is not a beta testing phase. Have fun with the new version!

Download: http://openmpt.org/openmpt-1-20-02-00-released
Changelog: http://openmpt.org/release_notes/History.txt
Title: Re: OpenMPT 1.20.02.00: Final version available
Post by: Harbinger on July 13, 2012, 13:47:24
Love this build and the application is really coming along now that the devs have handled a lot of the major persisent bugs. I do have a few questions and comments:

Quote [New] <js> Middle-clicking or Ctrl-right-clicking a channel header shows the Quick Channel Settings dialog, where channel name, volume and panning settings can be quickly set up.

I LOVE this new feature!! I would still love to see a Mass Channel Attributes dialog as i put forth in a
FR (http://bugs.openmpt.org/view.php?id=46) many moons ago, but this is definitely a nice addition until then. Among the many reasons for the Mass Channel request was so i can set up all my channels without any mouse clicks, just use keys to navigate between fields and controls to set up everything. This is one step closer to that realization, but the problem here is that normal keypresses (outside of a text field) still activate ModPlug even when the Quick Settings window is open. For example, i TABbed over to the Next/Previous arrows (actually TABbed back -- not very intuitive), and pressed SPACE to use the control. Unfortunately, i have SPACE set up for the Play/Pause function in the Patterns tab, so pressing space played the track. So in order to use this feature with only keys, i have to reconfigure my keys. Not cool...
But in a general sense i do like it and intend to use it instead of the Channel tabs on the General page.

Quote [Mod] <js> Interpolation and transpose context menu items have been moved to sub menus.

Now that you've moved the Transpose amounts to a more efficient submenu (good idea), can you now reconsider implementing a long-time
FR (http://forum.openmpt.org/index.php?topic=3186.msg25711#msg25711) to include Transposition by sevenths and possibly fourths? This would save us keypresses.

Quote [Imp] <js> Improved AIFF support (added support for loop points, up to 32-Bit sample resolution, AIFF-C files and little-endian sample data).

Thanks for this. As you may remember, Macintosh was my primary computer before 2008 (this program motivated me to switch to a Windows machine), and so i still have a lot of AIFFs from then (AIFF was the usual sound file format for Mac Classics). Now i can safely bring in those files.
Still have a persistent problem: AIFFs still cannot be previewed from the Folder Tree when double-clicking them, which has always been SOP. Should i submit a FR to get that "fixed" at some time in the future?

Quote [Imp] <js> If zooming the current selection is impossible, the zoom context menu item is greyed out.

Don't know what you're talking about. I selected one sample as well as all the samples in a typical waveform, right-clicked in a selection of the sound data, and selected "Zoom". The first time it adjusted the zoom to the right level to show the whole selection. Right-clicking again and selecting "Zoom" did not change anything and was not grayed out.

Quote [Fix] <js> The Korg Wavestation GUI did not work until song was played for the first time since OpenMPT 1.20. This may also fix the GUI of older versions of the Korg M1 plugin.

After extensive testing, i see you have completely fixed this. This fix gets my biggest thanks, as i do a lot of work with the M1 (it was after all my first synth). The fix also applies to the other VSTs in the Legacy bundle, including, as in my case, the original, unpatched package, version 1.0.0, which included the WaveStation VSTi and the MDE-X VST FX). I'm so glad i don't have to revert to an older build of MPT. Thanks again...

Quote
IT::Compatible Playback Mode
[Fix] <js> Combinations of Qxy (retrigger) with a pattern delay at Speed y should now work as intended.

XM
[Fix] <js> Set Finetune (E5x) should now work as in Fasttracker 2.

XM::Compatible Playback Mode
[Fix] <js> Tremor (Txy) behaves more like in Fasttracker 2 now.

MOD
[Fix] <js> Set Finetune (E5x) should now behave more like in ProTracker.

Many times you mentioned that fixed functions now work as the original tracker intended them to. Only problem is, how do we know what they were "supposed" to do? Can you clarify what they were doing and what they do now, in these particular cases?

QuoteXM
[Fix] <js> Volume column command gxx did not have memory (tx coda). Combinations of gxx and Gxx in the same pattern cell work correctly now.

So for XM tracks, do these apply fully when used at the same time? Does one get processed first or are their values added? Does the fact that Volume Column commands use decimal and Channel FX commands use hex have any bearing on the portamento processing?

QuoteIT::Compatible Playback Mode
[Fix] <js> Combinations of SDx with x greater than the current spend and tick delay commands didn't work.

For IT-compat tracks, does this mean that, for example, calling SD8 for a row that has 4 ticks to it will delay the note for two full rows?

Quote [New] <js> Keyboard Settings: Shortcuts can now also be found by pressing the sought-after key combination. Click search field to reset filter.

Now we're talking! How many of us have always been afraid to make any changes to our shortcuts for fear of entering a key combo that's already used for something else? (Old combos were overwritten and you'd have to go back to fix it.) Now we can make a quick check to see if a key combo is already assigned. The only thing better would be a "Print" button to print out a paper copy of our keys setup or a TXT file of it.

Overall, an excellent build worth its separate release. Keep up the good work! 8)
Title: Re: OpenMPT 1.20.02.00: Final version available
Post by: Saga Musix on July 13, 2012, 14:09:53
Quoteactually TABbed back -- not very intuitive
I forgot to set up the correct tab order for this dialog - will be fixed in the next release. Also the shortcut issues should be resolved at some point.

QuoteNow that you've moved the Transpose amounts to a more efficient submenu (good idea), can you now reconsider implementing a long-time FR to include Transposition by sevenths and possibly fourths? This would save us keypresses.
See my response here: http://bugs.openmpt.org/view.php?id=269

QuoteStill have a persistent problem: AIFFs still cannot be previewed from the Folder Tree when double-clicking them, which has always been SOP. Should i submit a FR to get that "fixed" at some time in the future?
This seems to happen with only a few types of AIFF files, I'll look into that shortly.

QuoteDon't know what you're talking about. I selected one sample as well as all the samples in a typical waveform, right-clicked in a selection of the sound data, and selected "Zoom". The first time it adjusted the zoom to the right level to show the whole selection. Right-clicking again and selecting "Zoom" did not change anything and was not grayed out.
This fix only applies to overly long sample selections which would previously not have adjusted the zoom at all. This fix does not apply when zooming into the same selection twice as in your example.

QuoteAfter extensive testing, i see you have completely fixed this. This fix gets my biggest thanks, as i do a lot of work with the M1 (it was after all my first synth). The fix also applies to the other VSTs in the Legacy bundle, including, as in my case, the original, unpatched package, version 1.0.0, which included the WaveStation VSTi and the MDE-X VST FX). I'm so glad i don't have to revert to an older build of MPT. Thanks again...
Good to hear that it's working. This was actually not a bug in OpenMPT but an oversight by the Korg programmers, which was apparently fixed in the latest versions of the M1 plugin, but not the Wavestation plugin. I have sent an email to Korg, but I have not received an answer as of yet...

QuoteMany times you mentioned that fixed functions now work as the original tracker intended them to. Only problem is, how do we know what they were "supposed" to do? Can you clarify what they were doing and what they do now, in these particular cases?
Qxy fix: The note would not have been retriggered if for example the current speed is 6, and there is a Q06 command and a pattern delay on the same row. Now the note is retriggered as one would expect it to do.
E5x fix: The correct finetune values were already displayed in the note properties dialog, but other finetune values were applied. E5x is now only applied when found next to a note.
Txy fix: The most notable difference is that the tremor on/off state is preserved even after the effect has been applied. So if the volume is turned of on the last tick of a row and there is no tremor command on the next row, the next row will remain silent.
E5x fix: Contrary to the XM E5x fix, this one is really just a very small difference that you won't notice under normal circumstances and I'm not even sure how to explain in words (been a while since I fixed this).

QuoteSo for XM tracks, do these apply fully when used at the same time? Does one get processed first or are their values added? Does the fact that Volume Column commands use decimal and Channel FX commands use hex have any bearing on the portamento processing?
From the manual: "Combining the effect [gxx] with 3xx will double the effect parameter (e.g. g01 would act like g02) and ignores the 3xx command."

QuoteFor IT-compat tracks, does this mean that, for example, calling SD8 for a row that has 4 ticks to it will delay the note for two full rows?
No, SDx never affects upcoming rows. As an example, SD5 didn't do anything at speed 5, even if there was a fine pattern delay on the same row (which would have increased the row's length by a few ticks).
Title: Re: OpenMPT 1.20.02.00: Final version available
Post by: Pizearke on July 25, 2012, 08:19:01
Quote from: Harbinger on July 13, 2012, 13:47:24
FR (http://bugs.openmpt.org/view.php?id=46)

I was just thinking about something like this today. I think it's a great concept.
Title: Re: OpenMPT 1.20.02.00: Final version available
Post by: .^o on July 31, 2012, 18:27:42
could it be possible to sync at tempo all instruments in only one time ?
Title: Re: OpenMPT 1.20.02.00: Final version available
Post by: Saga Musix on August 01, 2012, 00:01:01
This is not the place for feature requests or general questions.
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: Saga Musix on August 06, 2012, 13:11:55
Quote from: Harbinger on May 06, 2012, 16:46:24
UPDATE: Never mind, i found it. Now my question is, is only MIDI data being transmitted for any instrument set to Mapped (any associated samples are ignored/not played)? I'm guessing that the channel in the pattern editor where the instruiment is called defines which MIDI channel to send; "modulo 16" means the PE channels are divided into groups of 16, so that a MIDI-Mapped instrument whose note is played in PE channel 16 will transmit on MIDI Ch. 16, and for PE channel 17, it's MIDI Ch. 1, and so on? Can you fill us in on how we can best use this feature, and what you initially intended with it? Is it designed mostly for calling external MIDI hardware?

Now I can finally show you a good example of how this feature can be used: http://coda.s3m.us/2012/08/06/creation-of-the-ilisa-demo-soundtrack/
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: Harbinger on August 06, 2012, 19:42:31
Wow, OpenMPT is finding its way "inside the shop". I hope to see more of this with computer (sound) engineers... 8)
Title: Re: OpenMPT 1.20.01.00: Final version available
Post by: Saga Musix on August 06, 2012, 19:52:46
Well, coda has been using ModPlug Tracker / OpenMPT for a long time. He's not new to this at all, he's even contributed loads of example modules that are shipped with the tracker! And of course it has been used for many demoscene productions like this one in the past already.
Title: OpenMPT 1.20.03.00: Final version available
Post by: Saga Musix on September 04, 2012, 21:44:32
Since I haven't done much on the OpenMPT code lately, I'd like to put out the changes made since the last version pretty soon, maybe this weekend.
As always, you may have a look at the current version, test it for stability, regressions, etc.
So far I haven't found any new bugs, but of course that doesn't say anything. :D

Download: http://openmpt.org/openmpt-1-20-03-00-released
Changelog: http://openmpt.org/release_notes/History.txt
Title: Re: OpenMPT 1.20.03.00: Preview version available
Post by: Rakib on September 05, 2012, 20:58:07
*downloading*
Title: Re: OpenMPT 1.20.03.00: Preview version available
Post by: Saga Musix on September 08, 2012, 15:56:38
As I first want to straighten out some incompatibilities between the Exx/Fxx and Hxx portamento range for VSTis, I will be delaying this release a bit more.
Title: Re: OpenMPT 1.20.03.00: Preview version available
Post by: Saga Musix on September 19, 2012, 15:32:32
Ok, new try: Now that the pitch wheel stuff has been implemented correctly, I'm targetting this or next weekend for a new release. Happy bug hunting!
Title: Re: OpenMPT 1.20.03.00: Final version available
Post by: Saga Musix on September 22, 2012, 14:54:32
OpenMPT 1.20.03.00 is now officially out!
Title: Re: OpenMPT 1.20.03.00: Final version available
Post by: .^o on September 22, 2012, 17:31:24
thx, one more time.
Title: Re: OpenMPT 1.20.03.00: Final version available
Post by: Mr.Eagle on September 27, 2012, 00:32:06
Quote from: Saga Musix on September 22, 2012, 14:54:32
OpenMPT 1.20.03.00 is now officially out!

Yes, thank you very much indeed, jojo.... :)
Title: OpenMPT 1.20.04.00: Final version available
Post by: Saga Musix on October 27, 2012, 19:13:24
I hereby announce the immediate availability of OpenMPT 1.20.04.00.
This version doesn't bring many new exciting features, but rather addresses some bugs found in OpenMPT 1.20.03.00.

Download: http://openmpt.org/openmpt-1-20-04-00-released
Changelog: http://openmpt.org/release_notes/History.txt
Title: Re: OpenMPT 1.20.04.00: Final version available
Post by: .^o on October 28, 2012, 06:48:12
it's always a good day when I discover a new version.

many thanks jojo!!!!!