Ingen window size

[therealkitman]

Hi all,

I have loaded an ingen patch as a LV2 synth. However I can't resize the ingen window in any way (dragging corners or clicking FULLSCREEN in its toolbar) to see the patch contents. I can load the same patch in Carla and resize ok so I'm guessing a MusE issue? Apart from that the patch works fine in MusE.

PS. On a slightly off topic, the patch is just a test of drone/whooshy sounds. Is it possible to load a short sound file into a track and loop it while the other tracks plays or do I need to use an external loop machine?

Thanks, Chris.

[danvd]

Hi Chris,

What version (release or git/vcs) of ingen do you use? I can check it today, I think.

Regards, Andrew.

[therealkitman]

Hi Andrew,

It's 0.5.1.

Thanks...

----- Original Message ----- From: "muse-sequencer/muse" To:"muse-sequencer/muse" Cc:"therealkitman" Sent:Sun, 25 Oct 2015 04:12:11 -0700 Subject:Re: [muse] Ingen window size (#379)

Hi Chris, 

What version (release or git/vcs) of ingen do you use? I can check it

today, I think.

Regards,

Andrew.

—

Reply to this email directly or view it on GitHub [1].

Links:

[1] https://github.com/muse-sequencer/muse/issues/379#issuecomment-150912608

[danvd]

Hi Chris, Fixed in b966736.

Not closing - waiting for you to check if it works right.

[therealkitman]

Hi Andrew and thanks. However, do I assume I need to download the master zip first to get muse3? I am still using Muse2.

Regards, Chris.

[danvd]

Yes, this fix only in current master.

[therealkitman]

Hi Andrew,

I have downloaded the new master.zip and installed. Muse3 runs ok with all lv2 plugins except my ingen based ones. Log attached.

PS. Am confused; is muse3 a new released version? All labels still say v2.2 as well no news on the web page or forum?

Thanks, Chris.

muse3.log.zip

[therealkitman]

Uh oh. The issue with the pianoroll that was fixed in #376 has returned in muse3. That is a drawn A4 note plays back as A5 (in the first bar in the attached image) but recorded A4 notes in bars 2 and 3 play back ok as A4 but are labelled as A3.

[danvd]

Hi Chris!

I'll look at your log file soon, but now about pianoroll labelling. Yes, I had to revert my changes back, because original labelling was right. See #397. According to midi specification middile C is C3.

[therealkitman]

Hi ,

I find the issue of middle C most peculiar. I learned that middle C is C4 at ~256Hz and corresponds to midi number 60 and A4 at 440Hz midi 69.

https://en.wikipedia.org/wiki/C_%28musical_note%29

It shouldn't matter what synth is used because as long as key 60 is sent that should correspond to C4.

Most odd.

Regards...

On 15/12/15 13:01, Andrew wrote:

Hi Chris!

I'll look at your log file soon, but now about pianoroll labelling. Yes, I had to revert my changes back, because original labelling was right. See #397 https://github.com/muse-sequencer/muse/issues/397. According to midi specification middile C is C3.

— Reply to this email directly or view it on GitHub https://github.com/muse-sequencer/muse/issues/379#issuecomment-164645487.

[danvd]

Wow :) I did not chacked wikipedia. Now digging other resources, and the information seems to be very different. So let's consider middle C is C4. I'll redo changes soon ;)

[therealkitman]

I've got Ardour4 and LMMS. I repeat my pianoroll tests (tonight) just to see if they conform as well.

On 15/12/15 13:23, Andrew wrote:

Wow :) I did not chacked wikipedia. Now digging other resources, and the information seems to be very different. So let's consider middle C is C4. I'll redo changes soon ;)

— Reply to this email directly or view it on GitHub https://github.com/muse-sequencer/muse/issues/379#issuecomment-164648078.

[danvd]

Ok. thanks! Waiting for the info.

[danvd]

Just checked yout muse.log. More info needed. Can you made this steps?:

1. Recompile with CMAKE_BUILD_TYPE=Debug
2. Run under gdb
3. When it will crash, type bt and send output here.

[danvd]

Just checked note numbers in Logic Pro X. They are the same as in MusE. Cheked: 60 is C3, 72 is C4 in both apps.

It seems that it'll be better to leave labels as is.

[therealkitman]

That is puzzling. I just used both LMMS and Ardour3 (Ardour4 died on me) and their respective pianorolls. Both recorded midi (note 69), as well as hand edited notes, play the correct note and are labelled correctly in the pianoroll - that is a real A4 @ 440Hz is labelled A4.

I don't know about Logic Pro X (LPX) but I am willing to bet that it is wrong. I nearly got caught out with LMMS as you can calibrate the keyboard scale and I made sure A4 was 440Hz. I wonder if LPX has a similar calibration trick? Ardour3 works out of the box.

[danvd]

Hmm, I also made tests with ardour (version 4 in my case): drawed notes C3 and C4, exported midi part imported it into muse - they appeared as C2 and C3.

Perhaps it can be done by adding "Middle C" config option which can be changed at runtime...

I also want to know Tim's and Robert's opinion before making changes.

[danvd]

Some UPDATE: also imported this midi file into logic pro - they appeared as C2 and C3 as in MusE.

[therealkitman]

I have recompiled muse3 with DEBUG and ran things as requested under gdb. New log attached. muse3.log.zip

[therealkitman]

If you drew notes C3 and C4(middle C) and exported the midi file (midi 48 and 60) but the Muse and LPX plays C2 and C3 then something has to be amiss?

[danvd]

Recorded video with logic midi+list editor. Here it is: https://youtu.be/sfyYbMKZePE

[therealkitman]

OK, so you typed midi number 72 and it gave you C4 and midi 60 gave you C3. That has to be wrong?

See chart at https://newt.phys.unsw.edu.au/jw/notes.html

On 15/12/15 19:49, Andrew wrote:

Recorded video with logic midi+list editor. Here it is: https://youtu.be/sfyYbMKZePE

— Reply to this email directly or view it on GitHub https://github.com/muse-sequencer/muse/issues/379#issuecomment-164740545.

[danvd]

Yes. MusE also displays 60 as C3 and 72 as C4. But ardour displays 60 as C4 and 72 as C5

[danvd]

Hi Chris,

I examined your log file. The error is triggered when plugin instantiated, not when gui opens. Does this error apper in other hosts or not?

[spamatica]

Hi Chris,

2015-12-15 4:47 GMT+01:00 therealkitman notifications@github.com:

Hi Andrew,

I have downloaded the new master.zip and installed. Muse3 runs ok with all lv2 plugins except my ingen based ones. Log attached.

PS. Am confused; is muse3 a new released version? All labels still say v2.2 as well no news on the web page or forum?

No, MusE 3 is the coming version. the master.zip file you downloaded is a snapshot of the master branch, essentially what we are working on currently and now are trying to stabilize as MusE 3. Be aware that there has been a number of changes and improvements so there may be new bugs, but we try to fix the bugs as soon as possible, especially Andrew ;)

To explain further, incase it is unclear. MusE is developed in the open so what you see on the github page is the sourcecode as we are working on now. Some development may occur in separate branches but when it approaches being stable it is merged to the master branch, when we make a release we take what is in the master branch and package it up and put it on the download page. The idea ofcourse being that this is more stable than using MusE built from github. Usually access of sourcecode is not through the master.zip file but with the git tools which allows to download and manipulate the files in a more developer friendly way. This goes for users who want bleeding edge too, you have the same source folder and just update with the latest changes. E.g

1. download the full source code git clone https://github.com/muse-sequencer/muse.git
2. perform an update to get the latest changes git pull (standing in the muse sourcecode folder)

Hope that clears it up, if not ask away :)

/Robert

Thanks, Chris.

muse3.log.zip https://github.com/muse-sequencer/muse/files/62029/muse3.log.zip

— Reply to this email directly or view it on GitHub https://github.com/muse-sequencer/muse/issues/379#issuecomment-164636760 .

[spamatica]

2015-12-15 12:27 GMT+01:00 Andrew notifications@github.com:

I also want to know Tim's and Robert's opinion before making changes.

Not really my area, sounds very odd that there are two different variants of this. I found this (which seems closest to the spec which I believe you still have to pay for): http://www.midi.org/techspecs/midituning.php where it mentions middle C as 60, no number for middle C though.

/Robert

[terminator356]

On December 15, 2015 05:29:41 AM Andrew wrote:

Yes. MusE also displays 60 as C3 and 72 as C4. But ardour displays 60 as C4 and 72 as C5

This is mildly amusing how much confusion it's causing.

According to The SoundBlaster Book, note # 00 is supposed to be C0, and it goes up from there to #127 = G10.

Which makes sense, you don't need a book to realize that.

However, at one time I had written in pencil on the page, that the octaves go from -2 to +8. IIRC that was because that is how the MusE piano displays it, so I wrote it down to remember it.

Therefore, now I am recalling more about this problem: There are at least two possible reasons for the disparity:

1) Someone was trying to shift the octave numbers so that two digits are not required. Octave 10 would require a wee bit more horizontal space to display, and it is possible we were trying to minimize that, given the limited space in the piano. This may be the more likely scenario.

2) Someone was trying to align the octaves so that C0 is "middle C". I can't recall what the official Midi middle C is... Ah yes, The SoundBlaster Book says note #60 is "middle C" which is our C3, which is way off, so I guess theory 2) is wrong.

Make any sense?

Tim.

---

Reply to this email directly or view it on GitHub: https://github.com/muse-sequencer/muse/issues/379#issuecomment-164765035

[therealkitman]

I have only used Carla as a host and that worked fine with the same ingen patch.

On 16/12/15 02:49, Andrew wrote:

Hi Chris,

I examined your log file. The error is triggered when plugin instantiated, not when gui opens. Does this error apper in other hosts or not?

— Reply to this email directly or view it on GitHub https://github.com/muse-sequencer/muse/issues/379#issuecomment-164853133.

[therealkitman]

Thanks Robert, it sort of clears things.

I am a newbie when it comes to git (haven't found a decent youtube video yet :) ) but then I don't do a lot of compiling as I usually use opensuse's packages. I might clean my PC up a bit and put the git development stuff in a virtual environment.

Regards...

On 16/12/15 03:50, spamatica wrote:

Hi Chris,

2015-12-15 4:47 GMT+01:00 therealkitman notifications@github.com:

Hi Andrew,

I have downloaded the new master.zip and installed. Muse3 runs ok with all lv2 plugins except my ingen based ones. Log attached.

PS. Am confused; is muse3 a new released version? All labels still say v2.2 as well no news on the web page or forum?

No, MusE 3 is the coming version. the master.zip file you downloaded is a snapshot of the master branch, essentially what we are working on currently and now are trying to stabilize as MusE 3. Be aware that there has been a number of changes and improvements so there may be new bugs, but we try to fix the bugs as soon as possible, especially Andrew ;)

To explain further, incase it is unclear. MusE is developed in the open so what you see on the github page is the sourcecode as we are working on now. Some development may occur in separate branches but when it approaches being stable it is merged to the master branch, when we make a release we take what is in the master branch and package it up and put it on the download page. The idea ofcourse being that this is more stable than using MusE built from github. Usually access of sourcecode is not through the master.zip file but with the git tools which allows to download and manipulate the files in a more developer friendly way. This goes for users who want bleeding edge too, you have the same source folder and just update with the latest changes. E.g

1. download the full source code git clone https://github.com/muse-sequencer/muse.git
2. perform an update to get the latest changes git pull (standing in the muse sourcecode folder)

Hope that clears it up, if not ask away :)

/Robert

Thanks, Chris.

muse3.log.zip https://github.com/muse-sequencer/muse/files/62029/muse3.log.zip

— Reply to this email directly or view it on GitHub

https://github.com/muse-sequencer/muse/issues/379#issuecomment-164636760 .

— Reply to this email directly or view it on GitHub https://github.com/muse-sequencer/muse/issues/379#issuecomment-164875338.

[terminator356]

On December 15, 2015 06:03:26 PM you wrote:

On December 15, 2015 05:29:41 AM Andrew wrote:

Yes. MusE also displays 60 as C3 and 72 as C4. But ardour displays 60 as C4 and 72 as C5

This is mildly amusing how much confusion it's causing.

According to The SoundBlaster Book, note # 00 is supposed to be C0, and it goes up from there to #127 = G10.

Which makes sense, you don't need a book to realize that.

However, at one time I had written in pencil on the page, that the octaves go from -2 to +8. IIRC that was because that is how the MusE piano displays it, so I wrote it down to remember it.

Therefore, now I am recalling more about this problem: There are at least two possible reasons for the disparity:

1) Someone was trying to shift the octave numbers so that two digits are not required. Octave 10 would require a wee bit more horizontal space to display, and it is possible we were trying to minimize that, given the limited space in the piano. This may be the more likely scenario.

2) Someone was trying to align the octaves so that C0 is "middle C". I can't recall what the official Midi middle C is... Ah yes, The SoundBlaster Book says note #60 is "middle C" which is our C3, which is way off, so I guess theory 2) is wrong.

Make any sense?

Tim.

An old description is found in piano.cpp, by an original coder. (Note the German h for b.) The range matches what I wrote in my book:

/* 0 1 2 3 4 5 6 7 8 9 10 c-2 c-1 C0 C1 C2 C3 C4 C5 C6 C7 C8 - G8

  Grid ьber Oktave:

       +------------+ ------------------------------
   11  |            |
       |         h  |         7
       +------+     |
   10  |  a#  +-----+ ..............................
       +------+  a  |
    9  |            |         6
       +------+     |
    8  |  g#  +-----+ ..............................
       +------+  g  |
    7  |            |         5
       +------+     |
    6  |  f#  +-----+ ..............................
       +------+  f  |
    5  |            |         4
       |            |
       +------------+ ------------------------------
    4  |            |
       |         e  |         3
       +------+     |
    3  |  d#  +-----+ ..............................
       +------+  d  |
    2  |            |         2
       +------+     |
    1  |  c#  +-----+ ..............................
       +------+  c  |
       |            |         1
    0  |            |
       +------------+ ------------------------------

*/

The actual characters are bitmaps found in piano.cpp

I've cleaned up the bitmaps a few times in the past and the code. But I never changed the octave numbers.

I'm not sure if there was some other reason why it is like this. Trying to recall. There's always something...

But I'm wondering if there's no reason why we can't change the number range to a proper 0 through 10. Personally I've never paid much attention to the octave numbers so I probably never considered changing them.

There /is/ enough space to draw the "C10" in the bitmaps on the piano. It's only the top octave so it won't be too ugly.

Also I think we should mark middle 'C' specially. For example with brackets or a star or something.

I just /know/ I'm forgetting something. Am I?

T.

---

Reply to this email directly or view it on GitHub: https://github.com/muse-sequencer/muse/issues/379#issuecomment-164765035

[therealkitman]

Out of curiosity, if there is a final consensus that C4 is midi 60 and should be labelled as such, as it is in LMMS and Ardour3/4, should someone have a word with the Logic Pro X people too?

On 16/12/15 11:33, Tim wrote:

On December 15, 2015 06:03:26 PM you wrote:

On December 15, 2015 05:29:41 AM Andrew wrote:

Yes. MusE also displays 60 as C3 and 72 as C4. But ardour displays 60 as C4 and 72 as C5

This is mildly amusing how much confusion it's causing.

According to The SoundBlaster Book, note # 00 is supposed to be C0, and it goes up from there to #127 = G10.

Which makes sense, you don't need a book to realize that.

However, at one time I had written in pencil on the page, that the octaves go from -2 to +8. IIRC that was because that is how the MusE piano displays it, so I wrote it down to remember it.

Therefore, now I am recalling more about this problem: There are at least two possible reasons for the disparity:

1) Someone was trying to shift the octave numbers so that two digits are not required. Octave 10 would require a wee bit more horizontal space to display, and it is possible we were trying to minimize that, given the limited space in the piano. This may be the more likely scenario.

2) Someone was trying to align the octaves so that C0 is "middle C". I can't recall what the official Midi middle C is... Ah yes, The SoundBlaster Book says note #60 is "middle C" which is our C3, which is way off, so I guess theory 2) is wrong.

Make any sense?

Tim.

An old description is found in piano.cpp, by an original coder. (Note the German h for b.) The range matches what I wrote in my book:

/* 0 1 2 3 4 5 6 7 8 9 10 c-2 c-1 C0 C1 C2 C3 C4 C5 C6 C7 C8 - G8

Grid ьber Oktave:

+------------+ ------------------------------ 11			h	7 +------+	10	a# +-----+ .............................. +------+ a	9		6 +------+	8	g# +-----+ .............................. +------+ g	7		5 +------+	6	f# +-----+ .............................. +------+ f	5		4

+------------+ ------------------------------ 4 | | | e | 3 +------+ | 3 | d# +-----+ .............................. +------+ d | 2 | | 2 +------+ | 1 | c# +-----+ .............................. +------+ c | | | 1 0 | | +------------+ ------------------------------ */

The actual characters are bitmaps found in piano.cpp

I've cleaned up the bitmaps a few times in the past and the code. But I never changed the octave numbers.

I'm not sure if there was some other reason why it is like this. Trying to recall. There's always something...

But I'm wondering if there's no reason why we can't change the number range to a proper 0 through 10. Personally I've never paid much attention to the octave numbers so I probably never considered changing them.

There /is/ enough space to draw the "C10" in the bitmaps on the piano. It's only the top octave so it won't be too ugly.

Also I think we should mark middle 'C' specially. For example with brackets or a star or something.

I just /know/ I'm forgetting something. Am I?

T.

---

Reply to this email directly or view it on GitHub:

https://github.com/muse-sequencer/muse/issues/379#issuecomment-164765035

— Reply to this email directly or view it on GitHub https://github.com/muse-sequencer/muse/issues/379#issuecomment-164978699.

[danvd]

Some more tests: just installed Cacewalk pro audio 9 (remember it?). Tests are more and more interesting :): Originally drew notes in ardour and exported midi file.

In Ardour: C4 and C5 In MusE: C3 and C4 in Cakewalk: C5 and C6

Brrr.

[terminator356]

Just to clarify, middle C note 60 is C5, not C4, given that C0 is note 00.

[therealkitman]

This is so confusing.

https://en.wikipedia.org/wiki/C_%28musical_note%29#Middle_C Piano Keyboard

On 18/12/15 09:53, Tim wrote:

Just to clarify, middle C note 60 is C5, not C4, given that C0 is note 00.

— Reply to this email directly or view it on GitHub https://github.com/muse-sequencer/muse/issues/379#issuecomment-165636398.

[terminator356]

Yes, just been reading up on it. Guess what? Sonar defines middle C as 60 C5, just as I proposed. They said they are simply going logically by the midi notes, the same as I proposed. However - they said there is a drop-down option where you can alter the base octave number to whatever you want, like -2. It affects nothing else whatsoever but the drawn octave numbers.

It seems that we may just have to do this, since there's no consensus and people may have different uses, eh guys? Consider that some may use a treble or bass clef and some may use a grand staff. Each user might want the middle C and/or the octave numbers at different places. We may need to provide a shift for both.

As for anything else - played notes, recorded notes, step record etc, do not touch them. Just the octave drawing. Our 'transpose' track info box will do any required real note shifting.

Whacha think?

Tim.

[terminator356]

Further: Since different physical instruments might define middle C and indeed the entire octave range differently, I think we should add a couple of new parameters to our Midi Instrument class. That way the user can define them in each instrument, either directly in the xml .idf files or using the Instrument Editor. Then these parameters are applied while drawing the piano keyboard.

Sound reasonable? Tim.