Re: [Rosegarden-user] Eating chickenshit rests

2020-08-20 Thread krsgrdn



On Thu, 20 Aug 2020 16:33:54 -0400
"D. Michael McIntyre"  wrote:

> On 8/20/20 1:40 AM, david wrote:
> 
> > Also, RG experts correct where I'm wrong - but aren't rests calculated 
> > rather than being actual events?
> 
> Under most circumstances, rests are just calculated to fill empty space 
> on the fly.
> 
> I've been way too overworked to do more than glance at this thread, but 
> I'll mention in passing that I'm a terrible keyboard player, and making 
> usable notation out of an imperfect live performance is something I've 
> spent many hours doing over the years. There is no magic wand. You just 
> have to attack the problem section by section. Sometimes you can get 
> Rosegarden to make the tweaks automatically with some combination of 
> quantizers, and sometimes you just have to manually rewrite those 
> sections, and then try to computationally re-humanize them a little to 
> blend those sections in. I'm quite sure I would have talked about this 
> kind of thing in the tutorials I wrote years ago, which would still be 
> more or less valid today, but I don't even remember where they are now 
> off the top of my head.
> 
> Sorry I can't take more time.

I wasn't tossing rocks at RG, not at all, just wondering if the mess created 
(maybe) by a midi keyboard could easily be fixed in RG. Actually as far as I 
know the guy who laid the midi track for me may well have set the 'precision 
knob' to 1/64 if such exists. This image shows what I received at the top, in 
the middle it shows how I edited it in RG and which it so happens sounds even 
more the way it should, and finally the doctored human copy which as per my 
policy leaves rests of 1/16 and less right out. 
  
https://tinyurl.com/y2enjucc

So without a midi 'precision-knob' on the keyboard it'd be the hardware that 
makes the mess, something the manufacturers might wanna add some control 
features. A midi track is created to get a sound and a timing and for that I 
see nothing wrong with even four tied notes and rests that look like grapes. 
It's mission is not to create sheet music, that however could be provided for 
with a precision-knob.  





___
Rosegarden-user mailing list
Rosegarden-user@lists.sourceforge.net - use the link below to unsubscribe
https://lists.sourceforge.net/lists/listinfo/rosegarden-user


Re: [Rosegarden-user] Eating chickenshit rests

2020-08-20 Thread D. Michael McIntyre

On 8/20/20 1:40 AM, david wrote:

Also, RG experts correct where I'm wrong - but aren't rests calculated 
rather than being actual events?


Under most circumstances, rests are just calculated to fill empty space 
on the fly.


I've been way too overworked to do more than glance at this thread, but 
I'll mention in passing that I'm a terrible keyboard player, and making 
usable notation out of an imperfect live performance is something I've 
spent many hours doing over the years. There is no magic wand. You just 
have to attack the problem section by section. Sometimes you can get 
Rosegarden to make the tweaks automatically with some combination of 
quantizers, and sometimes you just have to manually rewrite those 
sections, and then try to computationally re-humanize them a little to 
blend those sections in. I'm quite sure I would have talked about this 
kind of thing in the tutorials I wrote years ago, which would still be 
more or less valid today, but I don't even remember where they are now 
off the top of my head.


Sorry I can't take more time.

--
D. Michael McIntyre


___
Rosegarden-user mailing list
Rosegarden-user@lists.sourceforge.net - use the link below to unsubscribe
https://lists.sourceforge.net/lists/listinfo/rosegarden-user


Re: [Rosegarden-user] Eating chickenshit rests

2020-08-19 Thread david

On 8/19/20 1:02 AM, krsg...@trixtar.org wrote:

On Tue, 18 Aug 2020 19:19:00 -1000
david  wrote:


On 8/18/20 3:26 AM, krsg...@trixtar.org wrote:

Whenever a midi instrument is used to lay down a midi track (next imported into 
RG) there are the inevitable inconsistancies like an 8th note repeated in the 
next bar becomes a 16th plus a 16th rest. Is there a way to clean this up by 
asking RG to feed rests of a certain duration or less to adjoining notes to be 
absorbed?  Or, if that's impossible then simply not show them although this 
would lead to quantization issues? (I'm trying top produce sensible sheet music)

I hit that all the time. I'm not as precise a player as the timing
resolution MIDI offers and don't keep tempo very well. ("You've got a
natural swing," my classical piano instructor told me, and it took a lot
of work to overcome it!)

Same about the 16th note + 16th rest. MIDI timing says we released the
16th note before it should have ended (for a 16th note at that tempo).
My main MIDI keyboard is pretty dusty, so I wouldn't be surprised if
that's sometimes  being caused by accumulated dust in the switches.

I get mine already digitized into a midi, then I work it over in RG. The guy 
who does the midi track is a pro but still, I think it's near  impossible to 
consistently hit poerfect 16th except with a slow tempo. This shouild maybe be 
a job for the midi instrument, to use only durations X-Y for notes/rests. I 
mean sheet music is usually rudimentary and doesn't have to be microsecond 
precise, that's left to execution and interpretation (i think).


Definitely not. When I do hymns, I usually only quanitze to 1/8th notes.

Also, RG experts correct where I'm wrong - but aren't rests calculated 
rather than being actual events?



RG people - isn't there a setting that tells RG NOT to tie notes across
bar lines?

I think there is, somewhere in the prefs labyrinth, ran into it the other day

Yes.

When quantizing, maybe switch the Quantizer Type to Grid? I don't know 
if it would help, but maybe the heuristic notation quantizer tries to be 
sophisticated  and guesses more?


--

David W. Jones
gn...@hawaii.rr.com
authenticity, honesty, community
http://dancingtreefrog.com
"My password is the last 8 digits of π."



___
Rosegarden-user mailing list
Rosegarden-user@lists.sourceforge.net - use the link below to unsubscribe
https://lists.sourceforge.net/lists/listinfo/rosegarden-user


Re: [Rosegarden-user] Eating chickenshit rests

2020-08-19 Thread krsgrdn
On Tue, 18 Aug 2020 19:19:00 -1000
david  wrote:

> On 8/18/20 3:26 AM, krsg...@trixtar.org wrote:
> > Whenever a midi instrument is used to lay down a midi track (next imported 
> > into RG) there are the inevitable inconsistancies like an 8th note repeated 
> > in the next bar becomes a 16th plus a 16th rest. Is there a way to clean 
> > this up by asking RG to feed rests of a certain duration or less to 
> > adjoining notes to be absorbed?  Or, if that's impossible then simply not 
> > show them although this would lead to quantization issues? (I'm trying top 
> > produce sensible sheet music)
> 
> I hit that all the time. I'm not as precise a player as the timing 
> resolution MIDI offers and don't keep tempo very well. ("You've got a 
> natural swing," my classical piano instructor told me, and it took a lot 
> of work to overcome it!)
> 
> Same about the 16th note + 16th rest. MIDI timing says we released the 
> 16th note before it should have ended (for a 16th note at that tempo). 
> My main MIDI keyboard is pretty dusty, so I wouldn't be surprised if 
> that's sometimes  being caused by accumulated dust in the switches.

I get mine already digitized into a midi, then I work it over in RG. The guy 
who does the midi track is a pro but still, I think it's near  impossible to 
consistently hit poerfect 16th except with a slow tempo. This shouild maybe be 
a job for the midi instrument, to use only durations X-Y for notes/rests. I 
mean sheet music is usually rudimentary and doesn't have to be microsecond 
precise, that's left to execution and interpretation (i think).
 
> RG people - isn't there a setting that tells RG NOT to tie notes across 
> bar lines?

I think there is, somewhere in the prefs labyrinth, ran into it the other day



-- 
Artificial Stupidity will never be competitive


___
Rosegarden-user mailing list
Rosegarden-user@lists.sourceforge.net - use the link below to unsubscribe
https://lists.sourceforge.net/lists/listinfo/rosegarden-user


Re: [Rosegarden-user] Eating chickenshit rests

2020-08-18 Thread david

On 8/18/20 3:26 AM, krsg...@trixtar.org wrote:

Whenever a midi instrument is used to lay down a midi track (next imported into 
RG) there are the inevitable inconsistancies like an 8th note repeated in the 
next bar becomes a 16th plus a 16th rest. Is there a way to clean this up by 
asking RG to feed rests of a certain duration or less to adjoining notes to be 
absorbed?  Or, if that's impossible then simply not show them although this 
would lead to quantization issues? (I'm trying top produce sensible sheet music)


I hit that all the time. I'm not as precise a player as the timing 
resolution MIDI offers and don't keep tempo very well. ("You've got a 
natural swing," my classical piano instructor told me, and it took a lot 
of work to overcome it!)


Same about the 16th note + 16th rest. MIDI timing says we released the 
16th note before it should have ended (for a 16th note at that tempo). 
My main MIDI keyboard is pretty dusty, so I wouldn't be surprised if 
that's sometimes  being caused by accumulated dust in the switches.


I usually inspect the recorded notes for extraneous things that simply 
don't belong. Like, say, a note that was supposed to be a 16th note on 
the second beat that actually started a bit before the second beat. That 
might get turned into a 1/32nd+1/16th note tie instead, so now the 
nominal 1/t6th note runs longer and maybe throws following notes off... 
Untie the tie, delete the 1/32nd note. Then I run quantization on it and 
fix whatever else needs fixing.


I work a lot with sheet music. For me, it's much faster to put in notes 
using Rosegarden, vs trying to record them. RG particularly seems to 
have difficulties with polyphonic scores and printing. I usually have to 
do workarounds to make things print properly. For example, suspended-4th 
chord where the only note changing is the 4th. If the rest of the chord 
is half notes and the 4th is a quarter note over them, the only way I 
get them to print properly is to turn the half notes into tied quarter 
notes. Otherwise, notes get thrown off when printed.


RG people - isn't there a setting that tells RG NOT to tie notes across 
bar lines?


--
David W. Jones
gn...@hawaii.rr.com
authenticity, honesty, community
http://dancingtreefrog.com
"My password is the last 8 digits of π."



___
Rosegarden-user mailing list
Rosegarden-user@lists.sourceforge.net - use the link below to unsubscribe
https://lists.sourceforge.net/lists/listinfo/rosegarden-user


[Rosegarden-user] Eating chickenshit rests

2020-08-18 Thread krsgrdn



Whenever a midi instrument is used to lay down a midi track (next imported into 
RG) there are the inevitable inconsistancies like an 8th note repeated in the 
next bar becomes a 16th plus a 16th rest. Is there a way to clean this up by 
asking RG to feed rests of a certain duration or less to adjoining notes to be 
absorbed?  Or, if that's impossible then simply not show them although this 
would lead to quantization issues? (I'm trying top produce sensible sheet music)





___
Rosegarden-user mailing list
Rosegarden-user@lists.sourceforge.net - use the link below to unsubscribe
https://lists.sourceforge.net/lists/listinfo/rosegarden-user