Left side wave extension #1211
Replies: 39 comments
-
Moving the left side to the right doesn't behave the same as before either. The wave is moved instead of being hidden. |
Beta Was this translation helpful? Give feedback.
-
Hi! I'm here, I'm awake, I'm watching. It is no longer possible to have events hidden or partially hidden before the left border. However, as a compromise of sorts, when dragging the borders, press Ctrl |
Beta Was this translation helpful? Give feedback.
-
Oh, hey there :) But it did work. Was it not possible to keep that? I don't remember what the issue was that you fixed.. |
Beta Was this translation helpful? Give feedback.
-
I explained it in a recent post in detail. Yes, it appeared to work, but under the hood it was not correct. |
Beta Was this translation helpful? Give feedback.
-
Yeah, I recall the mails. It appears I didn't grasp the consequences. We need some other way to solve this then, that feature was very useful. After scratching my head a bit I think I remember my original idea, which I likely never got around to since I got it to work by having negative events. It should still work. |
Beta Was this translation helpful? Give feedback.
-
Yes I saw exactly what you were trying to do. |
Beta Was this translation helpful? Give feedback.
-
That said, yes we could use something in the Wave Editor to manipulate a wave's 'spos'. |
Beta Was this translation helpful? Give feedback.
-
I was going to try not only an 'spos' numerical adjustment box but also the ability |
Beta Was this translation helpful? Give feedback.
-
Are you saying to not do what I proposed on the account of we should not manipulate events in the arranger? |
Beta Was this translation helpful? Give feedback.
-
But as I mentioned before the way it was done was causing some problems. That wave parts have events is what I am trying to make clear to the user. |
Beta Was this translation helpful? Give feedback.
-
If you give me a few days I promise to try and make the hidden events at left left border thing work. I am starting right now. |
Beta Was this translation helpful? Give feedback.
-
I would appreciate that.
I see what you are going for with exposing events so it's evident that
parts have been glued together, but I still defend that the event at pos
zero is special and it is logical that it would reveal more wave, if it
exists.
Btw, regarding glue I noticed that gluing waves still caused doubling of
the audio when I tried.
Cheers
Den sön 4 apr. 2021 21:58Tim ***@***.***> skrev:
… If you give me a few days I promise to try and make the hidden events at
left left border thing work.
Possibly by tonight.
I'm just saying it's not correct and there will be problems down the road,
especially if people try
to add features that support it.
I am starting right now.
Please let me handle it and commit?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<https://github.com/orgs/muse-sequencer/teams/musedevteam/discussions/85/comments/11>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABCFAN3NYIE52AGQVS5ZCPLTHDAFRANCNFSM42LOHCHA>
.
|
Beta Was this translation helpful? Give feedback.
-
If I can make it work decently, I believe it may accomplish what you were trying to do, I have it working right now with full left and right border sizing, and left and right The midi is working just fine at the moment. |
Beta Was this translation helpful? Give feedback.
-
Breakthrough! This screenshot shows a wave part that has been split into two and a midi part. When I play this, it works. All the the waves are properly offset, and sound in sync with what's on the screen. It's working quite splendidly, I must say, all the tests pass so far, loading, saving, There are still some graphics things to fix, very similar to these sound fixes. |
Beta Was this translation helpful? Give feedback.
-
Alright, that looks great!
I'm grateful for you working on this.
/Robert
Den tors 8 apr. 2021 kl 06:39 skrev Tim ***@***.***>:
… Breakthrough! This screenshot shows a wave part that has been split into
two and a midi part.
All have had their borders adjusted. You can see the left and right
'hidden events' arrows are active.
Note that the right wave part has an intrinsic offset (the wave event's
'spos' member) due to the split
and this is automatically taken into account by the low level wave class.
When I play this, it works.
All the the waves are properly offset, and sound in sync with what's on
the screen.
*Most importantly*, when the wave parts are placed at bar zero, that is
when the wave events
(which are hidden before the parts' left borders) have *negative*
absolute time values.
This was crucial to get right since, when the parts are placed some
distance into the song
the absolute time values of the wave events are still not negative yet
although they are negative
with respect to the parts left edge.
It's working quite splendidly, I must say, all the tests pass so far,
loading, saving,
various part border sizes and placements etc. Man, it took a few night to
get one
crucial statement correct, where the wave's seek offset is calculated, so
that it
starts and plays correctly no matter where the play-head is placed.
There are still some graphics things to fix, very similar to these sound
fixes.
Please hang in there, I'll try to ensure the thing is usable.
[image: left border fixes 1]
<https://user-images.githubusercontent.com/5422362/113968275-c3f57f00-9800-11eb-8502-1fd389ecb73e.png>
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<https://github.com/orgs/muse-sequencer/teams/musedevteam/discussions/85/comments/14>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABCFAN7HH5GNSVUZYCS6INDTHUXOVANCNFSM42LOHCHA>
.
|
Beta Was this translation helpful? Give feedback.
-
I must agree with Tim. I think it's not a good idea to do a major release with basic bugs like that. I guess the new release will raise some interest, people will try it out, but when they run in such bugs, the reactions could be very negative. |
Beta Was this translation helpful? Give feedback.
-
Den ons 14 apr. 2021 08:27kybos ***@***.***> skrev:
I must agree with Tim. I think it's not a good idea to do a major release
with basic bugs like that. I guess the new release will raise some
interest, people will try it out, but when they run in such bugs, the
reactions could be very negative.
On the other side, we can't go on like that indefinitely. What about a
deadline next week on the weekend? Hoping that Tim can fix it by then and
Robert will have time to do the release?
Alright, good to know the general feeling. Sounds reasonable, I'll check
again second half next week.
A related matter if we, as you say, hope for raised interest.
Apart from uploading the release, updating our news pages and posting in
Linuxmusicians forums, should we include some other channel? Possibly make
some kind of event?
New videos would be great also.
Any ideas?
—
… You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<https://github.com/orgs/muse-sequencer/teams/musedevteam/discussions/85/comments/25>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABCFAN27BJXDQFRTVXSC4FTTIUYQ7ANCNFSM42LOHCHA>
.
|
Beta Was this translation helpful? Give feedback.
-
What comes to mind:
|
Beta Was this translation helpful? Give feedback.
-
You know, I could swear the looping was actually working. Unfortunately, due to deprecated Qt code and so on, I must merge the latest master code to my branch. |
Beta Was this translation helpful? Give feedback.
-
Eek! False alarm! It IS working. With Jack only. Technical: Jack has a transport mechanism. When our 'use Jack transport' button is off or So long story short, use Jack (and no latency correction) and it will be OK. |
Beta Was this translation helpful? Give feedback.
-
Good news! But is also means that it will never work without Jack? If this is the case, it should be clearly stated and documented IMO to avoid false expectations.
And there is an open issue on it as well, declaring it as a bug: |
Beta Was this translation helpful? Give feedback.
-
Well, not 'never'. I just need to correct my internal transport to mimic the number of cycles that Jack uses. Actually I am much more interested in getting my much more complete latency_looping branch to run, |
Beta Was this translation helpful? Give feedback.
-
Done. Looping: Fixed silent wave tracks when Jack transport off or non-Jack driver used.
RTAudio support could ever really have 'feature parity' with Jack, because RTAudio is a wrapper - So... Now to concentrate on the latency looping bug, in my branch. The app's looking sharp, kybos. You're a hell of a trooper for sticking with it. Stunning transformation, really. |
Beta Was this translation helpful? Give feedback.
-
A final thought tonight, Probably more important than the remaining latency looping bug is the |
Beta Was this translation helpful? Give feedback.
-
That's good news again, thanks! I've closed the ticket.
Sure. It's about the basic functionality (like the looping).
Thanks. I'm still convinced that there is a lot of potential in MusE (even if the "competition" is getting stronger, too). |
Beta Was this translation helpful? Give feedback.
-
Hey guys, wow, reading up on the progress and this is looking good :) No time tomorrow but I will work on the release on Sunday. Including some of the work Kybos suggested above, updating screenshots etc, I don't think I will get it done but it's closing in :)
Not sure what you tried but there must be a way to get git to help you. Regarding feature parity of RTAudio, it is true that some features are simply not possible to achieve. But freewheeling should still be possible to implement. Apart from simply not sending any audio to RTAudio I guess it would require the internal transport you mentioned to allow for freewheeling. |
Beta Was this translation helpful? Give feedback.
-
So, I got inspired, and maybe to prove my point, I went ahead and tried to implement this. Of course it turns out I got it partly backwards, but still it was pretty easy to get something working. |
Beta Was this translation helpful? Give feedback.
-
Wow, that's inspired alright. Roll up the ol' sleeves and try. Hm, why do you use the RTAudio driver? Seems a bit odd. At some point we should add automatically restarting MusE when users change the driver, or anything |
Beta Was this translation helpful? Give feedback.
-
About a native ALSA driver: Old MusE had one but it was removed. Could be a place to start. Notice that some heavyweights are leaning towards ALSA instead of Jack now. |
Beta Was this translation helpful? Give feedback.
-
Hmm, could you? I mean... I guess it's technically possible but you are completely in the dark as you can't hear what you are doing prior to the mixdown. Still it would be easy to implement for the dummy driver so it doesn't matter :)
Jack does make some things harder, like listening to reference material during mixing. And for whatever reason it looks like I can get more stable playback under very high cpu when using the pulse backend. Thinking about it this is probably mostly because the pulse backend tends to have very high latency.
Yeah, jack was a very interesting idea at the time but it turns out I personally hardly ever utilize the multi application connectivity, and probably very few people does. Going directly for the bare metal, alsa, should allow a little bit lower latency as less code is executed in the realtime scope. |
Beta Was this translation helpful? Give feedback.
-
Hi Tim,
I saw you push your fixes now, nice! So I thought I would take it for a spin before finalizing the release.
I did however found a regression as the title says. If I have a wave file that I have been cutting in it is no longer possible to extend a wave part to the left.
Or rather the part is extended but the extended bit is empty.
Could you take a look?
I don't know how extensive your changes are but I can try and have a peek too.
Beta Was this translation helpful? Give feedback.
All reactions