Hello!

I renew a question I asked some time ago - would it possible to implement some kind of a time axis in Hydrogen where I could define a new BPM tempo value at a given time?

Other programs I use have such a feature (Rosegarden, Ardour, MusE). It would be very convenient to have it also in Hydrogen.

B. Regards!

Hi!

We're working on it at the moment. You can try wolke's fx_and_samplefun branche if you want to try our timeline, it's implemented there.
- Sebastian

Ha, GREAT! :)

But how can I download this fx_and_samplefun branch? Somehow I cannot see it on the download page nor any patch... Besides, is it stable?

On the other hand I could wait a few days/weeks to have it inside a regular svn version. What's the estimated time for the feature to appear in svn?

Thanks!

Hi!

Don't know how long it takes. Maybe weeks, maybe 2 months. I really can't say.

You can fetch the code with:

"svn co http://svn.assembla.com/svn/hydrogen/branches/new_fx_rack_and_sample_fun sample_fun sample_fun"

This creates a dir called "sample_fun" with the sourcecode..
Every feedback is welcome!

OK :)
Hopefully will try it during the weekend.
Is the source code a full-and-patched Hydrogen, or just a patch to apply to a regular svn source or some plugin to compile and use?

If there are instructions how to get it working in the source, don't answer (at the moment I'm at work, and check it).

Cheers!

Hi!

This is a branch which holds wolke's experiments, so you can think of it as trunk + patches. You can compile it like any other version of hydrogen, just run scons in the root folder.

hiho daemonic_myst,
the branch is stable enough for producing. our band use the branch as a band member. and next time also to play live with it.
in moment, the timeline especial works not so good as i expected. its important to use small buffersize if you need exact tempo changes. i plan to rewrite the timeline backend function in next time.
anyway, there are some nice other functions in this branch. but not documented :(. i am very interested on feedback. we plan to move some features into the trunk version.

Hello!

Sorry for long delay in responding, but I've been quite busy recently :/
Btw, I've just donated you with $20; I know it's not much, but I'm only a 'home studio' user :) Any way, you have helped a few times so far, and your software is really GREAT, so at least in this way I could have helped a little in Hydrogen development.

And according to the features - I'm still very interested in BPM change working nice'n'smoothly. The first impression was, when I checked it, that it seemed to work OK, but it was a very small test and I haven't tried to use the feature in a real project yet. But in next few weeks I'm going to record a new song, so maybe it will be a good opportunity for more detailed testing.

And what other undocumented features did you mention? I'm willing to check them too, but it would be nice to have a brief description of what I could achieve with them and how.

Thanks!

hi,
important, the sample_fun branch with all new features is copied to trunk. all develop on the new features will happen directly in trunk. so don't checkout the obsolete sample_fun branch.

3 days ago i have make some changes on time line function (trunk).
now it works better witch bigger buffer size values.
but also, we get sometimes double hits. this mean a note on pattern position one will played two times at ones. you can hear this because the output volume is higher in this moment.

this is a general h2 transport problem. each tempo change in hydrogen needs a recalculation off all note positions inside a song or a pattern in pattenmode.
so it can happens that a already played note will played a second time. this happens so fast that we can notice only a double hit.

this problem only happens if you change tempo from fast to slower values.
the other direction (slow to fast) is save.

Hi,

thanks for this info. I'll try to give it a test today.
Btw, the issue with a note played twice when changing the tempo to a slower value, is it something that could be fixed in a near future or is it rather something that can't be overcome?

Cheers!

>Btw, the issue with a note played twice when changing the tempo to a slower value, is it something that could be fixed in a near future or is it rather something that can't be overcome?

good question! maybe its possible that we find a easy way to solve this without many changes on transport. but its a real complex problem.

Hi there!

OK, the first impression is very positive :)
About the issue - this double note it's not a critical problem I think; I suppose one can avoid this by simply lowering the notes' volume in the first position in the pattern right after the tempo change. Of course it would be very nice to not have the issue at all, but I think I can live with it :)

But one suggestion for correction; let me describe it on an example:

let's assume the following:
1) at the 5th bar BPM is set 140
2) at the 9th bar it is set to 100

When I play the song, and stop it after the 9th bar, the BPM stays at 100 (correct).
Next when I move the cursor to start playing, let's say, from the 8th bar, the BPM still stays at 100, while in my opinion it should be set to 140.

So my suggestion is to set the tempo after every cursor movement to a value that can be found in the first tempo change preceding the cursor.
I hope it's logical, and I clearly described what I mean :)

Can you make such an update please?

Cheers!

you are right. this is something to improve.

would you please let me know when it is corrected?
thanks!

now its better :).
jump by clicking on position ruler or by using ffwd / rwd buttons set the correct tempo.
midi and jacktransport will follow.
checkout rev >1488

edit:
midi ffwd, rwd and stop works now.
checkout rev >1489