BPM 1.5 Problems & Suggestions only.

[joker77]

What is the MAX ppq resolution in BPM?

[ssp]

hey b-righteous

no its not taken personally or am I getting angry as such.

I am just trying to keep this section on a good standing these days, and trying to build a good response report for MOTU and UVI.

I can only do this with all your help, its hard for me to keep you guys in the loop and not burn myself in the process.
Comments like that do not help is what I am leading to.

Yes there are other issues and they will be dealt with in time.

I do know that another very important project is underway at the moment and that is taking up everyones time, (do not ask for details as you will not get any).


b-righteous yes you can do that if you want, im trying to get in here as often as I can, but my time is so limited at them moment and I feel bad because im not giving you guys as much time as I should.

Bear with me a for a week and I will be in here more regular, I really need to finish the work on my next hardware project first.

b-righteous you have been and I did state one of the more constructive posters in the bpm forum, and as always I respect your comments.

we will leave things there and carry on with the thread as it should be.
If i get time later on I will remove all references and the original post as well.

[BrianL]

the grid settings should make no difference on real time recording to the step sequencer. As long as the quantize button is off. Just like it works in the piano roll. This is how it works on Maschine and everything else. With quantize off, it should record whatever you play just as you played it at max resolution.

I looked at this a bit more closely in Maschine, the Grid is basically doing essentially what I suggested BPM do, it actually allows for multiple events to occur in a given measure. The thing is, although Maschine appears to look like a Grid, it isn't. In fact it is a Piano Roll with Snap to grid settings. BPM on the other hand, is a true Box Grid, forcing event containment, but again, in theory, increasing BPM's resolution should allow for the same occurrences as Snap to grid in a Piano Roll. It likely needs need a bit deeper thinking but that's the gist... Close attention will also have to be given to assure that mixed Note repeat settings are allowed. Perhaps to that extent, the Grid should be always at a resolution of 64 allowing for all Note Repeat resolutions within that setting.

The Photoshop images are to provide visual representation of this, let me know what you think?

[BrianL]

Noting this issue further, while the increase in resolution appears to be a solution to fixing Note Repeat, there are still other issues with the Drum Grid.

Being that it is a true grid, there's no provision for nudging notes, which brings me back to the 'Step [Grid Mode]/Free [Piano roll]' toggle option I mentioned before.

The only way to allow such actions (such as nudge) lies in being able to work in a Piano Roll mode. So ultimately the Devs are likely going to have to eliminate the Grid altogether and or bring both solutions to fruition as to accommodate such usage.

Although, under such circumstance, there will end up being a need to protect events; preventing the switching from one mode to the other from doing things like relocating/pocketing notes into Grid Squares when switched from Piano Roll to the Grid Mode. I've seen that happen with other apps and that could get messy. (yeah this is getting deep)

[ssp]

Some things that you are asking for will require a seriously large amount of new code to be written and old code changed.

This in turn would take quite some time to do then would take further time with the testers again. It would mean a wait for the same period if not longer than you had to wait for this update.

I cannot give you any info on when another update is likely as even I do not know. All we can do at the moment is just add to this section and clarify any issues that are present.

My priority for BPM is MIDI implementation, this is the main feature I have tested with different controllers. However the reports you are putting up mean I can also live test some of the issues you are having and confirm them.


Once again thanks for all the input you are all putting into this thread.

[b-righteous]

Why not just have the have the step sequencer behave just like the piano roll in this regard? The racks handle this perfectly. No need for anything complicated. We just need the step sequencer to not quantize to the grid res when quantize is off. This will solve the note repeat problem and also allow us to play our drums from a controller without quantize.

Don't get why it is not already like that?

Anyway, I'm not sweating it. The sequencer is just inadequate and I just use my MPC for that. Would actually prefer them to work on other aspects like more controller assignments and fix the clip window /sampler bugs.

[b-righteous]

Here is my updated list.

Bugs/issues!
-Fix clip window crashes with slicing and editing.
-Fix random loss of sampler input options
-Allow quantize button state to be saved with project.
-Fix CC assign to quantize button not working on rack parts.
-stop sequencer playback from triggering MIDI select.
-Allow step sequencer to record incoming notes without quantize to grid just like the piano roll.(as long as quantize button is off)This will fix the note repeat issue not being recorded right.

More Mapping options.
-Pad copy, paste, clear
-Line copy, paste clear
-Start end adjust of samples
-zoom for sample edit
-map selection to pad and play buttons in clip window
-load to load selected browser item.
-volume/pan of selected pad or part. (Just one pad/part volume and one pad/part pan CC that affect whatever pad or part is currently selected.)

Open pads sample in clip window button (assignable to cc)

Audition/MIDI select mode for slicesto select and audition slices from pads. In fact, just put the entire LoopLab slice feature set in BPM.(BPM calculate, snap to zero, auto scroll for cursor etc.)

Loop edit tab for pads with New IRCAM time stretch algo from Ethno 2!

Chromatic pads

Bank/rack part groove settings. It could be made like the new bank aux fx where you could toggle the global groove on or off to override the bank grooves.
-A render option to render the banks groove to pattern and zero the banks pos and vel controls.

Save & load individual rack instruments in file menu.

Global pad edit to edit filter cutoff etc. of all pads in bank at once. All pads would jump to same absolute value. Shift+click bank.

Update step sequencer with right click menu tools from racks sequencer.
-select/pencil tool
-quantize
-snap to grid on/off
-add select line and select all
-add nudge function.

Add duplicate pattern and nudge to piano roll menu.
Add graph automation to piano roll. (Graphs affect whole pattern vs single notes.)
Allow toggle to piano roll sequencer in banks.

Redo! Still one level to keep it simple. Use the same undo button to toggle like on the MPC. It should keep both versions and let you toggle. Once you record something new it flushes.

Allow undo to set undo points with mouse edits.

Automation read write system with sequencer mute/solo automation (don't mute note off events so sounds can decay naturally.)

Pad mute mode button similar to the live button. (assignable to cc) This turns the pads into sequencer based mutes.(not audio based)

Option to record live scene changes to song in real time.

MIDI out

Dedicated controller with bidirectional communication and LCD feedback.

[joker77]

And a hardware controller seriously, btw b that list is exactly where I want to see bpm at version 2, hopefully by then at least.I see you you kept a backup

[b-righteous]

Alright, I updated with that. I agree most of this would have to be a BPM v2 The bugs and issues should all be resolved in 1x ASAP.

[BrianL]

Why not just have the have the step sequencer behave just like the piano roll in this regard? The racks handle this perfectly. No need for anything complicated. We just need the step sequencer to not quantize to the grid res when quantize is off. This will solve the note repeat problem and also allow us to play our drums from a controller without quantize.

Don't get why it is not already like that?

Anyway, I'm not sweating it. The sequencer is just inadequate and I just use my MPC for that. Would actually prefer them to work on other aspects like more controller assignments and fix the clip window /sampler bugs.

Well to point it out again, the Step Sequencer is a true Box or Segmented Grid. Because of this, it is impossible for it to behave exactly like a piano roll. The only thing that can be done with a grid is to adjust its Box/Segment sizes or better said the quantity of Segments within its blocks or measures.

Technically, because of the way BPM is structured, the easiest fix for Note Repeat would apparently be to increase the resolution, allowing that greater number of segments.

This is pretty much what it does even now, only the Note Repeat resolutions exceed the available resolutions of the Grid Sequencer.

The only real complications come in when you start to think about those who use features like 'nudge' to "Humanize" a pattern. At it is, there can be no humanizing because of the structure of the program. Ofcourse we could play with the start/end times of samples we use, but that's not the same as nudge, nor is it as convenient. We can ofcourse use swing settings, but that's also not quite the same.

Really, as far as I'm concerened, nudge is not important as I myself never use it. I like to create tight rhythms, but at the same time, we're talking about a program that should lend to varied workflows so I kept in mind, the idea of nudging; which is why I brought it up in the first place.

Yes both sequencer types in BPM are lacking, but we can't use the fact that we have alternate Sequencer options as a rationalization or compromise; there will be others who only have BPM as their sequencing solution.

For this reason, Motu/UVI is going to need to assure that the technology used in BPM becomes more transparent, working/flowing with the user as oppose to clashing with them due to its shortcomings.

[b-righteous]

Well to point it out again, the Step Sequencer is a true Box or Segmented Grid.

Then just change it. Edit: I see what you are saying but still.
What if they just up the selectable resolution? You will still have an issue when you play beats and record them. (BTW I don't input beats with a mouse.) What if you have the grid set to 16 and you play 32 triplets? The way it is now you have to guess what you are going to play before hand. This is a flawed and inferior design. It really needs to be changed. This goes deeper than note repeat. Just playing a simple drum pattern it jacks up what you played.

What is needed is to change step sequencer resolution to 480 ppq for all live input.
-Delete the code that links the grid to pre-recorded material so it is independent. For example, changing grid does not change notes in pattern.
-If quantize is on, then it will quantize (new input only) to selected grid resolution.
-Get rid of the double resolution function since it will no longer be needed.
Now that I look at it there is a bit of change but I would rather they fix the issue at the root. Either way, I'm glad I still have my MPC

[davidb63]

Problem with MIDI input from Roland GI-20.

Anyone else using one of these or other guitar MIDI interface? I have problems where notes are getting hung and strung together.

EDIT:

I've done more testing... I can open any project (or start a new one) and put an instrument on a rack and record some notes. There will be one or two notes during the course of playing that will not receive (or record) the 'note off' when the note stops. I can see on the keyboard map that the note goes off but you don't see the note displayed in the sequencer at that point. It's not until you play that same note again, you see the end of the previous note (note off) and the new note start at the same point. for this instance of the project, the notes will always be the same, but if you close it, open it again, and record new data.. the 'hanging' notes will change to some other random notes. I have made every adjustment possible to the roland gi-20 to no avail. Other sequencers (Sonar & Live) record data form this unit with no problem.

This pretty much makes BPM unusable for me, since my primary means of playing MIDI is via my GI-20 and guitar.

Any ideas?


EDIT:

I attempted to record a video of this issue. The sound wont record from my Profire2626 to CamStudio (never used this before) so there's no sound, you you can see the NOTE OFF getting sent via MIDI OX monitor.. and you can see the problem on BPM.. not recording it. It's hard to see unless you use 720 and pop out to orig size. Sorry for the quality!

http://www.youtube.com/watch?v=WeikJ9oLpH4" onclick="window.open(this.href);return false;

[b-righteous]

Not sure on this. This is a unique way of using BPM so it's hard to say what the issue is. It could be your controller is sending messages that BPM can't deal with. Is it using polyphonic pressure or pitch? That is my wild guess. Is quantize off?

[davidb63]

Hey B. Yep, probably not too many peeps using BPM with a MIDI guitar.. lol. hard to say. The unit is pretty common though. It does pitch to MIDI conversion. When a string stops vibrating, a NOTE OFF gets sent. I've been playing with MIDI Ox, filtering differnt msg's, to try and pin down what could be causing it, but no luck so far. Quantize off or on, doesn't make any difference. I've tried going from the GI-20's USB MIDI to it's standard MIDI cable connection and made no difference. I hope I can get this figured out, at least a work around, cause I'm not much of a keyboard player I've got my guitar setup pretty nice with the GI-20, works great in Sonar.

[ssp]

david the other issue is MIDI buffer overflow. this can cause the symptoms that you are experiencing. When the buffer has too much data coming through it will just cause an overflow of too much data and cause these problems. When i stress test a controller with my software i can have a large aount of realtime data thrown at the buffer and sometimes it handles it, sometimes it collapses and lags and hangs notes.

make sure your not forwarding incoming MIDI events to the output port as this will create a feedback loop and which will cause a MIDI buffer overrun.

check on all of that first, also it could be a tracking issue, this would be the most likely culprit.

i am sure there are other people on here who have similar equipment.

edit (thread here): http://www.motunation.com/forum/viewtop ... 87&start=0" onclick="window.open(this.href);return false;

they talk about gr10-20 etc.