Odessa (LIVE - IT FINALLY HAPPENED!)

mschenkel.it

@Samu said:

@brambos said:

For now they go into the input slot, but I think Jonatan is thinking about alternative visualizations.

Ok, makes sense. Separating Audio and Midi AUv3's in the UI feels like something that needs to be done.

For example the AUv3 Midi Generators (since they can not be 'routed' like Audio) could be just round circle-icons on either side of the screen. As for the 'midi-routing' tap'n'hold source icon then tap destination icon to connect them.

Who knows, maybe GarageBand will get AUv3 midi support some day too
(GarageBand does many things 'right' imho already, but some features are still missing arghhhhhh).

Talking about midiAU and AUM and their representation:
Slide down the mixer page, so you can only see the generator slot on the bottom of the screen, while in the rest of the screen you see, like the audio fx chain below, the midi chain mirrored upwards. And this would also sort routing

oat_phipps

Haven’t done this in a looong time, at least since mood came out, and even then I had the beta. Gettin the app DTs from not having this yet

Wrlds2ndBstGeoshredr

She's a cute kid and talented for her age. Here is a professional adult concert pianist taking theme suggestions from the audience and improvising on them:

oat_phipps

@Wrlds2ndBstGeoshredr said:
She's a cute kid and talented for her age. Here is a professional adult concert pianist taking theme suggestions from the audience and improvising on them:

I have to say after listening to a couple of the audience ones, I prefer the simplicity and melody of the kid’s piece

DatGood

Peeps into thread to see if app released yet.
Sees nothing.
Closes browser window and goes back to work.

ExAsperis99

@ph8aerror said:
Peeps into thread to see if app released yet.
Sees nothing.
Closes browser window and goes back to work.

https://thumbs.gfycat.com/FantasticFluffyAmericanbulldog-size_restricted.gif

Daveypoo

brambos

It has been ‘in review’ for over 7 hours now. Usually the review takes 30 to 60 minutes.. so I don’t know if this is a good sign. We shall see in the morning (midnight around these parts now).

lukesleepwalker

"I keep pressing the button but I don't hear anything."

ExAsperis99

@brambos said:
It has been ‘in review’ for over 7 hours now. Usually the review takes 30 to 60 minutes.. so I don’t know if this is a good sign. We shall see in the morning (midnight around these parts now).

I can just see the bafflement of the review committee as you explain how this magic will work.
"You know. For kids!"

JohnInBoston

While we're all sitting around waiting, I'll throw one more idea in about including tuplets in sequences. For those who've been around here for a while, one of the early iOS drum machines, Molten, had an excellent way of adding tuplets to sequences. It defaulted to a typical 8-channel, 16 step sequencer, presenting a grid where you could toggle hits. But if you selected any beat, a knob allowed you to change the subdivision of just that beat for that channel. So you could trivially pick the snare drum and add a triplet fill (or a septuplet fill, if Frank is in the house) without having to set the entire grid to some other set of divisions. Was always my favorite feature of this app, which unfortunately hasn't gotten any love in the last several years. But it's a UI idea worth stealing if you want to take a simple sequencer and quickly enable odd subdivisiions.

gusgranite

@JohnInBoston said:
While we're all sitting around waiting, I'll throw one more idea in about including tuplets in sequences. For those who've been around here for a while, one of the early iOS drum machines, Molten, had an excellent way of adding tuplets to sequences. It defaulted to a typical 8-channel, 16 step sequencer, presenting a grid where you could toggle hits. But if you selected any beat, a knob allowed you to change the subdivision of just that beat for that channel. So you could trivially pick the snare drum and add a triplet fill (or a septuplet fill, if Frank is in the house) without having to set the entire grid to some other set of divisions. Was always my favorite feature of this app, which unfortunately hasn't gotten any love in the last several years. But it's a UI idea worth stealing if you want to take a simple sequencer and quickly enable odd subdivisiions.

ooh I like this.

skoptic

@JohnInBoston said:
While we're all sitting around waiting, I'll throw one more idea in about including tuplets in sequences. For those who've been around here for a while, one of the early iOS drum machines, Molten, had an excellent way of adding tuplets to sequences. It defaulted to a typical 8-channel, 16 step sequencer, presenting a grid where you could toggle hits. But if you selected any beat, a knob allowed you to change the subdivision of just that beat for that channel. So you could trivially pick the snare drum and add a triplet fill (or a septuplet fill, if Frank is in the house) without having to set the entire grid to some other set of divisions. Was always my favorite feature of this app, which unfortunately hasn't gotten any love in the last several years. But it's a UI idea worth stealing if you want to take a simple sequencer and quickly enable odd subdivisiions.

That was the main reason I bought Molten, that bit was so easy and great

CracklePot

@JohnInBoston said:
While we're all sitting around waiting, I'll throw one more idea in about including tuplets in sequences. For those who've been around here for a while, one of the early iOS drum machines, Molten, had an excellent way of adding tuplets to sequences. It defaulted to a typical 8-channel, 16 step sequencer, presenting a grid where you could toggle hits. But if you selected any beat, a knob allowed you to change the subdivision of just that beat for that channel. So you could trivially pick the snare drum and add a triplet fill (or a septuplet fill, if Frank is in the house) without having to set the entire grid to some other set of divisions. Was always my favorite feature of this app, which unfortunately hasn't gotten any love in the last several years. But it's a UI idea worth stealing if you want to take a simple sequencer and quickly enable odd subdivisiions.

I think Seek Beats just added this feature. You can select a beat and access a slider that goes from 1-100 for # of hits.

chris_foster

@brambos said:
When the 'conventional bases' are sort of covered I certainly want to look into more far-out concepts.

I think ITVL (semi generative) has some good ideas

Will

A dedicated fugue sequencer would be fun. FM and Quantum have this territory covered but as a simplified AU MIDI plugin, it'd be welcome. Set some notes and then set N pitch/speed variations of those notes.

busker

@JohnInBoston said:
While we're all sitting around waiting, I'll throw one more idea in about including tuplets in sequences. For those who've been around here for a while, one of the early iOS drum machines, Molten, had an excellent way of adding tuplets to sequences. It defaulted to a typical 8-channel, 16 step sequencer, presenting a grid where you could toggle hits. But if you selected any beat, a knob allowed you to change the subdivision of just that beat for that channel. So you could trivially pick the snare drum and add a triplet fill (or a septuplet fill, if Frank is in the house) without having to set the entire grid to some other set of divisions. Was always my favorite feature of this app, which unfortunately hasn't gotten any love in the last several years. But it's a UI idea worth stealing if you want to take a simple sequencer and quickly enable odd subdivisiions.

The new Garageband beat sequencer has a somewhat similar feature, you can divide a single note in the grid for an instrument from 1 to 8 repeats.

Will

Moar modules:

1. Rhythm generator. Sort of like the arpeggiator but a simple xOx style set of 'on/off' buttons. Hold a chord (or a single note) in, and get that rhythm out. Quantized to note on or free running (quantized to AU clock). Well both, quantized to the AU clock (phase?) but one would reset to the start of the pattern for each MIDI Note ON (like LFOs often work).
2. Similar idea but using the Euclidian drum sequencer where track 1 = lowest note, track 2 = second lowest note... This alone could make the most amazing arpeggios.
3. A note splitter/distributor. Please. Please. Pretttty super duper please? Play a chord into it and it will send individual notes out to N destinations. Because, chords in Ripplemaker. Or chords spread across Ripplemaker, Troublemaker and Viking. Or... Chords spread out to transpose three different Odessa instances individually. Or... Shit, I just peed. brb
4. CC step sequencer. Basically like the bass sequencer but instead of notes, you can send variable CC amounts. Think this would largely cover the hand drawn LFOs @Samu was talking about. Would be great if this worked like the drum sequencer with a zoom so that it wouldn't require an instance per CC
5. MIDI effects like delay, transpose, scalar key lock, etc. Probably more trouble than they're worth to set up at this point but if @j_liljedahl continues with his plan to come up with a clever way to stack these on a single channel and that could be translated to making these work simply in series, suddenly very very interesting. Just the ability to offset (vs a repeating style 'delay') by musical amounts would be fun and interesting to throw in.
6. Again, if these ever land as something that can be inserted on an AUM channel strip outside of the Input slot... A couple of AUM specific 'automation' LFOs because yunz are working together. Essentially the current LFO module (or a simplified version) that allowed for automagic mapping to Volume, Pan or Sends 1 - 4 of the current channel. This can obviously already be done with the current LFO module and AUM's mapping facilities. The idea here is the automagic bit. Insert it on a track and pick a destination on that track. Done. Hello tremolo, hello autopan...

re: 3 above. Basically still after this on iOS http://polymer-app.com/help/

oat_phipps

I have a hunch somebody got rejected because their app has no audio.

Will

@oat_phipps said:
I have a hunch somebody got rejected because their app has no audio.

Or it's been passed but the Good Mr. Bos has yet to awake to the good news. I'd watch this space: https://twitter.com/bram_bos

[Deleted User]

@syrupcore said:

@oat_phipps said:
I have a hunch somebody got rejected because their app has no audio.

Or it's been passed but the Good Mr. Bos has yet to awake to the good news. I'd watch this space: https://twitter.com/bram_bos

Something about screenshots in his last post, maybe that's the holdup.

I dunno why I'm so excited I can't even run it yet, I'll have to do an iOS 11 update first look.

brambos

Yep, screenshots.

j_liljedahl

@syrupcore said:
Moar modules:

...

5. MIDI effects like delay, transpose, scalar key lock, etc. Probably more trouble than they're worth to set up at this point but if @j_liljedahl continues with his plan to come up with a clever way to stack these on a single channel and that could be translated to making these work simply in series, suddenly very very interesting. Just the ability to offset (vs a repeating style 'delay') by musical amounts would be fun and interesting to throw in.

Just to make sure you're not expecting something I won't deliver: I'm not planning to make a simplified stack of MIDI plugins attached to a single channel. The reason is that you can connect the same MIDI source to several MIDI destinations, or merge multiple sources into a single destination. To do this in a similar way that the AUM audio mixer works, I'd then need to add something like mixbusses for MIDI. I don't like that. Instead, my plan is to replace the cumbersome MIDI matrix in AUM with a more intuitive table of A->B connections, and make this faster to access than the matrix. That means that to express an A->F->B chain, you'd add two connections: A->F and F->B.

gonekrazy3000

@j_liljedahl said:

@syrupcore said:
Moar modules:

...

5. MIDI effects like delay, transpose, scalar key lock, etc. Probably more trouble than they're worth to set up at this point but if @j_liljedahl continues with his plan to come up with a clever way to stack these on a single channel and that could be translated to making these work simply in series, suddenly very very interesting. Just the ability to offset (vs a repeating style 'delay') by musical amounts would be fun and interesting to throw in.

Just to make sure you're not expecting something I won't deliver: I'm not planning to make a simplified stack of MIDI plugins attached to a single channel. The reason is that you can connect the same MIDI source to several MIDI destinations, or merge multiple sources into a single destination. To do this in a similar way that the AUM audio mixer works, I'd then need to add something like mixbusses for MIDI. I don't like that. Instead, my plan is to replace the cumbersome MIDI matrix in AUM with a more intuitive table of A->B connections, and make this faster to access than the matrix. That means that to express an A->F->B chain, you'd add two connections: A->F and F->B.

Please don't remove the matrix though. I love it to be honest if we could group channels together to hide them it would let us have neater projects as well. Pretty much the only reason most of us wanted to put them in one channel was to reduce horizontal scrolling once we have really complex midi chains.

Samu

@j_liljedahl In my mind I can see the the AUv3 Midi Plug-Ins as circle objects without a channel strip on either side (left or right) of the screen that can be easily connected to either to the sound-producing or midi-processing objects with a "tap&hold on the source circle then tap a destination circle" to make the connection. The connection could be identified by a faded line between the objets behind the channel-strips.

That's just an idea and I feel you'll come up with something even better

Carnbot

A more advanced midi matrix would be great. I'd always thought an audio routing matrix would also be useful as an overview to be able to change connections quickly in large sessions as an alternate way to arrange the routing in the main view.

j_liljedahl

@gonekrazy3000 said:

@j_liljedahl said:

@syrupcore said:
Moar modules:

...

5. MIDI effects like delay, transpose, scalar key lock, etc. Probably more trouble than they're worth to set up at this point but if @j_liljedahl continues with his plan to come up with a clever way to stack these on a single channel and that could be translated to making these work simply in series, suddenly very very interesting. Just the ability to offset (vs a repeating style 'delay') by musical amounts would be fun and interesting to throw in.

Just to make sure you're not expecting something I won't deliver: I'm not planning to make a simplified stack of MIDI plugins attached to a single channel. The reason is that you can connect the same MIDI source to several MIDI destinations, or merge multiple sources into a single destination. To do this in a similar way that the AUM audio mixer works, I'd then need to add something like mixbusses for MIDI. I don't like that. Instead, my plan is to replace the cumbersome MIDI matrix in AUM with a more intuitive table of A->B connections, and make this faster to access than the matrix. That means that to express an A->F->B chain, you'd add two connections: A->F and F->B.

Please don't remove the matrix though. I love it

This new connection table would do anything the matrix can, except that it would be much easier to see what is connected where, and easier to read (larger, non-rotated text, etc). The matrix is too cumbersome when you have lots of MIDI endpoints, since you need to scroll in two dimensions.

to be honest if we could group channels together to hide them it would let us have neater projects as well. Pretty much the only reason most of us wanted to put them in one channel was to reduce horizontal scrolling once we have really complex midi chains.

Well, that's another issue as far as I can see. I'll probably add some kind of bookmarks/pages to quickly go somewhere without having to scroll and find.

j_liljedahl

@Samu said:
@j_liljedahl In my mind I can see the the AUv3 Midi Plug-Ins as circle objects without a channel strip on either side (left or right) of the screen that can be easily connected to either to the sound-producing or midi-processing objects with a "tap&hold on the source circle then tap a destination circle" to make the connection. The connection could be identified by a faded line between the objets behind the channel-strips.

That's just an idea and I feel you'll come up with something even better

My idea is quite similar. MIDI plugins could fit several vertically in the same space as an audio channel strip. Each node would have a menu button (like on audio channels) that shows a menu of how the node is connected. For MIDI-output nodes, this would contain two columns, one for its connected MIDI sources and one for its connected MIDI destinations. Drawing MIDI-connection lines behind the whole thing going into destination nodes (synths etc) would be cool, but that would be mostly visually pleasing. The MIDI connection table (and per-node connection menus) would be where to actually manage the connections.

Carnbot

@j_liljedahl said:

@gonekrazy3000 said:

@j_liljedahl said:

@syrupcore said:
Moar modules:

...

5. MIDI effects like delay, transpose, scalar key lock, etc. Probably more trouble than they're worth to set up at this point but if @j_liljedahl continues with his plan to come up with a clever way to stack these on a single channel and that could be translated to making these work simply in series, suddenly very very interesting. Just the ability to offset (vs a repeating style 'delay') by musical amounts would be fun and interesting to throw in.

Just to make sure you're not expecting something I won't deliver: I'm not planning to make a simplified stack of MIDI plugins attached to a single channel. The reason is that you can connect the same MIDI source to several MIDI destinations, or merge multiple sources into a single destination. To do this in a similar way that the AUM audio mixer works, I'd then need to add something like mixbusses for MIDI. I don't like that. Instead, my plan is to replace the cumbersome MIDI matrix in AUM with a more intuitive table of A->B connections, and make this faster to access than the matrix. That means that to express an A->F->B chain, you'd add two connections: A->F and F->B.

Please don't remove the matrix though. I love it

This new connection table would do anything the matrix can, except that it would be much easier to see what is connected where, and easier to read (larger, non-rotated text, etc). The matrix is too cumbersome when you have lots of MIDI endpoints, since you need to scroll in two dimensions.

to be honest if we could group channels together to hide them it would let us have neater projects as well. Pretty much the only reason most of us wanted to put them in one channel was to reduce horizontal scrolling once we have really complex midi chains.

Well, that's another issue as far as I can see. I'll probably add some kind of bookmarks/pages to quickly go somewhere without having to scroll and find.

The connection table sounds like a great evolution. Shortcuts would be cool. How about a double tap on the mixbus icon to jump to the that mixbus for example.

Having collapsible and expandable channels seems like it could be cool for arranging/organising midi and audio channels though.

ExAsperis99

@brambos said:
Yep, screenshots.

Yeah, well, I didn’t even want these stupid old midi sequencer apps anyway.
slinks very slowly toward the door; hesitates with hand over knob; casts backward glance into the room…