In reply to:
I like the 'combine-the-small-apps' idea more if I compare the preferences apps R5 versus Zeta. A good idea would be a more user friendly Cortex with some basic building blocks available right from the start (can be 3rd party).
I've got two main concerns with the small apps idea. The first is that it leads to lots of windows open, and multi-window interfaces, beyond a certain point, are really bad. Cubase being one of the worst offenders there.
The second is that it may be difficult to start up and shut down everything. It's okay in a real hardware studio, because cables stay put. But with software, you'd have to completely re-wire your virtual studio each time you start your machine.
The latter problem might be fixable by having the applications auto-connect each time you start them, though clearly the fewer applications involved, the better. A separate mixer and hard disk recorder app might be okay, but a separate app for each effect plugin, etc., would be far too messy.
One thing I did wonder about is whether it's possible to replace BeOS's Media Kit Mixer. At the moment, it's awful -- no calibration markings, and only volume and mute for each channel strip.What if each channel strip in the Media Kit Mixer were to have EQ, aux sends, and group assignments, along with 8 busses like its hardware namesake? Using that, and either an external app to act as an effects unit (or slots in the mixer for effect plugins), it'd be possible to do a mixdown from several audio sources using just the Media Kit Mixer.In this case, the multitrack recording application would output several "channels" of audio, which appear in the Media Kit Mixer in the normal manner. Naturally the mix settings could be saved to disk, and maybe basic MIDI control could be added.Since it'd be part of the Media subsystem, the mix window wouldn't need to stay open; if you change the pan position in the mixer and close it, you still hear the panned audio even with the mixer window closed.
Whether this is possible, I've no idea. I'm not sure if it's possible to replace the Media Kit Mixer like that. It'd certainly be possible to supplement it with a similar application, but then applications that didn't know about it (e.g., SqueekySynth 1.1, Soundplay, etc.) would still connect themselves to the normal Media Kit Mixer.
In reply to:
make it visually easy to adjust notes or re-record a selection (maybe multiple times in a row and later choose the best option).
Adjusting notes will probably work in the same way as most other sequencers, using a grid editor / piano roll, unless you had a better suggestion in mind?
Recording multiple takes is a difficult one to get right. If the transport is in loop mode, and you're recording, I suppose it's possible to make it automatically advance to the next free track on the second pass. On the other hand, it wouldn't take more than a couple of seconds to stop recording, manually move to the next track, and start again, so it's not much of a gain.
The transport will be controllable via MIDI, so you can start/stop recording while sitting at the keyboard, rather than the computer. You may also be able to select which track is currently "active" by sending MIDI messages. For instance, if I knew I wanted to record four different takes of a piano part, I could make four tracks and assign them to piano, select the first track, and record it. Then, I could stop recording, move to the next track, and start recording again, simply by using the assignable controllers on the master keyboard (or the top octave's keys, etc.).
In reply to:
Also I'd like to see what I'm hearing that means the time-line should be displayed in sync with the notes being played (and be adjustable).
I'm not completely sure what you mean here, but the current transport position will be displayed clearly on the screen, as some kind of line, on both the grid editor and the arrangement windows.While playing back, you'll see the line sweep across the screen. You'll be able to pick the line up with the mouse and move it, to change the transport position. When it hits the edge of the screen, the editor window will flip to the next page, unless you've turned "follow mode" off. The pages might overlap a bit, to ensure you never get lost when it flips.
In reply to:
I like 3Dmix globally, you can do more with it then you might think (like realtime pitchshifting and timestretching!) witch is a bad thing because it's somewhat hidden,
3DMix can do pitch shifting and time stretching? I hadn't found this feature before, and even now you mention it, I still can't find it.
That, along with the lack of transport controls in 3DMix, really illustrates what I meant about existing applications not being "well-rounded". 3DMix gives you a really fancy 3D mixer, phase-adjusted panning, timestretch and pitch shift, but it doesn't even have proper transport controls and (as far as I know) can't record. Almost the perfect example of how feature creep can kill an application before it can do anything useful!
I don't want the same thing to happen with the applications I'm developing, so I'm strictly against feature creep. The idea is to match the basic functionality that standard audio hardware and software (on other platforms) provides, rather than to break any new ground (though that would be nice if it doesn't come at any price in terms of development time).
In reply to:
'Everything' should be recordable and later editable, the 3D-mixing is what I'm thinking of. Would be nice if it would listen to midi controllers and smoothing those 127 steps (or use 14bit).
What do you mean by 3D Mixing here? Do you mean the visual representation of channels as columns on a rotatable 3D grid, or shifting the phase of the channels when you pan an audio source?
The 3DMix application does both of these. The former is a novel idea, and done properly it might be quite a good way to mix, especially if you can visualize whether any frequencies are clashing, to help with EQing. However, it is a bit of a novelty in that it isn't actually necessary to produce a good mix.
I'm not too impressed with the latter though; on a reasonable pair of monitors it sounds rubbish. Perhaps done properly it would be quite good, but I'm not convinced. A panpot might not produce a very accurate stereo image, but at least it doesn't produce any weird phase distortions.
|