The Perfect MIDI Controller ? Help us design it !

[PeterB]

I'm not sure where you're buying them but that's very expensive. Thomann sell them in Europe for 345 Euro (£286): http://www.thomann.de/gb/akai_apc_40.htm

The price I spoke of is the Manufacturer Suggested Retail Price. I agree that have come down in pricing.
I see them being sold at 299 at times.

Yet the official pricing is around 449,- I believe.

Anyway, this is the top limit of the price, anything below that is ok, depending on the quality of the product.
Higher quality, more sturdy, better controls = more costs.
Low quality = low costs....I know

[PeterB]

I may be dreaming, but my ideal controller would have a memory so you could "pre-load" groups of q's and transitions etc then trigger them all simultaneously by hitting one button...

I've been thinking about this for a while now and am sure it must be possible using Beyond and Pangoscript... but haven't had time to play much recently...
But if this was built into the control surface, maybe as a number of programmable 'macro' buttons, that would be awesomely useful! I'm sure everyone has sequences of buttons they just can't hit fast enough sometimes to match the change in the music... or just sequences they use A LOT that would be nice reduced to one button press?

Also, while we're talking control, why not build it around a small form factor PC and make a dedicated Laser Desk, much like a lighting desk or digital sound desk?
After all, the PC's spec doesn't need to be too immense to run most software well and by cutting all the unnecessary cack from the OS you make it much more stable... using a PC for show related stuff makes me nervous at the best of times... so I (probably not alone in this) tend to keep a "Laser Only" PC just for this purpose, with almost no other software installed and no internet connections to keep it as clean as I can... so the cost of my isolated PC could happily be moved to the budget for a controller and would also save me 5-10 minutes of set up time plugging in all the relevant cables, keyboard, mouse, MIDI gadgets, monitors etc etc etc... also, it'd look a lot less 'hobby jobbing' when I took lasers to a gig!

Anyway, just floating that as an idea...

But as others have said:- No touch screen thanks. Big buttons that 'click'. Space to write labels with a sharpie if needed. An XLR connection on the back for a 'littlelite' would be nice. Lots of pots, sliders and a big red "oh $#1t!" button. A joystick would be nice, but not essential.
Also, why MIDI and not USB? Or will it be USB using MIDI and I'm just getting lost in the technical backwater?

Anyways, I'm sure there's more for the "perfect" controller, but that's my wish list at the moment.
As to price, if it had the macro memories, I'd happily pay a few 100's more than the average APC40 price.
If it had all that and the built-in PC making it a fully dedicated "Laser desk"... well...

Wookie, you've given it some thought which is very good, but your are straying off the track here.
We're talking about a control interface only, so think buttons in hardware style only.

What you are saying ( and you've got a point ) is that you would want to map a multi-cue action under 1 button.
This already can be done using LivePro, not sure on Beyond.

So in the end the software would be the thing that would do the cue playing and it would be up to the software to start 1, 2 or 4 cues at one time by hitting 1 button.

I love the " big red "oh $#1t!" button" idea . lol

[dnar]

I use a Korg controller now, and one feature I rely on is the ability to configure scaling on faders. For example, I scale the animation fader to provide a range of 25% - 101% in QS, not the full range. I want to slow down animation only. The other bug bear I have and I raised this with Justin last year in Shanghai, is that QS does not let you aceive exactly 100% via MIDI. From memory the steps size results in 98% then 101%. Its a pain, as I must be sure to select the effect reset button before I play any programmed show, otherwise the cues drift from the music.... In fact it would be great if the timeline and shows saved to cues are NEVER effected by the animation speed adjustment! Simply no need for it!

[dnar]

I love the " big red "oh $#1t!" button" idea . lol

I do this already, I have a button mapped to "Blackout". Its also handy to reset master effects as setup in the QS config.

[White-Light]

I may be dreaming, but my ideal controller would have a memory so you could "pre-load" groups of q's and transitions etc then trigger them all simultaneously by hitting one button...

I think you may be over complicating this. Beyond and QS allow you to capture a multi-cue image using the "quick capture" button and the resulting capture can then be dragged and dropped from the QuickFX preview window into an empty cell in the workspace. If you've applied FX, these are captured too.

If you want multiple cues like this all in one place, you could set up your own tab and drop them all to the same page. Now the point behind saying this is the procedure is literally select the cues / FX, click "capture" and drag and drop. Its takes maybe 3 or 4 seconds per multi-cue.

I would suggest that to actually programme cues in groups into the buttons or script them would be a far longer and more complex process and somewhat duplicating what you can do anyway, because once you have your multi-cue captures in the workspace, you can then trigger them using the cue selection buttons in the normal way.

The only thing you can't do in this manner is create groups of FX without attached cues. However, you can create your own FX and then drop them back into the FX area as custom FX which again alleviates the need to do this separately.

This does raise two interesting points though. For the FX area, I'm not sure how this can be managed easily via button as the FX area has 4 rows of 99 cells now in Beyond, so potentially that would require 396 buttons! (Although I doubt most people will have more than a few custom FX).

For Tab selection, you need to have far more buttons than there are workspace tabs to allow for custom workspaces where the owner has added their own pages, although I think this is easily manageable. Also, the Beyond workspace will change. ATM it uses the QS Workspace. I understand it will have its own workspace when 2.0 is released.

[The_Doctor]

Yep, two parts to the MIDI standard, the hardware interface and the protocol.

SysEx is an add on that really caused a lot of compatibility between some equipment.... An after thought it certainly was...

Well, the use of F0 and F7 wasn't an afterthought, there was always an intent to allow expansion room in binary data. It only looks like hindsight because of the immense numbers of ways people ended up using it later. The same case can be made for 'General MIDI' and other frameworks based on control change numbers.

Btw, I second your later comment on scaling a fader response. On the other hand to avoid resolution loss this needs to be done in the receiving software unless you really do need a limited number of 'detents' well below 128.

I looked at the Wikipedia entry on OSC late last night (probably a bad time to try that, but never mind..) It allows some amazing things, regexp pattern matching, URL like addressing, but there is a huge amount of data and processing requirement there compared to MIDI whose patten matching is simpler than a brute force firewall. In the case of a controller MIDI wins because it's like robotic Lego, it allows people easy mixing and matching of existing stuff. The processing requirements for OSC make it harder to understand, and likely more consuming of power and money. Nothing wrong with it, it's just that if two functions can do the same task in code, it's usually best to pick the simplest one for speed, low cost, and less chance of something going wrong.

[PeterB]

Guys,

Just to let you know all remarks are being processed.
Not all will be able to be used of course.

The system will need to adhere to the sheer basics of MIDI if we want to be able to use it cross-OS and cross-device or cross-software.
It's obvious tactile faders, buttons, knobs with various sizes depending on the usage, will be preferred.

Now please start thinking about a layout.

I myself would think a layout with cue buttons in the top half
Below that adjusters ( knobs ) per channel with size/zoom being one and
Below that faders per channel

To the main right top an " Oh S#!t" button
Just below that a mini joystick with small buttons that can match up to 4 zones for quick zone adjustments.
Right bottom master faders for speed and intensity , or would intensity be enough ?

Just something I deducted thus far from your replies.

Those of you who are more crafted then I am in designing a layout please do so and share it with us.

Thanks of all the input and support.

[norty303]

Can I suggest looking at the layout of some lighting desks for button/fader placement.

One of the worst things about the APC is having to reach over the faders to get at the FX buttons (using a Beyond map)

For common action/quick bashing of things, a row/rows of buttons beneath the fader are quite useful. Just like flash buttons on sub-masters on a lot of desks.

You don't often rest your wrists whilst using faders, but you do often use buttons from a fixed hand position.

How about a nice padded rest at the bottom?

[White-Light]

If you just take the Beyond Interface, move the position and rotation controls (including the Joystick control) to below the sliders then I think you'll have a layout that works for both Beyond and QS, and given Beyond has more functions than most other software, other software's as well.

Thinking bout the limitations re the FX buttons numbers above, this could be overcome with the addition of range buttons ie 1-15, 16-30, 31-45 etc. It would also be good (although it may be too expensive), to have a small screen below each button containing the label of contents. Great for all buttons but essential for those like FX where the content might change. OLED would be great but at $90 per button, definitely out! I guess its going to depend on how easy it is to relate buttons to what's on screen without labelling.

I actually think this idea could work really well as it would fit alongside the APC40 rather than compete against it given the price difference.

[norty303]

But what if you don't use them for fx or even Beyond??
Let's not limit the design by software we're currently using.
Joystick needs to be well away from the faders and with an area to rest your palm for precise control.
TBH I've never yet needed to move the beam in that fashion so I could take it or leave it.
Have the people asking for a joystick actually used one in the real world or just think it's a cool idea?