Hi everyone,
 
thank you all for the replies. First I want to point out that I really like the efforts some other users invest in order to get a nice controller environment. I thought about some of the facts you pointed out (like the bad numbers of sold controllers).
From my point of view there are two challenges:

• Cheap or mid-prized Controller which are not doing what I am buying them for
• Great controllers with a way to short product lifecycle

The gear less than 1K typically has bad faders (too small, bad haptic, no automatization) and encoders which don’t reflect the current status. Typical examples are the controllers build within Masterkeyboards. As a result, the customer has a bad experience. Even better, he won’t be able to control more than a subset of the needed functionality. Why should the invest in a greater controller, when the first impression is bad and he always read that it is his fault that the gear is not working properly…
The other side of the spectrum is the expensive and really nice gear. Problem: five years from now you get into trouble… driver etc. If I am not able to monetarize the investment in this time span, I lose a lot of money and my friends will make fun of me (Zune early adopter syndrome …)
And here is the business challenge: If I am buying a mid-price guitar, I will get 6 strings (or more), the fretboard and all the other features will be accessible. If I switch to something more expensive (a nice Gibson for example) I KNOW that I will be able to connect it to my amplifier in 2036. Meaning the guitar is not expensive, it is an investment… Controllers are the “instruments” of mix engineers – so this comparison is valid (jm2c).
My conclusion: the user interface has to follow a paradigm which is easy to access by a “simple” (to build) controller. Check out the MPC controller/software combo: for a reasonable price (<400 street) you get a controller which has an acceptable quality level and can reach easily most of the functions of the software. Yes, it is not a full blown DAW, but it is the concept I am speaking about. Plugins, Sampler voices, track mixer, Sequencer parameters – everything is organized in 4x4 matrices and easy to modify by one set of encoders. Even I got the idea behind it – and I am old ;-).
And this is why HW controllers always will be greater than virtualized controllers (aka touch): In front of a computer, you don’t need a keyboard. The virtual one on the screen does the job. Wait – this is not true! I don’t have to look to my physical keyboard, because I can feel where I am. With an on-screen keyboard, I am slower multiple times. The next level would be a VST instrument. It is less intuitive than any HW (OK, exception: machines with minimized user interfaces like the fizmo). If I know my hardware, I don’t have to look to reach the filter cutoff – I can feel the control knob. At the mixing stage, this is even more important. I don’t want to check via parameter if the mix is right, I want to listen. Then I will move the first two faders as well as fader 13.  I can’t do that with a mouse or a little touch screen. If I put my fingers slightly wrong, I got channel 12.
 
@Gibson/Tascam/CW: Put your heads together. Gibson knows how to build stuff that lasts. It is the same instrument jack since 50+ years. There are better solutions, but compatibility is important, too. Tascam knows how to build great electronic hardware. CW can explain their mixer with the pro-channel concept and together there must be a way to build something which can do the job. How great would be something like a Faderport with an additional 4x4 endless encoder matrix tuned to access the Pro Channel Modules? From the software side I could switch to a user interface theme where the controls follow that paradigm…
 
The success of Ableton is based on the availability of great controllers – think about that.
 
Oh - I forgot: As a Lifetime-Update Customer I am happy to spend all the saved update money on additional HW, if you find a way to make it work 20 years from now... (class-compliant, drivers developed and maintained inhouse...)

I was just getting VERY tempted to purchase the Behringer X-Touch. After reading this, I am not so sure. I currently use the Presonus Faderport, but I want more functionality as of late. I was hoping the X-Touch would allow me to control EQ and Compression plugins and sends as well as the normal stuff. Would this not work due to ACT or something?
 
Are Control Surfaces just dying out? Is it a waste of money to go for the X-Touch? Sorry guys for all the questions. I was just getting a bit excited about the X-Touch and now I feel a little rain coming...ha ha ha.

vdd
@Gibson/Tascam/CW: Put your heads together. Gibson knows how to build stuff that lasts...[snip]...Tascam knows how to build great electronic hardware. CW can explain their mixer with the pro-channel concept and together there must be a way to build something which can do the job.

 
From what I understand, one of the reasons Roland acquired Cakewalk was because they felt the combination of Roland hardware coupled with Cakewalk software, as embodied in the VS-700, would be unbeatable. Didn't quite turn out that way 
 

The success of Ableton is based on the availability of great controllers – think about that.

 
This is an area where I'm pretty knowledgeable, having used Live since Day 1 and edited the manual for Live 2.0. I would turn that phrase around - the availability of great controllers is based on the success of Ableton Live. For the longest time I had to use a Peavey PC-1600 live as a controller. It wasn't until later that the APC controllers appeared, after AL's user base reached critical mass for justifying a controller. Furthermore, at least for me Ableton Live without a controller is no fun at all. When people ask why I use SONAR but Ableton Live for live performance, my answer is that SONAR is a million-dollar recording studio disguised as software, while Ableton Live is a musical instrument disguised as software. As such, it really really needs a controller, and with Push, Ableton has moved Live even more into the "musical instrument" domain.
 
I'm really hoping that touch matures to the point where it does what we need, because removing the need for extensive hardware development and tooling would remove a barrier to making a controller.
 
 

Sylvan
Are Control Surfaces just dying out?

 
I tend to think it's not so much that they're dying out, but that they were never all that alive in the first place. By and large, I think controller sales in general didn't meet most manufacturer's expectations. The biggest problem is that creating hardware is expensive and time-consuming, so it has to be amortized over a large user base which by and large, doesn't really seem to exist.
 
However, the Softube looks promising and the X-Touch is Mackie control-compatible, which is basically what the industry has settled on as a control surface protocol. 
 
What kind of tells you all you need to know is my sig, which is going to go away soon because I decided that the lack of interest was a sign from the gods that I shouldn't sell it . I said "best offer gets it" because I expected there would be multiple offers, but only three people asked over the past year or so, and none of them made an offer - all of them wanted to know what I'd be willing to sell it for (which was lower than the lowest price I could find on eBay or Reverb.com).

I don’t want to repeat myself, but there are two aspects I need to point out:

1. One of the reasons why a device like the VS-700 is hard to sell was written before. Expensive, but great hardware – without proper support. There is no driver for W10 at the Roland download page. Why should somebody wants to spend money on that? Companies like RME show clearly that it is possible to maintain drivers of discontinued hardware without going down and out. If I spend a couple of grands for a piece of hardware, I don’t want to deal with DIY driver stuff. DIY is what my modular synth is for…
2. I am very excited about the Softube Console 1 integration. I really hope that it will work with the existing PC modules. I am a little bit nervous about the lifecycle of it, because right now it is not within the product catalog of Thomann anymore…

Probably it is a (cheaper) solution to use a digital mixer like the X32 Producer. It fits into a rack, has 32 i/o (USB) and the knobs doing what is expected. Than the DAW is a better ADAT, with the ability to automate the mixer. I don’t think that is the best solution, because the audio engine of Sonar is great.
 
Where did I start? The question was "Is ACT depricated?". I have to modify the question: Is CW willing to accept the de-facto standard HUI in a way, that the really great stuff they are developing will be better controllable by it? Endless encoders, embedded displays, ...
 
 
 
 

I doubt if the market failure of VS-700 can be generalized. I think its main purpose was to get Cakewalk more into small to mid-sized studios, which obviously didn't succeed. Maybe just because that market declines anyway and/or those studios had no reason or were not able to ditch PT (and the specialized controllers). According to its size and price the VS-700 was also no option for bedroom producers. So who should have bought it? The faderport on the other extreme is still available. It is cheap and small enough to keep it on your desk even if you don't use it that much ;-)
 
What has to be considered while inventing a new hardware controller IMO: Be aware that it is one of the last things people buy for their (home) studio. Then they buy one and have to recognize that it is not that easy to setup. So it fails their purpose to make working with a DAW easier. Instead it is a frustrating experience way too many times.
 
Take the Mackie Control protocol as an example. It is a widely used but in reality it is seldomly working correct out of the box with your personal setup. One example: Since three days I got a Behringer X-Touch here. But it has problems: The first V-Pot submits broken (jumpy) values when turned to the left in any DAW I tried. Except for Mixbus and the AZ Controller plugin for Sonar for whatever reason. When playing back fader automation the fader moves are  "shaky" (https://www.youtube.com/watch?v=DtFwmEwhUHo shows a tool that provides a workaround for this). Using it with Sonar is also frustring since Sonars Mackie Control plugin does not provide the possibility to edit EQ or Plugin parameters. Only the ProChannel Compressor can be edited AFAICT. Using the great AZController instead, these things work in general btw. It does not always do what you thought it did 2 minutes ago, but every behaviour and assignement can be edited at least. IF you are a nerd and do not want to waste your time with making music.... So if I send the X-Touch back it will be because both hardware and software don't work together like they should. It is not because I didn't want to use a hardware controller and or there is no market for it at all. 
 
And these Mackie protocol problems exist even on touch apps (like DAW Remote HD) ;-) Using ACT is even worse, since it is not bi-directional which it needs to be especially since touch controls are here to stay (although I don't like them). Instead there is no native support for OSC either. There also does not seem to be a dedicated iOS/Android touch app for Sonar on the horizon. But nevermind, in Sonar you can not even rename a controller plugin, so what...

vdd
There are some generic templates I have to customize or I can use (expensive) Mackie controllers. But as I learned, I will not reach my busses that way...
 

I think you're saying that the Mackie controllers won't allow you to control busses in SONAR. I now have this working just fine.  The essentials are that the Mackie Control protocol in SONAR uses MIDI note G#5 (80) to select the bus (busses) and MIDI note E5 (76) to select the track(s).  
 
The discussion thread that has the background, and the link to my victory post within it is here:[link=javascript:void(showMsgNum(3376318))]permalink[/link]
 
I hope this is helpful.