Thank you very much for kind words and informative details!
You're right, our marketing efforts now are close to zero and we're far from being well known. We'll be working more on this.
Quality comparisons with most competitors you mentioned is available for a long time though not so known to the public:
http://www.thedeemon.com/articles/video_upsize_methods_comparison.htmlIt's a nice idea to charge more money.

Although many people still find VE too expensive (we see this in uninstall feedbacks all the time). I guess a good solution will be to have two versions: one very easy to use and working fast for most home users and one with maximum quality and settings for professionals, the second being more expensive.
Some gross algorithm details are present in this article:
http://www.thedeemon.com/articles/what_is_super_resolution.htmlYes, motion tracking is used extensively. We're not sure about releasing too much details - don't want cloners. There is a more serious paper for 2008 IEEE International Conference on Image Processing:
http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=4711763where I am one of the authors. The paper describes further development of the method which provides even better quality but much slower. VE uses a simpler method.
You're right about names "Dee Mon" and thedeemon sounding bad. Historically, "Dee Mon" is just creatively spelled form of name Dmitry in Russian language and has nothing to do with deamons. Domain thedeemon.com appeared way before it started to be used for selling software. Last year I opened a company and it got a better name: Infognition.
www.infognition.com is currently an alias for thedeemon.com and later will become the main domain (after it's one year old, around this October or November).
Although it's been a while since last version of VE, its development continues. With funding from one big Korean company we developed a realtime super resolution implementation working in YV12 colorspace. It's going to be used in next version of VE as a third mode of SR. It also has some deblocking and scene change detection, the latter increasing quality on cuts. We also plan a big redesign of user interface, making it more interactive and WYSIWYG (that's in version 2.0). The main problem of VE now is its dependency on system codecs which often work not so good and crash a lot (infamous ffdshow!). Probably we'll use a more self-contained approach to make VE more stable.
As for SR plugin, it's waiting to be adapted for newer versions of Adobe software. Making such a plugin for other host applications is also planned. Here we're just limited with our manpower because the company is still quite small (at least I'm not alone now, as it was a year ago).
So, thanks again for bright ideas and information. I hope we'll be moving more in this direction in near months.