It's pretty obvious that you get more detail if you combine two pictures into one, assuming both pictures aren't identical. I guess the skeptics watch movies with the movie paused. They do the same thing with everything, they just pause it and cluelessly try to measure it. They don't know the big picture.
What you're talking about is called "interpolation," which is a method used, in addition to anti-aliasing, in order to make a low resolution source look more passable (interpolation helps correct for movement-based artifacting, for one). If you look at the source web site for your images, you will see they run several AA passes on the image first, as well they should. AA on video is fairly processer-friendly, and yields decent results when using modern anti-aliasing alogorithms.
Whether or not interpolation makes the picture look better will depend largely on the scene being processed (it does little to nothing for very still shots, for example). My point has consistently been that you aren't going to get the same results from using a low-res source and upconverting as you will from simply using a high-res source.
The VE web site talks about "converting" SD video to HD, but this is misleading. They are filtering the video to make it look better at HD resolutions, but that is a far cry from actually playing an HD mastered source.
You are not creating more "detail," you are essentially averaging the detail both within the frames and between frames to make up for the lack of resolution close up. You will notice that Ms. Jolie's face in your upconverted sample is quite blurry. If you were to also display a high-res comparison image in addition to your scaled (but not filtered) set and your interpolated/AA set, you would see a much sharper picture.
You seem to mistake making the picture look better for adding detail. The greatest level of detail you can possibly have are the native pixels of the source image. That isn't the same as the viewer perception when the image is scaled. If you think I'm saying they are, you're not reading a word I write.
I'm posting this picture so they sane people can see what skepticism is all about. For the skeptic this upconversion has the "same detail", and they don't bother upconverting because "their eyes cannot physically discern a difference" because they choose to run away from the screen instead of sit closer to the truth.
Again, you appear to not be comprehending what I am saying. I have been quite clear on the distinction between the detail of the source image and the perception of the viewer. I have also been quite clear that it is preferable to scale the video using anti-aliasing (or interpolation, although the results can be quite similar).
You're also divorcing real-world viewing conditions from the process. If you're viewing, say, an HD-DVD mastered at 1920x1080, whether or not you will see a benefit from upconverting will depend greatly on viewing distance.
The human eye stops being able to resolve the distinction between 1920x1080 and higher resolutions on a 37" display at about 4 and a half feet. With a 42" display, it's a bit over 5 feet. If you sit closer, then it will make sense to filter the image (although there are very few players that can handle HD-DVD filtering directly), and I have never said differently. Those are quite realistic viewing distances for a small room, and have to do with display size choice as well as video resolution.
But don't mistake that for adding detail, you're really covering up the loss of detail inherent in a low-res source!
Nothing wrong with that, I do it all the time when scaling up low-res videos (standard DVDs, for example) on my computer. I also bought a 1080p display because I sit less than 12.5 feet from the display, which is the cutoff point for discerning the difference between 720p and 1080p. I can clearly see the difference if I switch resolutions.
I'm not sure where you got the idea that I don't ever upconvert video, or that I think it looks "the same," as I have never said those things. I have said that at certain viewing distances the images look the same, and it doesn't matter if you upconvert or not, as your eye won't be able to see a difference. If you are closer, and your display supports a higher native resolution than the source, it makes sense to upconvert. Although, with your display it would look sharper to simply not scale 1920x1080 video.
I don't upconvert 1920x1080 sources because that's my display's native resolution. I upconvert everything else because it's smaller, and I don't like the blocky look of unfiltered video, although the blurriness of filtered SD video isn't always that much better. I'd much rather simply view an HD source in the first place, but that's not always an option.