Your eyes can deceive you. Don't trust them.

In the first image I saw both as matching grey pills the whole time - neither looks even slightly tinted to me. Is anyone else getting that? I have no color blindness at all and the cube illusion works just fine. Weird.

Bar out, fan.

I had the same thing happen... they looked like two grey pills. But I went back and paused it on that frame and noticed the longer I looked at it, comparing the color of the pills, the more they became tinted reddish and bluish.

exactly the same for me - all the other illusions "work", except this pill one - I saw them both as gray immediately, never red or blue.

(and I have seen The Matrix movie)

OK. Looking extremely closely and using paper to block out the image, I have to say they fudged things on some of them.
I saw two grey pills the whole time.
The colored tiles fade to grey as they "mask off" the other tiles, they start no where near the shade of grey they end up as, their color has faded a lot in the process.
The grey tiles on the floor also change shades as they are 'masked off' quite clearly. I went 1/4 speed, and also tried masking them off myself, they clearly faked this one.
I put a straight edge on the checker board and sure enough, those lines are slightly curved....just barely but they are.
The two table tops are NOT the same size at first, I measured and the vertical table is definitely longer on the long side. That one's obvious. (EDIT:I'm wrong about that)
The spinning dots does work for me, as do convex images and auditory illusions.
So I'm not ready to call 'fake' on this, but IMO it's fudged badly.

Sorry, newt, but that's simply inaccurate.

I saw two grey pills too, but you're completely wrong about the others. I screen shotted all the images into paint.net to verify them.

The rubix cube image is 100% real. The RGB values for the blue and yellow tiles are identical (127,128,129).

Same with the the tiles under the table. They are are off by a small amount (rgb 70 68 71 vs rgb 70 68 70), but I'd but that down to the video encoding.

Ditto with the checkboard; zooming in with paint.net the lines are pixel straight (there is some anti-aliasing at the edges, but it doesn't affect the "straightness of the checkerboard").

The tables too, are the same size. I rotated the vertical table.

If you don't believe me, try it yourself.

You went farther than I did then.
All I can say is when I cut out squares in a piece of paper, I could see a difference until they 'masked off' the image, then it had changed. I know these can work without fudging, which is why I was disappointed.
Did you note the difference between the 'colored' image and the 'masked off' image? It sure seems like there's a difference to me, if I stop it 1/2 way through and cover all but 2 squares, one is slightly lighter than the other on my monitor. That went for both the cubes and the floor tiles. Maybe it's 'eye memory' or something, but it sure seemed to me that the center tile was noticeably lighter until the 'masking off' happened.
I used a piece of paper against my monitor to measure the table,....I must have moved it when marking it, because now when I do it, it seems the tables ARE the same size. Damn touch screen, kept starting the video every time I touched it.
If those lines were really pixel straight, my paper is cut with a curve or my monitor has a problem.
Again, you went farther than I did to prove it, so I'll defer to you and accept I'm seeing things, even when I mask them off myself.

EDIT: Just a thought why I may have seen it differently, do you think it's possible that 'light bleed' or 'color bleed' on my monitor has anything to do with it? I mean, since the pixel next to the 'grey' block might be glowing bright yellow, it could color the grey slightly yellow, while the RGB value would not change?

The pills look yellow and turquoise to me... I dunno what they were going for in that one. The yellow-handed one has a yellow highlight, the turquoise-handed one has a turquoise highlight, so I inferred they were yellow and turquoise...

@newtboy - just hit print screen, paste it in in Paint or similar and mask off the colors there. A much better test.

the bar / far illusion at the end is the best i've ever seen

@newtboy - I'm blown away at how certain you are it's all fake. I suggest you do what I did: Instead of using paper on your screen, just take a screenshot and insert into an image editor and inspect things there.

I cut the three tiles out and pasted them side-by-side and they are in fact the same color: http://i.imgur.com/e5lcV5P.png

I dragged straight lines on the checkerboard before and after the dots were added, and it has only straight lines.

I copied/pasted the blue tabletop, rotated it and it fit perfectly on the other one: http://i.imgur.com/QzT8nc8.png

Nothing was fudged in the video. It just shows how powerfully your brain is latching onto what it believes it is seeing.

It's like that dress photo from a few weeks ago. "Is it white and gold or purple and black?!" Many people were hardcore in one direction or the other.

The only one that left me confused is the pills. 1) He said they were red and blue, but they were yellow and turquoise. 2) They had holes in the pills allowing the background color through; it was only there that they looked colored, otherwise they were just gray. I suspect they were just trying to shoe-horn in a red pill blue pill Matrix reference.

? I'm confused. I admitted my 'test' was lacking, and deferred to @ChaosEngine who didn't trust his eyes but measured...so I am no longer certain it's 'fake' at all, in fact I was careful to NEVER call it 'fake'. Perhaps you only read my first post in this thread and missed my admission that I was wrong?

As I said, I think there's a reason it LOOKED like it did to me...probably the surrounding color reflecting off the white paper I used to mask off and 'coloring' the grey squares....and likely the paper I used was not perfectly straight to use as a straight edge....and measuring anything on a screen is less than perfect, which is why I poorly measured the table and lines...the second time I measured it I did better, and I admit they are the same size contrary to my original statement.
I also only saw grey pills.

The reason I was skeptical is I've seen these same 'illusions' faked many times. For instance, the last time I saw the Rubik's cube, the tiles had dark shaded colored edges they removed when 'masking off' which obviously changed the color. It's not that I don't 'believe' in optical illusions, I just think people are cheaters more often than not these days and fudge things they don't need to fudge. This time it was my method of 'masking off' that seems to be my issue...I don't have an image editor that will do much for me here....sadly...but it doesn't mean I don't trust Chaosengine who did and set me straight.
OK? ;-)

Indeed, I didn't see your follow-up comment. I was aghast upon reading your first comment and I immediately had to respond.

I like how he points out that the color "illusions" are not actually a brain failure, but an amazing visual processing feature honed by evolution. If the rubix cube were a real object photographed in yellow light and then in blue light, those tiles would have to be yellow and blue (not grey). It's like our brains calculate the true color under white light, and that's what we see.

Just think about how useful that is, without that feature we could never build up knowledge about what color different plants and animals are, because their color would seem to change drastically with the lighting conditions.