15th May 2005, 12:10 AM
Well it looks unrealistic, namely noting the parts where the hair comes out of the face, but it doesn't give me any strange feeling.
But, let them continue working on it.
For all we know, it was just the preexisting knowledge that it is CG that let us "realize" it. Don't even bother saying ANYTHING at all that says it's just plain obvious. Scientific testing is the only way to show that WASN'T the cue.
So, A TEST! I DEMAND A TEST!
...
Um, really though. We need someone to gather together a selection of images of real people and a selection of images of the latest and greatest CG (it MUST be stuff we have never seen before). This person will label each picture with a number (in random order, no patterns at all) and make a chart stating which numbers are for real pictures and which are for fake ones. Then, a second person who doesn't know which is which, the first person is the list keeper, will be the one to present the images to us in some thread. The reason the second person is there is why double blinding is called "DOUBLE", it prevents us from somehow gathering which is which by noting exactly how the person posted, or how they respond to our guesses. Anyway, we all make our guesses on which ones are real and which ones are false. The bunch of it all is counted up for each individual picture. Then, the first person counts them all up and gives us a final chart showing what was real and what wasn't. If it's photo realistic, we'll do no better than random chance (in this case, the odds are 50 50 if photoreal). If it's not photorealistic, we will do significantly better than chance that it becomes reasonable to assume we could tell the difference.
This is the test I propose. Currently, I suspect we will be able to detect with near perfect accuracty, but I may easily be wrong for the reason I started with at the top, thus the need for this test. At any rate, the results should be saved and sent to people who can do something about it as proof either one way or another, they can then duplicate the test themselves and boom, we have a confirmed scientific proofiness showing if we are there yet or not.
But, let them continue working on it.
For all we know, it was just the preexisting knowledge that it is CG that let us "realize" it. Don't even bother saying ANYTHING at all that says it's just plain obvious. Scientific testing is the only way to show that WASN'T the cue.
So, A TEST! I DEMAND A TEST!
...
Um, really though. We need someone to gather together a selection of images of real people and a selection of images of the latest and greatest CG (it MUST be stuff we have never seen before). This person will label each picture with a number (in random order, no patterns at all) and make a chart stating which numbers are for real pictures and which are for fake ones. Then, a second person who doesn't know which is which, the first person is the list keeper, will be the one to present the images to us in some thread. The reason the second person is there is why double blinding is called "DOUBLE", it prevents us from somehow gathering which is which by noting exactly how the person posted, or how they respond to our guesses. Anyway, we all make our guesses on which ones are real and which ones are false. The bunch of it all is counted up for each individual picture. Then, the first person counts them all up and gives us a final chart showing what was real and what wasn't. If it's photo realistic, we'll do no better than random chance (in this case, the odds are 50 50 if photoreal). If it's not photorealistic, we will do significantly better than chance that it becomes reasonable to assume we could tell the difference.
This is the test I propose. Currently, I suspect we will be able to detect with near perfect accuracty, but I may easily be wrong for the reason I started with at the top, thus the need for this test. At any rate, the results should be saved and sent to people who can do something about it as proof either one way or another, they can then duplicate the test themselves and boom, we have a confirmed scientific proofiness showing if we are there yet or not.
"On two occasions, I have been asked [by members of Parliament], 'Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?' I am not able to rightly apprehend the kind of confusion of ideas that could provoke such a question." ~ Charles Babbage (1791-1871)