Comparing Encodings via Screen Shot

[majik]

Hi. I would like to compare my different encoding options and I think the best way to do that is to take screen shots and view them side-by-side. I have PowerDVD that will take screen shots, but I cannot get the same frame. If I pause PowerDVD, then click "Next" it starts playing again. I have tried pausing then stepping through until the second marker ticks, for example, pause it at 1:59 then click step forward until it reads 2:00. But if I step forward again, the time will jump back to 1:59??

What is the best way to guarantee *the same frame* while doing a screen shot? What tool would you use?

[charleski]

MPC (http://sourceforge.net/project/showfiles.php?group_id=82303&package_id=84358) is a good choice that allows direct access to a specified frame.

[majik]

Thanks, it works perfectly :)

[majik]

I think I might be doing something wrong, or a piece of software I have is limiting my quality. My screen shots from the original and my 42% reduced encoded versions look almost exactly alike! I see barely any quality loss.

My source was Star Wars: Attack of the Clones. I went through and using MPC I used "Save Image" to save .bmp files of screen shots of certain frames from the original source. I tried to pick frames on both ends of the spectrum: good lighting with little action and dark with high action. I then ripped it with DVD Decrypter and encoded with Shrink. Shrink said it had to reduce the file size by 42%. I then played this encoded version (mounted the ISO with DaemonTools and played it in MPC) and saved the same frames in .bmp. Using MS Paint I opened the original screen shot then the Shrink encoded screen shot. Hitting Alt+F, Alt+2 repeatedly will switch the images so I could get a good look at the difference. In the good lighting scenes I could not see *any* difference. In the dark, high action scenes I could see small areas of the screen that were a little fuzzier, but I wouldn't have been able to see the difference if I hadn't been swapping the pictures quickly and looking for quality loss.

I am suspicious of my experiment because I would expect a considerable quality loss when reducing the file size by 42%. Can anyone point out a flaw in my experiment? For kicks, I pulled out the same frames of a CCE 7-pass encoded version and I could see no difference between that one and the original. The thing is, the Shrink encoding took about an hour and the CCE encoding took 9 hours.

I would post some sample screen shots, but the files are over a meg each. If I reduced the resolution to make the file size smaller, there is no way you would notice a difference in quality.

To sum up: I see no benefit in encoding with CCE when Shrink does an admirable job, unless my experiment is flawed in some way. Opinions?

Michael.

[CWR03]

You're more likely to notice differences in quality by comparing the video rather than screenshots. There are problems inherent to certain types of compression which won't be apparent from a single frame - slow camera pans or fog/smoke tend to bring out the worst in any encode.